SD 76-SA-0028-1 CR-144966 SPACELAB USER IMPLEMENTATION

SD 76-SA-0028-1

CR-144966

SPACELAB USER IMPLEMENTATION ASSESSMENT STUDY

(SOFTWARE REQUIREMENTS ANALYSIS)

VOLUME I

EXECUTIVE SUMMARY

L. R. HoganStudy Manager

Prepared under Contract NAS 1-12933for

Langley Research CenterNational Aeronautics & Space Administration

Page intentionally left blank

Rockwell International

Space Division

FOREWORD

The Software Requirements Analysis Study was conducted to definitizerequirements and implementation approaches for the Advanced TechnologyLaboratory (ATL) Spacelab payloads of Langley Research Center. The effortconsisted of an expansion and in-depth analysis of ATL software requirementsidentified in the basic study, Spacelab User Implementation Assessment Study.The study was conducted by the Space Division of Rockwell InternationalCorporation under Contract NAS1-12933. Mr. F. 0. Allamby was the technicalmanager for the Langley Research Center.

The final report consists of two volumes: an executive summary, and atechnical report of all the analyses/trades conducted during the course ofthe study. A succinct summary of the study objectives, principal conclusions,tradeoffs, recommendations, and future related efforts is presented in theexecutive summary. The technical report includes the development of the studydata base, synthesis of implementation approaches for software required by bothmandatory on-board computer services and command/control functions, and identi-fication and implementation of software for ground processing activities.

iii

SD 76-SA-0028-1



Space Division

CONTENTS

Section Page

1.0 INTRODUCTION AND SUMMARY 1-11.1 STUDY OVERVIEW 1-11.2 SYNOPSIS OF STUDY RESULTS . . .. . . 1 - 3

2.0 SIGNIFICANT STUDY RESULTS 2-12.1 ON-BOARD PROCESSING REQUIREMENTS DATA BASE . . 2-12.2 ON-BOARD COMPUTER SERVICES 2-22.3 ON-BOARD COMMAND AND CONTROL 2-102.4 GROUND PROCESSING REQUIREMENTS DATA BASE . . 2-172.5 GROUND PROCESSING IMPLEMENTATION COST FACTORS . 2-212.6 PROGRAMMATIC EVALUATION . . . . . . 2-262.7 PALLET-ONLY CONFIGURATION SPECIAL CONSIDERATIONS . 2-35

3.0 PROGRAMMATIC SOFTWARE-RELATED RECOMMENDATIONS . . 3-1

4.0 RELATED FUTURE EFFORT . 4-1

SD 76-SA-0028-1



Space Division

ILLUSTRATIONS

Figure Page

1.1-1 Study Approach . . . . . . . . . 1-22.2-1 On-Board Processor Flight Hardware Definition . . 2-22.2-2 Software Development Approach . . . . . . 2-32.2-3 Experiment Flight Software Development . . . . 2-42.2-4 On-Board Processor Flight Hardware . . . . . 2-62.2-5 Mini/Micro Processor Software Preparation/Validation . 2-72.2-6 Programmatic Cost Evaluation . . . . . . 2-72.2-7 ATL Payload 3 On-Board Processing Requirements . . 2-82.2-8 STIL Function with Mini/Micro Processor Approach . . 2-92.3-1 Data Management Design Concept Comparison . . . 2-102.3-2 ATL Experiment Panel Components . . . . . 2-112.3-3 Comparison of Aerospace and Mil-STD Costs . . . 2-112.3-4 Level III/II Integration Configuration (Typical) . . 2-122.3-5 Computer-Aided Hardware Cost Factors . . . . 2-132.3-6 Typical Measurement Display . . . . . 2-132.3-7 Computer Display (Aided) 2-142.3-8 Automated Control Logic . . . . . . . 2-152.3-9 Command/Control Concept Comparison (Per Experiment) . 2-162.4-1 Optional Processing Documentation . . . . . 2-182.4-2 Ground Processing Size Estimates . . . . . 2-192.4-3 Ground Processing Software Availability (MSFC) . . 2-202.5-1 Interactive/Automatic (Batch) Concept . . . . 2-212.5-2 - Ground Processing Tutorial Requirements . . • - . - - - . • • 2-222.5-3 Remote Terminal Alternatives 2-242.6-1 Typical ATL Payload Hardware Complement . . . . 2-282.6-2 Software Programmatic Complement . . . . . 2-282.6-3 Programmatic Costing Criteria . . . . . . 2-292.6-4 Cumulative Recurring Cost Summaries . . . . 2-332.7-1 AFD Panel Allocation for Orbiter Payloads . . . 2-35

vii

SD 76-SA-0028-1



Space Division

TABLES

Table Page

2.1-1 ATL Experiment Definition and Payload Grouping . . 2-12.4-1 Essential Computer-Supported Activities . . . . 2-172.5-1 Batch Processing Considerations .. . . . . 2-232.5-2 Implementation Cost Comparison . . . . . 2-252.6-1 Man-Machine Interface Grouping . . . . . 2-272.6^2 Program Cost Summary - Baseline ATL Traffic Model . . 2-302.6-3 Program Cost Summary - 2 Flights/Year Limit . . . 2-312.6-4 Program Cost Summary - 1 Flight/Year Limit . . . 2-322.6-5 Contingency Evaluation . . . . . . . 2 - 3 42.7-1 ATL Pallet-Only Payload-Required Hardwired Panel Space . 2-362.7-2 Cost Comparison of Command/Control Approaches for

Reference Pallet-Only A T L Payload . . . . 2-37

ix

SD 76-SA-0028-1


Space Division

1.0 INTRODUCTION AND SUMMARY

The tasks of the basic Spacelab User Implementation Assessment Study (SVIAS)definitized alternate integration and checkout approaches for ATL Spacelab pay-loads. One of the more significant factors in the processing cycle that wouldaffect the cost and efficiency of the activities was the software required by theon-board and ground operations. Not only could software become a pacing item inthe design, development, integration, and operations activities, it could alsobecome a significant cost factor. Thus, this effort, the Software RequirementsAnalysis Study (SEAS) was conducted to establish requirements, assess alternateimplementations, and identify programmatic costs of software and related hardwarefor the flight and ground operations associated with ATL Spacelab payloads. Inthis section, an'overview of the study, and a synopsis of the results and recom-mendations are presented.

1.1 STUDY OVERVIEW

STUDY OBJECTIVES

The primary objective of SRAS was to develop an integrated approach for the'development, implementation, and utilization of all software that is required toefficiently/cost-effectively support ATL flight and ground operations. It wasrecognized that certain aspects of the operations would be mandatory computer-ized services; computerization of other aspects would be optional. : Thus, theanalyses were to'encompass not only alternate computer utilization/implementa-tions but trade studies/evaluations of the programmatic affects of non-computerizedversus computerized approaches to the operations.

•A principal criterion in the development of the ATL software definition wasto maximize the autonomy of individual experiments and define each experimentsystem as self-contained as practical. The intent of this criterion was to max-imize the flexibility and PI control in both flight and ground operations byavoiding the synthesis of interdependent experiment-to-experiment, experiment-to-Spacelab, or experiment-to-Orbiter hardware/software systems. Independenceof experiment systems was a goal that would be limited only by factors externallyimposed. For example, constrained Spacelab-Orbiter interfaces impose the require-ment to integrate experiment housekeeping services such as telemetry, caution andwarning, and annotation data within the control and data management subsystem(CDMS) of the Spacelab.

A second criterion in the development of the ATL software definition was toderive an approach to flight and ground operations that will enhance/promote theusability/accessibility of the ATL Spacelab to a broad/diverse spectrum ofprincipal investigators. The selected approach must reflect direct access andinvolvement of the Pi's in an understandable format. Admittedly, a limiteddegree of standardization in the techniques for PI participation is required.But this standardization will provide clear-unambiguous definition of PI respons-ibilities and greatly assist in avoiding interaction and interdependency betweenPi's and other program elements.

1 - 1 . ' • ' •SD 76-SA-0028-1


Space Division

STUDY APPROACH

The approach used in SRAS is illustrated in Figure 1.1-1.

l.O

ON-BOARDPROCEDURESREQUIREMENTS

2.0

GRMD SUPPORTPROCEDURESREQUIREMENTS

3.0

ALTERNATIVEIMPLEMENTATIONS

k.OSELECTED ANDRECOMMENDED ATLIMPLEMENTATION

' ! / \r—'---,j COST REDUCTION II ALTERNATIVES I• STUDY j

PROGRAMCOSTS

CONTINGENCYANALYSIS

Figicpe 1.1-1. Study Approach

On-board procedures or operations were identified for 25 candidate ATLexperiments in Task 1.0. These operations were segregated into mandatoryon-board computerized services and command/control/monitor services that couldbe accomplished either by manual (hardwired), computer-aided, or automatedtechniques.

In Task 2.0, the mission analysis/planning and project management taskspertaining to ground operations were evaluated to establish required/desiredcomputer support.

Alternate implementation concepts for both the on-board services andground operations were synthesized and evaluated in Task 3.0. One task of asecond adjunct study to SUIAS, the Cost Reduction Alternatives Study (CEAS),was conducted in parallel with SRAS and significantly influenced the effortsof Task 3.0. In SRAS, the original intent was to assess the optimum use ofthe Spacelab CDMS. • The CRAS effort broadened the scope of evaluating alternateon-board accommodation of services to include the use of dedicated mini/microprocessors for mandatory computerized services of ATL experiments.

Programmatic evaluations were conducted in Task 4.0. Commonality of pro-cedures and reuse of software and software-related hardware were the primedrivers in the development of a recommended ATL implementation approach forflight and ground operations. An assessment of the potential impact of notimplementing the recommendations and/or the unavailability of certain NASAagency 'resources was conducted as part of a contingency analysis.

1-2SD 76-SA-0028-2


Space Division

1.2 ^ SYNOPSIS OF STUDY RESULTS . . ' '

ON-BOARD OPERATIONS ' . .

The results of the CRAS effort indicated that maximum utilization ofexperiment-dedicated mini/micro processors was the most cost-effective approachto accommodate mandatory computerized on-board services. Mini- and micro-,proce'ssors that could perform the, required services and consisted of commerci-ally available equipment were synthesized. In addition a software developmenttool, called the flight software support system (FSSS), was conceptually definedto maximize the. reuse of software and minimize the mission-unique softwarerequirements. " • , . ' < • • - •

The CRAS concept for accommodation of mandatory computerized on-board serv-ices was adopted in this study. Specific application of the concept to threereference ATL payloads indicated that a typical payload. would require 6 mini-processors, 14 micro-processors, and 2500 mission-unique statements, of software.

..Command, control, and monitor functions of the reference-ATL experimentswere evaluated to determine the impact of alternate implementation approaches.If.these functions were hardwired (dedicated control panels) the average costper experiment was about $5K. Accommodation of these functions by a computer-aided approach required actuation hardware, software, and an interactive term-inal (CRT/keyboard/micro-processor)., The actuation hardware costs were nominal(about $600 per experiment). Both non-recurring and recurring software wasrequired. The non-recurring software was a delta to the FSSS for preparationof measurements and operating procedures in the interactive terminal/dedicatedmini-rprocessor of the experiments and was estimated ,at $75K. Recurring softwareto code the measurements/procedures was estimated at $6.3K per experiment. Twointeractive terminals ($3K each), which could .support two flights per year andwould be utilized as common work stations in the Spacelab, were also identified.

Complete automation of command/control functions' of typical ATL experi-ments was discarded. The inclusion of decision algorithms in the softxâreprecluded the use of the FSSS software development approach. The estimatedcoslt for the automatic command/control approach for the typical AT,L experimentwas $155K. . . . . .. > ' .

Although the computer-aided approach was more costly than the hardwiredapproach, it was recommended for use on ATL experiments that did not requiremanual dexterity and/or visual acuity. The flexibility and versatility of thecomputer-aided approach will be more compatible with the anticipated reflightand modifications inherent in an evolving technology program such as Langley'sATL. Also, ATL experiments will be flown in pallet-only as well as habitable-module Spacelab configurations. The available panel space in the Orbiter. aft-fllght-deck precludes dedicated/individual experiment cpmmand/control/monitorpanels for the desired complement of experiments in each payload.

GROUND OPERATIONS . . . .

An assessment of the mission analysis/planning and documentation tasksassociated with ATL ground operations identified 29 essential and 5 highly

1-3

SD 76-SA-0028-1


Space Division

desirable computerized services. New development of the attendant softwareprograms would be in excess of $13M. Coordination with MSFC indicated thatan on-going software development program at MSFC would result in computer pro-grams that appear to fulfill Langley's requirements. The MSFC effort includestutorial software that will enable personnnel relatively unskilled in computerprogramming to utilize the programs. The primary emphasis at this time is touse remote terminals to exercise the computer programs. However, an initialattempt has been made to compile some of the software for use in a mini-processor.In, this study, SRAS, the compilation of the MSFC programs tailored to Langley'sneeds and augmented with a few additional programs was called the ground soft-ware support system (GSSS). As the MSFC effort is an on-going activity, thefinal status/results are not known. For purposes.of programmatic costing, aworst case estimate of $700K was assumed to convert the MSFC program to a GSSSsuitable to Langley's requirements.

There .are two basic alternate implementation approaches for implementationof the GSSS. One approach, batch processing, was eliminated because of theinherent time-delay, inconvenience, and documentation requirements associatedwith remote exercising of computer programs' based upon written requests/datasheets. The second basic approachînvolves interactive terminals that permita GSSS user to corrmmicate directly with software programs. By the means ofa CRT/keyboard a user queries the program which, in turn, in real time, providesresults and leads the user step-by-step in the exercising of the program.

