JPRS ID: 8490 TRANSLATIONS ON USSR SCIENCES AND TECHNOLOGY PHYSICAL SCIENCES AND TECHNOLOGY

• Document Type: CREST
• Collection: JPRS
• Document Number (FOIA) /ESDN (CREST): CIA-RDP82-00850R000100060001-3
• Release Decision: RIF
• Original Classification: U
• Document Page Count: 139
• Document Creation Date: November 1, 2016
• Sequence Number: 1
• Content Type: REPORTS

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R00010006000'I -3
~ ~ -
CFOUO 3il79~

i JUNE i979: _ i OF 2
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~
FOR OFI-ICtAL USE ONLY
JPI2S L/8490
1 June 1979 ~
~
TRANSLATYONS ON USSR SCYENCE AND TECHNOLOGY
~ PHYSICAL SCIENCES AND TECHN~LOGY
CFOUO 31/79)
U. S. ~OINT PUBLICATIONS RESEARCH SERVICE
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850ROOQ1 QOQ6QOQ1-3
NO'T~
JYIt5 public~Cions conCain informnCion primarily frnm foreign
newsp~pcrs, periodic~ls ~nd books, but ~lsn frdm news ~gency
rransmissions and bro~dcasCs. Materi~ls from foreign-langu~ge
sources ~re tr~nslated; ehose from ~nglish-l~nguage 90UrCe5
nre transcribed or reprinted, wiCh Che ori.ginal phr~~tsing and
othcr characeerisCics rcr~~ined,
Headlines, ediCori~l reporCs, and maCeri~l enclosed in br~ckeCs
~nre supplied by JPIt5. Processing indic~tors such ns [TexC]
or (~:ccerpe ] in the �irat 1 ine of e~ch item, or followittg Che
l~~st line of ~h brief, indic~Ce how the original informaCion 4185
~ processed. Wt~ere no processing indicator is given, ehe ~nfor-
- maeion co~ys summarized or exCr.aceed.
Unfamiliar names rendered phonetically or transliCer~ted are
enclosed in pareneheses. Words or names preceded by a ques-
tion mark and enclosed in parentheses were not clear in the
original but have been ~upplied as appropriatc in context. -
Other unaetributed parenthetical notes within the body of an
item originate with the source. Times within items are as
given by source.
'I'he contents of this publication in no way r.epresent the poli-
cies, views or attitudes of the U.S. Government.
~
CCPYRIGEIT LAWS AND REGULATIONS G0~/ERIVING OWNERSHIP OF
MATERI.ALS REPRODUCED HEREIN REQUIRE THAT DISSEMIDIATION
~F TE{IS PUBLICATION BE RESTRICTED FOROFFICIAL USE OM.Y.
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850ROOQ1 QOQ6QOQ1-3
~
FOR OFFZCZAL USE ONLY
JPRS L/8490
1. June ~9 79
rRANSLATIONS ON USSR SCIENCE AND TECHNOLOGY
~ PNYSICAL SCIENCES AND.TECNNOLOGY -
. (FOUO 31/79)
CONTENTS PAGE
GEOPHYSICS, ASTRONOMY AND 5PACE
Main Problems af Space Walking by Man
(Yu. N. Glazkov, et.al.; NA ORBIT~ VNE KORABLYA, 1977).. 1
One Principle for Measuring Angles of Attack and Slip During Space-
craft Flight in 'Near' Space and the Atmosphere
(A. B. Krymov; DATCHIKI I VSPOriOGATEL'NYYE SISTEMY
KOSkiICHESKIKH APPARATOV. ROBOTY I MANIPULYATORY.
TRUDY IFAK, 1978)...........o 33
A New Method of Syntheaizin~ Artificial Motion and its Applica-
tion to Locomoting Robote and Manipulators
(M. Vukobratovic, et.al.; DATCHIKI I VSPOMOGATEL'NYYE
SISTEMY KOSt4ICHESKIKH APPARATOV. ROBOTY I MANIPULYATORY. -
TRUDY IFAK, 1978) 47
Semiautomatic Manipulator Control Systems and Computer Investi-
gation of Their Dynamics
(V.S. Kul~ahov, et.al.; DATCHIKI I VSPOMOGATEL'NYYE
. SISTEMY KOSMICHESKIKH APPARATOV. ROBOTY I MANIPULYkTORY.
TRUDY IFAK, 1978) 58
Algorithms for Combination and Supervisory Contro~. of Manipu-
lator-Robots
(Ye.P.Popov, et.a1.; DATCHIKI I VSPOMOGATEL'NYYE SISTEMY
. KOSMICHESKIKH APPARATOV. .ROBOTY I MANIPULYATORY.
' TRUDY IFAK, 1978) 72
The Problem of Perception and Synthesis of a Model of the
ExCernal Environment by a Self-Contained Planetary Rover
(L.N. Lupichev, et.al.; DATCHIKI I VSPOMOGATEL'NYYE
SISTEMY KOSMICHESKIKH APPARATOV. ROBOTY I MANIPULYATORY.
TRUDY IFAK, 1978) 78
-a- [III -USSR-23S &T FOUO)
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOR 0~'FICIAL USE ONLY
CONT~NTS (ConCinued) P~ge
Inveskigaeing the Algoritihms for ConCrolling the Motion nf u .
Seli-Contain~d planetary Rover by the Mathematical Modeling
Method
(V.F. Vasil'yev, et.a1.; DATCHIKI I VSPOMOGATEL'NYYE
SISTEMY KOSMICHESKIKH APPARATOV. ROBOTY I MANIPULYATORY.
TRUDY IFAK, 1978)...~ 89
Automatic Control of g Moving Planetary Rover Complex
(J. Benes, P. Kolar; DATCHIKI I VSPOMOGATEL'NYYE SISTEMY
KOSMICH~SKIKH APPARATOV. ROBOTY I MANIPULYATORY. TRUDY
IFAK, 1978) 97
PUBLICATIONS
Computer Structures and Software
(L. N. Korolev; STRUKTURY EVM I IKH MATEMATICHESKOY
OBESPECHENIYE, 1978) 109 =
Satellite Coa~unications in the lOth Five-Year Plan
(A.D. Fortushenko; SPUTNIKOVAYA SVYAZ' V DESYATOY
PYATILETKI, 1979) 125
New Book on Man in Open Space -
(Yu. N. Glazkov, et.al.; NA ORBITE VNE KORABLYA,
1977) 127
Sensors ana Auxiliary Systems of Space Apparatus, Robots
and Manipulators
(B.N. Petrov, V.Yu. Rutkovskiy; TRUDY 8-OGO MEZHDL'NARODNOGO
SIMPOZIUMA IFAK PO AUTOMATICHESKOMU UPRAVLENIYU V PROSTRANSTVE:
DATCHIKI I VSPOMOGATEL'NYYE SISTEMY KOSMICHESKIKH APPARATOV.
ROBOTY L MANIPULYATORY, 1978) 130
-b-
- FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR OFFICIAL USE ONLY
GEOPHYS~CS, ASTRONOMY AND SPACE
MAIN PROBLEMS OF SPAC~ WALKING BY MAN
Moscow NA ORBITE VNF KORABLYA in Russian 1977 siqned to press 12 Sep 77 -
pp 5-44
(Section 1 from the book "Na orbite vne korablya" by Yu. N. Glazkov, L. S.
IQ~achatur'yants and Ye. V. 1Qiruno~, Izdatei'stv~o Znaniye, 100,000 copies,
176 pagea]
('Text~ The experience of mnnned space f].iqhts in the USS1t and the United
States convincinqly dea~onstratea that man on board a manned epacacraft (PKK) _
or a mnnned orbital sLation (POS) is one of the main elementa which enaure
effect3ve completion of the fliqht program. The required hardware, h~gh
profeseional training and extensive capab~lities in control of the spacecraft -
systeans pexmit the cosrtwnaut to actively aff!:ct the course of the flight. _
For exatnFle, during the flight of the Mercury MA-9 P1IX, a failure occurred
in the su~romatic device which determinea the aequance of instructions during
reentry. The qround services and the astronaut manaqed to effect reentry
by usinq manual oontrol. This is not a sinqular case. Failures in the on-
board equi.pment of other PiaC have also occurred. Due to the active operatf.ons
of the crew, the flighta of the Voskhod-2, Gemini-8 aind Apollo-13 PIQC were
completed aucceesfully. -
A space walk by man wae neeessa~y in son~e fliqhts for successful completion _
of the plenned proqram. For example, the apacecraft had to be unpressurized
and work had to be carried out in pressure suits to correct malfunctions in
the dxkinq mechanism of the l~pollo transport spacecraft during the first
expedition of the Skylab POS. The sepair opexations wero carriad out and -
docking was succeasful. M emerqency situation of the Skylab POS after orbi- �
tal injection vras also eliminated due to the operationa of the crews in open
epace.
Thus, the fliqhta of Soviet oesmonauts and American astronauts showed that
man can perform maintenance, aesembly, repair and transport of cargo beyond
preseurizad conrpartments diractly in outer space. Therefore, the problem
of future reeearch in space ia closely related to operations beyond the
pressurized coapartments of Plat and POS. Specifically, operations outside
the spacecraft will comprise a significant fraction of the total w~ork of
1
FOR OFFICIAL USE ONLY
I
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOR OFFICIAL USE ONLY
coamoru~ut,~ cYuring flighta of the muitiuse traneport spacecraft (MTKK) . The
number and duratiion of space cvalka, presgure auit requirementis and require-
ments on the life support and other systems were cletermined according to tihe
MTKK progrnm.
Approxima~ely 1,073 space fiights are planned dur3ng the period 1979-1990.
Theae include:
~pace etati.one of the Spacelab type 56
larg~ apace observatories 36
modular stations 46
automatic satellites and stations of NASA* 147 -
miscellaneous satellitee (tielevision, connnunicationr3 and
so on) 224
satellite~ of other aqencies 305
assembly in orbit 259
Completion of this proqraa? can be provided by 597 flights of MTIQ( with dif- -
ferent payload. The follcwinq main tasks of human activity in open space
are planned during thie:
- l. Operatioa of larqe space telescopes (replacement of mirror modules, re-
, moval of contamination from the instrumente, mcvf�q solar panels and so on).
2� Mainte.nance m! spaoa�laboratories ~to studp the earth's natural resour~:es
(asseaablp of parabolic arftennas 9 m in dia~w~e~;'installation of film ca3settes
aad so on). -
3. Putting the onboard systems into aorkinq condition (connecting electrical
and hydraulic pluqs and main lines, tranaport and attachinq of solar panels
and antennas, reaavinq protective devices used during orbital fnjection and _
~ so.on).
4. Inapection, debuqqing and repair of the external devices of orbital objects
(hatches, mechanisms for openinq optical sensors).
�
5. Scientific reaearch and so on.
Thus, the problems have been postulated. It is no less important to determine
the neceasary number of space walks and their frequency. For example, main-
tenance of the spacelab unite may require 118 sp~ce walks to complete the
tasks enuaierated above usder Nos. 2, 3 and 5. But 788 space walks are required _
to carry out all the tasks of maintaininq space objects in orbit. The distri-
bution of the number of walks outside the apacecraft, which are planned before-
hand for a single fliqht, ia shown in Fiqure 1. Figure 2 demonstrates the =
duration of one space walk, calculated for completion of operations by a single
peraon.
*NASA Nati~nal Aeronautics and SpacE Administration (United States).
� 2
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR OFFICIAL U5~ ONLY ,
0. , ~ ~
- b d
, ~
~e
(1) ~ s ' (2) '
~ 4 Obraxtuona~~wA
~ ' -
~ ~ ,
~ /lna~upyena~( 3
O
0 � p 40 60 e0 !00
apeurane,vax nanoto~ (4 )
Figure l. Diatribution of Number of Planned and Potentially
Possible Space Walks Per Flight
1~Y : ~
1. Number of possible operations 3. Planned
in open space per flight 4. Percent of orbital flights
2. Potenkial
x
m
. ~
~o ~ z .t ~ s s ~
v~~ 1)
Fig~:re 2. Duration of Oparations in Open Space, Calculated for
Completion of Operations by a Siugle Cosmonaut
KEY: 1. Hours ~
Which hardware can support planned and unforeseen nperations? I� was decided
to conduct the space walk by the airlock method, using the airloc:k chamber -
(ShK). The length of the airlock chamber was selected at 3 m and the length _
of the tether was selected on the order of 21 m. This lenqth permits one to
- complete 95 percent of all types of operations. The resulta of investiqations
and tests and, specific,~lly, study of the proposed travel routes of the cos-
monauts convinced us that it is feasible to recruit two crew members for
operations outside the PKK. The total number of cosmonauts: 20 (GOmmander
3
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
I~OR OFFICIAL U5~ ONLY
and pilot) and 40 (payload apecial.ist~), was firet determined to complete 788
space walks during 597 flightis.
Tt was established during simulation of the cosmonaut's activity in open
space tha~ the upper part of the preseure suit is the most mobila (207,000
cyclea during a simulated 12-year period). The presaure suit should have
movable joints in the region of the shoulder, elbow, wrist, waisti, knee, hips
- and ankles. This indicated analysis of the cosmonaut's body motions while
performing maintenance on objects. Possible movements of the heat must also
ba provided. The mobility of the ax~ms in the pressure suit and o� the wris~
in gloves has special significance.
Requirements on the life support system (SZhO), its type and composition were
determined on the basis of s3.mulation and the tasks of the space walk and
requirements on th~ systems for supporting the cosmonaut's activity outside
the manned spacecraft and manned orbital station were refined. TYae experience
accumulated in the-USSR and the CTnited States in work by man in open space
made it posstble to determine the range of tasks entrusted to the cosmonaut
outside the PKK and POS:
maintenance and repair of the PKK, POS and of automatic space objectss
installation-diamantling and assembly operations in orbiti
maintenance of scientif+c research apparatuss
rendering assistance to crews experiencinq distress in orbit and also re-
- placemen't of PKK and POS crews;
transport of various types of cargo;
~ experimental lnvestigations in open space.
This list of t~asks is naturally generalized and incomplete. Th~ tasks will
be corrected and expanded as investigations in space progress.
� Brief Characteristics of Manned Space Objects Which Support Space Walks by Man -
Manned spacecraft and manned orbital stations which support space walks by man
. are distinguished in design from manned objects which do not support human
activity outside the craft. They primarily have additional equipment for
depressurization and subsequent pressurization of the corresponding compart-
ments, special pressure suit design, life support systems and so on. -
A space walk was accomplished during the flights of the Voskhod-2, Soyuz-4
and Soyuz-5 PKK, a nwnber of Gemini and Apollo PIQC and the Skylab POS. Let
us familiarize th~ reader with th~ brief characteristics of these and future
manned objects, having especially determined the means for supporting space
walks outside these objects.
~ 4 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850ROOQ1 QOQ6QOQ1-3
FOR UFFICIAL USE ONLY
Voskhad-2 was a manned spacecraft designed �or orbital flight around the
. earth. Unlike the three-place Vos}:hod PKK, the Voskhod-2 spacecra�t was made
in a~wo-place variant. This increasQd the freQ space which Eacilita~ey
movement of the cosmonauts in the cockpit durinc~ preparation for a epace
walk. A"saft" type airlock chamber, rotated in~o the working posit:ton in
orbit, was d6signed for the spacp walk. '1'he ShK was connected to the cos- .
monauts' cockpit by a presaurized hatch. The ShK was also equip,ped with a
pressurized hatch on the ~ide turned toward space. Thus, the Voskhod-2 had
- two pressurized spaces which were separated from each o~her and a~ the sr~me
time could connect one to the other.
This design made it possible to c~mplete the space walk beyond tt~?e spacecraft
without depressurizing the cosmonauts' cockpit. The hatches were controlled
remotely from the console installed in tne cosmonauts' cockpit or manually.
Two movie cameras, a light system and a control console were placed inside
the ShK and tanka with an air supply for repressurization of the ShK and with
an emergency oxygen supply are placed outside it. The cosmonaut ia supplied
with oxygen from a self-contained backpack SZhO while working outside the
PKK.
i ~ ~ i ~ ~ ~ ~
o ~
~ ~ i ~ S ~
~
~
~ ~ ~ ~ ~
~
Figure 3. Drawings of A. A. Leonov During Preparation for a Space
Walk
S
FOR OFcICIAL USE ONLY
i
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000100060001-3
P~[~ UCr~L~.Lf\l, tl5~ ~NGI
c?n 18 March 1965 cosmon~u~ A. A. Leonov of tihe USSR m~de tihe world's �irst
apace walk. The walk lasted for 23 minutds 41 secnnds, of which 12 minute~
9~econds wero outside the PKK. The aoscn~naut completed the p7.~nndd pro-
gram of inveatigations, having separated himself from the ~p~aecra~'t to a
tether length o~ 5.35 m. One can gain some idea of the experiment from
~'igure 3.
The Soyuz ia a multiplace PKK designed for flights fn geocentria orbi~,
maneuvering, approach and docking. Z'he weight o� the ~pacecraf~ is approxi-
ma~ely 6.5 tons. Due to ~he moclular design and th~ capacity �or moderniza-
tion, the Soyuz PKK is used �or differPnt purposes in space flighta. Auto-
nomous and group fl~.ghts, an astronomi.cal laboratory, a transport spacecraft _
for sup~lyinc~ ~he Salyut POS and modification for the joint flighr in the
Soyuz-Apo11o international program this is an incomplete list o� tho _
tasks per�ermed on the Soyuz PKK.
7."he spacecraft c~nsists of an nrbital modu].e, a descent vehicl�~ where a crew
t~p to cosmonauts is located and an instrument-service module. The orbi~al
module and dascent vehicle are connected by a pressurized manhole, which
pezmits the use of the orbital mo~ule as an ShK. Tn fihis case the orbital
- module is equipped with a depressurization and pressurization syst~m and
also a hatch for emerging into o~en space.
A�ter development of the experimental orbital station consisting of docked
Soyuz-4 and Soyuz-5 PKK in 1969, two c~smonaut~ transferred from one space-
craft to another across open space. The PKK w~re eqiii.pped with special
devices for moving along their outside surface. The cosmonauts employed
self-contained SZhO ~nd pressure 3uits fo.r, operatio~s outside the gpacecraft. -
A diagram of the experimental POS is shown in Figure 4.
2 3. a � ~
,
. . '
.
i ^
_ ~x-~ ~ ~ ~
%
- , ~
- ~
~ ` SU t';.~ f ~ ~ , ~ I }l~ ~ ~ ~ _
` ~
~ i /i-,:= ~
_ 7 7 6 5 7 3 7 5
1 =
1 -
Figure 4. Diagram of Experimental Orbital Station ~ssembled From
Soyuz-4 and Soyu~-5 Spacecraft: 1-- solar panels; 2--
instrum~ent-service modules; 3-- orbital modules; 4--
_ exit hatch; 5-- approach anteniias; 6-- docking assembly;
7 windows
6
FOR OFr IC?AL i1S(: ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000100060001-3
~
. FOR OFFICIAL US~ ONLY
The~Gemini was a two-place PKK o� the United States, designed �or �lights
in geocen~ric orbit. The weight of the apacecraft is 3.13-3.80 tons. The
spacecraft consists of the arew compartmen~, radar and orientation sygtem -
modules and also an auxiliary module. An overall view of the Gemini PKK
is shown in Figure 5. Unlike the variant using an ShK, the enti.re space- -
- craft �,vas depresaurized in the given case during the space walk.
,1
,
-I , ~ ~j I ~~''~II
. , , � I''I~II~~~
~ ~
. ,
~ I' r''� -
i;~
. ~ ~ .1:+..~~�' ~I -
� ~ y � ~
~
_ ~f . : I` I'~ ~ ,
~ ~ ~ -
~
~ ~
y.:
.fi ~ ~
. ~ ~ I . G
n , ~ ~S~
. Fic~ure 5. Gemini Manned Spacecraft
The Apollo is a three-place PIQC of the United States, designed to deliver
two astronauts to the lunar surface. The spacecraft is made in the form of
a modular design consistinq of a crew compartment, engine compartment and
lunar lander (Figure 6). Depending on the flight goals, the spacecraft may
consist of the crew compartment and engine compartment the main unit for
flights in geocentric or selenocentric crbit. When an exit onto the lunar
- surface was planned, besides the main unit, the spacecraft contained a lunar
lander which made a soft "landinq" with two astronauta on board. The weight
of the spacecraft, equipped for landing on the M~n, comprises 43-47 tons.
The astronauts made a space walk in geocentric orbit, on the Moon-Earth
route and also walked on the lunar surface. The space walk was accomplished _
through the hatches of the crew compartment and oE the lunar lander and the
wslk on the lunar sufface was made through the front hatch of the lunar lander.
All the spacecraft compartments were depressurized in this case.
The Skylab is~an experimental POS of the United States (Fiqure 7). Three
expeditions of three astronauts in each were sequentially delivered by the
Apol3o tranaport spacecraft to the atation. The weight of the strstion in
orbit without the transport spacecraft is 77 tons. Skylab consists of
~
, ~ 7
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000100060001-3
FOR OF~IC~AL USE ONLY ~
~
. 3 ~
I
~ c- ,1/J , ..i I, -
, ~ C ~ -
~ ` + ;
r K. T:~~ ,'~T1 ~ ~ ~
.
t si; 'ro"n~'''~'.r''� ~ e ~ r ,
~ - - ~ ~
,:1~. i
~ , /~r~V ~ ,
' ~:f'.r"s~ ~ � ~:r~~
~ ~ ~ ~~r~~~ -
I
Figure 6. Apollo Manned Spacecra~t in the Veraion for Landing �
- on the Lunar Sur~ace: 1-- engine compartment~ 2--
crew com~artment= 3-- l~:.isr lander
laboratory and housekeeping compartments; a waste disch:irge compartment, _
docking structure for docking of the transport spacecraft, a aet of as~rono- ;
mical instruments and an airlock chamber. 2'he ShK is hermetically sepArated ,
from ~~he rQmaininq compartments o� this station and permita two astronauts
to make a apace walk simultaneously. During the flight the astronauts made !
repeated walks outside the pressurized compartments of the ntation.