Five interactive terminal approaches were evaluated. Two of the approachesutilized a remote terminal approach. The computer or data processing center iseither on site with the user, or remotely located. The primary shortcoming ofthese approaches were the host machine-time user costs. The other three inter-active terminal approaches utilized dedicated mini-processors. The differencesbetween these approaches are the mechanizations .of the memory systems; central-ized disc,'dedicated tape, or dedicated disc. The mini-processor dedicateddisc approach was preferred because of the flexibility and versatility of thismechanization even though it was slightly more costly than the other two dedi-cated processor approaches. Each mini-disc system will cost $37K and couldsupport two flights per year. """"

Other than the disc memory system the mini-disc concept could utilize thesame equipments as the on-board mini-processors. Although not specificallyaddressed in this study, the Pi's must individually conduct a significantmission planning/analysis effort. The Pi's dedicated on-board mini-processors,combined with a disc memory set and the GSSS, could be used for this effort.

PROGRAMMATICS

As the ATL program is comprised of a broad range of experiments the on-boardservice requirements vary significantly. Some experiments require primarily man-ual dexterity and/or visual acuity and are not adaptable to computer-aided com-mand and control. Thus, a representative ATL payload was synthesized in orderto derive programmatic costs. This payload consisted of six experiments thatincorporated computer-aided command and control and four experiments that hadhardwired command and control. Reuse of hardware and software was postulatedas follows.

1-4

SD 76-SA-0028-1


Space Division

ITEM

INTERACTIVE TERMINALSMINI -PROCESSORS

MICRO-PROCESSORACTIVATION HARDWAREHARDWIRED PANELS

DEDICATED PROCESSORSOFTWARE

CDMS SOFTWARE

REUSE

100%

4.0*

25%

0%

RATIONALE

"SHARED; EQUIPMENT COMPLEMENT WILLSUPPORT 2 FLIGHTS /YEAR.

EQUIPMENT INTEGRAL PART OF EXPER-IMENT;' REF1I GHT..OF EXPERIMENTSASSUMED TO BE 40% ,. ;

PROCEDURES WILL PROBABLY CHANGESIGNIFICANTLY ON EACH "REFLIGHT

NEW EXPERIMENT MIX— MISSIONTIMELINE EACH FLIGHT

Three- ATL traffic models were evaluated; baseline or Yapdley model,two flights per year maximum, and one flight per year. Application of theappropriate cost factqrs and reuse criteria indicated that the recurring soft-ware and software-related hardware costs averaged slightly more than $200K perflight. Minor variations between traffic models were due to different utiliza-tion/amortization of the software-related hardware items.

The recommended approaches for ATL software implementation are dependenton two elements—the FSSS and the GSSS. An assessment of the impact of nothaving these two software preparation tools was conducted. If the FSSS is notavailable the delta recurring software development costs that would accrue fromfour ATL flights would equal the development costs of the FSSS. If the GSSS isnot available the remote-terminal/remote-DPC concept could be utilized to linkLangley to MSFC. Implementation of the Langley GSSS is not time-critical asthe cost impact is of the order of $30K per flight after the first flight. Itis recommended that implementation of the Langley GSSS be postponed until theMSFC software program development activity is continued for at least anotheryear.

The reference ATL pallet-only payload was specifically evaluated. Theavailable panel space in the Orbiter aft-flight-deck (AFD) will not accommodatededicated command/control/monitor panels for the reference ATL payload. If thecomputer-aided command/control approach were adopted for seven of the 12 experi-ments in the reference payload, the AFD panel space is marginally adequate.Specific panel layouts and consideration of potential equipment interferencesmay eliminate the margin and dictate either restricting the experiments onpallet-only payloads to those adaptable to the computer-aided approach or inte-grating the panel requirements of multiple experiments. It is believed thatintegrated panel costs will greatly exceed the adaptation costs associated withthe computer-aided approach.

SUCCINCT SUMMARY

The primary results/conclusions/recommendations of the SRAS are as follows:

1. Maximize the use of dedicated mini/micro-processors.(from CRAS)

2. Develop the FSSS 'for both mandatory on-board services andcommand/control functions.

1-5

SD 76-SA-0028-1


Space Division

3. If appropriate (manual dexterity/visual acuity not required),'implement the computer-aided command/control approach.

4. Develop a GSSS tailored to Langley's application.

5. Implement the GSSS with a dedicated mini-processor/disc system.

6. Emphasize experiments adaptable to the computer-aidedcommand/control approach on pallet-only Spacelab, flights.

1-6

SD 76-SA-0028-1


Space Division

2.0 SIGNIFICANT STUDY RESULTS

A summary of the major trades and analyses conducted during the study, andthe primary conclusions'of these efforts are presented in this section.

2.1 ON-BOAKD PROCESSING REQUIREMENTS DATA BASE

In order to assess the potential use of computers and software to supportthe on-board experiment operations, a definitive set of data for each of the25 representative ATL experiments was developed. These data were derived fromTM X-2813, Study of Shuttle-Compatible Advanced Technology Laboratory (Langley,September 1973), the Shuttle Sortie Payload Description Study (MSFC), and dis-cipline specialists/surrogate Pi's on the Rockwell Space Division Staff. Therepresentative ATL experiments and their groupings into three typical ATL payloadsare indicated in. Table 2.1-1.

Table 2.1-1. ATL Experiment Definition and Payload Groupings

EXPERIMENT IDENTIFICATION

NAVIGATION

NV-.I MICROWAVE INTERFEROMETER

NV-2 AUTONOMOUS NAVIGATION

NVr3 MULTIPATH MEASUREMENTS

EARTH OBSERVATIONS

EO-1 UDAR MEASUREMENTS

EO-2 TUNABLE LASERS

EO-3 MULTISPECTRAL SCANNER

EO-4 RADIOMETER

EO-5 LASER RANGING

EO-6 MICROWAVE ALTIMETRY

EO-X SEARCH AND RESCUE AIDS

EO-8 IMAGING RADAR •

EO-9 , RF NOISE

PHYSICS .AND CHEMISTRY

PH-2 BARIUM CLOUD RELEASE

PH-3 AEROSOL PROPERTIES

PH-4 MOLECULAR BEAM LAB

PH-6 METEOR SPECTROSCOPY

MICROBIOLOGYMB-I COLONY GROWTH

MBr2 MICRO-ORGANISM TRANSFER

MB-3 BIOCELL ELECT FIELD OPACITY

MB-4 BIOCELL ELECT CHARACTERISTICS

MB-5 BIOCELL PROPERTIES

ENVIRONMENTAL EFFECTS

EN-1 MICRO-ORGANISM SAMPLING

EN- 3 NON-METALLIC MATERIALS DEGRAD

COMPONENTS AND SYSTEMS

CS-2 ZERO-G STEAM GENERATORCS-X CONTAMINATION MONITOR

EXPERIMENTSSPDA NO.

XST-OO.I

XST-004

XST-007

xsr-oioXST-OII

XST-012

XST-002

XST-003

XST-005

XST-006

XST-008

XST-009

XST-015

XST-016

XST-017

XST-019

XSr-020

XST-021

XST-022

XST-023

XS7-024

XST-027

. XST-02*•

XST-026XST-040

PAYLOAD 1MODULE + PALLET

X

X

X

X '

X

X

X .

X

X

X

X

PAYLOAD 2MODULE + PALLET

X

X

X

X

X

X

X

X

X

XX

PAYLOAD 3PALLET-ONLY

X

X

X

X

•

X

X

X

. X

X

X

X

X

2-1 SD 76-SA-0028-1

Rockwell InternationalSpace Division

Data packs were prepared for each of the ATL experiments and submitted toLangley in September 1975. The following documents were included in the exper-iment data packs: (1) a narrative description identifying the experiment'spurpose, sensors, and implementation approach; (2-) an equipment list of theexperiment assemblies and their potential Spacelab location; (3) condensedequipment performance and operations descriptors; (4) command and data signalflow paths; (5) a measurement list including form, rate, source, and destinationrequirements; (6) control requirements; and (7) an experiment data managementrequirements matrix that reflected the entire flight regime.

2.2 ON-BOARD COMPUTER SERVICES

The initial scope of'this effort was significantly expanded as a result ofa parallel adjunct SUIAS effort, the Cost Reduction Alternatives Study (CEAS).Part of the CRAS effort was to evaluate the use of dedicated mini/micro-processorsas well as the Spacelab CDMS for on-board experiment computer services.(Originally the SRAS effort was limited to establishing the preferred CDMS util-ization.-) The res.ult of the CRAS effort was to recommend the maximum utilizationof dedicated processors. As this result directly influenced the SRAS effort, theCRAS results are summarized herein and then applied to the representative ATLpayloads.

In CRAS, five representative Spacelab design reference missions were analyzedby creating experiment definition data packages similar to but not to the depth ofthe data packs discussed in Section 2.1. Each experiment of each payload was ana-lyzed to determine required/mandatory on-board computer services. Figure 2.2-1summarizes this analysis. Twenty-seven services were identified. Not only were

_LJ_

ALLLif t S C I E N C E

COMBINED AS RONOMYM A P S

(XFCIIMtMT

CH-701 I" l»»l«0"l»'tn-70* e»ti« TVT•r-0] UtlOKTtl

nnuni

ilS

3 J

'HYBRID CONCEPT "ALWAYS" REQ'D

'MAXIMIZE

•CENTRALIZED APPROACH

•DEDICATED APPROACH

•MULTI-USE SOFTWARE DEVELOPMENTCONCEPT COST-EFFECTIVE

•SYNTHESIZE STANDARDIZED SOFTWARE/HARDWARE APPROACH

Figure 2.2-1. On-Board Processor Flight Hardware Definition

2~2 SD 76-SA-0028-1


these services required by several experiments within a payload, but multiplepayloads required the same services. In addition, an evaluation of theseservices coupled with an assessment of the Spacelab CDMS and the Spacelab/Orbiterinterfaces indicated that some services must be implemented in dedicated proces-sors, some in the CDMS, and others could be implemented in either the CDMS ordedicated processors. , That is, a hybrid on-board computer mechanization approachwas always required. In order to establish a preferred approach, softwaredevelopment and implementation concepts were synthesized for a maximized central(CDMS) approach and a maximized dedicated processor approach.

, ; Because of the repetitive nature of the on-board services' the first step inthe synthesis process was to. define a software development technique that wouldmaximize the reuse of software. The technique was called the flight softwaresupport system (FSSS) and is illustrated in Figure 2.2-2. The 27 identifiedon-board services comprise the library of the FSSS. After initial development,only new initialization data (data tables) are required for subsequent applica- .tion. The remaining elements of the FSSS (tutorial, source code editor, compiler/assembler, etc.) are also software. These elements are -also a one-time develop-ment and provide the. capability to link library routines to a mission-unique

• flight applications program. Admittedly, some of these elements are machine(computer) unique. However, an industry trend is to develop software and hardwarethat is vendor-interchangeable. Only a minor level of standardization will berequired.

Even with the conceptually defined FSSS, mission-unique software is required.Estimates of this software for each of the five CRAS payloads were established toprovide the basis for software preparation costs.

NON-RECURRING

FLIGHT SOFTWARESUPPORT SYSTEM

LIBRARY]

MISSION-UNIQUE

ON-BOARDCOMPUTER SERVICES

MOTH*

OAT* ACQUISITIONDATA ANNOTATIONTIM/RCO FORMATTINGRfCMOER CONTROLOISTLAV FORMATTINGsci DATA CONTROLCfCMTIONS STATUSUNIT CHC CMDATA COMPACTIONINMI SIC ANAL

-

KWTIANst<m«*w

10050

100*50*50

ISOJ5*2550*50,

FSS SYSTEM

TUTORIAL PROGRAMSSOURCE CODE EDITORCOMPILER/ASSEMBLERCODE GENERATOR(S)LINK-EDITORDEBUG-TRACEOPERATING SYSTEMUSER'S MANUAL

EXPERIMENTAPPLICATIONSOFTWAREPROGRAM

APPLICATION PROGRAMS

COMBINEDASTRONOMY

AMPS

LIFE SCIENCES

'3500^ -ff—>[6,00d

MAPS

Figure 2.2-2. Software Development Approach

2-3SD 76-SA-0028-1


Space Division

The preparation of the mission-unique software for the two approaches isillustrated in Figure 2.2-3. In the mini/micro approach each experimenter,with minimal programmer support, develops his own flight applications softwarefor experiment operations. Measurements data must be integrated at a centralactivity because of the constrained Spacelab/Orbiter interface. The productsof this approach "are a series of application programs (shown as tapes) for eachexperimenter's mini- and micro-processors plus one program for the CDMS.

MINI /MICRO APPROACH CENTRALIZED APPROACH

r- -*-•» r--1—iMEAS. _ OPER.

~txpivvr

k\

i|t fc.._

--A-1

•'-JUSER

•

--T r —i ii i

-.4 U. -

B

---JFSSS

r ,

STIL (FSSS)

nii

-_iN

-' |

1

CDMS FLIGHT TAPE

LEGEND; MEAS.OPER.A.B.NSTILCOM3

H

MEASUREMENT REQMTS (TELEMETRY, CAUTION/WARNING, ANNOTATION)EXPERIMENT OPERATIONS IMPLEMENTED BY COMPUTERS

INDIVIDUAL EXPERIMENTSLEVEL III INTEGRATION FACILITYSPACELAB EXPERIMENT COMPUTERSOFTWARE FOR SINGLE-FUNCTION MICROCOMPUTERS

FLIGHT TAPES - SOFTWARE FOR MINICOMPUTERS OR CDMSSTATEMENTS - FORTRAN OR OTHER HIGH-ORDER LANGUAGE

Figure 2.2-3. Experiment Flight Software Development

In the centralized approach only micro-processor software is developedwith the user FSSS by the experimenter. All other experiment operations mustbe rigidly specified and controlled as the flight applications software will bedeveloped at a centralized software test and integration laboratory (STIL) forincorporation into the CDMS. The significance of the documentation/control ofthe operations is reflected in the cos't estimates for software preparation.Based upon previous/similar software development activities at Rockwell the per-statement software preparation' cost is $62. The ASSESS program at Ames ResearchCenter is more characteristic of the mini/micro approach. Estimates of

2-4

SD 76-SA-0028-1


Space Division

the software development costs of the ASSESS program plus record-type documenta-tion (which is adequate for the experimenters) and contractor overhead indicatedper-statement costs of about $31.