~ ~
~
- ~ 2 ~rr .G.G+-,f~.~ 3 4 -
' ~
~ 'a,.;--^~ :~j~ ? / -
~ ~ i1 ~''~~j; , i 1~~I~
~'~~`f� ~ ~!1;_ ti~riD , I . . ~
~ �.1~~~
~l�~.'~ _ 1~4 � i`'~~,
..n'i:~~y~ ~ _
~ r
� 6 5 -
: ,
Figure 7. Skylab Manned Orbital Station: 1-- laboratory compart- _
ment; 2-- bulkh~ad; 3-- housekeeping compartmenti 4--
_ station enqine; 5-- airlock chamber; 6-- docking
structuret 7-- Apollo transport spacecraf'c
= The MTKK (design) is a multiusE~ transport spacecraft of the United S"_atea.
- The pro~osed crew is three men and up to four specialists. A diaqram of -
the MTKK at the moment of payload injection into orbit is shown in Figure 8.
The MTKK includes firs"~-stagc booster units, second-stage fuel tank
_ 8 .
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~dx o~xicrnL us~ orr~Y
(f~titiiAOnable componenti~) ~.hfl tiha epacecraft itigelf witih p~yload. 2'he pur-
po~e of tihe spaceara�t ie to in~ear varioue payloads intio orbi~ (autonuitiia
~~tellt:~e, POS unitg and eo on), maintenance and repair of unmanned apace
gfetieme, r~ecue of PKK and POS orewa who have experienced distreps in orbiti
and so on. The maximum payload weight in~ected into orbiti is 29.5 tons and
~ thati retiurned tio earth i~ 11.3 tona. The ta~ks enumereted above require -
_ th~ active participation of aetironaute outaide tihe pre~~urized compartments.
,
, ~
, ~ fi
. .
I
~ I~ _
b ~ ii
. .
'~ir,.ti~ ~~r~ t'i ~
f � ~ , ii~
' 1~~i`~''^~~ ~ ~
J~at.~ ~'a ~ >
w ~ ��c"" ' i�~ `a
T,_L'1~"~ s
~ ..tl--'' 1.~, s (I,, ~
:
~ , .
1 ~
tiJ
. ~ 11,,
/ ' ' ti
j/ \
/
i
~
~
Figure 8. Multiuse Tranaport Spacecraft With Open Carqo Compartment
at Moment of Payload Injection
Spacelab (desiq~n) (Figure 9) is a manned orbital unit with a crew of 2-4
persons, desiqned for multiple use. The unit is delivered into orbit and
is returned to earth by an MTKK. The maximum weiqht of the unit is 11 tons.
ita purpose is to carry out gcier~tific-technical experiments in q~ocentric
orbit without separation from the MTKK. The unit may contain preesurized
and tu~preasurized compartments. Zt is planried to conduct routine and un-
planned operations in open space. The design providcs for an ShK with space
exit hatch.
The lonq-term unit orbital station (design) is a manned orbital atation with
an operatinq period up to 10 yeara. The station is assembled in orbit from
9
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007102/09: CIA-RDP82-00850R000100060001-3
. ~Ott tl~~ICIAL USC ONLY
? . J ,�i~ .
. r ~ .
~ ~ ~ ~
- -~?y/~ ~c:
I � L: ~:S:i~
r~~~_~~ ~ ;
r . , 7y` ^ a
�c~:3 .
.i:
i,~R.i.sa
YY J/
y_.~"y -?=z
~ ~
I
F'i,qure 9. Spacelab Unit: 1-- epace exit hatch~ 2-- electrical
and hydraulic communications linea of MTKK syetemst
3-a presaurized con~artmen~
individual unitis delivered by the MTKK. The number of unitis may be differents -
~ from 3 to 6, and accordingly there may be 6-12 crew members. The stations
~ are designed to carry ~uti a complex of scientific inveatigdtiona, to study
the Earth'g naturai tiesources, for operation of sateilite syetems and ~o on.
Obviously, so~~e tasks and specifically operation of autometic apace objects,
will be relatied tio epace walks.
Thu~, dne niay ~onclude that hwnan activity outside a manned space object
occupies one of the central positions in coamonautics.
Zn the following sactions we will discusa the practical operntion~ of cosmo- ~
nauts in open apace accordinq to the USSR and United Etatea programs, methoda
- of perfnrminq them and hardware.
Maintenance and Repair of PKK and POS
Unmanned space apparatu~s (KA) automatic artificial Earth satellites (IS2)
and stations (AS) consist of a set of assemblies and systems havinq �
specffic deaignetion and different deqree of complexity. Man is located
on board PKK or POS. Therefore, special requirements are placed oa their
onboarc~ systems and, moreover, auxiliary systems are installed. Thus, a
life support system which supplies the cosmonaut with oxyqen, water and
nourishment and which removes various types of harmful fmpurities (carbon
dioxide, anmania and so on) from the pressurized cockpit, is absolutely
necessasy. The requirements on the tiemperature-humidity conditiors of the
qaseous m~adium, vibrations, noise, q-forces and on the dependable functioninq
of onboard systems are increased.
PKK and POS are used for national economic and scientific research purposes.
These are, for example, study of the Earth's natural resources, comnunica-
tfons, televisioa and astronomical, astrophysicdl and meteoroloqical investi-
- gations. They are also used to develop the design of various; aystems snodf-
fications, som~ production operations and production processes directly under
space fliqht conditions.
10
FOR OFFICIAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
- ~dtt d~I~ICYAL U5~ ONtY
Liati u~ pree~n~ an approxitnate liet o� tihe onboarfl sye~ems of PKK and POS:
,
th~ etiructure and itie componentst
power aupp].y eyetiemgt
orientiatiioa and etabilization systiem:
aerv~ member syatem (engines, contirol gyroacopes, flywheels and eo on)s
radio control and tra~eatory measurement eyekem=
navigation and control aystem=
ecientific reaearch apparatue (payload)=
life supporti eyatiems
temperatur.e requlating syatemi
reentry and landing support systemt
emerqency rescue system=
radio communications syatems
approach and dockinq support system and so nn.
2'he combination of these aystems may vary as a function of the fliqht qoals.
_ 2'his is primarily related to payload. For example, the multipurpoee Soyuz
PKK may be assembled in difLerent versions as a function of designaLion.
- The self-contained manned astronomical obaervatory (Soyuz-13) was equipped
wi~h astronomical instruments of the Orion-2 system, but there wae no
approach and dockinq system on it. The spacecraft wa~ cquipped with a
weldinq apparatus operating under space vacuum conditions in the experiment
with space welding. The experiment with dockinq two spacecraft and transfer
of two cosmonauts from dne spacecraft to another (5oyuz-4 ar~d Soyuz-5) re-
quired that an approach and docking system, depressurization and pressuriza-
tien systems, pressure suits and other equi~,ment be supplied. The Soyuz
manned spacecraft can be'used as a txansport spacecraft to deliver crew and -
carqo to orbital stations and to return the crew and necessary research
materials to Earth (Soyuz-11, Soyuz-14, Soyuz-17, Soyuz-18, Soyuz-21 and so
, on). In this case the spacecraft is equipped with an approach and dockinq
system, system for transfer to the orbital station and so on and a place is
provided for storing the carqo delivered to Earth.
The operatinq efficiency of onboard systems and their dependabf.lity are
determined by an optimum combination of human activity and operation of
automatic systems and by optimum distribution of functions between the auto-
matic systems and the cosmonaut. Wa feel that it is erroneous to postulate
11
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~Ok O~~tCtAL U5~ ONLY
_ tihe qaeetion itigeif of which is betitiers an P.utom.ation or man. Man hag ~
number ~f l.ndieputable advanti~ges over an automaton, buti ~ti ~he eame tima an
autiomatic eyetem hae a number of quali~iee which are nwre valuabie cort~pare8
to man. For example, automa~ic ~y~tems retain their ef�icienay durinq acce-
leratiior?g ~everal orders higher than doe~ man~
Automatong netn op~arate at anibien~ al.r tiemperatiures on the ord~r of 1000 �C,
whereas overheatiing of the humar~ organism makes him compietely ine�fic~+ent.
The tiolerancea ro ambienti ai,r pres~ure ar~ incomparably more rigid �or man ~
than for ~echnical eyetemg. The radiation resistance of ~lectronic systecn~
is Lene of tiimes higher tihan the permiasible radiation doses for man. How~var,
man with his experience, intuition and akill to combine unrelated events into
a unified loqical system at first qlar~ae and with the akill to separate more
significant inFormation for a given moment and with many other qualiCies
' is an irreplaceable link in apace research. it is obvious rhat the main
engineering-psychological task o� designing space b}~~8Cti9 conaists in
developinq a complex "man-automaton" ayst~m in which the deficienoies of one
link are compenseted by the sdvantages of the other.
7'he scales of inveatiqations in epace will increase and capital investments
in space proqxams will also increase. Thus, the expenditures for the Morcury
Program comprised 0.275 billion dollars, those for the Gemini Program com-
prised 1.290 billion dollars and those for the Apollo Program comprised on
the order of 25 billion dollnrs. St is natural that these capital invest-
ments require an increase of economic effectivenesg of the implemented ~pace -
progrart~. And this in turn is most closely related to the operating depen-
dability of the onboard equipment and with maintenance of it directly during
fliqhti.
The problem of maintenance is rather complex and is detezmined primt~rily by
the fact that the dependability of onboard equipment of space objects depend~
on many flfght factors, the level of development of science and t~chnology,
production and so on. The conditions of space and of the space medium them- -
salves (deep vacuum, cosmic radiation and meteorite hazard) reduce the depen-
dability of space object systems. These factors espPCially affect the assem-
blies and systems located on the outaide surfac~s of space ob3ects and outside
pressuriZed compartments.
This naturally requires periodic inspection, preventive maintenance, replace-
_ ment and repair of systems directly in open space. For example, the prolonged
effect of cosmic radiation may lead to deterioration in the properties of
organfc materials. Solar panels used as power supply sourc4s lose their
efficiency due to the effects of sclar radiation. As a result the energy
required for the onboard equfpment is reduced. A deep vacuum is harmful to
s.aterials, including "volatile" elements, which considerably alters their
operating characteristics.
2he lenqth of a flight is on~ of the main factors affecting the der~~n~tabili.ty
of onboard systems of space objects. The longer the spaceflight, ~he more
difficult it is to ensure the efficiency of the onboard equipment at s given
12
FOR O~FICIAL iJ5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
t~oK c~rrtr.tnt, crs~. c~N~.Y
levpl nf d~p~ndability. Ar~d after a11 crew e~fety depends an ~h~ oper~,~ing
dependability of ~he ~y~tiems and agsemblies of ~he m~nn~d ~pacecr~yft or
etiatiion.
7'he length of mann~d flightig is ever increa~ing: 424 hour~ for 5oyuz-9,
1,512 hourg for the autonomous flight of the 5oyuz~l8 and ~h~ ~light with
th~ 5alyut-4 orbital acien~i�ic station and �rom 143 to 29~ haura for lunar
expeditiione. 7'he length of three crews remAining onbo~rd the Skylab manned
orbital sta~ion comprised 28 d~ys, 59 days and 84 days, respecttvely. The
leng~h of the flight of the firat expedi~ion on ~he 5alyut-4 PO5 comprised
30 days and that of the second comprised 63 days. Thus, development of
orbital n~ar-Earth, near-lunar and lunar stations of long 3ura~ion r~nd
organization of expeditions to the plan~ts requi.re dependable funci:ioninq
of the onboard systems for many months and aven years.
One cgn vividly dpmonstirate the dependenc~ of the probability of successful
manned flight on its duration at the modern level of production technology:
flight to Mars (length of flight approximately 2 year~) 0.006t
flight to Venus 0.03j
lunar landing by the Apollo proqram (length of flight 200 hours) 0.95t
flignt of the Mercury spacecraft 0.95.
_ F~suring the probability of a successful flight of 0.95 may be taken as the
required dependability for manned flight.
, The dependability of a space rocket complex can be determined by t~e expres-
sion: -
P ~ P,P=P,P~,
(1)
where P1 is the dependability of the complex during launch; P2 is the depen-
dability of the complex during orbital injectionj P3 is the dependability _
of the complex during orbital flight; and Pq is the dependability of the
complex during reehtry and landing.
If the value of P is close to 100 percent (for example, 95 percent), one may
be confiden~ of a successful space flight. 2n the remaining (1 - P) = R
cases, it is obvious that additional measures r.re required to ensure depen-
dable operation of the onboard equipment. If the ~3egree of dependability
of auxiliary equipment is taken as PdoP, then one can determine the probabi-
lity of a successful flight:
p~ ~ 1 R � R~on~
13
FOR OI~FICIAL U5E ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
xax o~~ici',AL US~ ONLY
where
~ Rnon ~ ~ paw~� t2 )
What additiional measures may increase the level a� operA~ing dependabili.ty
o� onboard equipmen~ and consequently of crew safety and the ~f�3.ci~ncy of
~he enrire space flight? One of the passible m~thods o� solving this pro-
blem ia maintenance (TO) of tihe onboard equipment o� the PKK and POS during
spac~ flight.
Mttintenance of~.the onboard syatems can be divided into automatic, remo~e and
crew servicing.
Automatic TO asaumes automatic identiftaation o� failure and switching to
reserve sy~tems or units. ~
Remote TO may be organized by using an unmanned eatellite equipped with
manipulators, television camera and other equipment. The operator can con-
trol the satellite from Earth, from the PKK or the POS. Crew maintenance
asaumes, for example, diagnosis, repair and postirepair checks with direct
participation of the cosmonaut.
Each of these types of TO has its own pasi~ive and neqative aspects. Auta-
matic T0, thQ most prevalent today, is related to an increase in the weiqht
and overall dimensions of space objects. Remote TO has i~s own limita~ions
the use of manipulators, delay of control signals and so on. Moreover,
each of the types of TO may be suitable or unauitable depending on the
phase of space flight.
For exartq~le, the most preferrable for the orbital injection phase is auto-
matic T0, which is related to transience of processes and characteristics
of crew disposition: the cosmonauts are located in the reentry vehicle and
are strapped to their seats and the flight proceeds under acceleration con- -
ditions. TO by the crew is reduced in this case to switahing to reaerve
units upon failure of automa~ic equipment.
ltemote and crew TO is feasible during orbital flight and during the stay on
the surface (for example, on the Moon or Mars). The possfbility of unfore-
seen failures, difficult access to repair points and many oi:her factors, we
feel, make crew TO more preferrable. Crew maintenance assumes the participa-
tion o� man in identification of the failure, determining the degree of im-
portance and proqram of repair, selectinq the hardware required for repair
and conducting the repair and postrepair ch~cks themselves.
The dependability of a typical life support system with different variants
of maintenance is shown in Figure 10.
Op~erations inside pressurized compartments have their own features related
specifically to weiqhtleseness. Hawever, the operating conditions in
14
FOR OFFICIAL USE OM.X
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
t~'OR nFFIC~AL USC (1NLY
~
~
P~
~
QQ i ( 2 ) fexNUVrc~r~
3~ o4,Gi~uiv~
M
Q 1
a~ DJ/O~upo6a~ver? I
' Q6 IQl1 17XMU40.fA`CtYiOdG1q� -
~ ff~ ~~..~~1.....~....~ AYf/6G~NU.O 7X~J/70M'PK ~ I `
~ Z ~ ~ Q4 � -..~...r
aJ ~4 ~
Bydeund9ynva u
Q? P 1P1Nf
/4PCM?D
Q/ ~JKUIXi.~YONUp
~I
I
I
0 ,70 X~ !SO ?LO 250 .l00 ,lSO 400 4S0 Y%iJ
~oona none~u (5) -
Figure 10. Dependability of SziiO Witih Different Variantis o� Main-
tenance
1~Y :
1. System dependability 4. Without duplication and without -
2. Crew maintenance crew maintenance
3. With duplication without crew 5. xime of flight
maintenance
_ pressurized compartments can be approximated to ground conditians ta some
degree. Universal attaching devices, equipping the operator's posi~iona,
spare parts and tools, design of onboard equipn~ent convenient for performinc~
repair-maintenance operations and so on have been created for this.
Another matter of working outside the space~cra�t or station. The specifics
- of ~he medium, the need for donning a pressure auit and in some cases a
weightless state all create some characteristic features and limitations
during maintenance and requires solution of a number of problems.
As we have already said, man will perform a complex of engineering-technical
tasks and specifically of repair-preventive maintenance operations with the ~
onboard equfpment and maintenanre of it outside the spacecraft or station.
A significant part of the apparatus-assembly composition of the onboard
systems of PKK and POS fs related to one or another d~qree to open space,
i.e., to the space outside the pressurized compartments of the space object. ~
For example, the temperature control system is based on dissipation of heat
by a heat carriers in radiat~ve heat exchanqers located on the outside sur-
face of the spacecraft. The solar panels (SB) of the power supply system,
for example, on the Soyuz PKK and on the Salyut and Skylab POS are also
located on the outside surfaces. The main SB of the Skylab station have
dimensions of 30 m on the station unit. Both SB consist of three sections
- mounted on cantilevers. The cantilevers with panels are folded and are
attached to the station hull by explosive bolts until orbital injection.
After orbital injection, the panels are unfolded into the working position.
In unfolded form, each panel has dimensions of 7 x 9 m and the total area
of the SB is 110 m2.
15
FOR OFFICIAL USE ONLY
I
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
.
FOR OFFTCIAL US~ ONLY
,
- The radio communications system also has a~et of various antenna~ in-
etialled into the working posi~ion after orbi~al in~ection of the obJeC~.
Thie ia ralgted, Eor exampl~, to the pencil-beam an~enna of the ~pollo
apacecraft, which occupiee the working position after separation of the ~
sp~cecraft from the third stage ef the 5aturn-5 carrier ro~k~t. The salar
panels and radio antennag of the approach and docking system of the Soyuz
pIC~ are also unfoldad af~er orbital in~ection. _
The orientation and stabilization syetem communicates with the surrounding
space by various types of eennors used in controllinq the spatial pasition
of the apacecraft. This is related, specifically, to optical sen~ors: the
optical sighting device of t.he Soyuz PKK, ~he optical periscope used by the
astronaut during control of the Mercury PKK, aetronavigation instrume:it~ of _
the Apollo PKK and so on.
2'he enqine units to provide forward motion ar?d rotary motion around the _
_ center of masa and also tanks w3th spare fuel, oxidizer and compressed -
gases are located on the outaide of the spacecraft in open space. It is
natural that situntions may occur durinq the space flight which require the
cosmonaut's participation in maintenance of the systems, i.e., a space walk.
Now familiarize yourself witr. Table l. The distribution of operations in
- open ~pace durfng maintenance of individual systems of PKK or POS is given
in it. ,
Table 1
Requirements on prevalent ~erations conducted '
System outside the PKK,
type of TO percent
Approach Replacement of components 50
and docking
Engines Elimination of leaks and replacement
of components a0
~lectzonics Replacement of components 10
Structure Repair of detected damage and elimina- ~
tion of leaks 75
Power supply Replacement of components 10
sources
SZhO Elimination of leaks and contaminants
and replacement of components 5
These are theoretical investigations and conclusions. But they have been
confirmed by the practice of space flights. Analysis of failures during
the flights of the Mercury, Gemini and Apollo PKK and of the Skylab POS
permits one to determine the most typical malfunctions which may potentially
cause or have caused the necessity of working in open space:
16
FOR OFFICZAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~ox o~rrcrnL us~ ornY
aepara~ion o� tihe me~~orite-heat shieldi instiallation o� the heat shield
cover in orbit (Skylab)~
�vroed unfolding o~ the solar panel directly in orbit (Skylab)s
weakening of the attachment o� individual structura]. assemblias and parts
(Mercury and Apollo)~
failure o� ahroud ejection machanisms (Gemini)=
clouding of the windows and the ob~ectives of optical instruments (Apollo
aad 5kylab)=
eng3ne failurea (Gemini and Apollo)s
leakages of gases and liquids (Apollo and Skylab);
faiZures in the docking system (Apollo).
The attachment of the heat shield to the satellite hull weakened during the
flight of the M~rcury-6 spacecraft. To refine tha nature of the malfunction~
a space walk could have be~n required but thia was not planned for the
Mercury program. The flight managers and the crew of the Gemini-9 spaoe-
- craft were forced to abandon docking since the main shroud~designed for
docking did not separate from the apparatus. The crew proposed a apace
walk to cut the electric wires holding the main shroud. However, the flight
managers rejected these operatiions. The windows became very clouded during `
the flight of the Gemini spacecraft. This interfered with observation and
- photography from the spacecraft. The clouding was removed by a~ astronaut
who emerged from the spacecraft.
The optical instrum.ents and windows also became clouded during the flight
of the Apollo spacecraft. During the flight of the Apollo-17 spacecraft,
astronaut Evans made a space walk. Inspecting the housing of the engine
compartment, he detected delamination of the upper coating burned by a stream
of hot gases flowing from the engines.
= Human capabilities in performing repair operations and maintenance in open
space were manifested more clearly during the flight of the Skylab orbital
station,. Analyzing the crew activity of the Skylab station, one can co~ile
a cyclogram of maintenance and repair operations in open space:
_ preparation to emerge beyond the pressurized compartments;
space walk;
moving to the locatian of preventive maintenance or repair;
inspection of the assemblies and systems and checking them;
17
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FO~t O~~TCTAL US~ ONLY
per�orming ~he r~quired operationss
checkir~g operations after completing the works
_ ,
- mov~.ng ~o ~he airlock chamberj
entry into ~he prasetucized compartments of the gpacecraft or station. ~
Zfiis diagram is considerably simplified. Et~ah specific case may have its _
own cha~acteristic features which aonsidarably change the sequence of events.
E'or example, one may be required to ascertain installation-disassembly
operatians, tranapor~ of spare components and so on wi~h reapect to the
maintenance object. W'hen moving along the outside surface of the spacecrafti, ~
station or other space object, the cosmonaut may perform such an important
- operation ae inspection of the outaide sur�ace. This is neoessary when
gas leaks from ~he pressurized compartments or leaks of liquids occur, for
- example,~�rom the circuit of the heat control system.
. Thus, inspection and repair are important constituent parts of maintenance
and speci�ically of maint~snanae related to operations outside the spacecraft.
Inspections, rep].acement or repair of solar panels, ejection or unfolding
and dockinq mechanisme, removal of contaminanta from optical aurfaces, cor-
rection of spacecraft assembly and subassembly failures, repair of antennas
and drives, determination of leakaqe points and inspection of the external
surface are far from all the maintenance tasks solved outside the space-
craft in open apace.
" Maintenance and Repair of Automatic Space Objects
Along with investigations in space using mar!ned space objects, wimanned
space objects are employed just as extensively and many more unmanned objects
- than manned objects are launched.
Automatic devices in space perform the role of scout, from the tracks of -
which man can operate. Thus it was with investigation of circumterrestrial
- space and thue it was with orqanization of expeditions to the Moon. Actually,
- the flight of Earth's first space ambassador, citizen of the Soviet Union
Yu. A. Gagarin, followed after launching of the first artificial Earth
satellite, biological satellites and after careful preparation and testing.
Automatic devices also laid the road of man to the Moon. The Luna and Zond
automatic stations provided study of circumlunar space and of the lunar
surface, a soft landing on the lunar surface, complex investigations from
selenocentric orbit and collecting samples of lunar soil with return to
Earth. The flights of the American sutomatic apparatus for study of the
~ Moon were Ranqer, Lunar Orbiter and Surveyor. And finally, landing of
J astronauts on the lunar surface oc aurred after manned circumnavigation of
the Moon. Venera, Mars, Pioneer and Mariner automatic interplanetary stations
are now periodically launched toward the planets of the solar system. Aufio-
matic devices will also be used in the future as the first investigators,
18
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
_ ~'0~ O~~TCIAL U5~ ONLY
especiaily in unstudied or littie studied environments and environmenta
unauitable for human viability.
~ ISZ and KA are r~ow used extensively for soientific research, to check new
technical ~dlutione (modernization of cor~troi eys~ems, testing of new
enqines and so on), to study the Earth's natural resources and o0 on.
ZS2 and KA may be classi�ied by deaignation in the following manner:
_ research sate].lites~ _
_ communications sate].liteat
meteorological satellitest ~
geodetic eatellitesj
sa~elli~es to investiga~e the Earth's natural re~ourcesj
satellites to test onboard and ground equipmentr
btological satellitesj
KA for investiqations of the Moon and planeta.
ISZ and KA carry on board apparatus designed for extensive and multifaceted
investiqations and observations. Thia ;s frequently unique, complex and ~
very expensive apparatus. Besides the payload (the apparatus of purpoaeful _
designation), ISZ have the same systems as any controlled spaae object, i.e.,
orientation and stabilization system, servo member system (engines, control
gyroscopes, flywheels, gravity stabilization com~onents and so on), electric
power supply systems, temperatur~ control aystems, radiu engineering system
and many othere. But the systems required to aupport man on board are absent
in them.
_ It is natural that various malfunctions may occur both in service systems
and in research apparatus. There were cases in the practice of space resee?rch
when the most insignificant failures on board the ISZ or KA either made them
generally unsuitable for operation or sharply reducad the efficiency of their
practical use. This is related primarily to components which begin to func- -
tion after orbital injection: the solar panels, antennas, aomponents of the
- orfentation and stabilization gravity system and so on.
An emergency condition of ISZ systems may require replacement and this is
related to a nonroutine launch with additional expenses. If it becomes
possible to conduct repair operations in orbit, these expenses will be reduced
and re~air in orbit itself may be economically effective.
Systems are created from automatic satellites in orbit which provide communi-
cations, naviqation and meteorological service. For example, the Transit
19 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
_
~'OR OFFIC]:AL US~ ONLY -
sys~em of �ive sa~ellitea aesis~s shipa o� th~ commercial �leet and oceano-
yraphio vessels to determine their coordinates. Among tihe totial number of
sa~sllites of this ciass, part were in~ected into orbit to replaae ~hose `
thati failed. Navigation and geodetic sate111tes and thoge for d~velopmnnt
oE onboard equi~ment utilize the qravity orientation and stabilization aystem.
The American ATS eatellite,~'deaigned to tes~ onboard and ground equipment of
future meteoro].ogical, communications and navigation satellite syatems and
for scien~ific research, is equipped with a similar system. 7~`hese orienta-
- tion and stabilization systems do not require cantiinuous fuel expenditure
(compressed gases and liquid fub.l) stored on board the satellite. There
are four hollow rods in tha system form~d after orbital injection of the -
satellite by "pulling" a tension tap~ lubricated from a drum ~hrough a
filler. The mds seemingly form a cross. The angular dis~ance be~ween rods
and their length are regulated. ZWo similar rods serve to damp librations.
These systems have a considerable number of movable ftl8Ct18I1~Cd1 subassemblies.
Failures in them may lead to a reduction of efficiency or may lead generally
to failure of the satell3te due to loss of orientation.
A characteristic feature of interplanetary spacecraft, like some ISZ operating
in high orbits (ag~proximately 11,000 km), is a second start after injection
' into the intermediate geocentric orbit. The onboard equipment is checked
' and prepared for the second start during flight in the initial geocentric
orbit for transfer to the specific trajectory. It is during this period,
the period when the apparatus is in the intermediate geocentric orbit, that
, humdn interference is possible. Man can check and rQpair failed onboard
equipment. Thus, for example, American specialists were forced to cancel
operation of a number of ISZ (ATS-IV INTELSAT-2,-3 and so on) due to injec-
tion into an uncalculated trajectory. If a manned spacecraft (for example
of the MTI~ type) had been in orbit around the Earth, the cosmonauts could
have transferred the satellite into the calculated orbit.
There were cases fn the practice of space flights when the propulsion unit
of the satellite itself was used to inject the ISZ iizto the calculated orbit
due to operational failures of the carrier rocket power plant. This was
reflected in the further flight program sinca it reduced the onboard fuel
and oxidant reserves. Moreover, leakage and total depletion of the fuel
reserves may occur while retaining the efficiency of ~ther onboard systems. _
Even in these ~ases ISZ and other unmanned space objects could be refueled
directly dur3.ng space flight.
_ The designs of future automatic space objects already provide during the
development staqe their maintenance by man directly in orbit.
- Thus, for example, it is planned to inject the LST satellite (observatory
for astronomical observations weighinq 11 tons and 3.6 m in diameter) into
an orbit 610 km high by using the MTKK. The satellite will be returned to
Earth for preventive maintenanca a�ter 2.5 years of operation also by using
the MTKK. Longer operation without repair is unprofitable and reduction of
- the operating period requires additional flights of the MTKK which is relat~d
20
FOR O~FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~ ~o~ o~r~cz~, Usc oNr~Y
tio high expendi~ures. Therefore, maintenance of the satellite by cosmonauts
tiransported to it by ~he MTKK is provided during the opera~ing period. Three
tiypes ot maintanance are planned:
1. Delivery o� the satiellite into the MTKK payload compartment. The aolar
panels, an~ennas and other components of ~he satellite struc~ure unfolded in _
orbit muat be folded.
2. Grasping the satellite with the MTKK manipulator and attaahing it at some
distance from the spacecraft.
3. Arranging the satellite on the MTKK work ~able.
_ All three types of maintenance require the active efforts of m~n in open space.
- According to the first design, the satellite hull should ba presaurized with
an arti�icial atmosphere so that the cosmonauts can work without pressure suits
when performing maintenance operations. The second desiqn (it is regarded as
more preferable) ~nvisions the manufacture of a satellite with an unpressurized
hull; the cosmonauts will work in pressure suits when maintaining it.
The active interference of 3nan may also be required during failure of
satellite-to-earth data systems on the ISZ. The cosmonaut may ~hen disassem-
ble the information carrier from the ISZ and return it to Earth. A similar
experiment was conducted during the flight of the Gemini-10 PKK when an
astronaut, making a space walk, transferred to the unmanned objec~ and dis-
mantled a device with scientific information. In ~his case the Gemini-10
PKK was at a distance of severaZ metera from the object.
- Maintenance of automatic objects and systems in orbit poses'new problems and
primarily problems of detecting the object in orbit, approach or docking
_ with it, transferring the cosmonaut to the object for maintenance, trans-
porting the equipment required for work and, finally, conducting checks,
preventive maintenance or repair operations. The problem of vital activity
under weightlessness conditions occupies a siqnificant place in this multi-
- faceted activi~y of :nan in open space.
Thus, one can conclude that maintenance of automatic ISZ, space t~pparatus
and satellite systems requires the active efforts of man in orbit. Preven-
tive maintenance, gathering useful information, inspection and estimation
. of operational suitability, unplanned repair, transfer to workinq orbits,
transport, refueling and so on may primarily be related to these operations.
- Installatiori-Dismantling and Assembly Operations in Orbit
If one traces the course of development of space research, one can clearly
see the trend toward an increase in the wetght and overall dimensions of
space apparatus.
I 21 -
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR OFFICIAL iJSL ONLY
This is quite explainable ai.nce the payload volume and wQight and conse-
quently the weight and overall dimensions o~ the spacearaft and carrl.er
rocket structiure i~crease wi~h expansion of research.
The tasks of the future require orbital assembly of even heavier space ob~ecte
and even larger etructures. These nbjects may primarily include intierplane-
tary spacecra�t, block orbital stations of long duration, refueling stations,
large (more than 30 m in diameter) radio and optical telescopes and larga-
� area solar panels. For example, one of the future NASA programs ie a epace-
craft variant assembled in orbit from separate units. Uz~it injection into
orbit and subsequenti supply of the :~pacecraft will be providad by the 'r1TKK~
Development of ari orbital atation of long dura~ion also providea assembly
~ of it in orbit from individual units delivered by the MTKK to the installa-
tion orbit (1-2 units of the station per flight).
Naturally, installation-dismantling and assembly operations are required
when performing a wide range o� tasks in open space:
maintenance and repair of manned and automatic space objects;
assembly of small and large space objects;
technical supply of manned stations;
crew rescue;
servicing scientific apparatus and experimental investiqations.
_ Actually, dismar~t.,ling operations of failed units and subassemblies with ~
subsequent installation of them may be required.
~
For example, antenna diameter in the future may reach 30-45 m. Such an
antenna would have to be assembled in orbit in open space. In this case the
mcst optimum is apparently a modular design of the system. Individual
modules will be assembled into one unit and this means that operations in
transporting the units to the assembly point, joining them and adjustment
will have to be carried out. For example, a 45-meter antenna can be ~
assembled from 240 panels. The technology of installation on this scale
is unusual since the "escalator" principle is employed. This prfnciple
inclu:~ies the fact that the cosmonaut "stands" on a platform (Figure 11) and
beqins ~ssembly with a round section, moving along a rail installation as
need rec~uires. Having finished one row in this manner, he transfers to the
second and so on. Simulation of this type of assembly operations in a
hydraulic environment shawed that the cosmonaut is capable of effectively
completing the required installation. During tests, an operator in a pres-
. sure suit should raise the panel, incline it at an angle of 90�, rotate it
by angle of 90� and secure it at the corresponding position on the round
panel. Some results of tests are presented in Table 2.
22
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000100060001-3
t~dlt O~H'IGiAL t15~ ONLY
Tab 1p x
_ ._.......~-o-
(1~ Cpestnee epe`~~enrram,~o�
Puiuep n~He~a~ ~ woe no ~~anrtNO ~nn~
. f,b x ~~5 0,54
- f,2 x 2,4 d~d' ~
2,25 X 3 0~~
it~Y:
1. Panel dimensiong, m 2. Averaqe time expended nn in-
~tiallation, min
These datia ere very valuable when ealculating the totial time to aae~mble
thig antenna. Uttlizing time expenditures, one can compile the total
cyclogram of the cogmonaut'e work during aeeembly.
.
�
i
~
�
d dJ
Figure 11. Sequence of Antenna Assembly: a-- beginning of
- assemblys b-- asgembly of firat rowt c-- aesembly
oi secand rows d-- end of assembly
The teata clearly ahowed that a pressura suit, life support system, ir?etalla-
tion for movinq the cosmonaut, systems for tetherinq the cosmonaut and re-
quired equipment, developed method and technique of assembly are required
when pi:zforminq these operations.
One c.f the most important assembly operatfons in space is transport of in-
divi~iual modules to the assembly point. When performing thia operation,
the cosmonaut must move in supported (with contact of some surface) and un-
supported space. Individual units for movinq the cosmonaut are required for
workinq in unsupported space.
23
FOR O~FICLAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPROVED FOR RELEASE: 2007/02109: CIA-RDP82-00850R000100060001-3
. ~UR ()1~'~ICIAL U5~ ONLY
' Mainten~na~ of ~ci~ntifia ite~@aroh App~~atu~
Mui~ilati~r~1 inv~s~lgatiion,~ durS,ng ~pac~ flighti r~quir~ vaYi~u~ typ~g of
gei~nti3,f~,c r~e~~rch ~quipment on board tih~ space ob~~eti. 1'hi~ ~quipmene i~
- plac~d botih ingi8~ the preg~uri~ed compartment~ of PKK, pdS ancl automatSe
iS~ ana on tih~ir ouCsid~ surfacee.
2"he appar~tu~ placed in open g~~c~ may include inetrument~ for studying
m~e~~r ~ra~i~n and rontamination o~ eurfacee, to l.nvegtiiga~e ~ogmic rey~
and ~t~liar radietion end in~tizuments to rske apectirographg of the ~arth'g
gurfac~. Mor~over, th~ ef�~ct of flight cond~tiona on various ~tructiurai
m~t~rial~ wag inv~~tigated durinq eome space fiights. Th~ scientiific-
t~chnical apparaCus ig placed in open ~pac~ if it cannot be locati~d in pr~s-
guriz~d compar~nCg or, for ex~mple, if the ir~strument muat be separatied
from the ~pace objeot hull to eliminate ~he epace object's effect on th~
result~ of investigations.
_ Anely~is of expercl~ntg pianned durinq spac~ fliqh~s show~ that part of
them requires active ~ffortis of man in open gpace. 2b 8etermine his func-
_ tions beyond the gpacecrafti, 1,212 scientific-technicel experiments planned
- by NASA for the period 1968-1980 for PKK makinq fiights in geocentric orbitg,
wer~ analyz~d. The iist of studied experimente included:
a$tronomical~ -
biologicali
investiqatione in the study of the ~arth'g natural resour~:es and of space
physirst
meteoroloqical inveetigations;
investigations of communications equipmentt
experiments related to tests of new space technology and the use of orbitfll
sys tems j
- investigations in the field of space medicine.
_ Of the 1,212 experiments, approximately 500 were analyzed to deteztnine the
functions of man inside the PKK and during space walks. The following
cosmonaut functions were determined:
installation and unfoldinq of scientific equipment=
testing and chackinq scientific equipmentj
levellinq and calibration;
controlling the operation of apparatus=
~ 24
~Oit OFFICIAL U5~ Otv'LY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~dit nt~'I~'ICYAL USC ONi,Y
mcvinq lnad~ and contiain~re wl.tih ~h~ re~uit~ of ~cien~if3c ob~ervatitoner
removal and replaoemanti of unitiet
repair. ~
Moreover, 56 percenti of tiha total number og opera~ton~ require work in open
spaee. 5pecifically, these are operations related ~o as~ronomical inve~ri-
qations attd tio stiudy of ti1t~ Ear~h'g n~~ura1 resourees. Medical and biologi-
e~x1 investiigatiions require only 5 and 22 pereenti, respectively, of operations
in open ~pace.
_ Scientifie-tschnical experimen~g related to operatinns in open space are
present~d in Table 3.
Only work with the X-ray focusing teleacope may reqt~ire several operations
in open epaee:
1. Correation of maifunctiions after unfolding the structurai components in
space and so on.
2. Adjustment by using ~ager eyatems.
~ 3. Connecting the monitoring apparatus, the data orientation syatem, solar
panels and other operations.
Table 3
Area of inveatigations ~cperiments
Astronomical One-u?eter telescopesf set for study of the
sun (90-centimeter solar telsscope)s X-ray
facusinq telescopet 10-kilometer interfero-
meter
Biological Capture of microorqanic cultures in circum-
terrestrial orbitt effect of space conditions
on bacteria sporeas effect of space condi-
tions on primatess multipurpoee bioloqical
installations
Communications and Assembly of 30-45-meter antennas and large
navigation enerqy units
Physfcal Investiqations of the physics of space and
of the luminescence of aunrises and sunsats
Development of produc- AssemUly of larqe structures; assembly of
tion operations in orbit orbital stations around the earth
Experiments with a aet of scientifi~ apparatus for studyinq the Sun and -
other celestial bodies were conducted extansively durinq the Skylab proqram.
The complex was controlled from the inner pressurized compartments of the -
25 -
FOR OFFtCIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOR O~I~'ICIAL US~ ONLY
orbi~al etiaCion. Maintenance o~ the complex r~quired tih~ astiron~uti'e working
3n open ~paee ex~r~atiing caeeettieg with ~he reeultis o� photosurv~y~,
~ransporti nf tham tio ~h~ l.nner comp~r~nte of the atiatiion and inst~ilatiion
of new film casaetitieg. The aetronauta wer~ gupposed tio par�ornt tihe followSng
opera~ione:
pr~p~ratory opera~iona, depreeeuri~at.ion of the airiook chamber, openiag tih~
hetch and te~king a epaae waikt
n~oving to the work sitie by meane of hand holdat
prepara~ion of the work ei~e=
dismantiing of Qasaettest
moving to the airiock chamber~
entierinq the airlock chamber, olosinq the hetch and pressurizing the chamberr
~ransfer to the station aompartmenta.
Three main work sites: "centrai," "peripheral" and that near the ShK, are
provided at the site o� the planned op~rations. T,ro eatronauts ehould take
a space walk to replace cassettee. They exi~ throuqh the ShK hatch in turri
(oxygen is supplied throuqh the tether fmm the orbital station'e SZhO). One
astmnaut remaina ne~r the ShK hatch while the other moves alonq the station
atructure. He moves by usinq hand holds located on the outer skin of the
station. There are seauring devices for the feet at the work sites so that
the astronaut, having freed his hende, cnn extract the caseettes. The estro-
naut ahould rotate the set of astronomical instruments at the "central" work
site to replace the next caseette. Four casaettes are replaced here. The
astronaut~~climbs" tio the work area along a ladder to replace two caesettes
at the "peripheral" work site. The cassettes with the expoaed film nre
transported by the second astronaut by means of special manipulatora for
servicing the work aites. Cassettes with u~expos~~ s!'_- ~::,a txansforred.
- Trro aatronauts must oompiete 150 operatioYis durinq 2 hours (by calculation)
to replace six cassettes. The third crew member is located inside the POS
near the console during these operations to monitor the operation of the
o.~boara systems and to provide insurance to the astronauts w~rkinq in space.
The standby variant of transporting the cassettes provided for uainq a loop-
shaped rope retrdcted by hand.
The final step is to deliver the film with the results of scientific observa-
tions to Earth by n~ans of the tranaport eupply spacecraft.
An experiment for trapping meteor particles an8 microorganisms was planned
and partially conducted in the Gemini proqram. Tt~e scientific equipment
(a holder with various types of traps and also with bioloqical specimens
for investiqatian of the affect of space tliqht conditiions on them) was in-
stalled on the unmanned Aqene obiect ia the region of the dockinq assembly.
26 ~
FOK O~FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOIt tl~'~~CIAL USC dNLY
. The sohem~ for performl,ng ~he experim~n~ providad tiha~ tihe agtronauti tiak~e
a spae~ walk a�~ar tihe manned Qemi.nl. sa~elliti~ doak~ wit.h th~ unmann~d
Agena ob~eat ar~d opene ~he door~ of tih~ holder, �old~d durSng tihe orbl~tigi
in~ect3on leg. ~n on~ of ~he subeequent fli,ghtie, anoth~r aetronau~ d~~mantle~
the holdar after docking during ex~ravehicular uperatiione and da~ivers it Co
t.he apacearegt and ~hen ~o Ear~h. An experimenti wi~h this taohnique, buti
withouti dooking~ wae perforn?ed dur3ng tihe glighti of tihe C~mini-10 PKK.
Thue, one may conclude that man's working in op~n gpace n~cupies one of ~he
central posltiions bo~h in compie~ed ar~d in fu~ure scien~ifiC inveetiigations
using mannsd and unmanned equipment.
A~si~~ing PKK Crsw During Space Flightit Replacement of Crew~
Th~ etatis~ice of manned flightis shows thati, d~gpite tihe increase of the
dependability of onboard equipment, there ie the probability of malfunations
occurring whsch threataned the life of the cosmonauti. We have alrendy men-
tSoned failures during the maru~ed flights of the Gemini-8 anfl Apollo-13
spacecrafti. Malfunationa complicated th~ proc~~~ of 8ocking the main unit
of the spacecrafti tio the lur~ar module during some fiights of the Apollo PKK
~ of various typea. A leak of fuel component~ oacurred in the engine unit of
- the Apollo trangport spacecraft during the flight of the Skylab orbital
station. The malfunction was so aerioue in nature that a decision was made
to prepara a modified rescu~ spacecraft for launch ta evacuate tihe crew from
the atatiion. And thig decision was canaelled oniy after careful ch~ecks on
, the Earth and workinq out the mathods of reentry on eimulator~.
These examples show that operations of aseistinq pKK crews in orbit are not -
imaginary, but quite real. Such typical malfunctions as failure of the pro-
pulsion units, orientiation and stabilization aystesng, depresaurization of -
pressurized compartmente, failure of the life support and temperature cor.tirol
_ system of the PKK arid other malfunctions in transport auppl,y spacecraft
docked to orbital atations may requfre urqent measures to evacuate the crew
from the object and subsequent return to Eartt~. -
There are now several ooncepts which provide rescue of cosmonauts in flight:
1. The preaence of individual rescue devices on board.
2. The presence of a special rescue apacecraft in orbit near the orbi~ of
the PKK or POS.
3. Launch of a rescue spacecraft from Earth if an unplanned situation occurs
on the PKK or POS.
In the first case the rescue equipment is placed on board the manned orbital
object. T'he desigr~ of thfs apparatus is shown in Fiqure 12. The technology
of its practical use includes the followinq. If an unplanned situation
occurs in orbit, the cosmonaut in a pres~ure suit with SZhO ehould be located
inside the apparatus, should prepar~e it for return to Earth, should leave the
27
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOR O~~ICIAL USE ONLY "~~I
emargency ob~ea~, ehould orien~ iti, ,awi~oh on tihe retirorocketi angine to ~
provide raun~ry? ar?d ehould land with a p~rachutie. Resaue ueing r.hese ~ypeg
of apparatue requires tha~ tiha cnetnonauti fulfill operatiions 3n opan space.
i
: I
i~~ ,
,
~l ~ \
~ ~ A -
~ ~
/ II ~ .
~s~~~,,,,~i` I f. ~
i, ~t ~
~ .