An additional major consideration in' the evaluation of the two implementationapproaches was the hardware required. The CDMS is furnished with the Spacelab andthus no unique costs for on-board equipment were applicable. However, the soft-ware prepared in the centralized approach is remote from the experiment hardware.Thus, experiment simulation (hardware or software) is required to validate theflight software at the software development site, and CDMS interface simulationhardware is required to verify the compatibility of the experiment system withthe Spacelab CDMS at the Pi's site. Postponing the validation of the software/hardware interface until Level III integration was considered to be unacceptable.

Introduction of the mini/micro approach requires additional flight andground hardware. Processor models-consisting of commercially available equipmentwere synthesized (Figure 2.2-4). Appropriate peripherals were included so thatthe preparation of the flight software could be accomplished with the flightprocessors. As the PI is directly involved in the development of the softwareand uses the flight processor, neither simulation software nor hardware isrequired. The validation process with the mini/micro approach is illustratedin Figure 2.2-5.

Cost factors were derived for each element of both on-board mechanizationapproaches and applied to each of the CRAS representative payloads. Programmaticcosts were developed for each approach as illustrated in Figure 2.2-6. Equiva-lencies between the CRAS representative payloads and the other payloads in thetraffic model were established. Schedules of each mission type for the threedifferent traffic models were developed. Learning curves/software reuse estimateswere defined and costs by mission type were generated. Summations of the program-matic costs indicated that the mini/micro approach;would be about 60 percent ofthe costs of the centralized approach. Also, a gross evaluation of the impact :ofnot having the FSSS indicated recurring costs would increase by about a factorof 4^ • ': .

The maximized mlhi/micro-processor approach was incorporated in this study.Each'Of the. reference ATL experiments/payloads were analyzed to establish thecomplement of mini- and micro-processors and flight application program state-ments required for the on-board services as illustrated in Figure 2.2-7. Therequirements of the three'reference ATL payloads are summarized in Table 2.2-1.In order to subsequently develop ATL programmatic costs a nominal ATL payloadrequiring 6 mini-processors, 14 micro-processors, and 2500 mission-unique flightapplication program statements was identified. The CDMS softidare integrationactivity required for each ATL payload is summarized in Figure 2.2-8. New com- .puter software is not required for each mission. The CDMS programs for Spacelabsubsystem measurement functions can be utilized with the'experiment-unique initi-alization data.

2-5SD 76-SA-0028-1


EXPERIMENT (OR SA) DATA IU5»(SPACEUI)

PDI (OtIITER)

f

RECORDER ~lNO. i r*"<

(DUAL) '"RECORDER""$3400 NO. j *^

V»

PROGRAM

$1835 SIO"*<»E •—

DEDICATEDEXPERIMENTCOMPUTER f

(I6IC.I61IT) (DISCRETES)*!

$17.265

A! $56CINPUT/OUTP

SECTION

i

!

1C _

. ML_r

L.$3000

(DEDICATED) OF'IADISPLAY/CONTROL .

SYSTEM COST: $31,100

MINI-COMPUTER FLIGHT SET

f" b.MTlM»

IT — 1

BB

0,-Oc <^? C^ CBBB BE

. -. "• r . i "*"

•

iiTimr

OtV IEADMwotrwtin• mtiM • » — [>pttA||

ttADY tw~*uu« ^ H |/o»«n«w „ .1r IYNC ™ •<

'a »•

«m

nnr«

iV i

- •

SYSTEM COST $5,6*0

RECORDERS 5.235

TOTAL $10,875

MICRO-COMPUTER FLIGHT SET

s

INTERFCENTt

rfMA-l

T VMtOfcl ....

1

1

r~ ~

-^P. . •"*

1

• <L_L

-1 1 ^ mi1 1 ^ oumn1 ' - Ml

1 ^

-1 1 S MI7

• — ,, oumnC ""

1 ^J 1

I "*cr* ' i i/o1 - *"l_

CAv j

1 » ) EXPERIMENTDAC | rl CONTROL

(4 CHANNELS)"'

ZiT~te= EXPERIMENTMUX R OATA

P2 CHANNELS)*

OF CHANNELS CONSIDERED PARTSIC COMPUTER

OURCE FOR TIME AND STATE VECTOR.ACE TO CDMS EXPERIMENT COMPUTER.M. D< THROUGH CDMS COMPUT9.

US INTERFACE ADAPTER

bt->Tllw!

ni'L «» i...... >tOM 1

u 1' - 1

1 •«* •«• -1 ',. <l^ . i Tlot-it IMQO 1 1

1 || 1 ' -0,-ûxc. '• ((CUSTOM)1 1107 1_^ 1AM 1 /^ ,

• |f} \ ^ lAoni' l«'! ! | J HOIO

'~ ~ 1 AMP

J 1

', t1-..-.IJ"". 11 N |U.IB "— *l| »' AOC

: '¥ II i — i i

s^ lH^H> — ~% — r i£ • M . „., »• j

L » J""^j (OPTIONAL)

J

<~ ii7j k—4-x>-il i »w<cmro«41 - U3I

^ S "^

1 .J-i5~i . J> flOC

flN — i — Ti ^

• • -J

in nssmnjj; ^1* HK.« (OPTIOIUl)•< yj * '"-o«<« otnv

J

Figure 2.2-4., On-Board Processor Flight Hardware

2-6

SD 76-SA-0028-1


Space Division

PREPARATION

SEQUENCE

Of

OfOATIONS

TUTORIAL

DEDICATEDPROCESSOR

MANUALD/C .

EXPMTGROUP

IL_ _ —. — — — REWORK— —

-VALIDATION-

MEAS. LIST

TIM UST

QV»

LEVEL III

MANAGER

^ PI

J DEVELOP j ^TIL

— «1 SPACELAB

I INTERFACE

LEVEL IIIMTEGKATION

Figure 2.2r5. Mini/Micro Processor Software Preparation/Validation

MISSION EQUIVALENTSCM •nuuiinvt •

•in-wtuuncM

IOM M'llCSM0HM>Cr MMt'MIWVWT

• find MAMÂCtUfllNC

SCHEDULE OF MISSION TYPES

**"SM«M

• •• •»•»•*

J>ST- S

•

•

t—

;• •

ij .• i •

:l• ! :

7~t.~

1

"."

'

k

,

•> ?zs

COSTS OF EA. MISS ION TYPE

PREFERRED APPROACH

• MINI/MICRO PROCESSORS• *6Q% OF ANNUAL/CUMULATIVE

COSTS OF CENTRAL APPROACH

• FSSS DEVELOPMENT• MINIMIZES RECURRING COSTS

Figure 2.2-6.- Programmatic Cost Evaluation

2-7

SD 76-SA-0028-1

Rockw

ell InternationalS

pace Division

S1N

3W31V

1S

ÔN-BOARD COMPUTER SERVICES ^

SaO

SS

aDO

o-d -OD

DIW

^V*-

S»O

SS

3DO

cJd-INIW

NO

IlDV

dW

CO

V

IVO

SIS

A1V

NV

X

lJilVW

NO

IlDia

SD

d

cJ3aaO-O

H3Z

SISAIVNV a3i»nojS

WD

OdS

NV

Dl

'OclO

OD

NO

ISS

3tJdWO

D V

IVO

3D

N3

0O

3S

0

3in

a3

HD

S

ON

IlNIO

d

JJOS

N35

NO

IiVc)3

N3

O

AV

IdSIO

NO

IlV13c)clO

DO

inv

ON

ISS

3DO

Hd

SD

IHdV

cJO

SX

D3H

D 1

IWI1

ON

IllVW

DO

d A

Vld

SIO

ON

IiiVW

MO

d O

NIQ

MO

D3H

ON

IiiVW

a'O

d

Acli3

W3

T3

i

•VIO

NN

V viva

ivnoia

•sinoDv viva

ivnoia

noaiNO

D viva

BDNSIDSS

ISA

1VN

V

ON

3M1

SISAIVNV >o

oi-xD

ino

dH

XD

Vg

M/D

"IOD

1N

O3

W

bO

dlV

Id

NO

IiV«3N

39

Q

NV

VN

VV

OD

9NIS

S3D

OH

d 3

9V

WI

SniV

iS S

NO

IlV>

J3dO

9NIS

S3D

OS

d-3)Jd

lOU

NO

D «Q

c)OD

3»

o0o>-Q.

00C

N

IXCN

SinINTf

(NsCN

(NCN

8mCN3a

(N

^

0ro;

^2|jS

-1

2m2—o000

•X-om-w0CN-

1ccLUa.

XLU

uuu2u2•x.

0o:

LU

U_

ccJj

LUSooc.o—

41

oJQ-a22uuuu2

1 NV-2 AUTONOMOUS NAVIGATION §22uu

omo-2=3.

22

CN

-a22

U

o

uu,u

22

1 EO-1 LIDAR MEASUREMENTS 22

[ EO-4 RADIOMETER

uu2

oinCN

-a22uuu2

U! O

2

1 EO-7 SEARCH /RESCUE AIDS

2Q:

ocdO052

8CMaauuuLU

toLU1

LU

QooO.

8CN

—2a2a.2ou01 PH-4 NEUTRAL GAS PARAMETERS

omroaauuua1 PH-6 METEOR SPECTROSCOPY1 EN-1 MICRO-ORGANISM SAMPLING 8C

Nauu00a1 EN-3 NON-METALLIC MATERIALS uuu

I CS-X CONTAMINATION MONITOR

o<noo

O

LEGEND 58 o

2-8

SD

76-SA

-0028-1


Space Division

Table 2.2-1. ATL On-Board ProcessingRequirements

ATL PAYLOAD 1

ATL PAYLOAD 2

ATL PAYLOAD 3

NOMINAL ATLPAYLOAD

MINI

PROCESSORS

. 5

2

7

6

MICRO

PROCESSORS

14

12

15

. 14

SOFTWARE -STATEMENTS

3300

1000

2950

2500

COST FACTORS(CRAS DATA)

«33KEACH

$11K

EACH

WIPERSTATEMENT

PI/USER CDMSSOFTWARE

<i TO 12EXPMTS

DATA TABLES300K BYTES/FLIGHT@ $0.01/BYTE *

MACHINE TIME3 HR/FLIGHT@ $375/HR *

INTEGRATION MANPOWER3 MAN-MONTHS/FLIGHT@ $50K/MAN-YEAR*

INTEGRATION COSTPER FLIGHT

S16.6K

* Rock well-experienced rates.

Figure 2.2-8. STIL Function with Mini/Micro Processor Approach

2-9

SD' 76-SA-0028-1


Space Division

2.3 ON-BOARD COMMAND AND CONTROL

In addition to the mandatory computerized on-board services, alternateapproaches to perform the nominally manual command/control functions of exper-iment operations were evaluated. A basic manual (hardwired) approach was syn-thesized for each experiment. Computer-aided and automated mechanizations forthese same hardwired functions were also considered. The three approaches areillustrated in Figure 2.3-1. The hardwired approach reflects a typical laboratoryconfiguration with discrete switches, potentiometers, indicators, lights, and

MANUAL (HARDWIREAUTOMAT IC / COMPUTER-A IDED

CONTROL

SAftTY-RELATED I j_CONTROLS AMD j J"

DISPLAYS

CONTROLS

DfDICATtOOPERATING

DISPLAYS

DATA DISPLAYS

CORRELATION DATA I 1DISPLAYS I T

CORRELATIONDATA

ORIITERCOMMUNICATION «-iLINKS L

PAYLOAD

DEVICES

SAFETY-RELATEDCONTROLS AND

DISPLAYS

CONTROLS

DISPLAYS

CORRELATIONDATA

Figure 2.3-1. Data Management Design Concept Comparison

meters. In the computer-aided approach the majority of the discrete componentsare replaced by an intelligent terminal (CRT/keyboard). Some functions must behardwired for safety reasons (e.g., pyros, emergency backups, etc.). Instead ofdirect wiring from discrete components to remotely located experiment equipment,command/control/monitor is accomplished via an experiment command bus (independ-ent of the Spacelab experiment data bus) with the computer-aided approach.Actions are initiated via the keyboard; reactions/status are monitored on theCRT. The hardware configuration with the automatic approach is the same as thecomputer-aided approach; the difference is in the software. Decision-algorithmsand sequencing are done with software. The operator monitors the activities and(if necessary) overrides the automated commands/sequences. For each command/control implementation approach, the pertinent hardware and software cost ele-ments were defined and estimated.

2-10 .

SD 76-SA-0028-1


Space Division

HARDWIRED, APPROACH

In the case of the hardwired approach the design, development, and integra-tion of the control panel and the cost of the components were considered. Basedupon the definition data in the experiment data packs the discrete panel compon-ents for each experiment were identified (Figure 2.3-2). Note that only discrete

Experiment

L NV-1

L NV-2

J. NV-3

4. £0-1

5. 10-2

.6. KH

7. EO-4

I EO-5 J

Toggle Switches

ON /OFF

?•

9»(1">

Wt IA"\

6 d")

13 0")

8W)

4

^^^LJ4**I^

Positions2 3

2

3

5

L -.

TalkbacksPositions1 2

7 2

11

6

5

10 1

4 3

t

7

Rotary Switches• No. ofSwitches

1

3

1

1

5

U^

to. ofPositions

$

2B-pos,)

•

30-pos.)

3

75

9/2 gang

714235

5

— L^

Digit IndicatorsNo. of

Indicators

1

111

12

9

32 '

9

3

2

6-,

No. ofDigits

3

26

33

2

22

3

3

5

3.'

No. of Lights

Bit

9

11

9

9

7

1

jfj

Dis-crete

•

'— *•».