~ , ,
Figure 12. Individual Rescue Apparatus: 1-- solid-fuel retro-
rockett 2-- taii section of enginet 3-- enqine
orientation lever= 4-- windowi 5-- quick fastenerj
6-- inner shellt 7-- parachutet 8-- compressed
oxyqen ta~nk~ 9-- ablation coating= 10 inner coating
Equipment of the aecond type includes a rescue spacecraft designed to assi~t
the crew durinq fliqht to the Moon. The technology of uaing the reecue
apacecraft is explaiaed in Figure 13. The use of this design saves consi-
derable time frcm the moment the ur~planned situation occura until the crew
is evac~aated from the spacecraft. Evacuation can be accomplished throuqh
the inner pressurized passages formed after the spacecraft dock or across
open apace if docking ia impoasible (~ailure of the dockinq aubassembly,
wncompensated rotation and so on).
M example of a third rescue varian~ is the program for evacuating the crew
from the Skylab station. in the event of an unplanned situation, return to
Earth is accomplished in the Apollo transport spaceCraft docked to the station.
If the transport spacecraft cannot be used (m~lfunctions occur in the systems
of the transport spacecraft itself), a rescue apacecraft should be launched
fran Earth. The rescue s~acecraft may dock to the side (second) dockinq
assembly of the mooring structure. It is typical that the rescue epacecraft
is a nadificatfon of the Apollo transport spacecraft. Only 24 hours are re-
quired to reequip it to a rescue variant.
28
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~oK orrtr,t~r. us~ nrtLY
~ ,
,
~~~a o
~~.�r \
~ / e
~l~ J
Figure .L3. Diagram of Uaing Reecue Spacacraft: 1-- resoue
epacecraft in pArking orbitir 2-- launch of lunar
epacecraftt 3-- reacue apacecraft at moment of
- luner epaaecraft launch= 4-- lunar gpacecraft at
ntcment emerqen~y occura= 5-- rescue spacec~~aft at
moment of emerqency on lunar gpacecraftt 6--
encounter of regcue spacecraft and emergency lunar -
spacecraft
KEY: i
l. Earth ? Moon
There is a desiqn of a rescue apacecraft based on the Ge.mini PKK, which
ie named Big Gemini. Z'he aadified epacecraft ie deaiqned to rescue three
aetmnauta.
~ One of the taeke of the MTKK is to rescue the personnel of manned orbital
objects. The maneuverinq capability of the MTKK with variation of the
angle of inclination of orbit and brief periods of preparation for the next
launch (approximately 14 days from the momant of r~turn to Earth) considerably
increase the poseibilitiea of crew rescue. Specifically, the time the crew
ia in orbit after an unplanned eituation occurs iE roduced.
Z`z~anefer to the rescue apacecreft, accomplished without mutual docking of
the objects, envisions such an important operation as moving fn unsupported
. apace. But it may happen that the crew members are injured. In this aitua- -
tion the transfer operation is replaced by one of transport of the wounded
cosmonaut. Individual equipment ie used for movinq and transport. This
equipment will be necessary in the event of unplanned situations of a
technical nature or due to loss of the efficiency of cosmonauts (unconscious
state, disorientation and so on) located in open space.
As one can see, space walking is of important aignificance when performing
rescue operations in orbit.
29
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~o~ oH~~tcint, U5E ONLY
Cargo Trar~~port
Thus, the coemonautia perform the nwet diveree taeke in npen space. 7'hese
are both maintenence of mar:ned and unmc~nned space objects and ~.n~taLlatiion-
diemantling and dssembiy nperations, rendering aseistiance i.n orbit and
servi.oing the ~aientiific apparatus. Performing tihese tasks ie related no~
oniy to the movement of mart buti algo to tiranspor~ of various types of
equipment. For example, ecientific invQatigations in the field of astronomy,
bioloqy, ~he physice of tt?e cosmos, navigatiion and communicatinns, mainte-
nance of orbitial sta~ions and asaembly of large struatures number 53 opera-
tinng with cargo transpor~ to a distanae up to 18 m a~nd 39 operation~ wirh
cargo traneporte with meiae up to 40 kg.
The experiment in trc?ppinq me~eor particles in orbit and also in studying
. the eEfec~ of space condi~ions on microorganisms, conductied in the Gemini
program, required traneport o� scientific equipment with mass of approxi-
' mately 3.~? kg. Astronomical operations in tihe Skylab proqrdm were also re-
l~ted to tiranaport operations in open space, during which the asrronauts
- transferred approximately 700 kg of carqo.
Replaceable equipment, film cassettea, traps, holders, movin cameras, failed
unite, tools, inatallation parts and so on this is far from a complete
list of the transported equipment. Al1 this equipment has diffarent weiyht
and overall dimension characteristics and, consequently, different devices
for traneport, and in most cases requires special means of securing.
Cargo traneport is frequently most closely related to the movement of man.
Depending on the conditiong and ~he specific task, movement of man may occur
upon contact with the aurface of the space ~bject or in unsupported space.
The first experitnent of movement in tuisupported space was achieved during
the space walk of the Soviet cosmonaut A. A. Leonov. Maving along the
surface of the Plac by using hand holds was accomplished during the fXight -
of the Soyuz-4 and Soyuz-5 Pla( durit~g transfer of two cosmonauts from one
- spacecraft to another.
Carqo transport can be accomplished by threc methods:
directly by the coemonaut himself during movement in contact with the sur-
face of the space objectt
by means of various types of devices and manipulatorss -
by using jet devices for meving in unsupported space.
The first method is simple, but inconvenient since the cosmonaut's hands are
occupied with moving and the traasported cargo must be attached to the pres-
sure suit.
Transport by using a manipulator was used durinq transfer. of film cassettes
from the set of astronomical instruments to the Skylab station. Three mani-
- pulators with mass of 13 kg each were located near the ShK 'ltch and provfded
30
FOR OFFICItU. U5F. ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~u~t nrr~~rc;tn~, usr. c~Nt,~
mechanicAl ann~ac~ wiCh ~ha wo~k eitn~. ~i~he distianae of' i;r.~?n~por.t w~.~
approximataly 9 m~ The e~fect of forml.ng a rod witih unre~linc~ ~ pr~-~strec~~~dd
metal ~ape ie used in the design ~nd principle o~ tran~parti. ~riao ~ape~,
being unrenled �zom ~ drum, are wound into tubes upon em~rginq �rom thr~
caeseti~e, and one ~ube i~ inside the othc~r. The length oi: the rnd is
approximately 8.2 m and the process o� extension and r~tiri~val 1~~Ls gor
approximatiely 1 minute. The ~ranspor~ed aArgo i~ attaa}~ed to ~he and of
~he rod tapea.
Cargo tiranaport using ~et dovices considArably expands the range of cargo
movement and the maneuvert~bility nf delivering i~. 'i'hese devices wi11 be
used more in assembly of large gtructiures in drbit. Assembly oE modular
obj~cts or of constituent structures can be accomplished by in~~ction of
aepara~e unit in~o orbit with subsequent ~oining of them into a unified
structure. It is known that orbi~al in3ection using carrier. rockc~ts is
accompliahed with given accuracy. 2fius, the units in~ected into orbit wil.l
be locatad in a speci�ic zone. Assembly of the final structure obviously
requires detection of the units in this area, approach to ~hem and transport
of them to the asgembly site. This ia a complax engineering-technical tagk.
There were cases of losses of different equipment during operations in open
space and with the open hatches of PKK. i'or. example, t}~e astronauty lo:~t
' their movie camera during the flight of Gemini-10. The lost appara~us may
contain valuable scientific observations (cassettes, movie and still c~mer~.a
and so on), Transport operations using jet devices are also required to
retrieve it.
There is a large number of practically unnecessary objects in orbit, in
addition to useful space systems. These are, for example, fragments of the
- collapsed stages of carrier rockets and satellites. Equipment which has _
become unnecessary after conducting of one or another o~~erations hus been
dumped into space. For example, approxima~tely 32 kg of various objects were
dumped beyond the spacecraft during the flight of the Gemini-12 PKK. Being
located in sufficiently high orbits, they have a long lifetime. The threat
of collision with them in orbit will increase as the number of these types
of objects increases. Therefore, one should expect that the problem oi
"cleaning up" space occupies a specific place in cosmonautics and this task
is also closely linked to operations in cargo transport. It is quite obvious
that force and specifically, the tractor force of the zocket engi~e should
be applied to a body of one or another mass to move it in unsuppor~ed space. -
_ In this case the orientation and stabilization of the transported cargo must
be provided from the viewpoint of safety, installation requirements and so on.
Rather high requirements on the accuracy of orientation, stabilization and
the relative speeds of mooring will also l~e placed during organization of
loading-unloading operations in the "transport supply spacecraft-orbital
station" system acro3s open space.
Thus, organization of transport operations in open space is one of the most
important aspects of the cosmonaut's activity in orbit. _
31
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
' ~~oK or~icrnt, usr otvt,Y
~
Experimential rnvestigatiione in out~r Space
Space walka by man are necessary to condunt an entire complex o~ experimentai -
invea~iga~ions. Some of th~m may be enumer~ted: -
~esting pressure suiti~ af different modificatiionst
tiesting varioue tiype~ o� toola for working ou~side tihe epne~craf~i
test3ng thn repairability of varioua system~, their degign version which
facilitateg the ~ask of maintenance and evaluating the equipm4nt of "work
plati�orm~" for conduc~ing repair and maintenance operationsr
investigations and analyses of various types of installations and devices
for moving ~he coamonaut ~jet devices, tethers, hand holds, attaahing de-
vices and so on).
For example, methods of maneuvering ;.n unaupported space and approach with
a spaceczaft usinq a tether, methods of moving along the outside surface af -
~ the spacecraft using hand holds and attaching devices and procedures fnr
various methods of movement were evaluated during the flighti of the Voskhod-2,
Soyuz-4 and 5oyuz-5 PKK. Pr.essure suits for working in open space and life
support eystems were tested. Prior to landing on the Moon, a pressure suit
of "lunar" modification was tested in open space during the f].ight of the
Apollo-9 PKK. Tools for working in open space in gtandard and sFecial ver-
sion, equipment of the "work platform" for conducting repair operations
and devices �or moving the astronaut in unsupported space were tested in
- the Gemini program. The ,~stronauts of the Skylab PO5 installed a container
on its outside surface with various specimens of onboard equipment which -
could be disassembled and returned to Earth to study the prolonged effects
~ of space flight conditions. Specimens of heat shield materials in open
� space, where they were delivered by astronauts, were tested during the flight
of the station. _
Such complex and grandiose operations have been proposed in the future in
orbit that tests cf various systems and units under conditions of real sp~ce
�light will be simply necessary.
These types of experimental investigations include testing the welding unit
under conditions of a space vacuum on board the Soyuz-6 PKK.
An example of experimental investigations in open spaco is also the experi-
ment on board the Skylab orbital station to evaluate various types of units
for moving in unsupported space.
COPYRIGHT: Izdatel'stvo "Znaniye," 1977
6521
CSO: 8144/1276
32
FOR OFFICIAI. U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
- ~nR OI~rICI:AI, i15C ONLY
G~OPHYSIC5, ASTRONOMY AND SPACE
ONE PRINCIPLE FOR MEASURTNG ANGLES OF ATTACK AND SLIP DURING SPACECRAF2
FLIGHT IN ~NEAR' SPACE AND THE ATMOSPHERE
MosCOw DATCHIKI I VSPOMOGATEL'NYYE SI5TEMY KOS;tICH~SKIKH APPARATOV. ROBOTY
I MANIPULYATORY. TR[1DY TFAK in Russian 1978 signed to press 8 Aug 78 pp 39-48
[Article by A. B. KrymovJ
(Text? The considered principle of ineasurement is based on analyais of the
dynamic pressure distribution along ~he meridional cross-sections of the
spherical element, which is part of any spacecraft surface durinq atmospheric
fliqht,and on analysis of the ion flux distribution fixed by ion traps whose
- axes of senaitivity are arranqed in specific planes during space flight.
Algorithms for processing the data of primary sensors, cons~r~sc~ed by the
unified princfple bo~h for atmospheric and for space flight, are proposed -
which permit determination of the angles of attack and slip over a wide
range of variation.
Information about angles of a~tack p( and slip ~ mi~y be useful for purposes
of manual or automatic control at all stages of flight of circumterrestrial
spacecraft, especially multiple-use spacecraft.
This information is especially necessary a~ stages of orbital injection and
return to earth (atmospheric stages of flight). Its presence permi.ts solu-
tion of the problem of spacecraft stability and controllability and makes it
possible to construct systems for control of angular motion which provide
acceptable (close to constant) quality of transient processes, despite the
very wide variation of the dynamic characteristics of spacecraft as the
control object.
In view of the fact that the space seqments of the trajectories of many types
of circumterrestrial spacecraft are close to circular, information aY~out
angles o( and ~j is equivalent in most cases to the information ordinarily
used about pitch and slip lN angles. For example, the latter occurs
during manual control of the angular position of the longitudinal axis o~f
the spacecraft over a wide range of possible bank angles Whereas en.gines
which creatF a moment around the "lateral" axis of the spacecraft to control
angle ~ at angles of close to zero and engines which create a moment
around the "vertical" axis at angles of the engines exchange roles at:
angles of Y close to 90�.
33
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
rox d~r~icrn~. us~ ornY
7'his ~xahange nonsic~orably complicatQS bo~h the cosmnnaut's work during manual
contirol ~nd ~fte stiructiure o� ~he corresponding autiomatic contirol systems. The _
indicated exchange of role of ~he enqinQS does not occur when uning angles
d and ~ ae ~he control coordiaates.
Systems for measuring angles d and ~ for gtmoepheria �light, based on re-
produc~ion by using modols o~ preasure distribution on any apherical surfacQ,
were considered in (1). Z'hese nx~dels were equ3aectiional potentiom~~ers,
while the values of angles o~ and ~ were determined by the vol,tage distiri-
bution on the potentiometers by electromechanical trackinq syetems.
Fur~her investigations, somQ results of which are outllned below, indicatied
the possibility of using the ion flux distribu~fon �lowing around ~he spa e-
craft during ciz~cumterrestrial space flights to determine angles a and ~
and they also permitted development of algorithms for calculating angles
G~ and ~ on the basis of analyzing their corresponding distributions. Z'hese
algorithms are convenient �or real3zation both by using apecialized compuL�ers
and by using BTsVM (High-speed digital computerj.
The shape of the nose cone in the form of a spherical element is feasible
in multiuse spacecraft and their prototypes to provide the best thermal
- conditions during Rtmospheric flight. In this regard, the pressure distri-
bution on the indicated elPment may be used to determine angles p( and
~ P B
P~'O
ae ~ _
n6 -
1
0,6 ~ 1,7
_
6 `~4' '
O,Z `
1G ~
O
- O,Z.f 0,75 1,I,fb;pod
Figure 1. Relative Dynamic Pressure Distribution in Spherical
Element Cross-Section Passing Through the Critical
Point
As is known (2~, the relative dynamic pressure distribu~ion at high
supersonic and hypersonic flight speeds represents an axi.symmetrical surface
- with maximum at the critical point a point where the free-stream flow is
normal to the surface of the spherical element. Relative dynamic pressure
is understood as the ratio of the dynamic pressure component at the currQnt
point of the spherical surface located at angular distance e from the criti- -
cal point to the dynamic pressure component at the critical point. ~
34
FOR OFFICIAL. USE ONLY -
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
rott a~rzc:IAL USL' ONLY
Func~ions P(e) for different M numb~rs axa ehown in Figure 1.
Let us use ~he rectiangular coordinate system OXYZ (Figure 2), whose orig~n
coincides wi~h the center o� a sphare, and whose axea OX and OY are parallel
to the longitudinal and vertical axes, respec~ively, of the apaaecrai,:, as
the reference eystem.
- Let us additionally denote by e y and ~ Z the angu7.ar distances from axis OX
of the meridional pro~ections of ~he flowing point of a spherical surface
onto planes OXY and OXZ. _
~ .
a
_ ~ n -
_ -
~ a:
s
Figure 2. Adopted Coordin~tte System
If distribution P( B) at angles a and equal to zero is taken as the base
(the critical poi..,~t coincides with the ongitudinal axis of the spacecraft),
- the surface of P(8i will be shifted in coordinates e y and ~Z upon the
appearance of p( and ~ not equal to zero (Figure 3). In this case the angu- t
lar positions e~ and ~ ym of the maximum relative dynamic presaure distri-
bution curves for the cross-sectional lines of the spherical element by
planes OXY and OXZ are related to a and ~ in the following manner:
a = ~ym+ ~1~
C09 ~um 9l~ Osm
~ = ACCSIR , `
Yi - sin= Uum 91DS dZm ~ 2~
During atmospheric flight, usually ~ and accordingly e Zm < 10� and ins tead
_ of (2) one can use expressions
~zm COS ~ym or ~ s~. e:m. ~ 3)
As is known, besides neutral gases, the atmosphere contains a significant
_ quantity of ionized qases in the range of altitudes from 50-80 km to several
thousand kilometers. The charged particle concentration at the same altitude
depends on the geographic location, time of year and time of day, solar
35
FOR OFFICIAL USE ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
K7.'
~OIt OF~ICIAL U5~ ONLY
~
, a,
- r~r
z
By _
Figure 3. 5patial Relative Dynamic Pressure Distribution in
Coordinates BY and BZ
activity, the earth's magnetic state and a number of other factors and may
vary hunc~reds and thousands of times. An example range of variation of
positive ion concentration, obtained as a resulti o� generalizing data of
more than 20 investigations of different authors carried out during dif-
ferent years and by using various apparatus (H is altitude in km and N is
the number of positive ions in cm-3) is shown in Fiqure 4.
It should be noted that very significant variation of ion concentration (10-
100 times) may occur during orbital flight within comparatively short time
intervals (minutes and tens of seconds). Due to the smallness of spacecraft
dimenslons compared to the distance covered by them within an indicated time,
let us further assume that the ion concentration varies synchronously over
the entire spacecraft surface.
' h;,rw
- ~00 ;
40D
O ,
10t 10~ >D~N,crv~
Figure 4. Dependence of Positive Ion Concentration on Altitude
Since the mean velocity af the thermal motion of molecules and also the
- velocities of the jet streams (ionospheric driftsl are significantly less
~
~ 36
FOR OFFICIAL USE QNLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
t
- ~~Ok (1N1~'iC1~1L tJ51: C1N1~Y
ehan the ~paa~craft flighti ~peed, Che i.ono~phere may be aF~umeci "fixed" and
one Can judge ~.he angular po~ition of the spacecratt toward the velocity
v~ctior by th~ dir~ctilon nf t.h~ free-gtire~m inn flow with reypec~ eo tihe
gpacecrafe~bnunc~ ~xe~ (3]~
'I'h~ d~p~nd~nce of th~ numb~r of trapped partiicl~~ on tihe angul~r, poaition of
~h~ ion trap axig 9 td tih~ ~pacecraft velocity vectdr, the e~~me ~g P(B
is an axigymm~trical surfaae wftih m~ximum at a pointi corregponding tn B= 0.
If the Current of a cylindrical ion trap (TgIL) is dendted by i(0) with tihe
arrangement of its axig aldng the flow, and if the current during inclina-
tion of ita axia by angle 8 to the flow is denoted by I(6), the zel~rive
current of the Ts2L does noti depend on ion concentrati.on and is determined
by the expreasion
/ i0) ~cos0 c c e=
I ~ - n nrccos ~t - ~t ~ t ~t: l ~ f 4 )
i
where d is the TaIL diameter and c i~ the shift of the flow passing through
the input grid with respect to the cathode (Figure 5).
The value of c in the qeneral case depends in a very complex manner on the
design and potentials of thc TsIL electrodes,* angle 0 and also on the
- kinetic energy and charge of the ions.
_ in the cuse of the simplest two-electrode TsIL fthe input yrid pf the
_ cathode)
hT oin ~D . Ui,q 1 -
4 U;; ` 1~ Y' cos~ t! r ~
(5)
where h is the length of the TsIL (the distance between the input grid and
cathodel, T= mV2/2 is the kinetic energy af the particle (it is equal to
- approximately 7.2 ~~V at flight speed of 7.2 km~sec-1), q is the particle
charge and Uk is the cathode potential.
~ As will be shown belaa, traps with a static c:~aracteristic I(8 ) in the form
of two linear segments are preferrable for purposes of ineasuring angles p(
and Therefore, the use of a two-electrode trap is not feasible since
�unction I(B) is very curvilinear in this case.
*The potential of the input grid, usually equal to that of the spacecraft
surface, does not exceed 4-8 V and is essentially unreflected in the tra-
jectories of 0~ fons, which comprise the main part of the positiva plasma
ions. This potential will subsequently not be taken into account.
37
FOR UI~FLC111L USE ONLY
,
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
xd~ o~~icraL u5~ ort~Y
C ~ : J
,
` ~ , o
~ e
c `
. ~~i
f ~
. Figure 5. Dieqram of Cylindrical Yon Trep: 1-- Cathoflej 2--
controi gridt 3-- input (aperture) grid= 4-- refiec-
' ting qridi 5-- equipotential qapt 6-- overlap area
An acceptable form of the characteristiic i(ei can be found for multielectrode
TsiL wtth corresponding ~election of tiheir deaign parameterB and grid potan-
_ tials. The value of c is determined in thia case by the expression
+~.i
~ ~ r ze 9,Z .
q ~ i'~" 7'
c~
os~
t~
k~-i
"~i (6)
~ ,~x-~ ~~k 9 i~~ u~ ti,
b C~ uk ) i -f' p -
xs
- t
q Ui
h' tg i~ ~ ~tco~t e~
c
where n is the number of potential trap electrodes (the input qrid ia aesumed
to be zero), hi ia the distance between the grids of the i-th potential qap,
h~ is the distar.ce between qrids of the i-th equipotential qap (following the
i-th potential), Ui is the potential increment in the i-th gap and 1 is the
number of equipotential gaps (the equfpotential grids do not follow each of
the potential gaps).
Th~ formula expresses the total ion deflection in the potential and equipoten-
� tial qaps. The firat potential qap and so on is located behind the input
(zero) qrid.
Function I(8 ) for a three-electrode trap (qrid-grid-cathode) fs presented
as an example in Figure 6.
An increase in the number of electrodes permits control of the static charac-
teristics of the TsIL over a wide ranqe.