Miscellaneous

Slg. strength meter; receivertuner (2 knobs)

One analog potentiometer

Sig. strength meter; two poten-tiometers (align, adjust receiver

Spots

L- — — — JFigure 2.3-2. ATL Experiment Panel Components

components are indicated; stand-alone displays such as oscilloscopes, TV monitorsand spectrum analyzers will require additional panel space and are applicableregardless of the implementation approach. Based on empirically derived formulas(previous Rockwell programs), panel area and weight requirements were developed.Cost estimating ratios for aerospace-qualified ($1800/lb) and Mil-STD programs($170/lb) were applied. These data are reflected in Figure 2.3-3. As Spacelab

PAY LOAD

\

2

^—^

EXPERIMENT

NV-3EO-2EO-5EO-9PH-2PH-3PH-1»EN-1CS-XMB- 1/3 .

TOTAL

EO-3EO-6PH-3

•Jtt'S

UF 1 PUTwe i un i(LB)

. 30.7.1*0.6*3.526.071.823.lt •2<i.916.121.9

*

5«i.9W.9

_^ "™*™^ ^ * ^

COSTS (1976 $K)

AEROSPACE

55.373.078.346.8

129.242.1U.B29.039.4

*537.9

98.880.8

ks.42.L- .^ -^

MIL-STD

5.26.87.34.4

12.13.94.22.73.7

*50.3

9.27.5

J^— 1 •** >)

Figure 2.S-3. Comparison of Aerospace and Mil-STD Costs

'2-11SD 76-SA-0028-1


Space Division

payload controls are not crew safety provisions (They must be operationallysafe but do not provide crew survival/safe return capability.), the use ofMil-STD design criteria was considered applicable. The average cost of thedesign, components, development, test, and documentation for the ATL experimentswas about $5K each.

COMPUTER-AIDED APPROACH

In order to implement the computer-aided approach for experimentcommand/control/monitor functions, additional hardware is required at both endsof the transmission link. Figure 2.3-4 illustrates a typical computer-aided

oti DATAfXPBtlMENT DATA ACQUST1ON ILS

MICH DATAIATE

SWITCHING MATMX

t IDISPLAY

TOMINAI.NO. 1

otsnAvTBUMNAL

NO. 2

Figure 2.3-4. Level III/II Integration Configuration (Typical)

implementation for an ATL Spacelab payload. The experiment operator is providedintelligent display terminals (CRT/keyboard/micro-processor) to command/control/monitor experiment actions/reactions via the experiment command bus to dedicatedmini/micro-processors in each ATL experiment system. Overall control of experi-ment operations and the flow of data to/from the Orbiter is maintained by theSpacelab CDMS via the experiment data acquisition bus-RAU link to the experimentsand the CDMS MUX.

The interface hardware at the experiment end of the transmission link isillustrated in Figure 2.3-5. Actuators and decoder elements are required torecognize, interpret, and effect the translation of a digital signal to thedesired control action. The average cost of this activation hardware is about$600 per experiment. The mini/micro-processors can be the same hardware describedpreviously for the'mandatory on-board computerized services. The intelligent

2-12

SD 76-SA-0028-1


Space Division

MINI/MICROPROCESSOR

DISCRETE LINES

SERIAL LINES

PARALLELLINES

— —

DEHUX

H

DRIVER

mr.iTAL/lANALOG 1

1 DECODER— j

L: RE LAY

OP.AHP.

DRIVE

(POWER CONTROL)

(PROPORTIONALCONTROL)

1 REED

Figure 2.3-5. Computer-Aided Hardware Coat Factors(Example: Microwave Radiometer - EO-4)

terminal used during development of the experiment system can also be the sameunit used for on-board services software development. However, two intelligentterminals for flight use were allocated to Langley ($3K each).

Three alternatives were evaluated for the software associated with thecomputer-aided command/control approach. The first alternative used the intelli-gent terminal as a command generation and measurement monitor mechanism. Thecommand generation consisted of replacing the manual switch/potentiometer opera-tion with a digitally encoded signal in response to keyboard inputs. Requiredoperations would be listed in an experiment operations handbook. Software forthe mini/micro-processor would be required to decode the command signal (keyboard),energize the actuators, and acquire/process the measurement data for display onthe intelligent terminal. A typical measurement display page is shown inFigure 2.3-6. Each measurement display page is customized to a particularexperiment by adding the experiment-unique data tables. The basic software canbe developed by using the FSSS, provided certain action-analysis routines areadded. The delta non-recurring cost to the FSSS is estimated to be $62K (1000statements @ $62 each). The mission-unique data tables were estimated to averageabout 800 characters per experiment which at $0.01/character is only $8 perexperiment per mission.

MICROWAVE INTERFEROMETER

NO.

1

2

3

<t

1718

k.

NAME

PALLET POWER

CHANNEL A

CHANNEL B

CHANNEL C;

TAPE REMAINING

TRANS. OUTPUT

VALUE

115-135-122-150:

1200

20

UPPER

120

-100

-100

-100•

|

-100

D H

| .2 | 20

LOWER

110

-160-160

-160.

-10

M 5

|.3|05|

UNITS

VAC

dBm

dBm

dBm;

FEET

mW

J

Figure 2. 3-6. Typical Measurement Display

The second computer—aided software approach evaluated included a display ofoperator's procedures on the CRT and automatically displayed the resulting status

2-13

SD 76-SA-0028-1


Space Division

—an automated checkoff list. Figure 2.3-7 illustrates the procedures displayon the CRT. This capability would require another FSSS delta, estimated to be200 statements ($12.4K). The function of this delta would be to analyze thecontrol action requested and set up the communication from one intelligent term-inal to one of the several mini- or micro-processors in the ATL payload. In

M 1

i2

3*

56

7aU

9

101

2

3«.

5

' R O W A V E I N T F R F F R O M F T F * i n t i rv» " v w r^ » t. i n i t n r k n w r j t i c R J U n t

P R O C E D U R E

R E C E I V E A T P X . 5

TV C H A M 1 0 N

B O O M C A M E R A 0 N

T E S T L I G H T 0 N

A D J U S T F O C U S

V E R I F Y L I G H T V I S I B L E

T E S T L I G H T O F F '

E L E C T R O N I C S 0 N

B 1 T 0 F.F

T R A N S M I T T E R 0 N

R E C E I V E R C H A N S E L E C T

S I G N A L O N A L L C H A N N E L S

T R A N S M 1 T.T E R OFF

R E Q U E S T A T P X . 6

/COMPUTER)\MESSAGE /

( R E M A R K S )

L L

^*/***•s

1 N

T I M E

23: 1 3 : 0 6 N•

F 1

A 1 3

A 1 *

A 1 5

• OPERATOR ENTRY CDMS ENTRY I - GMT M - MET

Figure 2.2-7. Computer Display (Aided)\ '

addition to the data tables required for each experiment for each mission(700 alphanumeric characters for each of 16 orbital operation verificationphases at $0.01/character—$112), there is also some missions-unique computersoftware that must be developed. In the conventional remote terminal configur-ation, multiple intelligent terminals communicate with a central processor thatincludes the software to identify each terminal. In the proposed Spacelab con-figuration this situation is reversed—one remote terminal must communicate withseveral processors. Additional software is required in each dedicated processorto recognize its call sign and respond (integration software, commonly calledhandshaking). It was estimated that this function would require 200 statementsfor eachexperiment for each mission and cost $6.2K (200 statements at $31 each).

The third alternative evaluated was to utilize the Spacelab CDMS intelligentterminal for the computer-aided functions. Even if the simulation hardware andsoftware required for Level IV validation is neglected, the costs would be about$25K per experiment per mission. Because the software is being developed

2-14

SD 76-SA-0028-1


Space Division

remotely and will be Integrated into a single machine, the development costswill be $12.4 for the program (200 statements @ $62 each), $12.5K for the inte-gration (3 man-months of effort at $50K/year), and $750 of host machine time(2 hours @ $375/hr).

The second- alternative, which included the procedures display, was therecommended approach. The versatility, flexibility, potential reduction inoperator mistakes, and capability for automatic recording of all in-flightoperations appeared to warrant the additional $6.3K per experiment (as comparedto the alternative that only mechanized the control actions).

AUTOMATED COMMAND/CONTROL

All of the hardware and software of the computer-aided approach is requiredas part of the automated approach. In addition a software program to make thedecision to proceed to the next step is required with the automated approach.In order to develop the decision algorithm software, a logic flow diagram suchas that shown in Figure 2.3-8 must be developed for each experiment operation.The FSSS cannot be used to create this logic nor the software required torepresent the logic. There is no known tutorial method to do this task. Basedupon Rockwell experience in developing automatic'spacecraft subsystems thatinclude decision logic, it is estimated that about 5000 FORTRAN-type statementswould be required to automate a typical navigation or earth observation experi-ment of the ATL. Even at $31 per statement the per-mission costs would be$155K per experiment.

EXP. NV-1- MICROWAVEINTERFEROMETER

BLQCK X.5 VERIFY OPERABILITY

IIW

1

t

1

4

..

•

,

i*-v

IMMCATION

M1BCOM

TVMOMTOI

TVMOMrrai

UONT

»_»

4W4UM

TVMOMTCB

-*— > --«

coNitoi •

INTMCOM

TVMONirOiCHAN. Ul.

iNiacoM

UBMTIWICM

•

IMNTSWIKM

INTVCOM

-^^

r

1

((

(

O1I4IIN

IKUVIAtfXJ

Js

ANIINNAovucnoNINSIIUM(NT

alVMONON

CHAN. IT

If —OKXXTV

CAMIUON7

umr

1B ANIINN*

lAicnvaiMji

1

UOMT

1MOUOT L

s~

CJIQUNCI

NO iiijcrtvCHAN. 1

taUOUiST

MO lOOMCAM.~* ON, AND

tOl

MQ ACTlVAtlUOMW

KO^ - *~*~s^~s

TIME

Figure 2.3-8. Automated Control'Logic

2-15

SD 76-SA-0028-1


COMMAND/CONTROL IMPLEMENTATION COMPARISON

A comparison of the recurring costs associated with the three command/con-trol implementation approaches is presented in Figure.2.3-9. The software costsassociated with the automated approach preclude this alternative unless criticalperformance or safety is an overriding factor (e.g., emergency shutdown of ahigh-powered laser). Comparison of the average hardwired and computer-aidedapproaches indicates about a $2K difference per experiment. However, in lightof the increased flexibility, versatility, simplification of crew training.,simplicity of in-flight operations, adaptability to design change, and compat-ibility with the pallet-only Spacelab configuration (reduced panel space), thecomputer-aided approach is preferred where applicable. This recommendation isbased upon the assumption that a tutorial system such as the FSSS is developedby the agency and would not be uniquely assessed to the ATL.

HARDWIRED COMPUTER-AIDED AUTOMATED

COSTFACTORS

PANEL COSTS S5.0< HARDWARE COSTS S0.6K

SOFTWARE COSTS S6.3K

HARDWARE COSTS S0.6K

SOFtWARE COSTS S155K

NONCOST

FACTORS

• INDIVIDUALIZED PANELFAMILIARIZATION

• MANUAL CREW CHECKLIST/RECORDING

« IN-LINE MODIFICATIONS MAYREQUIRE NEW PANEL

« CONSTRAINED IN PALLET-ONLYMODE

• COMMON CONTROLSTATION

• AUTOMATED CHECKLIST/RECORDING

• IN-LINE MODIFICATIONSMAY BE ACCOMMODATEDIN DATA TABLES

*-COMFAT<«L€ WITH fALLCT-ONLY MODE

• SAME

• SAME

MODIFie*TIONSCOUiD REQUIRESQiJtWARE CHANGES:-,,:**: m

VIABLE ALTERNATIVES NOT RECOMMENDED

Figure 2.3-9. Command/Control Concept Comparison (Per Experiment)

It is recognized that the computer-aided approach is not applicable for allATL experiments. Some ATL experiments require minimal control such as turn-on/turn-off action, or operator physical dexterity/visual acuity such as the micro-biology experiments. Therefore, both the hardwired and computer-aided command/control approaches are viable alternatives. The approach selected for eachexperiment should be based upon that experiment's specific control requirements.In general, if manual dexterity and/or visual acuity is not required, thecomputer-aided approach is preferred.

2-16

SD 76-SA-0028-1


2.4 GROUND PROCESSING REQUIREMENTS DATA BASE

The ground processing activities that were assessed included the analyses,planning, engineering, test operations, and management of the support needed foran ATL flight. The two sources for the identification of the required activitieswere the task WBS and the documentation lists developed in the basic SUIAS. EachWBS entry and document was evaluated to determine the potential computationalassistance required. A requirement was identified if the calculations were com-plex, the volume of data was large, and/or the process was iterative/repetitive.

The assessment indicated that 29 computer-supported processes were essentialfor efficient ATL ground operations (see Table 2.4-1). In addition, five of the35 types of documents were also identified as requiring computer support (seeFigure 2.4-1). The 29 essential processes were evaluated by the Rockwell program-mer staff to establish what application and library routines would be required,and the size of these routines. The results of this analysis are summarized inFigure 2.4-2. The composite total of 218,000 FORTRAN statements at $62 perstatement precluded consideration of a new or Langley-unique development ($13.5M).

A descriptive data sheet was prepared for each of the 29 essential processesand sent to JSC and MSFC.for evaluation. Although programs of a similar naturehave been developed at these centers (as well as at other NASA centers and aero-space contractors), MSFC programs indicated a high degree of correlation withthe required programs for ATL payload ground processing requirements (see Fig-ure 2.4-3.). A continuing program at MSFC is being conducted to develop thistype of software. Perhaps most importantly the MSFC programs include thedevelopment of tutorial software for each process. In addition, the programsare operational on their S/1108, are being recompiled for an S/360, and some areeven being converted for use on a mini-processor (a PDF 11/45).