38
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
I
~'n~ oF~t~iCtAL lJ5L nNLY
,7~~)
/
~B
20 dD d ~pdd
~iqure 6~ Dependence o~ Relatiive msIL Current on Angular Po~ition
of its Axis to Velocity Vactor
Two rows of drain openinga on the gpherical element may be used to meaeure
anql~s oC and ~ during atmoapheric fitghti (~'igure 7). 7'he op~ning~ arranged
on thp cross-sectiionai line of the element by plane OXY gerva to de~ermine -
the angle of attack, while thoge arranged along tih~ cross-sectional line by
the plane normal ~o plane OXY anc~ passing tihrough the center of the sphere
ati angl~ ~A to axi.e OX are uged to me~sure the slip angl~. Angle ~P is
melected from the conditiion
~ _ ~mac ~"~mm ~7~
~ ~ ~
where o(~ and ~Y ~,n are the maximum and minimun~ anticipated angleg of
, attack.
In the general case thare should be several planes of nrranqement of the
drain openings with large variations of an les oC and /3 . However, taking
into account the fact that usually anqle 10� and - A'~ S 20-25�
durinq atmospheric flight, the two planes indicated above a,re sufficient.
In this casa a sliqht mutual effect of angles o( and ~ on +:he accuracy of
their measurement and the corresponding pressure distribu*.ions in the
cross-sections are provided.
_
~
s
a'
Figure 7. Dfaqram of Arrangement of Drain Openings
39
FOR OFFICIAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
rOtt d~~ICIAt, U5C dNLY
A blo~k 8i~gram of an analog-1ogi~ compue~r and ~ngl~s aC(/g) is ~hown in
~'igur~ 8.
p numb~r ~f valu~~ corra~~onding Ca tih~ diacr~e~ valu~e di th~ diff~r~nr~
~~ntiic charaeterigtic P( 9)- P(~ *,~1) ,,d i.~ tih~ angulnr dis~ar~ce betw~en
tihe openings conn~c~efl to one Bangor, sel~cted in the r~ng~ df 60-75� ~ccor-
ding to (1), is tiak~n from th~ di~�erential presoure ~nngnr~ conn~c~ed by ~
channel~ to th~ tlr~in op~ninys. -
Char~cteristic P( A)� P(~- ig ghifted with varia~ion of ar?gle o~ 1n
coordinates 0~, and therefor~ ite zero corresponds to th~ anqle of attiack.
it is obvious thati the va1uQ of the anqle of attack can be found as ~he sum
o� values of the int~rvals of ~en~nr axrangement ~o the interval where th~
charact~ri~tiia pasees through zero end of Che addition i~tqure 9) _
~ ` U=!?
(n)
Uz~.i ~s~ '
wher~ ~ is th~ angular interval between the drain openings, u Ei+l is the
value of the sensor signnl correspondinq to the riqht boundary and U~i is
that corregponding to the le~t boundary of the segment on which the charac-
teristic paeses through zero.
In tihi$ case
a~ mk d, f 9)
where m~s ~he number of entire intervals to the seqment where the charac-
teristic passes through zero.
The logic part of the diaqram, conaisting of relay and logic elementg, deter-
mines the interval on which the characteristic passea threugh zera and con-
nects by msan~ of keys the sensor aiqnals correspondfng to thQ working
section to the computinq circuit 8 and a siqnal is fed simultaneously from
the v~ltage dfvider tn the output adder corresponding to Che whole number
of intervals t~ the interval where the characteristic passes through zaro.
The calc~latinq algorithm o((~) is realized in the followinq manner with
the presence of a BTsVM on board (Figure 10.
Siqnals from the sensors are convezted to diqital code in the ana~oq-code
converter (PAK) and are fed into the BTsVM, where the anqle o� attack (slip
enqle) is detesmined by the algorithm considered above.
During atmospheric flight, the dyramic pressure and during space flight Lhe
ion conrentration vary over a wide range and synrhronous automatic regulation
of amplification is required to maintain sceptable accuracy c},aracteristics
of the sensors and of the entirQ ~ystem. Thfs regulation is accomplishQd by
40
FOR O~FICIrV. US~ ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~ ~t~it brH' IC LAL U5~ ONLY
'
. .
~~t
I
.
~
_ F
f ~ f (
I r. ~ ,4. i: .~D.;r
~
IA'A~ ,Cr~ j(:7~ F�---- A:y~ A'�!~lI .C~~
~ :J i ~1
:
i.~~
~
~ t �I
l~J~
LS~C..l y~ .+f,
Tt.~
Figure 8. Block Diagram of Analog-Logic Computer of Angle o~t~ )
~
oco~ d
et~ ~ a~~ B
- Fiqure 9. Determination of Addition c~
feedback alonq the angle of inclination of the working section of the dif-
ference characteristic
K ~ Us~- ~s~.~
~ (10) -
The effect of the alqorithm on the boundaries of the workinq range is also
expanded where ariqles d(/j) are determined by the signals of the extreme
left and extren~e riqht sensora by interpolation.
41
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~
I
~Oit n~F'ICIAI, USt~ ONLY ~
in tihe ca~~ when abeolutie pre~aure geneore are uged, ~h~ circui~ mus~ be i
suppl~mented by aclding elementa which make !,ti pos~3ble tio obtain aignnls
propor~idnal ~o ~he differenCe pre~~ure ar?d Che anql~g o� at~ack (slip
anqles) ~re eub~~qupntly det~rmSn~d by the algorithm presanCefl ~bov~. ~
Due tio th~ velidl.tiy of Nawtion'g theory of flow daring ~p~ce flight (iE the
TeIL ig no~ ~hedad), the relatiive ion difltrl.butiion t(B) of the TetL ie
, d~termined, a11 tihings being equai, only by their ~ngular positiion to tihe
velocity vea~or. Therefore, during analysis iti i~ convenienti to asgume
thati the TgiL are located on eome sphere with axes of sansitivity directied !
tioward its center. '
,Qam vu,ru ~ ~
~ .
A'v.w,vyr.
fam~?p ( 2 ) ,
/!A-A' Ftt,~~~~t~e~'
?~3~
r..~~
~ r.._.~..~___
,/~al d~~ct(jf??/!r/U CalJf~iI00 ( ~ ` ~
I pl~l~~[y~ lil~,V t OOn/aii,rd/ ~ ~
I ~
I ,n6m P'~JJ e 0 a,~ ~ ~
( I
P(~�!f ~d. I
( ~j ~0 ~rt�s I
I ~ g~' I
( J r Xi?oswyPyv~'~�~/,~ I
( ~
dJ ~tn~ 1
~ ~
t-~
I da ~o.~' ~.rvl~ ~rm ~a (
~ i
.
I ' ~
~ ~ A~~d. - :
d. ~ ~
I r~l w~ ot~/.~ w %~'J~vqO ~a~ >~'re.~~ ~
._.~..y~_ _......__.~___.._._~J
a,,~
Figure 10. Algorithms for Determfning
.
KEY:
1. Sensors 6. No
2. Conanutator 7. Yes
3. Requlator 8. Left boundary
4. Sfgnal recording 9. Right boundary
5. From sensors
42
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
CnCt Ut'i~'ICIAL USC dNLY
Witih an incre~ge of tihe me~gured ranga of angle~ wh~n the mu~ual atfecC of
angles a and ~ on mo~eurament accur~cy i~ gigni�ican~, tha m~aeuring meri-
dian mu~ti be rota~ed immadiatiely after varietiion~ of angles D( (or to
eliminate the indica~ad effeot, which requiree both an incrn~?se in the number
� o� m~rL nnd in developing prob].emg related to their arrangement and connec-
tion. _
One can name a number nf inethods of arranging the Tsi~ on a spherioal eurface _
a~ anglea of di�ferenti grids: latitiudSnal-latii~udin~l, meridionai-~atitudinal,
meridional~meridiot~al, at the vertices of spherical tz~iangles ar?d so on, but
they ~ither have regions with nonuniform di~tributiion of sensors or compli-
cat~ ~he aignal processing circuit.
y
,
z
~
~
~ -
r
~ Figure 11. Diagram for Arranging TsiL
- We feel that the best variant of arranging the TaiL is at the nodes of inter- -
section of the meridional grids with the mutually perpendicular polar axea
(Figure 11). However, the entire sphere is divided into six sections corre-
spondinq to the sfde of a spherical cube and the center of each aide is
used as the pole for the meridians of conjugate to eliminate the
effect of the meridian convergence factor when the sensors are located near
the poles (Figure 12).
The problem of the computer in this case is to find the corresponding mea-
- suring quadrant, to select the measuring meridians in the found quadrant
and then to calculate angle OC(~) by the signals of the measuring meridian
sensors. The search for the qu drant (Figure 13) i:z which the maximum sur-
face I( B) is located can be carried out, for example, by reselectinq the
sum of siqnals of sensors located at the vertices of spherical quadrangles.
Selection of the measurinq meridian within the found quadrant can be organized
in two ways. The first method includes the fact that the measurinq meridians
for the slip anqle are connected by the signals from the angle of attack com-
puter. The second method includes the follawinq. Since the signal level of
the measurinq meridian sensors passing near the critical point is qreatest,
then, havinq omi.tted the sensor siqnals thr~ugh the diode matrix, we will
have siqnals on its outputs corresponding to the desired meridian.
43
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
I
b'Ult (1~~rIC1:AL USL ~NI.Y
~
i
Figure 12. Taice-Maridional Grid on Spherical Cube
, ~
Anglea oC and ~b �or the aignals of tihe measuring meridian sensore can be
found by the,~~lgorithm presented above �or tihe case of using absolute pres-
eure aensors.
~;1 ,
~ 1 ~ ~
~
1 / ~
~.~1/ ~ /
. -~~�1�~ i~
i~
_ , ` /
1 I ~ /
~~aL~~~ ~ i
~1 '
i'~ ~
! '^�,d.. ~ / \
` ~ _
, / \
Fiqure 13. Determination of Optimum Number of TsYL _
Since a minimum of 4-5 sensors is used in calculating the precise value of
the anqle of attack by the method indicated above, one or two of its workinq
sections are located in an adjacent quadrangle (quadrant) upon shifting of
the difference curve alonq the measuring mer~dian toward the boundary of the
quadrangle, althouqh the zero of the difference characteristic has still not -
passed the boundary of the measuring quadrant. Therefore, it is necessary
that the measuring meridian continue beyond the limits of the quadrant. It
is also necessary that the senaors on the measuring meridians be located at
approximately equal angular distances. The latter leads to installation of
, additional sensors located other than at the nodes of twice-meridional qrids
(they are noted by crossQS in Figure 13).
The desire to reduce the number o� TsIL required for measurement leads to the
necessfty of expanding the working range of their static characteristics,
- which in turn causes a reduction in the accuracy of determining angles 0( and ~
1' ' -
44
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~ I~0lt 01~l~'ICIAL US~ ONLY
As an example let us consid~r di�ferenti variAn~s oE dividing the sides a�
a spherical cube for tihe case of circular m~ttsurament of angles o~ e~nd _
rf ~he gides of the spherical cube are di.vid~d by measuring meridians into
thr~e, �our and �ive partis (see Figure 13), respec~ively, 104 TsIL (56 muin
and 48 auxiliary) are zequired in ~he firs~ case, 196 TsIL (98 main and 48
auxillauy) are required in t.he aecond case and 344 TsTL (152 main t?nd 192
auxiliary) are requiied in ~he third case.
Since the s~atic characteristics of I(B ) near 0 are significan~ly non-
linear, the moet linear parta shi�ted (by one-two intervals? from the origin
~ must be taken as the workin se
g gmenta. Moreover, no fewez ~han three sensors
must be located simultaneously on the linear segment. Thi.~ is necesaary so
that the two other sensors connected to the computer correspond to tihe linear
segment upon descent of one of the sensors onto the nonlinear segmtn~. Thus,
the wnrking range of characteristic L~. consists of a linear segment measuring
L= 2~ on the left branch of characteristic I(B), the same segment on the
right branch and a nonlinear segment having length I,n =(1-2)~
Lr = (5-6).l (11)
For the considered variants of dividing the sides of a spherical cube when
the angular distances betwoen nodes (at which the TsIL is located) comprised
30, 22.5 and 18�, respectively, a TsIL is required with value of the worki.ng
ranqe oE 90, 120, 67.5-90 and 54-72�, respectively.
If the measurement errors of the TsIL are taken as 1 percent, then a~ indi-
cated by analysis, the mean errors of ineasuring angles oc( caused by TsIL
errors comprise 0.7, 0.55 and 0.45�, respectively, at the indicated values
of the working ranges.
Thus, a sharp increase in the number of sensors in the third case yields a
very insignificant advantage in accuracy and the second variant of diviafon
is apparently more preferrable. -
- The outlined principle of ineasuring angles p( and ~ permits one to obtain
information about the indicated angles at practically any stages of flight,
both atmospheric and space, of circumterrestrial spacecraft. Moreovar, only
part of the general algorithm which permits circular oricntation during
space flight is used during atmospheric flight.
BIBLIOGRAPHY
1. Petrov, B. N. and A. B. Krymov, "Measuring the Angles of Attack and Slip
Angles by Using Models o� Pressure Distribution on a Frontal Spherical
Surface," IZV. WZOV, PRIBOROSTROYENIYE, Vol. 1G, No. 12, 1973.
45
FOR OFFICIAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
,I
~OR OFFICIAL US~ ONLY ~
2. Be~otserkovskiy, S. M. et al., "Supersonic Gas Flow Around Blunt Bodies
(Theoretiical and Experimental 2nveatigation)," TRUDY VTS AN SSSR~ 2zd.
VTs AN SSSR, 1963.
3. Swain, D. B. and W. H. Bennet,,"Atti~ude Sensor for Vehicle Orientatiion
in Space ~'light," AzAA JOURNAL, Vol. 3, No. 3, F'ebruary 1965. ~
COPYRZGHT: Izdatel'stvo "Nauka", 1978
6521 .
CSO: 1870
- i
_ i
i_
, ~
~
i
i
I
i
i
1-
~
(
_ '
I
46
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
I~OR OF~ICTAL U5C ONLY
GEOPHYSICS~ ASTRONOMY AND SPACE
A NEW N~THOD OF SYNTHESIZING ARTIFICIAL MOTION AND ITS APPLICATION TO
LOCOMOTING ROBOTS AND MANIPULATORS
MOSCOw DATCHIKI I VSPOMOGATEL'NYYE SISTEMY KOSMICHESKIKH APPARATOV. ROBOTY
I MANIPULYATORY. TRUDY IFAK in Russian 1978 signed tn press 8 Aug 78
pp 107-114
(Article by M. Vukobratovic, D. Hristic, D. Stokic and N. Glahazic, Yugoslavia]
- [TextJ. The main difficulties which arisa in solving problems of controlling
complex dynamic systems are related to their high dimensionality. The tradi-
tional trend here is to linearize equations of dynamics with subsequent de-
composition of them. In this case any differentiation of equations of dyna-
mics usually leads to introduction of additional conditions and relationships,
while variation of the operatinq mode of the system is accompanied by a de-
crease of solving accuracy.
When investiqating most control systems, main attention is devoted to study
of quite specific operating states or modes of both the system as a whole
and of its individual parts. Based on the qiven states of individual sub-
systems of the investigated system as a whole, one can approach solution of
the problem of reducing dimensionality in a new manner. In this case only
part of the dynamics of the system not encompassed by the characteristics of
the given state remains open ta regulation. The required compensating ef-
fects on the qiven (known) characteristics are then organized by using the
requlated part of the system.
Introducing the concept of nominal dynamic operating mode of the system
forcea one to consider the problem of the optimum solution (control) obtained
for a complex multiconnected system. Thus, for example, redundancy of seve:al _
mechanical systems is provided due to additfonal conditions of optimality,
realization of which, however, is not always possible. In most cases the
solution is found here in the range of "pure" control problems. A vivid
example of realizinq inconvenient conditions of optimization is various
- walking systems. A convenient mathematical mod~l for these systems is dff-
ferential equations of a complex pendulum aystem. In this case, specifically,
the motions of the supports are assumed strictly monotonic.
47
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOIt OFFICIAL USE ONLY
Optimizatiion oE tha mction of a walking sys~em from the condition of some
energy criterion (for example, a minimum moment o� mo~ion) presente a spec3.a1
problem o� synthesizing control of motion of the indicatafl aystem. The
attemp~ to realize a system o� some kinematic program which seemtngly realizes
the special criterion of some global ari.~erion of optimali~y as ~he law of
control. of motion is also quite appropr3ate. Hence, two stages can be deter-
mined in the procedure of synthesizing the con~rol of motion of a walk3ng
sys~em: selecting the aon~rol for the nominal dynam3cs o� the system and
correcting this control with regard to the ef�ect of large disturbances.
Both indicated levels can be determined in problems of controlling very
diverse biological systems.
Among the variety of various types of motions made by living organi~ms, the
walking method of looomotion is one of the most convenient for reproduction
in artificial systems. The possible mathematical model of this motiion is
description of it by means of some conditions of stability of the Qynamic
system in the �ield of forces acting on it. Moreover, these conditions vary
accordingly during motion. Analysis of the relationships between the force
field and the conditions of stable motion determines their ambiguous nature.
This is related to the fact that solution of the second-order differential
equation which determines this relationship flepends on two initial condi-
tions (integration constants). Therefore, information about variation of
integration constants must be used in the control law to orqanize atable
motion of the walking system. This relationship has been named the second
signal system.
Any variation of the initial conditions causes the required variations of
the forces of the servo organs with subsequent correction of the law of
~ control. The first level of the law of control (the level of nominal dyna-
mics) provides stable repetition of phases of the selected type of motion
in this case. The main difficulties of design are related here mainly to
_ developing a high-speed small control digital computer. And since a real
walking system (for example, a robot) is a nonlinear dynamic system of high -
- dimensionality, the level of nominal dynamics in the law of control is most
effectively represented by a set of alqorithms which determine the different
types of motion.
The second level of the law of control (adaptive) is called upon to work
out the neceasary correcting siqnals coming into the servo mechanisms upon =
variation of external conditions. Variation of the initial conditions for
n~minal modes leads to the appearance of additional relationships in the law
of control which stabilize the "nominal" motions. Upon variation of the
external conditions, similar to livinq orqanistns, any variation of accelera-
' tions, position or speed of motion of a walkinq system causes corresponding
compensation signals in the control system. The latter in turn create the
required controlling moments. ~
The general diagram of a syatem for controlling the motion of a robot is
presented in Figure 1. Let us write the equation of motion of this system
, in the following vector form:
48
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
_ A'Ott C)~'l~'IC1:AL llSl: ONI.Y
f~~~ t) B(~, t) U, ~ 1~
where is the n-dimena3onal vector of state, f(~, t) is the vec~or-~unction
o� dimension n x 1, B(~ , t) is ~he mntrix o� constant numbers oE dimen~ion
n x m and U ia the m-dimenaional vector o� the input ef�ecta. In this case
- le t us consider the general case of the control problem when the input and
outpu~ sigr~ale o� ~he system ara not completely determined. Le~ us denote
the known components o� vectors U and ~ by Up(m~ x 1) and ~`p(nl x 1), re-
spectiivaly, and 1et us deno~e the unknown components of these same vectors
by Ux(m2 x 1) and ~ x(m2 x 1), where ml + m2 = m and nl + n2 = n.
f~'~
f Z J
a~~r ~ -
s f ~
~ ~
Figure 1. Stabilization System: 1-- proqramming aevicet 2--
servo devicesj 3-- robot; 4-- sensors; 5-- criterion
of quality; 6-- calculation of correcting momentst
7-- new law of control
Let us then introduce the c~~nstant matrices P and R, related by the relationa
- [ poPx] J -US-1 _ U, ~ Rz'~ ~ _ ~ Ez 1 + ( 2 )
1 1
_ -
[ ~x, [B(~, t)) (1'o;!'xl = ~0_�:Bo: ~ (3)
Bxo ~ Bzx
where matrices BOO,xx,Ox,xO have the follawing dimensions:
Boo - ni X ml; I~ox - ni X~-"�z, Bxo - nz X mi~ B:x - na X ms.
Let us divide system (1) into two subsequent subsystems:
- {~o} _ {fo -t- ~Boo1 tUo} ~Qo:1 {U:}, (4)
{~s} _ {~x~~, (Bxo) {~o} ~~xx~ {Ux}, ~5~
49
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOR OFFICIAL USE ONLY
where ~ ,
~ ~ ~z.~f�, =
(6)
and fp(~, t) is a vector-function of dimension nl x 1 and fx(~ , t) is a
vector-�unction of dimension n2 x 1.
We note that solution (4) determines the significance of the unknown vector
Ux �or us
~Ux! _ ~BOx1T ~BOxJ~I ~~~x~T ~1~0~ {/O 1~r t~ ~B0o1 1U01~
provided that matrix [Spx]TIBpXI is not identical. Substitution of (7) into
(5) then leads us to the equation with respect to the unknown
' ~
ibx~ _~lx~~+ t~~ T~Bx01 lv0/ "t" lBaxl ~IBOxlT ~BOxI_1 X ~8~
x [Box1T ({~o} - {fo - fBool {Vo})~
and subsequent solution of algebraic subsystem (7) permits us to calculate .
the value of vector Ux. Thus, the system of equatioas (7)-(8) describes the
procedure for synthesizing the nominal dynamics. The nominal trajector can
be synthesized by using the procedure of. synthesizing a linear optimla requ-
lator [1]. It is obvious in this case that the synthesized linear regulator `
provides the required quality of stabilization processes in some vicinity of
space with respect to this trajectory, which we denote by ~ In similar
fashion we determine the working zone for the nonlinear regulator (E~), out-
side which the nonlinear characteristics already have a significant effect
on the regulated process. Let us now turn to some reqion of the space of
states ~
E" = n E�, -
for which
II ~ - (t) ii > o, c ~ (to) ~ E�~ -
and some txajectory of reverse transfer of eystem ~�A from a disturbed to a
nominal state.
As already noted above, we shall consider a class of systems with limi.ted
number of operating modes (trajectories). To supplement this, one can indi-
cate systems which accomplish direct compensation of large disturbances in
some region of the space of states i, ~ A(t) -~~(t) which permits
formulation of programmed trajectories from the conditions of stability for
- 50
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
[~Uk OT~'I~'ICiAi, U51: ~NI,Y
- tiham. Cnn~~quentily, Sf any di~tiurb~nce ~o~ing on tih~ ~yyt~m leadg i~ to c~n~
o~ r~~ion~ ~ti, ~h~ problem ~f the r~gul~tdr i~ raAuc~d in thig ~~ge rd nolec-
~ing th~ programmed tira~ectory which corresponds mosti to tihi~ ragion nf dig-
tiurbed ~tiatie. rn this ca~e ~he volume of required calcula~~.on~ i~ v~ry
modera~a (2, 3).
'i'he cri~~rion for s~l~cting th~ programmed ~.ra~~ct~ry, which enrrpgponds tio
gtate ~~(tp), is the minimum expreesi.on by numbQr ~
n
= i ~ " ~~n i ~ c~ ~~1~~ / = U, , . ~9~
i~~
if the sQt of programmed trajectories aan bQ represented in the function of
parametric vector p(pl, pr)T, wher~ r C n, th~ g~laction procass can
be Considerably factlitated by identification of the paramnt~rs of each tira-
jectiory. In otiher words, the pxobl~m reduces in this caso tio calculating
the vector during minimiza~ion o� our criterioa:
eria cio~
~~=o, ~,...,r,
The calculated vector pA also determines the new trajectory of corre-
sponding to the state of the aystem ~(tp). Since vector ~ contains n com-
ponents and since it is a function of different parameters, also includinq
the time here, diffic:ulties arise in Yememberinq all its poasible values.
Sumanarizinq the foregoinq, let us note the main features of the outlined
synthesis procedure. The state of the system should be described in the
function of its number of parameters and time. The state of the syatem ta
checked by two circuits:
1) the circuit for controlling the nominal dynamics by using an optimum
linear regulator for decomposition of the system into individual circuits
regulated by means of simple feedbacks;
~ 2) a circuit of large disturbances where some pr~qrammed trajectory most
closely corresponding to the new state of tr,e system is selected.
A number of examples of investi,gatinq co*,~iex syFtems is considered below.
. Controlling the motion of a two-support robot. Sy its nature, the problem
of synthesizing the law of motion of an artificial system is related to the
range of problems of combined control since 'rirst, one must determine the
controllinq moments for each supF.ort. Tti-~is fol].ows from the fact that the
dynamic reaponses of the supports may be r~garded as the external effects
, on the mechanism. The resultinq forces oi responses of the support and of
the locomotion surface are calculated at fixed points of the trajectory of
motion. These points are characterized by a zero moment of resulting forces
S1
FOR OFFLCIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
I~'n!t U~~'tCIAL U5~ hNLY
and also by alter~d form o� the ~quatiiong of dynamics. mhuo, we can salacti
tihe tra~eatorieg o� rtx~~ion of poi.ntis with ~cero r~gult3ng moment, al~o char~c-
tierized by zero moments witih r~ep~ct ~o tihe poin~e of drticuletiinn~ of tih~
� movabl~ elemente of the robot atruatur9 (Fl.gura 2). ~n thie CA88 the con-
trolling momanta Mp are also zero moments, while the diffarontiai equ~t~ona
of dynamice aeaume the form
t~:} = {/x (~~1 ~(~o:1T (~oxl�' I~u~l'' ({~n~ ~lo~E, t)})1. f11)
7'he ~p~ce of unknown componentis o� the equatiion of dynamicg x} deecribes _
th~ compeneatiing romponents required in tihis case t~ maintain a stabl~ stat~
and eo accomplieh equr~l motion or free movoment of individual aeetions. Con-
sidering eteady motion, we find that the parameters of atatie ah the initial
and final pointis o� the step are unknown. Theae parameters can be calcul~ted
by usinq the followino conditions: ~he value of the angular coordinaties and
their derivatives are equal Ati the initiel and final points o� the step, i.e.,
~ ~t) = ~ (4) ~ (12 )
~ ~ 7~) = ~ (0), ~ I 3 ~
where T is the period of the atep. Thus, equations (11)-(13) permit one to
calculate the compensating moments for the conaidered system.
z
~
2 ~
, Y
x
Figure 2. Points of Joining the Structural Elements With Zero
Moment of Rotation: 1-- structural elements; 2--
joints
The second important problem of stabilizinq the motion of a two-support a,?t.~ro-
pomarphic system is solved by using the adaptive level of the hierarchical
52
FOR OFFICIAL U5E ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~dlt ~rt~'IGIAL USL nNiaY
stiruc~ur~ of the 1~w con~roi. mh~ purpog~ of the ~~ebiliza~tom m~d~ i~
tio retiurn tihe di~tiurbing parame~~rg of ~he ey~ti~m tio ~ome n~mi.nal valu~~~
7'11~ ~peci�~.c charaatiarigtic of ~h~ problem of gtiabilizing the motion of a
two-supporti roboti ig alao the presence o� additiional d~gr~as of freedom in
- ~h~ circuiti o~ ~ha control ayst~m, which are two angleg c~f d~viation o~ tih~
~upporte i,n the longi~udinal and tranevozse plan~g, respectiivnly, wi~h r~-
specti tio ~he plane of motion. Additiional global ~e~dback, which ~nrompes~~~
tha ~xiatiing ~,ocal, �e~dbaak~, i~ org~r?ized ~o contir~l these coordinatee in
the gys~ean.
Since the mo~t interesting !.n the ~tabil3zatinn probl~m is th~ ea~e of 1arg~
distiurbin~ effects, let us dwell in somewhat mor~ 8~tiai1 on it~ When idgn-
tifyir.~ the current atatie of work, the qreate~t difficultieg ar~ related to
determining the parametric ve~:tior pA, by means of which the nominal mode may -
be apeeif~.ceZly deacribed. F'or example, the uniform motion of e robot may
be characl:erized ~ufftciently fuliy by varia~ion of two of itg par~cneters:
a-- the :~elative length of the step and T-- ths period of the walking motione.
Let us conaider the exampie of one gimple type of motiion with nominal tra~an-
torY Let ug characterize the seti of possible states by tihe equation
~i = S~,t f 14 )
where d is the index o� the lower (support) part of the s~ruc~ure. ey vgrying
parameters s and T, we find a set of curvea which characterize the etate of
the system as a function of theae parameters. The followinq coupling equation
~ between parameters s and T is valid for the remaining part of the stirurture:
~=~�-I-f3,(s--so~-f-~r~~--7'0)~ (].5)
where coefficients B8 and BT are calculated by ~he criterion of the minimum
quadratic error. Substitution of (15) into (9) with subsequent fulfillment
of conditions (10) leads ~o the following expressions: -
a~~s -t' a~a1' = bl? �s~s -f- vzsT = bs, (16)
where a11,12,21,22 are functions of coefficients Bs, BT, bl, bZ, i.e., func-
tions of the parameters of state. Equations (16) permit us to dete=mine
point A close to the true position of the depicted point in the apace of
states of the system. The analytical procedure indfcated above wae checked
by the digital. modelfnq method with selection of the nominal state.
Let the motion of a robot be along trajectory No. 1(Fiqure 3) and be sub-
jected at interval t/~ = 0.7 to the effect of a disturbing impulse calculated
by means of a gyroscopic system. The new values of parameters s and T are
then calculated by expression (16) with reqard to the coordinates of the
nominal sta~te of the system. These parameters also determine the disturbed
state of th~ system.
53 ~
FOR O~FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR O~~ICIAL US~ ONLY
f
, I~o~
,
a~ a~ ~ ~
i a~
l~d ~f
,
~ - - - - ~.d -
~~.r.~r.....a~"L `
/~L~
..,~r ,~,~~a~J
I ~
a~ I'/~ Lj~ e~ s~ 1. .
~ ,
.~1 p
. ~ . ~
~ -t~
'i= ~
D - `
_ ~s ioyr as ,t~yi'
- a b .
Figure 3. ideal Trajectorieg of Motion of Elemente of the Upper
Parti of ~he 3~ruature (Dashed Lines)t Transitiion to
New Tra~ectories With Diatiurbances (Solid Line)s a--
angles of deviation in transverss planet b-- anqlee
of deviatiion in longitudinal plane
The diqital computer-calculated values o� parameters e~ 0.4 and T~ 0.5 for
the considered example, which determined the atate of the syetem close 'r.�~
mode No. 4, ase preaented in F'igure 3. This txaneition of the systom from
one atate to another is noted by the solid line in this figure. Z'he coor-
dinate disturbancee used in our example are more illustrative in na~ure
aince they leave the ayatem in eome reqion of its permiesible st~ties. it is
obvious tihnt in practice we may also encounter aare severe situations when
tha aystem goes beyond the region of permieaible states.
Control of anthropomorphic manipulators. 2n this case the nominal mode is
that of reachinq the goal and maintaininq it, whereas the compensatinq motions
are accomplished to solve special problems of manipulation. Hence, the pro-
blem of controlling anthropomorphic manipulators can be ~olved in two corre-
aponding coordinate spaces, in the first of which the workinq zone is achieved
and in the second of which problems of manipulations themselves ar~ terminated.
According to the foregoing, the following hierarchi~al levels can be deter- -
mined in the control circuit of an anthropomorphic manipulator: the level of
nominal dynamics with motions of the system of minimum complexity and the
lsvel of disturbed modes with motions of the cort~ensating part of ~he aystem.
A minimum structure with the required kinematic and compenssting circuits
can be determined in the kinematic circuit of the structure. The latter
provides the degrees of freedom missinq in the minimum structure in the
system.
It is kt~own that a structure with three deqrees of freedom permits one to
achieve an arbitrary point of space whose position in the Cartesian system
can be characterized by three anqles. The trajectory of motion of a manipu-
lator in apace aiay have an arbitrary form, essentially unrelated to its
54
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~dtt O~FIC~AL US~ ONLY
kinemaCic circuit. Zn the general case the po~itiion of the work~ng pol.nti in
space can be detiermin~d by u~ing the fo118wing implicit �orm:
x /~~~i, ~1~a~ rra~~ U = ~si~~~~ ~i~a~ ~sl, = 1y ~J~~, d~s~ (17)
and ti.he equation of motion of tha extiremity for ~m~li incramen~s ie writiten,
respectiivaly, in tihe gorm
~
am ~ ~ ~ m~~' ~ d~D, ~ ~m m. ~ \
aw,, m~~,~ ~ ~q (~x, `~y, ~zjr, ~18 )
lAl ~0~~~ O~pa, ~~u) ~~0~+ ~J~ ~z)r~ a~! = a
p~ ~ 1= 2, 3. (19)
A rather wide alass af problema of manipulation can be adequately degcribed
by using these equations for some baseline a~ructure of a manipulator and
the neceeeary contmlling momentis are calculaCed (4-7~. Since ~he baseline
structure of the manipulator usually daea noti provide total solutiion of the ~
posed problem, additional deviaes are used in the structure which permit one
to achieve, for exart~le, operation with a grasped objact~ The dynamics of
these specific problems are ta~cen into account in the qeneral system of
differential equations by means of conditions of form
f ~~~o, A!o) ~ p, ( 20 )
where F~p is the force of inertia of the object and I~p ia the correspondinq
moment of inertia. It is obvious that the dynamic conditions of this type
may be reduced to purely kinematic limitations. Without regard to alipping -
. between the object and the grasping device, the expressions for determining t~
the inertial forces and moments include a aufficiently large nwN~er of para- '
meters
. Fo = ~i w+ ~U'~ ~U
Ate - Q),(q~, U~, tD, ~U�, Qi�, d~*, ,'~IN~,, (21)
ti
where ~ are the qeneralized coordinates of a structure of minimum complexity,
are the qeneralized coordinates of peripheral devices and ~E end F1RE are
the resulting external forces and moments. Continuous solution of equations
(20) and (21) for measured values of ~E and ~1RE permits one to calculate the
required compensating forces and mcments, i.e., to synthesize the adaptive
level of control. In th{s case the indicated synthesis procedure is consi-
' derably simplified if the parameters of the right side of (21) are measurable
and controllable. This qoal is achieved,for example, by installation of the
necessary sensora on the qrasping device. The conditions of mutual compensation
55
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
- ~OR d~f~'ICYAL it5~ nNLY
(~qul.librium) of fora~~ and momen~g $atiing on ~he ob~~c~ ar~ tihd foilowl~ng
~lgabrsic ~qua~ion~:
H
n ~o ~ ~~t ) ~ ~k ~ 0, ~~o =1~ ~ ~o~ X ( ~ l~r ) :lf - p,
~22)
whex~ R1 i~ the �orc~ record~fl on th~ 1-th s~nsor, ~pl ie the grnvitiy compo-
nant directed toward thig senaor, ~'p i~ the gravity of tihe ob~~at and 1 is
tih~ n~ ~ber of eensorg. i~ ie than aimple to calCUlare tih~ valuee oF ~p and
I~p ana also the correeponding incrementis p~p and ~t~j for tihe measar~d
vnlue~ of ~1. They can be compensat~ad by additiional connections in the con-
tirol circuiti.
Leti ug consider the dynamice of the manipulator during "drinking" as an
example o� one manipulatiion problem. Providing motion of the glass parallel
tio the vertiical axig, we tihus achieve a zero moment of inertial forces. A
number of gimple con~traints in this case ensures nonapillinq of the liquid.
F'or example, conditions (20) are written here in the form
-O~IHoI~Ih~I+ }rrol,:(--~1~~)-p, (23)
~
where ~~I is some maximum value of force. The second constraint concerns
that of the rate of moving the glass. The third condition pr~ovidn~ a zero
vniue of the resulting angular velocity and acceleration of the glass, where
ia the requirPd reaotion force durinq graspinq of the qlass. If condi-
t~on (23) is ur~fulfilled, it is simple ta celculate the neCessary condition
of compensation.
The considered problem was investigated experimentally by mathematical
modeling methods. Specific difficulties in synthesizing the controlling
moments are related here to the need to strictly maintain a vertical poaition
of the glass during Sts equally accelerated and equally retarded motions.
The minimum value of the controlling moment is limited by the value of the
inertl.al force of the glass with the liquid. It is difficult to achieve a
satisfactory solution of this sgecial problem with a simple configuration of
the mxnipulator.
Conclusions. A method of synthesizing the nominal dynamics of an object
having excess degrees of freedan durinq partial uncertainty of it is outlined
in the paper. The term "qiven dynamics" may b~ interpreted here as the con-
ditions for simple efficiency of the system. A large number of interesting
analogs to these conditions can be found in biology. Described in mathemati-
cal forr,~ as constraints, these conditions are essentially some criterion of
optimizinq the motion of the considered anthropomorphic mnr~ipulator. Thus,
by studying the motion of a two-support robot, the condition of its stability
can be written in the form of zero equality of the total nament at the pofnts
where the supports touch the surface of motion. Although the formulated pro-
blem as a whole is known in control technology (this is especially true of
56
FOR O~FICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~QIt n~I~'ICIAL USL ONLY
contirol of mobile ob~ectis), tihe die~3,ngui~hing fea~ura o� the coneidered
postiula~l,on is, we f~el, the partiial uncerbain~y of ~he nominal dynrunio~ of
~ha ob~eoti and of tihe controllinq moments. A significant fea~ure o~ ~he
conaidared poetiulation is a1~o tihe need tio ident3fy the workinq mode ~,n caee
of itie larqe daviatiions trom nominal values. To determine tihe rnnge of
daviatiions of pare~metiers, continuous variatiion of the veator of staties per-
mitia one to edmpare the real trajeatiory of the syetiem witih tih~ closeg~
nominal tira~eatory, whose parame~ers are placed in~o the memor~,es of tihe
contirol ~omputier. We fsel thati tihe approaah conel.dered here ~o syntihe~is
of control of a compiex mechanicai ~yatem using a digital computer is pro-
miaing.
BIBLIOGRAPHY
1. Letiov, A. M., "Malytical Con~roller Design," AUTOMATZCS AND REMOTE CONTROL,
Vol. 21, Nos. 4, 5 arid 6, 1960~ Vol. 22, No. 5.
2. Vukobratovic, M'. and D. S~okic, "Dynamia Control of Unstable Locomotion
itobots," MATtiEMATICAL BIOSCIE~iCES, Vol. 24, 1975.
3. Vukobratov~c, M. et al., "Algorithmic Control of Assistiive Devices for
Severely Handicapped Persons," Proceedings of the Fifth Interne~ional
Symposium on External Control of Human Extremities, Dubrovr~ik, 1975.
4. Vukobratovic, M. et al., "Development of Active Exoakeletons," MEDiCAL AND
BIOLOGICAL ENGINEERING, JanuaYy, 1974.
5. Vukobratovic, M. et al., "Analysis of Energy Demand Distribution Within
Anthropomorphic Systems," TRANS. OF THE ASN~, JOURNAY, OF DYNAMIC SYSTEMS,
N1E;ASUREI~NT AND CONTROL, December, 1973.
6. Vukobratovic, M., D. Stokic and D. Hristic, "Dynamic Control of Anthropo-
morphic Mar?ipulators," Prnceedinqs of Fourth International Symposium on
Industrial Robots, Tokyo, 1974.
7. Vukobratovic, M., D. Hristic and D. Stokic, "Algorithmic Control of
Anthropomorphic Manipulators," Proceedinqs of Fifth International Symposium
on Industrial Robots, Chicaqo, 1975.
COPYRIGHT: IZdatel'stvo "Nauka", 1978
6521 .
CSO: 1870
57
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOtt OF~ICZAL US~ ONLY
_ GEOPHYSIC3~ A3TRONOMY AND SPACE
SEMIAUTOMATIC MANIPULATOR CONTROL SYSTEMS AND COMPUTER INVESTYGATION OF'
THEIR DYNAMICS
- MOgCOw DATCHIICI I VSPOMOGATEL'NYYE SISTEMY KOSMICHESKIKH APPARATOV. ROHOTY
S MANxPULYATORY. TRUDY =FAK in Russian 1978 signed to press 8 Aug 78 ,
pp 115-123
(Article by V. S. Kuleahov, A. G. Leskov, V. S. Med�~edev and A. S. Yuahchenko,
USSR)
(Text~ l. Methods of semiautomatic control of manipulatora. Selecting the
method of manipulator control is detiermined by the class of working opera-
tions for performance of which it is designed. The most widely used for
semisutomatia control systems is the high-apeed method tl]. Control is
accomplished by using special levera having 3-6 degrees of freedom (2;. The
deviation of the lever determines the speed of rotation or the forward motion
- of grasping and tha related object of manipulation. In combinntion type
systems, the operator can assign the required orientation and poaition of the
grasp in space and also some of the simplest operations from the computer
control console. .
The force-vector principle of control may be used in those cases when it is
required to control the extent and direction of forces and momente applied
to the object of manipulation [3, 4j.
When performing operations which require precise positioning of the object
of manipulation or of the working tool, the copying princip.le may be used in
semiautomatic systems. However, thE~ use of a computer significantly alters
its realization. The kinematic diagram and qeometric dimensions of the con-
trolling member are now determined by the dimensions of the workfnq zone and
by e~~onomic requirementst the kinematic diaqram of the servo member may be
_ differen~. Unlike copying type manual control systems (5]~ reproduction of
the positions and orientation of qrasp in semiautomatic systems does not re-
quire reproduction of the relative coordinates in slave member mobility. To
emphasize this difference, let us call these systems semiautomatic position
control systems. If there is a force reflecting channel, the computer per-
mits highly accurate reproductfon of the forces on the control member acting
on the object of manipulation and relieves the operator ~f perceiving the
forces causes by the mass of the manipulator.
58
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOIt O~~ICIAL US~ ONLY -
Eaah of ~he enumerated metihode of contirol ie ef�ec~ive only when per~orming
a apec~gia olaes oE opera~iong. Zn thie regard, dif�eren~ me~hode may be
combined during developmenti o� eemiautiomatia manipulator con~rol systems~
t~
~ ,
~ RN
6'~ R~j ~
R>
G e?
t ~ R~a
Fb
e~ M
I d
I
G~
� Figure 1
2. The dynamics of manipulator control systems. Let us consider the equations
of manipulator dynamics and aome algorithms of aemiautiomatic control using -
computers and let us compare them.
The slave member of the manipulator consists of sections connected in kinemr~-
tic pairs which form an open kinematic chain (Figurg 1). Let us denote the
unit vectors of the rotational pair and sliding pair axes by ei and hi, the
moment or force developed by the servo drive of the i-th pair as a function
of the type of pair by Qi and Pi, the weight by Gi, masa by mi, ~he vector
of the main tnoment of inertial forces of the i-th section during rotation
around the center of mass by Mi and the vector o� total acceleration of the
center of nv~ss of the i-th section by wi. The vectors of the external force
F~ and of ~he external moment M~, are applied to the center of mass of the
last n-th section. Let ua consider this section jointly with the object of
manipulation.
The motion of the mechanism is describad by the following squations of dyna-
mics (6) in projections onto the rotational pair axes:
n
Qip r~~e R~p, n X Fo ~j (Mm R{p~ n? %C ~C+m ~~mwm~~J eip �
L mslp
p - i, . . l;
59
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
' FOR OF'~ICIAL U5~ ONLY
and in pro~ectiions ontio ~he sliding pair axess
p!~ -f' C~''r -4- ~i ~Cm - mm~m)~ hir = r =1~ , . k~ 1? l ~ n.
m��fr
Let us subaequently make use of the matirix no~ation of these equationa
IQ p1-}- (Fg Jn,nW~~ G'n ; M~ MnJ A-I- (G mu~ j A'1�) B a Q~ ~l ~
where [QP~ e[Q~,1� Qilp j i. P jk~ is the [1 x n] vector, [F~ - mn, wn +
~ ~n;Mv Mn] is ~he ~1:6] veator and (G1 - m1w1���Gn-1 ' mn-lWn-l;M1���Mn-1~
is ~he (1 x 6(n - 1)) vector,
- ~I~ X Rti~ e1~ X R~~~ n~.. e~~ X Rt~~ n hii hl~ h1k
A=
~t~ et~ o o... o '
B - I Bu B~l
~B9i ~
' 811~ B12 and B21 are matrices consisting of the follawing columns:
. 0 0 0
. `
0 0 0
B e{ X R~ ~ ~ B e~ B h�~ p= i.,..., l~~
11= P P P ~ 11= P+ 14= ~ r ~
e~ X Ri ,{~1 h~ r=1 k,
P A P P r
e{p X R{p n_~ etp , h j*
l, _ l if ii~Gn-1, k_ k if IKSn-1.
,
l-1 if i~ = n, {k -1 if jx - n;
_ the number of zero components in each column o� matrices B11 and B21 is equal
to iP_1 and that in the columns of matrix B12 is equal to (jr - 1).