Table 2.4-1. Essential Computer-Supported Activities

EXPERIMENT GROUPINGSSYST/PROGRAM COST ANALYSISGROUND TRACE GENERATORTARGET OPPORTUNITY GENERATORCOMMUNICATIONS COVERAGESOLAR/MISSION GEOMETRYRADIATION ENVIRONMENTORBIT CONTAMINATIONATMOSPHERE MODELORBIT DECAYORBIT MANEUVERSORBIT ERROR ANALYSIS 'SUBSATELLITE MOTION ANALYSISMISSION CONSUMABLESORBITER ATTITUDE MANAGEMENT

INSTRUMENT POINTINGMISSION TIMELINE GENERATIONEXPERIMENT DATA PROCESSINGORBIT EPHEMERIS/TIMELINE UPDATECONTINGENCY OPERATIONS PLANNINGSUBSYSTEMS PERFORMANCE MONI TORGROUND TRUTH COORDINATION/CONTROLPAYLOAD (EXPERIMENTS) STATUS MONITORTHERMAL ANALYSISMASS'PROPERTIES ANALYSISLOADS/STRESS ANALYSISELECTRICAL POWER ANALYSISDATA MANAGEMENT ANALYSISSTABILIZATION CONTROL ANALYSIS

2-17

SD 76-SA-0028-1


Space Division

'Only two essential programs (stabilization and control analysis, and sys-tem/program cost analysis) were not included in the MSFC list. Langley (FlightDynamics and Control Division) has initiated the development of a stabilizationand control analysis program with tutorial software. Rockwell developed a sys-tem/program cost analysis model as part of the Radiometer and basic SUIAS effort.Conversion of the operations manual to tutorial software would be a relativelyminor task. Specific correlation between the MSFC programs and those programsrecommended for documentation tasks was not achieved. However, MSFC's IntegratedMission Program (Figure 2.4-3) should suffice for the mission plan document; thepayload activity, scheduling program (or Langley's MASS program) can be used forresource scheduling documentation; and commercially available programs can beused for the remainder of the documentation tasks.

Adaptation of the MSFC and commercial programs to run at Langley involvesconverting, translating, or recompiling the source program (a FORTRAN listing)to fit whatever machine Langley would use. In the optimum case, the cost wouldbe for .only the machine time to compile the program. In the worst case, anS/360 to S/unknown translator would be required, which is estimated to cost $700K.

1. MASTER PROGRAM PLAN AND SCHEDULE2. CONFIGPRATION MANAGEMENT REPORT (MONTHLY3. LOGISTICS PLAN4. INVENTORY REPORT5. EXPERIMENT REQUIREMENTS6. RESOURCE ALLOCATION PLANS7. MISSION FLIGHT PLAN8. EXPERIMENT OPERATING INSTRUCTIONS9. GROUND SUPPORT PLAN10. MISSION'TURNAROUND AND REFURBISHMENT PLAN11. DATA REDUCTION REPORT12. TRAINING PLAN AND PROCEDURES13. INSTRUMENTATION LIST14. EXPERIMENT RESOURCE REQUIREMENTS15. EMC (TEST REQUIREMENTS) PLAN16. SPACELAB USER'S GUIDE17. EXPERIMENTER'S DESIGN MANUAL18. TEST REQUIREMENTS19. GSE AND FACILITIES PLAN20. SYSTEM REQUIREMENTS MANUAL

(X)(X)

(X)

(X)'

(X)

EQUIPMENT SPECIFICATIONSEQUIPMENT OPERATING INSINSTALLATION LAYOU^

35. FAILURE SUMMARY REPORTSA. EXPERIMENT INSTALLATION 5 CHECKOUT

' B. SPACELAB INTEGRATIONC. CARGO INTEGRATION

PERT MODELPERT MODELPERT MODELINVENTORY RECORDSEXPERIMENT-UNIQUEOPTIMIZATION PROCESSOPTIMIZED RESULTUSER DEVELOPSPERT MODELUNIQUE PER FLIGHTMISSION-UNIQUEMISSION-UNIQUEMISSION-UNIQUESUPPORT TIMELINE ANALYSISSTANDARDONE-TIMEONE-TIME,

-UNIQUEMISSION-UNIQUEMISSION-UNIQUE

EXPERIMENT-UNIQUEMISSION-UNIQUEMISSION-UNIQUE

ENGINEERING RECORDSMISSION-UNIQUEMISSION-UNIQUE

LEGEND:

H MANUAL

A AUTOMATED (BATCH)

I AUTOMATED, INTERACTING

(X) INITIAL FIVE PROGRAMS

Figure 2.4-1. Optional Processing Documentation

2-18SD 76-SA-0028-1

Rockw

ell In

tern

atio

nal

Space D

ivisio

n

. •

VI

NO

I1INI430

VIV

O

SJlU

OU

d N

OIS

SIW

NO

I1INI330 1

S3llH

3dOild

31IT131VS

ans

S30rU

INO

VW

i joanos uoaiuiNoiivanoidNoo t ssvwi

•DV

aVH

D8V

133VdS

/ 8311880

NO

IilSO

d 31IT131V

SN

OIlV

OIN

nWIV

OD

NO

IilSO

dT

VIlS

3T33nvi81S

38831

S3in03H

3S/S

NV

ldlN

3Wd013A

3Q lN

3WI83dX

3

All H

OI 8d

lN3W

IS3dX

3

SN

OIlV

lGdO

•»SNOiivanouNOD iwdxa

SN

OI1IN

IJ30!N

3WIJQ

dX3

SIS

A1V

NV

XI81V

W

.MttN

wuw

m.

1N3W

30YNVW

VIV

O

NO

IilSO

d X

V10S

3NI.U

108 ON

IUO

Id

(SIH

3W3W

3) NO

IilSO

dH

8W

3/A

a0103fV

ai

6

/i /

/

V*

/ P

8XXXX

1 EXPERIMENT GROUPINGS1xxXX

1 SYSTEWPROGRAM COST ANALYSIS 'xXXXX

[ GROUND TRACE GENERATOR* 'XXXXXX

[ TARGET OPPORTUNITY GENERATOR

8xXXXXXX

L COMMUNICATIONS COVERAGE

gxXXXXXXX

I SOLAR /MISS ION GEOMETRY •;

•XXXX

[ RADIATION ENVIRONMENT

§xXXXxXX

[ ORBIT CONTAMINATION •xXX

I ATMOSPHERE MODELIS) 'xXXXXHUJ

o—as iXXXXXX

| ORBIT MANEUVERS

§xXXXXX

| ORBIT ERROR ANALYSIS 'xXXXXXXX

I SUBSATELLITE MOTION ANALYSIS

§x '

XXX

| MISSION CONSUMABLES '

•xXXXXX

| ORBIT ATTITUDE MANAGEMENT ••xxXXXXXX

[ INSTRUMENT POINTING •xXXXXXXXXXXXXiUJ

ziI/II/I

X

§§XXXXXXX

[ EXPERIMENT DATA PROCESSING

ixXXXX1UJ

ZU4

Xa.

UJ

H-

CD

s 'XXXXXX

.XXXX

| CONTINGENCY OPS. PLANNING •xXXXXX

1 SUBSYSTEMS PERFORM. MONITOR

§xXX"xXXX

| GRND TRUTH COORD. CONTROL •xXXxXX

| PAYLOAD (EXPMTS) STATUS MONITOR •xxXXXXX

| THERMAL ANALYSIS •xXXX

I MASS PROPERTIES I ANALYSIS •xxX

[ LOADS/STRESS ANALYSIS 'xXXX[ ELECTRICAL PWR ANAL & REQMTS

§xxX</tw

t

^î

XXXX

1 STABILIZ. I CONTROL ANALYSIS ^mg2I*1-

§55§§§§§i§§§I§8i§881FORTRAN PROGRAM STATEMENTS(ROUTINES)

2-19SD

7

6-S

A-0

02

8-1


^x.

f

^X.

^SV^N. ROCKWELL - DEFI NED FUNCTI ONS

MSFC PROGRAMS "" ^^^

^^^^TASK

13tsih-

X^j§

^CD

g*•*

v>gp 5?i°{/> v!CD Q-

I5K-

Ul

^

PI/)

1/1UJ

H-

^S

—

g

EXVI

o5£o-

f0

viii»-VI S

K %

u 95QC 7< S

"1805 S

.

S

SgE2£»

iis

PROGRAM

WORLD MAP

ORBITAL TRACK OPTIMIZATION

PERCENT AREA COVERAGE EARTH RESOURCES

PHASING

RAPID OPERATION FOR MISSION PLANNING

GROUND LIGHTING ORBIT ORIENTATION

TIME OF ASCENDING AND DESCENDING SITE OBSER-

VATION

AMPS MISSION SIMULATOR

ASTRONOMY

LAUNCH TIME

AVERAGE VIEWING TIME

PSEUDO STRAY LIGHT ISOCHRONAL CONTOUR

CONTINUOUS SUN LAUNCH TIME

STRAY LIGHT AND VISING EVALUATOR

SITE ACQUI5TION AND LOSS

EPHEMERIS GENERATOR FOR ORBITAL EVALUATOR

AUTOMATIC SPACECRAFT OCCULAT1ON ROUTINE

SENSOR TARGET OCCULATION PROGRAM

(SLAVE)

(PSEUDO STRAY LIGHT ISOCHRONAL CONTOUR)

ANALYTIC SATELLITE EPHERMERIS PROGRAM

ORBITAL LIFETIME

RELATIVE MOTION

RADIATION MODEL

RADIATION CALCULATION

ORBIT ENVIRONMENT SUMMARY

CLOUD MODEL

ATTENUATION.MOOEL

INSERTION SURFACEINTEGRATED MISSION PROGRAM

PERFORMANCE CALCULATIONS

DEBOOST OPPORTUNITY PROGRAM

SOLAR PHYSICS AND EARTH RES EXP SCHEDULER

STELLAR TARGET SCHEDULER

AMPS MISSION SCHEDULER

CREW TIMELINE ,

GENERAL ACTIVITY SCHEDULER

TIMELINE INTEGRATOR

MISSION OPERATIONS SEQUENCE SUMMARY

DATA STORAGE PROFILE

ATTITUDE TIMELINE INTEGRATOR

FLIGHT PLAN SUMMARY

GRAPHICS LAYOUT

MASS PROPERTIES

SHUTTLE PAYLOAD INTERFACE LOADSTHERMAL MODAL NETWORKS ANALYSIS

THERMAL RADIATION ANALYSIS

POINTING CONTROL SIMUATION

SPACELAB CONTAMINATION MODEL

UNKNOWN CORRELATION

WMAP

OTO

PACER

PHASE

ROMP

OLOO

TADSO

AMS

AS TION

LAN TIM

AVTIM

PSLIC

30NSULT

SLAVE

SAL

EGORE

AS OR

STOP

(SLAVE)

(PSLIC)

ASEP

LIFTIM

REL

IADI

IADCAL

OES

CLOUD

ATTEN

INFACE

IMP

PERCAL

DEBOP

SPERES

SCREAM

AMPS

CREWTL

WAMPUM

TINT

MOSS

DAS PRO

ATI NT

fPS

AIDS

MASS

SPIL

S TIN DA

TRASYS

IPS

CONTAM

r*s38

33

29

23

23

2318

27

38

22

38

31

13

23

3132

28

32

23

31

32

32

17

30.

1730

30

2322

3223

U

32

32-

24

224«

13

32

38

24

32

23

3

4349

2033

1I

3

XXXXXX

XX

XXXXX

X

1

M/P

RO

GR

AM

CO

ST A

NA

1

2

X

a

Ul

iO

3

3

X

X

X

1

IT-O

PP

OR

TU

NIT

Y G

EN

EI

C

3

XX

X

X

XX

XXX

XXX

X

X

u.

AU

N (C

AT

ION

S C

OV

ER

AC

Iu

s

X•XX

X

X

X

t/M

ISS

ION

G

EO

ME

TR

Y

1

3

X

XX

XXX

X

XXX

TIO

N E

NV

IRO

NM

EN

T

c

2

X

XXX

X

CO

NT

AM

INA

TIO

N

i—S

8

5.!

X

ISP

HE

RIC

MO

DE

L(S

)

H

2.8

,

XX

XX

a

2

3

X

MA

NE

UV

ER

S

8

3

XX

X

X

X

XX

ERR

OR

AN

ALY

SIS

8

4

'

S£

JELLIT

E M

OTIO

N A

NA

L1

I

1.2

X

X

ON

CO

NS

UM

AB

LES

W"

3

3.5

X

X

XX

X

X

AT

TIT

UD

E A

NA

LY

SIS

8

]

X

X

X

X

X

UM

EN

T P

OIN

TIN

G

K

2

•

l.S

X

X

X

X

§

ON

TIM

ELIN

E G

EN

ER

AT

I

I

6

XXXXXXXX

X

0

IME

NT

DA

TA

PR

OC

ES

SIN

Of

tUJ

100

X

ULJ

EP

HE

M/T

IME

LIN

E U

PO

Al

t-S8

3

X

XXX

XXXX

X

XXXXXX

XX

X

o•J

'ING

EN

CY

OP

S P

LA

NN

II

Iu

s

X

X

X

X

s

•ST

EM

PE

RF

OR

M. M

ON

IT

•&

2.!

X

X

X

X

X

X

d

INO

TR

UTH

CO

OR

O

CO

N

O

8

8

X

X

X

XX

XX

X

XX

:PM

TS

) S

TA

TU

S M

ON

ITO

II

X

5

XX

AA

L A

NA

LY

SIS

^r

3

XX

PR

OP

ER

TIE

S &

A

NA

LY

SIS

i5

XX

S/S

TRE

SS

AN

AL

YS

IS

9o

5

X

PWR A

NA

LY

& R

EQ

MT

S

aUJ

S

XXX

XX

MA

NA

GE

ME

NT

AN

AL

YS

I

4

X

X

~

LIZ

& C

ON

TR

OL A

NA

LY

i

5

5

X

JOW

N C

OR

RE

LA

TIO

N

^

-

.X

•S STATEMENTS I

••W WORDS <K)

Figure 2.4-3. Ground Processing Software Availability (MSFC)

2-20SD 76-SA-0028-1


The 29 essential programs for ground operation processes and the fiveprograms for documentation comprised the data base for the subsequent analyses.The recompilation cost of $700K was used in the programmatic costing analysis.However, this magnitude of cost is quite unlikely. The trend in the computerindustry is to achieve hardware and software interchangeability between vendorswith minimal effort.

2.5 GROUND PROCESSING IMPLEMENTATION COST FACTORS

The alternatives for computerized ground processing activities are essen-tially the same as those for on-board services. A batch processing approachcan be .used wherein the user requirements are documented and exercised by acentralized data processing center (DPC). Or, an interactive terminal approachcan be used wherein the user communicates directly with what appears to be adedicated processor via a CRT/keyboard. These two approaches are illustratedin Figure 2.5-1, and the two formats of the initialization data are illustratedin Figure 2.5-2. (It should be noted that in the case of ground processing,existing computer software is being exercised—not developed; software isbeing developed in the case of on-board services.)

IJ PROGRAM-'x MER ;

INITIALIZATIONDATA

KEYBOARD KEYPUNCH

•

.

Figure 2.5-1. Interactive/Automatic (Batch) Concept

BATCH PROCESSING' iWith the batch processing approach, the user completes initialization

data forms similar to that shown on the right side of Figure 2.5-2. Theseforms are transferred to a centralized activity which keypunch/machine codesthe data and exercises the appropriate program. Printed results are returned

2-21

SD 76-SA-0028-1


Space Division

U

U

U

U

U

U

U

P

P

P

P

P

P

INTERACTIVE TUTORIAL

CREATE FLIGHT PLAN

.IS THIS A NEW FP?

Y

ENTER GST AT 0 HOURS UT TODAY

0, 57/5.805

ENTER GST AT 0 HOURS UT TOMORROW

1, 1, 2.364

ENTER CONVERSION FACTOR UT TO ET SEC43 .

ENTER OUTPUT DEVICE NUMBER

1 • TTY, 2 - CRT, Z - PRINTER,4 • MTAPE

2

ENTER VEHICLE ALTITUDE (KM)

210••••

ETC.

BATCH

FLIGHT PLAN NO --

GST-A - GST-B -

CF _

ALTITUDE _ INCL. _

LONG, ASC. NODE _

PERIGEE _ LAT _ LONG.

ETC.

U • USER PROGRAMP • PROGRAM

Figure 2.5-2. Ground Processing Tutorial Requirements

to the user. The factors to be considered with the batch processing approachinclude average number of computer runs per flight per program, data process-ing center personnel requirements, turnaround time, and host machine time.Table 2.5-1 summarizes these cpnsiderations and reflects estimates based uponprevious programs at Rockwell of a similar nature. Thus, with the batchprocessing, or centralized approach, the per-flight costs are about $82K. Ifthe data processing center at Langley' is used the turnaround time is at leastone day. Even this delay between data request and output data can be frustrat-ing to a mission analyst. Note that for the projected ATL program the DPCutilization rate is only 5.7 percent for two flights per year. It was assumedthat this rate can be .accommodated by existing DPC capabilities.

INTERACTIVE TERMINAL PROCESSING

With the interactive terminal approach the user communicates directlywith the processor via a CRT/keyboard by means of tutorial software. The leftside of Figure 2.5-2 illustrates a typical CRT display in the exercising ofthe interactive terminal"approach.

There were five alternative implementations of the interactive terminalapproach that were evaluated. Two concepts pertain to remote terminals andthree to dedicated mini-processors. The•remote terminal concepts are illus-trated in Figure 2.5-3. In one case the DPC was considered to be at Langley;in the other case the DPC was considered to be at KSC, the Level III/II/ISpacelab integrator. With the local DPC, initial .installation/purchase of aremote terminal is the only capital investment ($3K); the recurring costs are

2-22

SD 76-SA-0028-1


Table 2.5-1. Batch Processing Candidevatians

' ESSENTIAL PROGRAMS: 29' OPTIONAL PROGRAMS: 5' ESTIMATED AVERAGE RUNS PER FLIGHT: 20 TIMES EACH

3k X 20 = 680 RUNS

' PROGRAMMING TEAM1 LEAD PROGRAM ANALYST @ $50K/YEAR1 CODER (KEYPUNCH OPERATOR) @ $AOK/YEAR .

* ESTIMATED RUN RATE PER TEAM: VDAY

680 7 - 170 DAYS/FLIGHT

(50K + **OK) X = $61.2K/FLIGHT (250 WORKING DAYS/MAN-YEAR)

' TURNAROUND TIME PER RUN

LOCAL: 3 WORKING DAYS '(NASA SPEC. HANDLING)

REMOTE: 10 WORKING DAYS (U.S. POSTAL SERVICE)

' HOST MACHINE TIME

AVERAGE CLOCK TIME/RUN: 5 MINUTES

680 (RUNS) X 5 (MIN) 7 600 * 57 HR MACHINE TIME PER FLIGHT

* HOST MACHINE COSTS

AVERAGE COSTS/HOUR: $375 (S/360)

57 X $360 = '$2 IK/FLIGHT

' HOST MACHINE UTILIZATION

AVAILABLE HOURS PER YEAR: 2000

@ 2 FLIGHTS/YEAR ~-- X 100 = 5.7$

* PER-FLIGHT COSTS •

PROGRAMMING TEAM + HOST MACHINE

$61.2K + $21K - $82.2K/fLIGHT

2-23

SD 76-SA-0028-1


Space Division

REMOTETERMINAL

OK).-Co'uVLE,

•—

LOCAL

OKLINK

—mLINE

COUPLE!

MUX-MMUXINPUT/OUTPUTMJFFB

BASEDATAPROCESSINGCENTER

,

PROGRAM !FILE !

I1

TYPE 1 TERM IAS/MONTHPLUS $100 INSTAL1

TYPE II TERM

50 MILES

SWITCH CENTER1000 MILES

U.40/MILE PLUS S20/MONTH

IfO/lttONTKPLUS flOO

INSTALL

100 MILES

SWITCH CENTER

•TOTAL MONTHLY COST - $635 PLUS $3.2K INITIAL INSTALLATION!

•PER-FLIGHT COST - $3.8K (2 FLIGHTS PER YEAR) |

•DPC NOT INCLUDED

REMOTEDPCLINKCOSTS

F-igwce 2.5-3. Remote Terminal Alternatives

the host machine time and are the same as the batch processing approach—$21Kper flight. If the DPC is at KSC, an additional installation ($200) and monthlyservice charge ($635) .is involved for the Dataphone Digital Service linksrequired between Langley and KSC.

The three dedicated mini-processor implementations consisted of minor vari-ations in the memory configuration. One system, the mini-corrmon disa, utilizeda disc memory bank that was shared by all user.s. The two other configurations,mini-dedicated tape and mini-dedicated disc individualized the memory systemsfor each user. All three systems could utilize the same mini-processor andinteractive terminal ($18K). the memory systems were $28K, $15.4K, and $19Kfor the common-disc, dedicated-tape, and dedicated-disc configurations, respect-ively. The difference between the two dedicated systems is simply one uses atape memory; the other uses a disc memory. The common-disc memory configurationis a one-time investment. The other two memory configurations must be acquiredwith the mini-processors as the traffic/work-levels increase beyond two flightsper year.

SUMMARY

Programmatic costs for the five ground processing implementation conceptsare summarized in Table 2.5-2. Yearly costs reflect the ATI flight rate forthe baseline traffic model and are accumulated, arbitrarily, in the year priorto the scheduled flight. Batch processing costs are prohibitive and are thelargest even in the first year of the program. The costs of the remote-terminalconcepts and the mini-processor concepts are comparable initially, but rapidly .accumulate to significantly more than the mini-processor concepts. Thus, themini-processor concepts are preferred. Of the three, the mini-dedicated disc

2-24

SD 76-SA-0028-1

Rockw

ell InternationalS

pace D

ivision

iIi§oc

1oS2eoeoO1

,,,eo01

£01in01

S01«*»oo01

s~D.

s01

§Lb!

a.

inin-a-.

•f•ni«i

CM

^^

^^

§-1U.

%C3

•p5

eo v

'oJTO"*

OOf%oo»oovO-TCM

5CMCM

33a1 BATCH1'2

(NON- TUTORIAL)

oaeae — *

|o

i£ i

u!

Szi

CvK

C

M

5"

• itnoao* o0

0

-a-eoK

* ooS

1

ô\0«

vOin

« -a

-

JJ

* C

M

REMOTE TERMINALTO LOCAL DPC2*3

zoeQ X

i*J C9

nii

V*

VC

CM

in v>

I£

*2~CM0

1

• CM

in* 01

—^-T

* in0

1CM

« ?*

Ow

REMOTE TERMINAL 'REMOTE DPC2-3"1*

x:vOvO1

• '

* $

11

* S

11

'«* R

(

«^

inuvt5izx

invO

* i»\ii

* Ri,

* i»» '

* «•>,

« S1MINI-DEDICATEDTAPE5

inoo«vI•!*•.

* »

»l

11

* fl11

«s* «n1

* 5>1MINI-DEDICATEDDISC5

•t/\rx.

• xpi&«g,inx

n

t- vi

ta w

z

<.

—

S

xX

-I t-

X

0—

u.

<

V

oe

«>

ae oe

"i a.

Ul

< M

3K

•.

Ul

w

(—<

>

- z

^

t-

UJ

t- —

oc <

x

X

CN

U

J

• O

LU

1/V

fiv U

l O

C

•—

0

<

3-i

a

t- ae

ui

cy

U.

Ul

X

<

>•

UJ

«:

S

2

5

oe

£

U

<

-1

UJ

Za.

x

u.

ui

Ok

ui

u

u

<

x>-

oe

—

i- »-

ae u

ui

>

x

vi

<

a.

a.

oe

o

ui

Jo

uj

—

><

-1

V

I -1

Z

VI

0

<

U.

—

. *

z

< oe

_i

X

</l

X

t-

Ul

<<

3

2

— o

. t-

UJ

Ul

O

^>-

t/i K

—

K

a,

(-

O u

<ec

a.

uj

vi

uU

l U

l H

U

lX

o

o

z

—

aX

Z

X

O

Z

u

<

o

ul

x

— oc

3

o

oc a

. x

—L9

<

=>

S_l

Ul

>-

Ul

O>

-1

Z

<

Z

U4

o.

<

o

a

o

oe

CV

]IS

2-25SD

76-SA

-0028-1


Space Division

concept is favored because of the versatility and flexibility that is character-istic of this approach. Also, it is believed that more efficient operations(less manpower) can be achieved with the dedicated approach. These advantageswarrant the minor additional costs of $20K for a 10-year program. It should benoted that this recommendation is based upon the assumption that the GSSS/MSFCprograms will be compiled for an applicable mini-processor at nominal cost—notthe worst case estimate ,of $700K.

2.6 PROGRAMMATIC EVALUATION

ATL programmatic costs were developed, based upon the recommended approachesfor mandatory on-board computerized services (maximize the dedicated processorapproach), command/control operations (use computer-aided approach when reason-able) , and ground processing/mission planning activities (implement a dedicatedmini-processor/disc memory concept). A representative ATL payload model wassynthesized, and hardware and software cost factors for this typical payloadwere defined. Costs were developed for three different traffic models, consid-ering potential reuse and reflight of both hardware and software. As the pre-ferred approaches to software-related activities were dependent upon the avail-ability of an FSSS and GSSS, a contingency/criticality analysis was conductedto identify the impact of the lack of these two proposed software 'tools.

REPRESENTATIVE ATL PAYLOAD AND COST FACTORS.

The three reference ATL payloads indicated that multiple Spacelab configur-ations would be used, and significant variations in on-board operations andexperiment reflight would occur. In order to develop ATL programmatic costsa representative ATL payload model was formulated. The analyses of requiredo'n-board computer services indicated that a typical ATL payload would requiresix mini-processors, 14 micro-processors and 2500 statements of mission-uniqueflight applications software. However, the reference ATL payloads had 11, 11,and 12 experiments each. Therefore, control panel and computer-aided command/control equipment and software requirements were derived to reflect the totalcomplement of experiments on an ATL payload.

Analyses of the 25 reference ATL experiments indicated that not all ofthem could or would be controlled from a 'common work station. The experimentswere categorized into four groups according to their complexity of manned oper-ations and man-machine interface characteristics (see Table 2.6-1). The exper-iments in Group 1 required extensive command/control/monitor operations.Although the experiments in Group 2 were highly automated, a significant man-machine control/monitor interface was still required. The experiments inGroup 3 required only initiate/terminate control actions. The group 4 experi-ments required direct man-participation, involving manual dexterity and/orvisual acuity.

The operational characteristics of the first two groups of experiments arereadily adaptable to the computer-aided command/control approach. The fourthgroup of experiments are not adaptable to the computer-aided approach. Althoughthe third group of experiments could be-adapted to the computer-aided approachthe control actions are simple, non-repetitive, and require minimal monitoring;this group is not recommended for computer-aided command/control implementation.