Equation (1) permits one to compare the different methods of control and to
, determine problems of further investigations.
Let us begin with consideration of control by the force vector. The human-
operator develops force F and moment M on the control member, which repre-
sents a six-step lever as in high-speed control. During control, the 4perator
~ observes the motion of grasping by usinq a visual information system (Figure
; 2). It is suggested that the control forces (QP] be detezinined by the formula
~ [QPl = [Fl~il A - GB~ ( 2 )
60
, FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~'Olt OFP'ICIAL US~ ONLY
where [FM) ~ I1C1F'Ok~N10~ ~ k~ and k2 are scale coefficienta.
The second tierm on the righ~ eide of equality (2) can be introduaed if the
manipulator has no mechanical relief syetem.
By subatiituting the veotor (QP~ into tihe equatiion of dynamios, we find
(j'' `f' ~''s - rli~~tvn G~? ; ~1f -}-111 ~ ItiI n] ~ = N. ~ 3 ~
'I'he right side of this equation
N = (mm ; --1?1�1 B
determines the vector of the generalized �orce de~ennined by the inertia of
the manipulator. rf, spacifically, this term is close to zero, matrix A is
_ nondeqenerate and (FvM~~ @ 0, then the motion of the n-th section and the
related object of manipulation will approach the motion of a free solid due
to the effect of tihe controlling forces [FM~ controlled by the operntor
(R - ?n,~~� -4- G� ;1Vf M�1= 0.
lfi. ~i1
2
- q~
!~f'- OOION _ kvac.s ?/f pr/N ,~vo~an ~9'c X~l
q~~mplf'.Nl ~{�c/i% ~a~,e~
~6 L'ar~s/a~ o~p. tOa?~7~~
' JOumsn~K~ a~O c.t~~i
Figure 2
KEY: _
l. Human operator 5. Manipulator
2. Control member 6. Visual feedback
3. Computer 7. Power feedback
4. Drives
In the remaining cases the motion of the n-th section will be deflected from
that required due to the effect of inertial forcea N and of kinematic coup-
- lings. By integrating equations on a digital computer and by beinq given
the generalized forces [FM] characteriatic for typical working operations,
one can determine the value of the corresponding forces N. Z"he range of
applicability of the control method by the force vector for a given manipu-
lator'can be determined with regard to the characteristics of the operator's
indirect perception of the effect of these forces through the deviations of
motion caused by them. ~
61
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOR OFFZCTAL USE ONLY
i� the extiernal force+~ [F~M~) are unknown to the operator and alao i� there
is a signi�icdnt effect of inertial forces, it is feasible to have feedback -
which directly informs the human operator o� ~he effect o� ~he indicated
forces.
During control by the velocity vector, the operatior controls the speed o�
- sliding motion and the angular rotational apeed of graeping and the related
ob~ect oE manipulation in proportion to deviations of bxr and scpr which
con~rol the six-atep lever. The veloci~y vectior of relative motion 3n degrees
of manipulator mobility
(~Pto~Jo) _ [~Pt~o , . . ~Ptio~ j,o . . , ~jkoJ .
is selected as equal to
[${o~to1T = A`' (dxob~CoJT; . _f~oamol = [~ca~Pvl k,
k= diag[ki~, ki are proportionality constants.
Assuming approximately that the generalized forces developed by the drive are
proportional to the error between the given and current values of epeeda
with respect to motion, we find the following law of control:
(QP1T = kA'' [b~b~p~T - (G1~1T~ (4)
where [ SX S~ J= xp - ic p-~) ~(xp - x cp p- ~p l is the error between
the given [x~ c'P~] and current (ic ~ J valuea of the graspinq speeds of the
manipulator and k= diaq[ki) is the matrix of the proportionality constants. -
1 The equation of dynamics has.~he followinq1form: -
(by~~'I (AT j-~ l~ IFa - ?nnu~~ C� ? ~11. -F' ~1n1 A = N. ~5)
During control in the copying mode by using the computer, the control diagram
is similar to that in Figure 3j feedback is accomplished in this case by
position rather than by speed. With the previous assumptions, the controllinq
forces of the servo drives and the equation of dynamics have the followinq
form:
f QI'1 T = k,A''Aa ~b4'~a~ta~T - [CBjT, (6)
Ib~ta8x~a1 Aa (AT k~ -I- [FH - m�w� -1- C� ; ~11, -i- Mn~ A = N, (7 )
62
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~'OIt OT~'T~ ICIAL U5~ ONLY
where matrix AZ corresponds ta the con~rolling mem~er and [~Scp~,z d~x~x) is the _
- error vector between the given and current values of i~s rela~~.ve coordinates.
- Force feedback (the dashed line in Figure 3) may also be accomplished in this
system. Aesuming ~he posaibility of ineasuring the generalized �orces R in
each of the degrees of manipulator mobility, one can calcula~e the ~orces Itp
developed in the degrees of con~rolling member mobility by the �ormula
Ro = (R CB) A"lAe = (F'' n mnwn Cn ; Mn -f - Mnl A -I- N) A-iE(o
_ _ (Fn - m,nu~n -I- ~'+n ! Mg -I-1~1n] A~ NA-lAe~
The human operator will perceive the following generalized force on ~he lever
of the con~rol member
(Fs - m�w� -1- C,,;161H -I- Mn) NA-1~ -
The problem of analyzing the effect of the last term, determined by the mani-
pulator inertia, as in the f3rst case, requires apecial investigation. -
1f4. Ms!
- qe. Jl~~iY- OQldN - eiaac~n Mmra Xe!
ne~um.r~`~ npoBir. ~Amn~
~d',Zj c, xc ~
1 ' " y'a~
� ~ Ay ~
i L'!lAO~~X O ,O. ~'O.f?/ ~_r_~
. ~6~ -
.Ipu/rlG~OHQ/I 06~t C[~!d
Figure 3
I~Y :
l. Human operator 5. Manipulator
2. Control member 6. Visual feedback -
3. Computer 7. Power feedback
4. Drives
By comparing the considered control diagrams, we note that the method of
manipulator control by the force vector has the simplest alqorithm which �
does not require manipulatfon of matrix A. Unlike the other methods, it
permits the operator to control the force and consequently the acceleration
of motion of the object of manipulation, which is important when performing
work with increased requirements on the dynamics of the process. Since the -
main feedback in the control system by the force vector is realized through
the human operator, the main problem is to investigate this system as a
bfoloqical enginepring system.
63
FOR OFFICIAL USE ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOit OI~'FICIAL US~ ONLY
The ma3n distinction of ~he speed and poaition methods o� con~ml inc].udes
~ the presence of �eedbacks in the technical part of tihe syatem, which is a
_ multicoupling automatic control. system. Problems of investigating the ata-
bility of the technical part of the eystem, ~he poseibility oE autooacilla-
~ions oocurring and taking into account ~he dynamics of drives in each of the
degrees of mobility and of their mutual effect arise in this regard. Subae-
quent sections of the paper are devoted to investigation of these problems.
3. Linearized dif�erential equa~ions of manipulators. The equation of dis-
turbed motion of the slave member of a manipulator with n degrees of mobil3ty
has the form ( 7, 8]
H ~A) q ~t) Mu (t), H (P) = aP' -F' bP (8)
where H(p) is the operating matrix which characterizes fhe dynamic properties
of the slave membsr as the control objsct, a, b and c are matricas, q(t) is
the vector of the generalized coordinates of the slave member and Mp(t) is
the vector of moments developed b~ the drives of the degrees of mobility.
Together with the equations of the drive complex
Nw ~P) xi ~t) = A1 M ~P) ~~l - 9
Na ~P) xz ~t) = l~Ic ~A) ? ~t), :lla (t) = xl - xa (t) ~9I -
equations (8) describe the mathematical model of the manipulator dynamics
(p is the operator of differentiation).
Here N~ (p) , N$ (p) , Mm (p) and Ms (p) are matrices whose elements have poly-
nomials from p, u(t) is the control signal vector and xl(t) and x2(t) are
the maving moment and moment of resistance developed by the drives.
Let us solve equations (8) and (9) with respect to the vectors of the slave
member and control signal coordinates, having represented them in the form
of the input-output relations
- ~ (t) = X-1(n) WM c~) ~l (t) = cv (p) u
X ~P) = N ~A) tiVc (P) -j- WM (n), . cio~
WW ~P) = NMl ~P) .ll� ~P), Wc (n) = N~' ~P) :17~ ~P)~
_ where W(p) is 'che transfer matrix of the manipulator control system complex.
The diagonal elements of this matrix reflect the properties of separate
systems and the elements not be]onging to the main diagonal reflect the
cross-influence of external effects.
A block diagram of the system corresponding to equations (10) is pr~~r3ented
in Figure 4.
64 ~
FOR OFFICIAL USE ONfLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
xoiz nrirr.crnt~ cls~: nNLY
X, ~v ~ Q
H'~ /P~ -+1 /P~
l
- ~+'~/p~
F~igure 4
Study of the dynamic properties of manipulator control systems by ~heir mathe-
matical models is also a problem oE investigatir,a the manipula~or. as a multi-
dimensional automatic control aystem.
4. Analysis o� the stabillty of manipulator contirol syatems. Let u~ compile ~
_ thE operating matrix of system (10) and let us calculate its characteristic
det~rminanti. We find -
!I (p) ~ C;
~ ~p~ = ,~~C NC~/~~ ~ ~
.11M~P) ~ NM~P) ~11~
~ ~P) = det 1L ~P) = det N~, ~P) det Nc ~A) dut (f!o ~p) '4- wc ~P) ww ~P)l~
Let us assume that no transformations of any kind occurred which lead to a
reduction of the order of the equations during calculation of matrices t3m(p),
NS (p) , Mi,~ (p) and Mm (p) , which compriae the transfer matrices of the drive -
complex and also matrices W(p). Therefare, ayatem (10) describes the rela-
tions between all the variable states and all the componenta of the input _
ac~ions.
Let us make use of the Nyquist criterion to investigate the stability of thia
system. System (10) does not vary if the product Hp(p)q(t) is ~idded to the
left side of the first equa~ion in it and is subtracted. We have
11o~)4fr) = 111o(t), ,1to(t) - z,(l~.....ss(~)--xa(l)~
- NM ~P~'ZI ~t~ -/~I M(n) tu - q(t)1, N~ fn) ta tt) = Al~ fp) Q(t), -
z~ = f 11(P) - Ho (P)14 (12 )
I/o ~P) = dia6' I~lo~ (P) flos ~P) . . . Non (P~~~
Iloi ~P) = a~iP' c~~, t= 1, n.
The characteristic determinant of this system coincides with determinant (11).
Let us calculate the characteristic determinant of an open system by assuminq
the value of xg(t) = 0 in the second equation of system (12). Havinq per-
formed the transformations of the operatinq matrix of the ~erived system
sinilar to that presented above, we will have
Do (P) = det N� (P) del N~ (pj det [110 (P) Tvc (P) -i' WN ~P)1� (13)
65
FOR OFFICIAL iJSE ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR d~~ICIAt. US~ ONLY
The relatiionghip betiween the values of Q(p) and 0 p(p) ig ea~ablieh~ad by
the reintion
0~P~ ~ ~n ~p) d~t IE R~p)1~ (14)
where R(P) � tHp~P) + wgfp) * Wm(p))"lIE!(P) - HpfP)~. The determinanti of
matrix E+ R~p) 3e reduced to the form
det (E R (p)1 ~ 1 IC (P),
where K~p) is the eum of a11 the main 1-order minors (1 S 1 6 n) of tihe deter-
minant of tiranefer matrix R(p) and K(p) = L(p1/~p(p)~ L(p) ia eome ~oly-
nomial of p.
Let us aeaume that aystem (12) is asymptiotically stable in the closed atate
with respec~ to tihe mutual effect channele. Thie assumption conforms to
tJlat of the asymptotic stability of the complex of manipulator aystems tiaken
separately, i.e., sysxems determined without regard to interference.
~ti
For stability of manipulator control systems, it is then necessary and auffi-
cient that the hodograph of function K(j w) noti encompass point (~1, j0) upon
variation of frequency ~ from 0 to oo. This algorithm for analyzing atabf-
lity can bs accomplished on a diqital com~uter.
In the low-frequancy rang~e for elements of ;:ransfer matrix W(jcJ), the fol-
lowing relations are valid
C~t/Ctt at i ~ k,
1i~ik j~,~i (1~)lC{c Qt t= k. (15)
For the hiqh-frequency range, we have
det e j,ywl ~
_ lvt~ ~/W) - dec ai c~ . (16)
- It follows from expressions (14) ,(15) and (16) that ~ K( j c~ ) 0 in the low- -
frequency range and the approximate equality
n
K(fc~)zdelal~a~i-1.
+~i
is valid in the hiqh-frequency range.
Therefore, when approximating the construction of the hodograph K(jcJ), it is
sufficient to limit oneself to the low-frequency range (Fiqure 5).
66
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
F'dEt U~~ICIAI, U5~ OtJLY
J,p
j 0 ~1 w~ ,f'~
Q/1+a~
~'/~wJ
d~t.~/
v.~~
, Figure 5
n
Since the ~l~gnificance of the frection do~ a' ~ a~~ may be conaiderably lese
(a~
thnn unity, consideratian of the hodograph in the high-frequency range ie not
convenient. Therefoze, the hodoqraph of function Q(jc~) ~ 1+ K(jc~), which
may also se~-ve as an indicator of the contirol syatem interaction, may be con-
structed inFitead of hodoqraph K( j w).
- 5. Analysis of the dynamia propertiea of manipulator control systems based on
frequency m~thoda. T'he frequency cheracterietice of separate aystems corre-
spond to dia~qonal, wrhile those of crosg c.rouplirigs correspond to nondiagonal
elements of the matrix
w~~~~ ~ x-t (t~) tiy~ UW)� ~ 17 ~ -
Let us represent each of the matricea in the form of the sum of two matrices
with real and imaginary elementa. The siqnificance of the mtttrix elements
at different fixed values of frequency is deteraiined by addition, multiplica-
tion and conrreraio� of matrices with complex elements. _
Zb deterntina the deqree of distortion of the frequency characteristics of a _
system taken aeparately as a function of the values of the parametera of
other aystemei of the complex, it is suqqested that the follawing estimates
be used
b~ (t~) = 1-- ~ ~i (iw) Uc~) =.4ii (I~) Xir Uw)~ (18)
where Xik ( j W) are elementa of matrix X'1( j w). In this case the frequency
characteristic of the i-th aepr~rate system wii(j w) may be expressed by the
frequency characteristic of the correspondinq system Wi(jc~) taken separately
in the follow~ng manner:
(19)
Wti (t~) = O~ (f~) ~'i U~o).
61
~ F0~ OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OEt OF~ICIAI. U5~ ONLY
The v~lidity of the fo],lowing equali~y followe from tha propertiee of the
product of the direot and inverge matrtcee
n
U~~ ~u U~1= 1 - E' WIk ~I~I ~kl ~I~~~~ (20)
kM
where Wki w) nre elementie of the k-th colwnn of matrix W 1( jw
Comparing reletion (20) with expreesion (19), we have
w++ Uw) ~u UW) = Au (Iw) X~c = dt U~)~
hence follows ~he re~ation
I o~ (iW) I= ~ ~ W ~ Cv
k~ ~k (i ) tt (k~) I . (21)
x,~~
~ The followinq inequality is then vdlid -
n
- 8J = i- ~ e~ U~~) I~ E~ I WIk ~f~~ W k! UW~ I
k~.!
and one can write
n _
ai ~~J~ ~ ~ I jV1k ~f~J~ ~'Vki ~fW~ I� ~z2~
_ t,�i
Let us denote
I ti~'~k ci~) ~ti ~ ci~~) I= b~k
According to inequality (22), we find
. *
ai ~ ~ bte (23)
k.~l
Ra~i
It follows from expression (23) that the extent of dist:orting the Erequency
- characteristic of the i-th syatem taken separately is greater, the larqer
the modulus of each of the terms on the riqht side of inequality (23), and ~
the k-th system for which S ik (w) ~ di~ ( w), f, k, 1= 1, ...n, i~ k,
k~ 1 and 1~ i, will have the stronqest effect on the dynamics of the i-t:,
system. The "contribution" of the k-th following system to distorti.on o: .r~
,dynamic properties of the i-th system, which occurs due to interference of
the control systs::s through the slave member, can be judqed by the value of
~ik~w~�
68
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR OF~ICIAI~ USC nNLY
Re~lixing tihe propoeed algorithmg on e digl.tal cotqputi~r, nne ca~n a~oaompli~h
multiileterel inveetiga~ion of the dynamio propertiiee of manipule?tor contirol
~y~teme by parlorming ao~ione witih complex ma~ricses.
S`, q~'
qs
q~ q~_.. 4+
Figure 6
totg/
ca~ ^ wj
~ jA ~
~
-o-~o- � I
arg~t�-ieo ~ ~ ~~9(1
( ~y~
Figure 7 .
Zo ig /K'~r/
~oc9a;~~m~ y
2oig/w~/
2069/~~/
a~ w,,~-~eo
a~g w~
Fiqure 8
These algorithms are the basis of the developed library of standard computer
programs which permit calculation and subtraction of the frequency charac-
teristics of the chanriels of a multicoupled manipulator control syetem on
the graph plotter of a diqital computer and which permits analyeis of its
dynamic properties.
Exan~ple. The effect,iveness of the proposed alqorithms is illustr~ted in the
exauaple of investiqatinq the dynamic properties of a six-section manipulator.
The kinematic diaqram of the slave member of this manfpulator ia presented in
Fiqure 6.
69
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~d~ o~r~c~~u. us~ ortLY
F'unctiion k(~ta), calculatied on the dig~.tia1 compu~er for aeveral positione of
tihe kinematiic cha3n, ig pregented in Figure 7. t~ is obv~.ou~ that ti.he hodo- -
greph of funotiion K~~w ) for the conaidered manipulatiuz does no~ oncompae~
the poin~ (-1, ~0). Therefore, the menipulatior con~rol systams ara also
etiable with regard ~o intierferenee.
iti ia mor~ �eaeible to calculatie �unction Q(~~ 1+ K(~c~) w3~h re~pec~
~o manipula~or con~rol syatems. 5ince Q(~ w) directily links the value of the
frequency charec~eris~ics of complexes of separate syatems and those taken
aeparately, i.e.,
e U ~ Q (l ~o w)~
then consideration of function Q(~r~,) permits one to also judga the effec~
of system interaction on ~he properties of a compiex of syatems taken sepa-
. rAtely, begides determining ~he fact of the etability of separate syetems
(accordinc~ to the arrangement of curve Q(j w) with respecti to point (0, ~0)).
The resultis of calculating the frequency char~cteristics of aeparatie nu~nipu-
lator control syatems and aleo ~f functions d ik(w ) are preaented in Figure 8.
It is obvious from the fiqure that the interaction of control systems leads
to siqnificant variationa in the properties of separate control systems and
to the appearance of ~he croas effect of external actions. In thie
case functions Sik(c~) determine tha "contribution" of k-th cantrol
systems to distortion of the dynamic properties of the separate i-th ayatem.
The proposed algoxithms may be the basis for universal investigatio~i of the
dynamic properties of trackinq manipulator control systems.
BIBLIOGRAPHY
1. Whitney, b. E., "Resolved Rate Control of Manipulators and Human Prosthese~,"
IEEE TRANS. ON MAN-MACHINE SYSTEMS, June 1969.
2. Saenger, E. L. and C. D. Pegden, "Terminal Pointer Hand Contxoller and
Other Recent Teleoperator Controller Conaepts," Proceedings of the First
National Conference on Remotely Manned Systetc~, California lnstitute of
Technology, 1973.
3. Nevins, I. L. and D. E. Whitney, "The Force Vector Assembler Concept,"
ICiSM-JE'TOM Symposium on Theory and Practice of Manipulators, Udina, Italy, '
September 1973.
4. Maslova, N. K. and A. S. Yushchenko, "Some Algorithms for Manipulator _
Control Usinq Diqital Computers," IZV. WZOV, MASHINOSTROYENIYE, No. 2,
1975.
5. Kuleshov, V. S. and N. A. Lskota, "Dinamika sistem upravleniya manipulyatorami"
[The Dynaznics of Manipulator Control Systems], Moscow, Energiya, 1975.
70
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR O~T'ICIAL US~ ONLY
6. S~epanenko, Yu. A., "Algorithms for Analyais of the Dynamica o� Threa-
Dimenaion~l Meahaniems With Open Kinema~ic Chain," in "Mekhen3ka maehin"
(The Meohanics o� Mashines~, No. 44, Moacaw, Nauka, 1974.
7. Popov, Ye. P., A. F. Vereshchaqin, A. M. Ivkin, A. G. Leekov and V. S.
Medvedev, "Conetruc~ion of Roboti Control Sys~ems Using Dynamia Models o�
- Manipulating Device$," Report at the Sixth internationa~. Symposium of
ZFAK "Contirol in Space," Yeravan, ixd. Arm. NIINTi, 1974 (Rotaprint). ~
8. Leskov, A. G. and V. S. Medvedev, "Analysis of the Dynamics and Synthesis
of Motion Control Systieme of the Slave Members o� Manipulators," iZV. AN
SSSR, TEKHNICHESKAYA KIBERNETIKA, No. 6~ 1974.
COPYRIGHm: IzdBtel'stvo "Nauka", 1978
65 21
CSO: 1870
' 71
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~OR U~~'~CYAL iJ5~ QNLY -
~EOPHYSICS~ ASTRONOMY AND SPAC~
ALGORITHMS FOR COMBINATION AND SUPERVISORY CONTROL OF MANIPULATOR-ROHOTS
MosCOw DATCHIKI I VSPOMOGATEL'NYYE SISTEMY KOSMICHESKIKH APPARATOV. ROBOTY
I MANIPULYATORY. TRUDY IFAK in Russian 1978 signed to press 8 Aug 78
pp 124-127
[Article by Ye. P. Popov, A. F. Vereshchaqin, V. L. Generozov, S. L. 2enkevich
and V. B. Kucherov, USSR] '
' IText] The most modern manipulator-robota of today combine human intellec-
tual capabilities and the capabilities of high-speed computer equipment in
the control system. Man can solve a wide range of target designation pro-
~ blems both at the level of controlling the speed, trajectory or position of
the manipulator and at the level of forming subtargets of the computer in
the combination and supervisory control modes (1J, which transforma these
subtargets by m~ans or control algorithms into direct instructions to the
slave drives of the robot.
Let us consider the functions of control algorithms when the human operator
is introduced into the process of automatic control of the actions of a
manipvlator-robot at a high 1eve1 of the hierarchical control system by ,
- assigning the required technoloqical operation to the robot in generalized
- form.
The control hierarchy. The problem of controlling the slave member of ti~e
_ manipulator-robot from a digital computer when performing a qiven operation
in the automatic mode can be divided into three main steps determined by
three hierarchically related levels of the control system.
Generalized information which determines the content of the operation and
also the lacation of the required objects and tools is fed to the hiqher-
level input, called the strata~ic level [2j. This level, usually relying
~ on heuristic procedures, sepaxates the operation into a sequence of elemen-
tary grasping motions which change position and ori~ntata.on. in this case
the given operation is cor�;erted by the higher-level algorithm into a se-
quence of macroinstructions, for example, of the following type:
A~Sl+ pl~i e~Sz~ p2~i ...i e~Sm pn~r
72
FOR OFF~CIAL USE ONLY -
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
A'OK UI~rIC]"AL USL ONLY
where s is the parame~ers oE the final sta~e of grasping and p~.s the para-
me~ers which determine the mode of ~ransition to ata~e s(�or example, motion
in a straight line with retention of orientiation, provision of given force -
by grasping and so on).
The problam of the nexti 1eve1, called ~he ~ac~icaL 1eve1, ie convorsion of -
the macroinstructions r~� elementary gzasping motions to laws of matched
variation of generalized coordinates of the slave member in which grasping
changes according to the given mode of motion from the current position to
the gaven Einal position.
_ Sf motiion during each stage of mobility is provided by a self-contained
drive, tihe output of ~he tac~ical level of con~rol is the input control
- signals of these drives the desired laws o� mo~ion in the movable 3oints
of the mechanism.
~ The last, slave level of control is formed by the combination of the tracking
c3rives. The input of this level is the desired laws of motion in the movable
joints Y~ith regard to the permissible speec3s and accelerations of the drives
- and other reatrictions.
These laws may be calculated by two methods at the tactical level.
- The control signals of the drive in the entire range of control which realizes
the macroinstruction are determined in time and calculated for any macro-
instruction of type 6(s, p) in the first method. Moreover, the leng~h of
execution is a function of the permissible speeds and accelerati.ons of drive E
tracking [4]. The digital computer then generates these signals, usually
represented in the form of spline-functions at the drive input for execution.
The second method assumes sequential correction of the grasping position by
calculation and realization of the sequence of small increments of the
generalized coordinates of the slave member [3). Linearization with subse-
quent solution of linear equations and inequalities is used extensively when
calculating small increments of generalized coordinates.
The tactical level of control should have the capability of usinq both the
first and second method of control. The first method is convenient when
performing complex operations which permit preliminary planning. Z'he second
method is especially convenient when performing one-time small movements,
J k~ut also permits one to perform complex grasping motions.
Elementazy ~roblems of planning trajectories in those cases when performance
of one or another motion is standardized or when the higher level of control
does not place any restrictions an the method of transferring grasping to
given positions are also solved at the tactical level.
Thus, the digital computer solves three main problems at the tactical level:
planning of elementary grasping motions, separation of goal-oriented grasping
73
FOR OFFICIAL JSE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
,
FOit 0~'~TCIAL USE ONLY
motion to matched motions of self-con~ained dr3.ves and, finally, forma~ion
of drive contral signals with regard to the technical aapabilitiies, One of
the forma of realizing this control ideology is described in (4~.
The inverae operator problem. The result of a digital computer opera~~.ng at
the stage of planning the elementary motion may be a sec~uence of vectore
Sl~ S9~ . . Sm~ '
which determine the posi~ion and orientation oE grasping at sequential p~ints
of the required trajectory. If each drive controls varia~ion of one genera-
lized coordina~e of the slave membez, it is primarily of intereet to describe
the trajectory of motion in terms of generalized coordinates. A system of
nonlinear equations de~ermined by the design and current confiquration of the
slave member must be solved for this purposQ for each node o� the trajectory
- S{ _ ~ ~4{)~ t = 1, . . m, (1)
where qi =(qi, q2, qn~ is ~?e vector of generalized coordinates.
The effec~iveness of control depends~primarily on how quickly and clearly
this inverse problem is solved on the digital computer.
The matrix of generalized coordinates is found as a result of solving a se-
quence of inverse problems
~4i), ~ =1, 2, . . ~;t; 7' =1, 2, . . n,
corresponding to the required grasping trajectory.
The problem of constructing i:hese continuous control signals, which take into
account the technical capabi`.ities of the drives, and of the following se-
quence of moments of time now occurs
~1 C ~2 < . . . ~ ~M-1 \ trnr
to which the following conditions are fuiiil3.ed
{
4i (tt) = 9i
and the control time tm is minimum.
In control by the second method the method of small sequential changes of
the position of the slave member, the inverse problem reduces to the need to
solve linearized equations
74
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
. ~'OR 0~'rZCLAL USL' ONLY
_ ~S{ s++~ ~ s{ ~ ,l (q{) ~q~, (2 )
where J is a Jacobian matrix for the right sides of equations (1) to deter-
mine variation of ~q by tihe given increment ~ s of the vector of ata~e of
qrasping. -
~ Sf n= 6 and ~he kinematic diagram of the mechaniam is selected in a suitable
manner, the inverse problem can be solved precisely both in the nonlinear and
in the linear vdriant. However, the~absence o� a solution at some points of
the tra~ectory, degenerate situations, ambiguity and restrictions on the
- generalized coordinatea force one to seek special methoda of realixing the
planned trajectories. The problem becomes considerably more complex a~ n~ 6,
since in this casa additional problems related to the need to overcome tha
excesa degrees of cnnbility occur. For example, what minimum qroup of drives
realizes the given motion by the number of drives? If the drives are dis-
tinguished by priority, which drives and to what extent ehould they partici-
pate in motion?
One can answer these questions if the sequence of linearized problems (2) is
solved durinq the control process by the linear proqramming method [3].
Linear programminq of motion. Let us represent the desired values in the
form of the difference of two new nonnegative variables:
~4~ = x~ - , xJ ~ 0, x~ ~
0.
Let us find the nonnegative vector x of dimension 2n + 1, which minimizes
the linear function
n
~n+t ~i Yt (x'i - x~ Y~ ~ ~
i,i
with linear restrictions
n
- xn+t \ ~ ?ii 1~) - z~ ~ - OSi ~ .~Tn+1~ I = 1, . . s
)=1
Gk -[jk ~.xk - 2'k G Ck - 9A, 2A ~ ~r z'k ~ ~r ZMI ~ U,
where Ji~ is elements of matrix J and (Gk, Gk) is the lower and upper bounds
of values of the k-th generalized coordinate.
Let us note the characteristics of the given linear model. If max ~i 00 s
Figure 5. Fragment of Layout of Restored Section of Terrain
form of coordinates of a finite r~:imber of points of its suxface. Restrictions
on the height of location of the rangefinder under conditions of severely
broken terrain leads to extremely nonuniform distribution of these points in -
_ the area of the photographed zone and significant surface reqions may qenerally
be inaccessible to observation since they are ahielded by other sections of
the same surface. This circumstance makes it very complicated and in eome
cases makes it impossible to investigate terrain in remote zones. However,
the ~act of the occurrence of screened zones contains specific and rery sig-
nificant information about the presence of one or another features of relief.
- Based on the "languaqe" approach, it is possible to establish the correapon-
ding semantic rela~ionships and subsequently to operate with them when
plotting a model of the relief and when determining forbidden zones. In com-
bination with the gradi~nt method, which takes into account the quantitative
relationships between elements, introduction of semantic relationships permits
one to represent the description of the model in more compressed form and to
83
FOR OFFICIAL USE ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
I~'Oit Ui~H'ICI:AL U5~ ONLY
determine all the features of relief on it. Since r~stir~.ctione in ~ho
,~tiorage capacity usually do not permtt storage o� ~he en~ire file of photo-
grnphed poin~s of tihe relief, compresainn and encoding of inEormation should
be conduetod in real ~ime para11e1 with its ~rrival. ~rhis is acr~iaved by
using the,forma~~.on of wo intie madiate extreme matrices of dimenaion n x m:
matrices ~Y~j max~ and ~Yi~ mi.n~, where i~ 1, n and j Q 1, m.
The colls of the extrem~ matrices are filled as coordina~es x, y, z o~ relief
points oome in from the surveying system. The number of the line i and
column j of the matrix where the number y should be directod is determined
- by coordinatea x and z, bu~ coordinates yi~ are recorded in the corresponding
cell only after comparison with the contents o� thia ce11. With this method,
the number of atorage cells is completEly determined by sel~cting the dimen- -
sion of the matrix and f.s inciependent of the number and distr3bution Yaws in
the area o� the photographed relief points, in other words, compreseion of
any in�ormaL�ion flow to a previously given volumQ ia guaranteed.
Besides information compression, the algorithm provides separation of screened
zones ~ince the corresponding cells remain unfilled. The process o� synthe-
sizing a relief model terminates with formation of a matrix of inean heights
Yijsr}~ where yij8r = 0.5 (yijsr max + Yijsr mit~ ~d of a logic matirix
Lij}p of zero step from the extreme matrices, to which information abou~
the presence of screened zones and intormation about the presence o� dan-
' gerous drops of heights inside the square ~D xi x A z~, i.e., the informaL�ion
which would be unavoidably lost with averagin of heights by area, is re-
recorded. '?'he process of forming matrix ~Lij~ and o� the language used in it
will be considered below.
The relief model synthesized by the propased method is sufficiently simple
and compact, takes into account the characteristics of obtafning information
about relief by the oblique-angled scanning method and specifics of opera-
tion of onboard digital computers, does not require large storage capacities
and pezmits a search for dangerous zones and laying out trajectories without
= construc~ing any additional networks and reference gra hs, reducing th.ese -
_ operations to sequential transformation of matrix {Lij~O�
Determining forbidden zones. Let us relate the sections of surface not
achievable to the apparatus and sections, entry into of which may lead to
tipping over, to zones forbidden to traffic. The synthesized relief model
and the presence of a resolver of data about the apparatus itself in the
memory permit one to consider the problem of determining forbid3en zone as
comparison of tne model of the apparatus capabilities with that of objectively
existing external conditions.
_ The considerpd method of determining forbidden zone permits one to solve this
problem by seq~~antial transformation of the relief mode:l. The concept of
_ the shift operator Sitk,j�l~ which permits comparison of the contents of the
i and j cell with the contents of the i� k and j� 1 cell of the sFUne matrix,
, is introduce3 for this purp~se. Operator S can be given in matrix form and
the shift operation is then ~arried out by mult~.plication of the matrices cz�
it can be realized in the fonn of a comparison proqram. Let us call the
84 ~
~OR OFFICIAL USE ONLY -
I
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
rox o~rtcinr, vs~ oNLY '
o~erators which compare adjacent cells first-s~ep operators. There wi11
obviously be aight of these opera~ors.
If one assumes tha~ k= 0 and 1 e 0, we �ind the zero-step operator Sp which,
un].ike tha zemaining operators, compares ~he contenta of ~he cella with
identical coordinates of different matrices. Operator Sp is used in formation
- of ma~rix ~Li~}p from matrices ~Y~, j~max ~d ~Yi3 m9.n~ ~~e technicai charac-
~eristics of the apparatus are modeled by a�unction of the estimate of
passability Q~ f(0(1, p~n)? which takes into ~ccount the range of
operator 51, the geometric dimensions of the appara~tus, i~s kinematic and
dynamic characteristica, the capability of overcoming elementis of relief of
varioua type and so on. If the characteri~tics of the apparatus are known,
then Q=�(p~l, p(n) may be representad in the form Q= f(oC), where
o( is ~he range of operator S. The range of variation of oc ia limited by
the geome~ric dimensions of the apparatus in the layout. The function of the
estimate of passability represented in Figure 6 permits translation of the
quantitative relations determined by the e�fect of operator~ Si~k on
- matrix ~Y~,j~ to semantic rela~ionships which are recorded in matr~x {Li~~�
Q1',~l
Qk ' -
Qj ~ I
~ ~ ~ ~ a
_ "t~ �~r �rt �r~
Figure 6. Function of Passability _
The following glossary is used to describe the results of comparing the
relief model with the capabilities nf the apparatus in matrix {Lij}:
passable cell code "0," screened cell code "1" and forbicjden cell
code "2." Expansion of the glossary by taking into account the diretctions,
for example, of the "cell covered in direction 5," divisions of zones with
incomplete information into screened zones and zones located outside the
visual field, taking into account the nature and stage of determining the
coefficient "prohibition by height of obstacles," "prohibition by slope
steepness," "prohibition of zero step" and so on, is pussible. All trans-
formations of matrix ~Lij ~ are made with regard to the semantics of the
- content of its celi.s. This is achieved by establishing the hierarchy of
codes. Thus, for example, when dEtermining forbidden zones in a three-code
~ system, the prohibition code ha.s the l~ighest priority, the screeninq code
has the second priority and the passa'nility code has the third priority.
In the case of synthesizing a model of the external medium by survey data
- from two different points, ~he code hierarchy will be different, apecifically, ~
the pass~bility code will be higher than the screen ~ode. Different codes of
the same rank are simply added.
85
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
~Olt OFI~'ICtAI. U5~ ONLY
!!/dt ( I) ~ > ~ 2 t~ 4
~t/QmdL~G yIm'aJ 1'l~Pf ~~iJr ~'y~~~? L`111t ~'y~~~~ ~1jcPJ
~ � � o 0
Oa~~p ~ O ii O ~ f] � G] �
o ~ o 0
O~lcN 1~ Bo Q~ _ QP _ QJ Q~
P~.~yn ~l
Jf, C`1l~ ~ ~ ~~p j~ ~ ~ i%~ l`~J`~~
Figure '7. Structure o�,Algorithm -
I~Y :
l. Step 4. Estimate ~
2. Matrix 5. Result
3. Operator
To Eorm �orbidden zones around small-scale local obstacles deternutned only -
- by short-range operators, the inducti.on operation accomplished by the action
of operators S~-S12 directly on matrix {Lij} is introduced. The total struc-
ture of the algorithm is presented in Figure 7. -
The considered algorithm permits one to determine danqerous zones for any
relief and is universal in this respect. The boundaries of the forbidden
zones are moved away from the obstacles by a distance which provides safety -
of the apparatus, regardless nf the class of the obstacle and of at which
step these obstacles were determined. The algorithm also takes into account -
the geometric ditnensions of the apparatus, which subsequently permita consi-
deration of the apparatus as a geometric point when laying out trajectories.
~ ~ ~ ,~.ti:
~
.
.
ti*'~i
~
C" ~
S/ /,:t~.:l~..
~ / j~^r�r~.e.a
Q ~ ~ ~1, e
i ~ "
; 4 ~�i ::~.1..
~ / ~ ,1;;.;=~:''~'~
/ i / / / ~ : i.~� ~,~.y
t
Z ~ ~ ~ / ~ ~ , vi~i~t'~
0 0 ~ ~ / / / ' 0':~5:7
_ ~ /
/ / _^'.r~.
,y~ / � ~
".f, -
~
~ . . ~ /
~ .~:�a.s~ ~
. \ ~ ..1 ~ . . o ~ .
4~; ~ i+.
Figure 8. Fragment of Relief
86
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
' ~ rox c~ricrA~, us~ oKLY
The +~1.gorithm is easy ~o realize in machine languages o~ di�ferent lovel. ~
The resul~s of applying it to the fraqmenti of relieE depicted in Figure 8
are presented in Figure 9. Procae~ing was carried out by the rCL 4~70
machine by a program writtien in FORTRAN language.
in cases where there is a priori information about the nature of the terrai.n, -
the algorithm oar, be truncated. Operation ~n even surfaaes with separa~ely
- arranged local obs~acles can be accomplished auccesefully by using short-
range operators and ir~duction of prohibitionst in "dune" ~ype ~errain, one
may, be 1imi,ted by use of only long-range third- and fourth-etep operatorst
if there are no obstacles, proceasing can be stopped immediately after the
zero step and ao on.
~
0
~
z'
w~�
,P
. ,O' / -
~
/
~ ~ , 1,
Figure 9. Results of Processing on Digital Computer
- The structure of the algorithm also provides the possibility of self-teaching
and adaptation of the robot. During prolonged functioning, th~~ apparatus may
move considerable distances and may find itself in sections mad~l up o� various
types of rock with different supporting capacity. This circumst,:~nce requires
operational reformation o~ the function of the estimate of passabi.lity, for
example, by data of instruments of the scientific complex about the meas~ired
soil characteristics. Moreover, the characteristics of the apparatus itself
may also change due to failure of individual subassemblies or reducti~n of
its energy suppZy. It is hardly possible to take into account beforehand all
variations of the external medium which may occur on a remote and insuffi-
ciently st~di~d planet; therefore, problems of orqanizing the process of
aelf-teaching and adaptation of the apparatus acquire special signifi~:ance.
- The considered algorithm permits the apparatus to utilize its own experi~ence
rather simply and efficiently. Thus, for example, if conditions have prin-
cf~ally changed, the apFaratus will find itsel~ in critical situations about
87
FOR OFFICIAL USE ONLY
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOR dFFICZAL USE ONLY
which the sensors o� tihe angles of inclination, tactile sensors, the motor
o~arload proteative system and ~o on will signnl. rf a time reduotion of
~he threshold (hor,izontal seation) of function Q corresponds ~o each caee
of response o� ~he taati.ie ser~sor arid redaction of angle y of function Q
corresponda to cases ot an inarease of the skidding coeEficient or angles
of inclination o� the apparatus above permissible levels, in the final
analysis the robot itself will determine ~e requ3.red measure of eafety and
will adapt to external cond~.tions, 3ncluding adaptation to variation of the
soil characteristi.cs in the previously unprovided range. -
' BIBLIOGRAPHY
l. Douqlas, A. and 0'Handly, "Scene Analysis in Support of a Mars Rover,"
COMPUTER GRAPHICS AND IMAGE PROCESSING, No. 2, 1973.
- 2. Mutch, T. A. and W. R. Patterson, "Examininq the Martian Surface Wi~h the -
Viking Lander Camera," Proceedings of the American Society of Photoc~ram-
metry, 39th Annual Meeting, March 11-16, 1973, Washington.
- 3. Tikhonov, A. N., "Solving Incorrectly Postulated Problems and Methods of
Regularization," DAN, Vol. 155, No. 3, 1963.
4. Koslov, B. L., L. N. Lupichev, E. N. Orel, J. K. Hodarev and I. V.
- Shamanov, "On a Method of Construction of Roving Vehicle Control System,"
Procee~ings .~f the 21st Con;~ress of the IAF, Constance, 5 October 19~70.
5. "Kosmicheskaya ikonika" [Space Ikonics), edited by B. N. Radionov, Moscow,
Nauka, 1973.
' COPYRIGHT: Izdatel'stvo "Nauka", 1978
6521
CSO: 1870 -
~
- 88
- � FOR OFFICIAL USE ONLY
. L
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
FOit Q~~ICTAL USE ONLY
�l
GEOPHYSICS~ ASTRONOMY AND SPACE
INVESTIGATiNG THE ALGORITHMS FOR CONTRiOLLING THE MOTZnN OF A SELF-CONTATNED
PLANETARY ROVER BY THE MATHEMATICAL MODELING MF:THOD
Moscow DATCHIKI I VSPOMOGATEL'NYYE SISTEMY KOSMICF~SKIKH APPARATOV. ROBOTY
I MANIPULYATORY. TRCJDY IFAK in Russian 1978 aiqned to preas 8 Aug 78
pp 148-153
� [Article by V. F. Vasil'yev, P. M. Gurvici~, L. N. Lupichev and I. V. Shamanov,
usSR]
[TextJ Developing a system for controlling a self-containad planetary rover,
moving by instructions from earth without direct participation of man ~.n
control, is related to solution of a number of problems [1~. The purpose of
the motion of this apparatus may be to achieve some terminal zone or to cover
a given route. It is assumed that the a�paratus is equipped with a scanning
system, compu~er, moving member control devices, navigation means, course,
bank, trim, covered-path sensorm and so on.
The onboard scanning sys~tems permit rec~ption of information only about a
~c;mparatively small, ":~ca.l section of the surface. A safe local trajectory
of motion of the planetary~ rover should be determined from the results of its
processing on the apparatu,s. The total (global) trajectory consists of the
- aggregate of local secti~n,s.
A system of interre'lated algorithms s}~ould be developed for onboard solution
- of problems of prncessing information about the surface, determininq a safe
trajectory and controlling the motion of the planetary rover along this tra-
jectory. Analysis of the efficiency and effectivaness nf these alqorithms -
during their joint operation can be carried out by usin~ computer mathematical
~eiodeling of the controlled motion of the planetary rover alonq the surface.
To do this, a mathematical model of a sufficiently large section of surface
should firet be,constructed in the computer and the initial location of the -
apparatus and the direction of motion or the coordinates of the ~arqet should
also be qiven.
Complex modeling of controlled motion includes the following steps:
determining the position of the axes of t:he apparatus (anqles of bank and
trim);
89
FOR OFFICIAL USE ONLY
~
APPROVED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3

APPR~VED FOR RELEASE: 2007/02/09: CIA-RDP82-00850R000100060001-3
_
FOR OF~iCTAI, USE ONLY ~
obtaininq informa~ion about the aection of surfac~ (local zone) in front
of the apparatusj
determining the passability in the local zone;
determining the trajectory of motion of the apparatus within the local
zone=
- modeling the motion of the apparatus along the trajectory.
As a result of sequential fulfillment of these steps, the apparatua "a~~vers"
some local section of the trajectory, after which all the steps are r~apeated,
the apparattus "covers" the next local section and so on. Motion along a
global trajectory is simulated in this manner.
Based on the outlined principles, a complex mathematical model, the block
diagram of which is presented in Figure 1, was conatructed.
( Mi~i~i N=/ ,
~2)
� /le/tprNO~mi
/,~o~ura~ "1
JadaHUt ~~u ( 3 )
d4aarcNU~ u
NGVORdN010 ~
/1LL70,aYYNd?
a~napamQ
iY!!�na~u- .wo.~~~~N�1 Mod~~sN,T .f~vdrniNt� MoJi~i N'3' Mo~ni N'B
1/NP r
Onp