2-26

SD 76-SA-0028-1


Space Division

Table 2.6-1. Man-Machine Interface Grouping

GROUP1

GROUP2

GROUP3

GROUP4

NV-1

NV-2

NV-3EO-2

E0-*t

HIGHLY

HIGHLY

• PH-6

* MB-)

* EH-3

DIRECT

• PH-2

• PH-3• MB- 2* MB- 3

EXTENSIVE CONTROL I MONITOR REQUIRED

MICROWAVE INTERFEROMETER* EO-5 LASER RANGING & ALTIHETRYAUTONOMOUS NAVIGATION EO-6 MICROWAVE LATI METERMULTIPATH MEASUREMENTS EO-7 SEARCH & RESCUE AIDSTUNABLE LASERS EO-8 IMAGING RADARMICROWAVE RADIOMETER EO-9 RF NOISE MEASUREMENT

AUTOMATED BUT EXTENSIVE MONITORING/EVALUATION REQUIRED

* EO-1 LIDAR MEASUREMENTS. • EO-3 MULTI SPECTRAt SCANNER

* PH-4 NEUTRAL GAS PARAMETERS

AUTOMATED, ONLY INI TIATE/TERMI NATE .ACTI ONS REQUIRED

METEOR SPECTROSCOPY * CS-2 ZERO-G STEAM GENERATORCOLONY GROWTH * CS-X CONTAMINATION MONITORNON-METALLIC MATERIALS

MAN PARTICIPATION, DEXTERITY, VISUAL ACUITY REQUIRED

BARIUM tLOUD RELEASE * MB- it BIOCELL ELECTRICAL CHAR.AEROSOL OPTICAL PROPERTIES * MB-5 BIOCELL SPECIAL PROPERTIESMICRO-ORGANISM TRANSFER * EN-5 MICRO-ORGANISM SAMPLINGBIOCELL ELECTRICAL FIELD

OPACITY

The experiments in Groups 1 and 2 correlate with those for which dedicatedmini-processors for the on-board services were identified. Thus, it was postu-lated that six experiments of. the nominal payload would include the computer-aided command/control concept.

As the average number of experiments in the reference ATL payloads was 11,an allowance for hardwired control panels was also made. Several of the exper-.iments in Groups 3 and 4 pertain to microbiology and require either no controlsor minimal controls. In order to reflect the averaging effect of this class ofexperiments-it was postulated that four hardwired control panels would be-required for each ATL payload and would cost approximately $3K each. (The aver-age panel cost if all controls were hardwired was $5K.)

Typical ATL payload hardware requirements are summarized in Figure 2.6-1.The display terminals and printers that are allocated to the Pi's are fordevelopment/validation of flight software. Two additional display terminalswere allocated to the lead center and would be the actual common-work-stationflight hardware. The remote activation systems are associated with the sixexperiments that would utilize the computer-aided command/control approach.The mini-disc sets are allocated to the lead center to support the missionplanning activities; ,

Software cost factors are summarized in Figure-2.6-2. The recurring costsinclude the preparation of the flight applications software for both the mini-and micro-processors for all the experiments on a payload. Command/controlservices reflect the use of the computer-aided approach in the mechanization ofsix experiments. The CDMS services are associated with the integration oftelemetry, caution and warning, data annotation, and mission timelines with theSpacelab and Orbiter.

2-27

SD 76-SA-0028-1


ATI REPRESENTATIVE PAYLOAD • 10 EXPERIMENTS

PI ALLOCATION

• MINI-PROCESSOR SYSTEMS

•6 MINI -PROCESSORS $I68K•6 DISPLAY TERMINALS $ 18 K• •PRINTERS $ 12 K

• 14 MICRO-PROCESSORS * 154 K

• COMMAND /CONTROL DEVICES

•6 REMOTE ACTIVATION SYS. $ 3.6 K•4 CMD/CONTROL PANELS * 12 K

LEAD CENTER ALLOCATION

• MINI-DISC SETS

•2 MINI-PROCESSOR•2 DISPLAY TERMINAL•2 PRINTER•2 DISC MEMORY

• RIGHT COMMAND 'CONTROL

•2 DISPLAY TERMINALS

*)IK* 6K*.4K$«K

$ 6K

SUPPORTS 2 FLIGHTS PER YEAR

"BRIGHT HARDWARE

Figure 2.6-1. Typical ATL Payload Hardware Complement

RECURRING ON-BOARD SERVICES(PTR PAyin*r»

MANDATORY COMPUTER SERVICES

2500 STATEMENTS • »!/ * 77.5 K

COMMAND /CONTROL SERVICES

1200 STATEMENTS • $31 / * 37.2 K

' 72K BYTES (DATA TABLES) 0$. 01 / $ < X 7 K

CDMS SERVICES

BOOK BYTES (DATA TABLES) 6 *. 01 / $ 3. 0 K

3 HR HOST MACHINE TIME 8$375/ $ L 1 K

3 MAN-MO. INTEGRATION $12.5K

NON-RECURRINGSOFTWARE DEVELOPMENT TOOLS

FLIGHT SOFTWARE SUPPORT SYSTEM

COMMAND/CONTROL DELTA TO FSSS

1200 STATEMENTS 0 $62 /

I680K

$74.4 K

GROUND SOFTWARE SUPPORT SYSTEM DELTA

•TOOK

AGENCY DEVELOPMENT

WORST CASE ESTIMATE: COMPLETION OF MSFC PROJECT COULDALMOST ELIMINATE THIS ITEM.

Figure 2.. 6-2. Software Programmatic Complement

Non-recurring software costs are for the development of the basic FSSSand the delta to the FSSS to efficiently utilize the computer-aided command/control approach, and to convert/mo.dify exist ing/in-work mission planningsoftware at MSFC to'Langley's specific use. Because of the broad applicationof the FSSS it is recommended that its development be sponsored by the agency—not uniquely attributed to ATL. Upon completion of the GSSS work at MSFC,this software development tool may also be directly applicable to ATL payloads.

2-28

SD 76-SA-0028-1


PROGRAMMATIC COSTS

In order to develop programmatic costs it was necessary-to establish guide-lines for ATL flight rates and reflight commonality for both hardware and soft-ware. Figure 2.6-3 summarizes the,selected guidelines. Three ATL trafficmodels were used: the baseline (Yardley traffic model), maximum of two flightsper year, and a one-flight-per-year program.

•TRAFFIC MODELSYEAR

•BASELINE

•2 FIT/YEARLIMIT

•1 FLT/YttRLIMIT

•HARDWARE REUSE

•PANELS•ACTUATORS•MICRO-PROCESSORS•MINI-PROCESSORS

40%4M40%

100%

811

1

1

821

]

1

832

2

1

843

2

1

8$3

2

1

863

2

1

874

2

1

884

2

1

894

2

1

905

2

1

915

2

1

4 OF 10 EXPERIMENTS REFLOWN

SHARED UP TO 2 FLIGHTS'YEAR

• SOFTWARE REUSE

•MINI-MICRO SOFTWARE - 25%(ON-BOARD SERVICES AND COMMAND / CONTROL)

•CDMS -M (NEW EACH FLIGHT)

Figure 2.6-3. Programmatic Costing Criteria

As panels, actuator hardware, and micro-processors are an integral part ofthe experiment equipment, sharing of these end items between experiments wasnot considered to be practical. However, experiment reflights are anticipated.Thus, it was assumed that the reuse of this type of hardware would average40 percent^during the course of the ATL program. Mini-processors are stand-alone end items and can be shared between experiments. Thus, 100-percent reusewas assumed. Each mini-processor can support two flights per year. It isanticipated that a limited amount of-f l ight applications software will also bereusable. A 25-percent software reuse factor was estimated. As the experimentmix of each ATL payload is different each flight, it was assumed that the CDMSintegration effort ($16.6K per flight) would be required for each flight.

Based upon these cost factors and reuse criteria, a compilation of theprogrammatic software-related costs for each ATL traffic model was developedas illustrated in Tables 2.6-2, 2.6-3, and 2.6-4. Cumulative recurring costs(basic FSSS, command/control delta, and GSSS mods not included) are plotted forthe three traffic models in Figure 2.6-4. The minor per-flight variationsbetween traffic models are due to different utilization rates of the mini- andmicro-processors and the intelligent terminals.

2-29SD 76-SA-0028-1

Rockw

ell Inte

rnatio

nal

Space

Divisio

n

•8oft.

to8

ononoonen

Sen

CO

ooon

GOon

00onIAooon

COon

COon

&on

COonoCOononon

COon^™Xo•VU.

U_0tcUl

i\A,.3.

»mtr»

r»>

CM

£X•a*.

3OT

gS

S

«M

M£

«0

*"

o o

on

r>—

CM ••

o o

en

^^ *

* ^

4•*»O

O

On

r^— C

Ml*\

<«••»

IA

c^ ^^ ooC

M —

•»•*• IA

«•»

— O

OC

M

—

r>. — eo

oo o-\\o

rs.ô

oco on NO

r*-— oo

oo onvo

r>«r» CM

poo —

f>-

CM

•*

M

> IA

£^5

1C

MS

oo —\O

IA

«or^r«.cMW

> o o

o —

\o f"» C

M

•* so IA

IAIA

-B-

oo —

CM CM

r>on

^ft >^ ^^

-» C

M •-

OO

f

t C

*c ^0

OO

OO

f«

VI

^«/>

zS

v,5

i

Jg

I

2S

5-S 5

B

O irt >

- J P

p

oc vi vi <

/i ui

L.

<

vi u

i z H

o

Z

Ul «

J Z

« Z

V

I<j o

8 a

. ui

0

O «

—

0 O

oe oe

a. »- ^ —

oe< a. i <

o _

i <o

i

o >

«e _

i o? Z

gp

|p- T

S x

«^§

: g

RY/CONTROL •

MANDATOCOMMANDCDMS

-a-onr-IAIA*^

IAIAÎAIAÎASOIAIASOIAon***

OOen

CO

eo~r^•*•g

^

-

o

•

^

^p ^^

^^

8en

l

<

31 1

i2

z

p

•-ae

2

U.

z<

U

l VI

O

O O

Z

K «C

V

I —

0

o

3 —

o

o

a.

a. *

a. v

»a.

vi a.

a.

vim£ s

§ ee

a0oJC*?oo"*g

«

^^

^^

^^

^^

^^

^^

m

P5

^^

^^

^>

O

^j

9\

C9

^^<M

CM

0 g^

CM

CM

O

OIA

IA

CM

CM

V)

i

S §.,

VI

—

<U

. H

0

— 0

vi

o l-

? *—

vt

z

vi

Z

3

\

onCO-IAIAÎAIACM

ÎAVOIAIANOIAOCOIA"*

ooooJtÎACM

vO0IAIAV)

I£VI

o£S

2-3

0

SD 7

6-S

A-0

028-1

Rockw

ell InternationalS

pace D

ivision

rM

oen|rMon

oo •—

on* m

OO

—

rMIA

IA-*

oo •—

OO

—

eoC

D

IAIA

-oo

•oo

ts-S•s,

1A

IA^-

CO

•—oo

oor-.

IA

IA-*

oo •-

voC

O

oo rMon

IAIA

JT

OO

•••

CO

fMCOen

OO

•—

-o- vo JT1A

COooon

eneoeo

oCOon

IA

—ooeo i->^s«

n —

fc 8

E:

^^

^9

^^

o

oIA

IA

IA«<«N

O

COr».

en§

SI S

IA

to•

CXI

5O

—

U

O I-

_l —

xc

^s

: §

I/I

fi-llCC

M

is £

i —

O

VI

oo

| §

O

—

—

o

2 IS

2-31

SD

76

-SA

-00

28

-1

Rockw

ell InternationalS

pace D

ivision

&oonen-»

C

M —

sooco

Sonr<on

CM —oo

ooco

oo00on

on•»

C

M —

cooo

coonCM

CM

f-

en-»

C

M —

CO

oo

NO

COon

onCM

—oooo

incoonon

CM

—oo

cooo

§ooon

CM I

enoo

cooo

COon

on•»

C

M •-

ooeo

CM

ooonen

•»

CM

—oooo

COCJ

OOon

•*

CM

•-co

ooco

SonIA

—

ooCMCM

onr*onCM

•

CM

1A

CM

•"31

to•03

•s

oorôn°

CMsCM

0IAtrv

ta

oc

X

—

I/I 111

l> I <

O -I

sg£

i£O

O

2

VI

.J=

o

a. «e a.

v* m.

£

iiS

VIM

12 it.

UI

O

5u,

-

P

u

—l/»

O

—

I/Ir ta

.

2-32

SD 76-SA

-0028-1

Rockw

ell Inte

rnatio

nal

Space

Division

RooCO

h >•

•tCO

CM

00

to

O•

CO

O•

M

2-3

3

SD

76-SA

-0028-1


Space Division

CONTINGENCY EVALUATION

The preferred approaches for on-board.services, command/control functions,and ground processing were based upon the development of the FSSS and the adapta-tion of the MSFC software programs to a GSSS. A contingency evaluation was con-ducted to assess the impact if the FSSS and GSSS were not developed. The impacton the initial ATI flights is summarized in Table 2.6-5.

Table 2.6-5. Contingency Evaluation

0-oo

oCD1

| GR

OUND

OPE

RATI

ONS

FIRST FLIGHT COSTSWITHOUT FSSS

•MINI/MICRO S/W $ 375 K

• CDMS SOFTWARE 17 K

• PI HARDWARE 397 K

•LEAD CENTER HDW 6 K

TOTAL $ 795 K

WITHOUT LOCAL GSSS

• BATCH PROCESSING $ 82' K

• REMOTE TERMINAL $ 32 K

WITH FSSS

•MINI/MICRO S/W $ 115 K•CDMS SOFTWARE 17 K• PI HARDWARE 368 K•LEAD' CENTER HDW 6 K

TOTAL $ 512 K

WITH LOCAL GSSS

•MINI-DISC APPROACH $ 36.5 K

COMMENTS

• FSSS = $755K

• SAVINGS USING FSSSFOR 3-4 FLIGHTS EQUALSFSSS COSTS

• GSSS

IMPLEMENTATION IS NOTTIME-CRITICAL

• ONLY PROGRAMMATICS" FAVORMINI-DISC APPROACH

Without the FSSS the PI must develop his flight applications software,including all the library routines, independently. This effort will increasethe quantity of required o'n-board services software for each mission by about9600 statements.• However, without the FSSS the computer-aided command/controlapproach would be impractical to implement, and the computer-aided software(1200 statements) would not be developed. Thus, the net increase in softwarewould be 8400 statements. Assuming adequate programmer support is availablefor working directly with the PI (informal relationship/minimal documentation)these additional statements will cost $31'each and add $260K to the cost ofexperiment software.

Even if the problems/complexity/costs associated with the required integra-tion of control panels for the pallet-only configuration are neglected, theaverage payload hardware costs will increase without the computer-aided approach.With the FSSS (and computer-aided approach), $16K of the hardware costs was foractivation system hardware and simple control panels ($352K was for processors).Without the FSSS all experiments will require hardwired panels at an averagecost of $5K/experiment, or $50K/payload. Therefore, the total software develop-ment and related hardware costs for the first ATI payload will be $293K greater

2-34

SD 76-SA-0028-1


Space Division

without the FSSS than with the FSSS. If the.same software and hardware reusecriteria were used for the ATL traffic model without the FSSS, the delta pro-grammatic costs after four ATL flights would.be greater than the developmentcosts of the FSSS.

Implementation of a local GSSS capability with the mini-disc computersystem is not time-critical. Use of a remote terminal approach results in,essentially, a recurring $32K/flight cost; the mini-disc approach is a non-recurring capital investment of $37K.

2.7 PALLET-ONLY CONFIGURATION'SPECIAL CONSIDERATIONS

In general, the analyses and trades of alternate mechanizations of on-board operations were conducted without consideration o,f the specific Spacelabconfiguration involved. ' If the pallet-only configuration is uniquely consid- ,ered, then the limited control panel space in the Orbiter aft-flight-deck (AFD)becomes a prime driver in the mechanization of experiment command/control/monitor functions. .The baseline Orbiter payload panel allocation in the OrbiterAFD is illustrated in Figure 2.7-1.. This area must be shared by the Spacelab(subsystem controls) and the experiments. The baseline panel space allocationfor operation of Spacelab systems utilizes 1432 in2 of the 3200 in2 (shaded area)available for Orbiter pay loads. Thus, only the cros*s-hatched area (1768 in2) isavailable for ATL experiment control panels.

ON-ORBIT STATION(531

MISSION STATIONIn2)

TV MONITORS

R-7 PANEL(160 In2)

PAYLOAD STATION(1815 In2)

««&«&'«! TOTAL AVAILABLE ORBITER PAYLOAD PANEL AREA

&3g$fl AVAILABLE ATL EXPERIMENT PANEL AREA

Figure 2.7-1. AFD Panel Allocation for Orbiter Pay loads

2-35

SD 76-SA-0028-1


Space Division

Table 2.7-1 summarizes the required dedicated command/control panel areasfor the experiments of the reference pallet-only ATL payload. In addition,dedicated displays (spectrum analyzers, oscilloscopes, etc..) are required byseveral of the ATL experiments and are also indicated in the table. Even ifthe TV monitors (see Figure 2.7-1) are shared between experiment, Spacelab,and Orbiter operations, the dedicated displays/monitors increase the requiredATL AFD panel area to 2161 in2, which obviously exceeds the available space.By adopting the computer-aided command/control approach for the first sevenATL experiments listed in Table 2.7-1 (these seven experiments required mini-processors for the mandatory on-board computer services), the required AFDpanel space would be reduced by 950 in2. Making an allowance of 342 in2 for adedicated intelligent terminal for experiment operations would result in atotal ATL experiment panel requirement of 1563 in2, which is compatible withthe available space.

Table 2.7-1. ATL Pallet-Only Pay load-Required Hardwired Panel Space(Thousands of Dollars)

EXTENSIVECOMMAND,

CONTROL,&

MONITORREQUIRED

MANUALDEXTERITY» VISUALACUITYREQUIRED

EXPERIMENT

NV-1 MICROWAVE INTERFEROMETERNV-2 AUTONOMOUS NAVIGATIONEO-4 RADIOMETER

EO-7/8 SEARCH S. RESCUE/IMAGINGRADAR

EO-1 LIDAR MEASUREMENT .

PH-4 NEUTRAL GAS PARAMETERS

PH-6 METEOR SPECTROSCOPYEN-3 NON-METALLIC MATERIALSCS-X CONTAMINATION MONITORPH-2 BARIUM CLOUD RELEASEEN-1 MICRO-ORGANISM SAMPLES

TOTAL

COMMAND/CONTROLAREA (IN2)

95171209

209133133

95133133247

95

1653

TVMON 1 TOR

***

**

A

• *

'DEDICATEDDISPLAYAREA (IN2)

152152

152152152 -.

508

*SHARE TV WITH SPACELAB AND ORBITER OPERATIONS.

In this analysis, only total panel areas were considered. If actualdedicated panel layouts and interference between top-mounted and front-mountedpanels are considered, the accommodation margin.will be reduced if not elimin-ated. Thus, in actual practice, it may be necessary to implement the computer-aided approach in additional experiments and/or limit the experiments on apallet-only Spacelab to those that are compatible with the computer-aidedcommand/control approach.

A cost analysis of the two command/control approaches for the referenceATL pallet-only payload was conducted (Table 2.7-2). Dedicated panel costswould be almost $58K. In addition to the dedicated panels for the last five

2-36

SD 76-SA-0028-1


Space Division

experiments of Table 2.7-2, activation hardware for implementation of thecomputer-aided approach for the first seven experiments would cost $4.2K($600 per experiment). Software and data tables for these seven experimentswould cost an additional $44K. Thus, the computer-aided approach would costabout $15K more. It is believed that this delta cost for implementing thecomputer-aided approach is warranted. If non-dedicated hardwired controlpanels were utilized to meet the AFD space limitations, it is anticipatedthat the costs of integration of multiple experiment requirements into a singlepanel and the duplicate fabrication of these panels for Level IV integrationactivities will greatly exceed the $15K figure.

Table 2.7-2. Cost Comparison of Command/Control Approaches forReference Pallet-Only ATL Pay load

EXPERIMENTS

NV-1NV-2EO-4EO-7/8EO-IPH-4

PH-6EN-3CS-XPH-2EN-1

HARDWIRED APPROACH

PANELS

2.85.37.07.07.04.2

4.02.23.7

12.12.7

57.9

COMPUTER-AIDED APPROACH

HARDWARE

X

>

\

A. 2

24.7

SOFTWARE

44.1

-

73.0

2-37

SD 76-SA-0028-1


Space Division

3.0 PROGRAMMATIC SOFTWARE-RELATED RECOMMENDATIONS

Throughout the analyses the primary objectives have been to maintainautonomy of individual experiments, maximize hardware and software reuse, andminimize programmatic costs. The results not only reflect these factors butalso indicate that they are compatible. In this section, guidelines for thedevelopment of ATI payloads that will assist in the achievement of theseobjectives are delineated.

FLIGHT OPERATIONS CONSIDERATIONS

Develop the basic flight software support system. Analyses indicate thatfor a continuing program such as the ATL, significant cost savings can berealized if a software development tool is used. The proposed concept will'minimize the mission-unique software that is required.

Maximize the use of on-board experiment-dedicated mini/micro-processors.The cost savings in software development that can be achieved if dedicatedprocessors are used warrants the slight weight, power, volume, and costs ofthese processors. The Pi's autonomy and flexibility of design and operationsare also maximized with dedicated processors.

Develop the delta FSSS for corrmand/control functions. Although thecomputer-aided approach is slightly more costly than the hardwired approachthe Orbiter AFD panel constraints (pallet-only Spacelab configuration) willnot accommodate a completely hardwired concept. Development of computer-aidedsoftware and hardware on an individual experimenter basis would be costly andinefficient.

When feasible3 implement the computer-aided command/control approach.The reflight nature of the ATL program indicates that even if experiments areinitially scheduled for the habitable-module Spacelab configuration, they maybe subsequently scheduled for a pallet-only flight. Because of the AFD panelconstraints, a hardwired panel used in the habitable module may not be usablein the AFD. Thus, a second development would be required. Initial developmentof the computer-aided approach for command/control functions would provide thePI and Langley the maximum flexibility in payload grouping/flight scheduling.Also, the computer-aided approach is more adaptable to changes than the hard-wired approach. With the evolving technology associated with ATL experiments,flexibility of design is extremely important.

GROUND OPERATIONS CONSIDERATIONS

Implement the GSSS. The consideration of the number of times that missionplanning analyses must be performed and the duration of the ATL program make italmost imperative that a tutorial software tool be utilized. Batch processingis not only cumbersome and frustrating, it is also costly. In this study, onlythe payload integrator was considered in the mission planning phase. However,

3-1• SD 76-SA-0028-1


Space Division

each PI must also do individual mission planning analyses that pertain to hisexperiment. Current Pi's may have the necessary software programs, but duringthe course of the ATL program it is doubtful if more than a small percentageof the Pi's will be so equipped. Implementation of a tutorial GSSS approachwill facilitate the participation of a broad segment of the scientific commun-ity and minimize the affect of personnel turnover in the payload integrator'sorganization.

Adopt the mini-disc mission analysis approach. Although the remote term-inal approach will suffice the convenience, flexibility, and programmatic costs,warrant the mini-disc approach. This dedicated processor approach becomeshighly desirable when the Pi's are considered. Again, if a broad segment ofthe scientific community is to participate in the ATL program, techniques tominimize the costs of the individual Pi's and maximize the accessibility todata banks must be implemented. -For example, providing a remote terminal linkbetween a PI at the University of South Dakota and a central computer at Langleyis unrealistic. Except for the disc-memory device this Pi's dedicated processorwould suffice. Disc-memory devices could be shared between Pi's in the samemanner as the proposed sharing of dedicated mini-processors. (Note: disc-memory devices for individual Pi's were not included in the cost analyses ofthis study.)

DESIGN CONSIDERATIONS

As both the Spacelab hardware and Spacelab operations are in a design/development stage, specific design requirements for ATL payloads are notidentifiable at this time. However, the following guidelines indicate thetypes of design requirements that will have to be met.

1. All potential hazards due to experiment operations or to crediblefailures of experiment equipment will be redundantly instrumented;these instrument signals, properly conditioned, will be direct-wired to the Spacelab and Orbiter caution/warning system. The PIwill be required to demonstrate to a safety review board theadequacy of his analysis and design to avoid or contain any hazardor haz.ardous condition due to his equipment or its operation.

2. Experiment-derived data that will be telemetered to ground via theOrbiter avionics system will be acquired,- formatted, and annotatedwithin the experiment system prior to transmission under controlof the Spacelab CDMS.

3. The PI shall provide the interface hardware within his equipmentto decode and interpret command signals from the CDMS RAU.

4. The CDMS will provide Orbiter-derived annotation data (time,position, attitude) on a periodic basis, via the data bus/RAUnetwork. The PI shall provide the interface hardware to acceptand process these digital signals within his equipment.

3-2

SD 76-SA-0028-1


Space Division

*5. The PI should consider a computer-aided implementation ofcommand/control functions when the operator's proceduresare complex and sensitive to proper sequencing.

*6. The PI should consider a computer-aided implementation ofcommand/control functions when the experiment is remotelylocated from the operator's work station (specifically forpallet-only missions).

7. The PI should consider an automated approach of implementingcontrol for (a) emergency sequences, (b) time-critical oper-ations, (c) repetitive sequences where the operator's judgmentis not required, and (d) operator reaction time may be exceeded.(Note: Automatic control may utilize a mini/micro computer,but more generally would be implemented by clocked timer-sequencers or other mechanical devices—>particularly (a) and(c)—or by sensor trigger mechanisms such as limit switchesor optical detectors.)

*8. The PI should consider a hardwired approach for implementingcontrol if (a) the experiment equipment requires no in-flightmechanical or procedural adjustments, or (b) the operator'sparticipation is limited to initiating/terminating automaticsequences.

*These guidelines should be considered in conjunction with Flight OperationsConsiderations—When feasible, implement the computer-aided command/controlapproach.

3-3

SD 76-SA-0028-1


Space Division

4.0 RELATED FUTURE EFFORT

It is recognized that the preferred approaches for ATL flight and groundoperations software are contingent upon two key software development tools—the FSSS and the GSSS. Because of the repetitive nature of ATL Spacelabflights, as well as other Spacelab payloads, a significant programmatic costsavings can be achieved if software reuse is maximized. It is believed* thatthe proposed/conceptually defined FSSS will facilitate the reuse of on-boardsoftware as well as expedite the preparation of mission-unique software. Amore detailed definition and synthesis of the primary elements of the FSSS,coupled with a demonstration with representative payload equipment, should be .accomplished before a Spacelab programmatic commitment is made. It is recom-mended that such an activity be initiated within this calendar year in orderto support the initial Spacelab flights in a timely manner.

The additions to the basic FSSS for command/control by an interactive dis-play terminal is also recommended. Pallet-only configurations are frequent.With the limited panel space for payloads in the Orbiter, experiment groupingflexibility will be constrained unless shared intelligent terminals are viable.It should be emphasized that unless the basic FSSS is provided, the computer-aided approach for command and control is not recommended. Without the FSSStutorial feature, each Pi/user would be forced to prepare this software .usingmore conventional methods, or use the CDMS capability. Use of the CDMS would,of course, recentralize a major effort with an attendant increase in costs.

A conceptual CDMS-dedicated processor interface was defined. As both theSpacelab and ATL payloads are at the hardware development stage, a definitizedinterface (signal characteristics, coding, timing, etc.) should.be established.This proposed effort consists basically of analyzing the specific characteristicsof the CDMS and the Spacelab data bus and determining the interface requirements/specifications that a dedicated processor must meet. This analysis is not recom-mended until after the preliminary design review on the CDMS later this year.

The current GSSS development at MSFC was primarily for remote terminalapplications. A detailed analysis of the MSFC programs is required to determinethe potential extent of modifications to MSFC programs for use on dedicated mini-processors. As the MSFC program is still in progress, a preliminary activitv toconvert the programs to at least one mini-processor is underway, and a commitmentto a local GSSS is not time-critical, this effort can be postponed for at leastanother year.

4-1

SD 76-SA-0028-1

Documents

SD 76-SA-0028-1 CR-144966 SPACELAB USER IMPLEMENTATION