Appendix C Papers Submitted By Opening Session Presenters

Contents

Remarks on Reducing Space Science Mission Costs, 23

Eberhardt Rechtin, University of Southern California

Perspectives on Small Spacecraft: Results of a Recent RAND Workshop, 30

Liam P. Sarsfield, RAND Critical Technologies Institute

Influence of Technology on Space Mission Costs, 42

Frank J. Redd, Utah State University

Summary of Techniques for Reducing Space Mission Costs, 48

Wiley J. Larson, International Space University and U.S. Air Force Academy

Mars Exploration Program Strategy: 1995-2020, 53

Donna L. Shirley and Daniel J. McCleese, NASA Jet Propulsion Laboratory



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 22
--> Appendix C Papers Submitted By Opening Session Presenters Contents Remarks on Reducing Space Science Mission Costs, 23 Eberhardt Rechtin, University of Southern California Perspectives on Small Spacecraft: Results of a Recent RAND Workshop, 30 Liam P. Sarsfield, RAND Critical Technologies Institute Influence of Technology on Space Mission Costs, 42 Frank J. Redd, Utah State University Summary of Techniques for Reducing Space Mission Costs, 48 Wiley J. Larson, International Space University and U.S. Air Force Academy Mars Exploration Program Strategy: 1995-2020, 53 Donna L. Shirley and Daniel J. McCleese, NASA Jet Propulsion Laboratory

OCR for page 22
--> Remarks On Reducing Space Science Mission Costs Eberhardt Rechtin University of Southern California Introduction These remarks have two objectives. The first is to highlight the areas of greatest opportunity for cost reduction. To do that, I will need to distinguish between system-level and component-level opportunities, because by far the greatest ones are at the system level. True, a 25-cent transistor can ruin a billion dollar flight, but the use of that transistor came from a value judgment of the acceptable risk level for the system as a whole. The availability of most component-level opportunities, as it turns out, depends on prior system-level judgments. The possibilities list in the Aeronautics and Space Engineering Board challenge posed to the working group includes some of both. (This list is contained in the Statement of Task in Appendix A.) I will add three more: the definition of success, the use of software to reduce hardware and mission costs, and the use of ultraquality parts to reduce testing costs. The second objective is to suggest a dozen or so guidelines—heuristics—for reducing or avoiding costs. System-Level Possibilities By far the most important decisions that affect cost—as well as performance and schedule—are made in the very beginning of a project. As a well-tested heuristic states: In architecting a new aerospace system, by the time of the first design review, performance, cost and schedule will have been predetermined. One might not know what they are yet, but to first order, all the critical assumptions and choices will have been made that determine those parameters. A number of studies of past projects have concluded that 75-90 percent of the end cost was determined up front. Thereafter, unless there was a major mistake, the most that was affected later was between 10-25 percent. This is not small, but is typically within the uncertainty of the initial cost estimates. So what are the system-level parameters? An amended list of possibilities follows. They are more or less in decreasing order of potential cost reductions.

OCR for page 22
--> Modified List I System-Level Opportunities For Cost Reduction The definition of success. The client' s acceptable risk level; for example, the utility of back-up hardware and prototypes. The nature of the effort. Is it a single project or a product line (a continuing program); for example, is it a few large or many small spacecraft, an upgrading, or a planned product improvement? Life-cycle costs and their annual cost profiles. Improving schedule to reduce costs. Upgrading or block changes? Every one of these requires a value judgment by the client; that is, what is worth doing, what is acceptable, what is good, and what is not. But the client cannot make value judgments, even relative ones, a priori without some idea of what is feasible. The Definition Of Success For example, ask a typical client prior to conceptual design, ''What is most important: cost, performance, or schedule?'' And the answer is most likely to be either "Yes" or "It depends." The determination of what is feasible is largely up to the architect, who, of course, cannot make that technical determination without value judgments by the client as to what is wanted! A truly chicken and egg dilemma. Therefore, the name of the game in the beginning, when cost is being determined to better than an order of magnitude, is a joint process between client and architect to create a conceptual design that is both desirable and feasible. A client who insists on defining the mission and its cost up front is likely to get neither in the end. Therefore, to really reduce costs, the mission and its cost must be adjustable. There is a powerful heuristic that defines the initial ground rules: Don't (either the client or the architect) assume that the original statement of the problem is necessarily the best, or even the right one. So when is the right one stated? When has success been defined? In realistic terms, it is when there is mutual agreement that a complete, passable set of acceptance criteria have been developed for a plausible system. Only then is it even worth talking about costs. It may surprise you, but clients do not really deal in costs, they deal in worth, which has psychological, political, social, and personal dimensions in addition to

OCR for page 22
--> the financial. If you do not believe it, talk to your real estate or automobile dealer! The only thing worse than the client starting with a fixed, immutable definition of success is having a fuzzy one, say a set of performance goals. Predictably, the clients will be disappointed, the builders frustrated if not bankrupt, costs will come in seriously out of line, and only lawyers will be happy. So decide on the set of acceptance tests early and as a guideline, "Keep It Simple Stupid!" For the scientists here, what you are willing to accept very strongly affects systems costs. That last 10 percent in capability or elaborate detail can be a back breaker. An important thing to remember about cost reductions is that they come in different flavors. Some are true deletions of activity or equipment. Some are avoidances of overruns, cost uncertainties, and failures. Some are preplanned savings in foregoing options and eliminating contingencies based on better-than-expected progress to date. But with all the best intentions, the estimated costs may still be too high. So downsize the mission. More exactly, Scope! Scope! and Re-scope! Make sure that the project is not so far-reaching that success is inherently costly. Carefully define what is necessary and no more. If in doubt, cut! Once that is done, aggregate similar problems and partition the subsystems carefully. In partitioning, choose the elements so that they are as independent as possible, that is, elements with low external complexity and high internal complexity. For example, look for the reasons behind "contingency costs," determine the degree to which the system as a whole benefits from them, and then partition wherever possible to eliminate them. Many are due to suboptimizations that different structuring could eliminate. Cost Risk So far, so good. Then there is the question of cost risk. What variation in cost is acceptable? What are the "tolerance limits" of the nominal cost? How much "insurance" is required to keep that risk in bounds? What are the sources of that risk (poorly characterized commercial off-the-shelf [COTS], immature technology, accident-prone launch vehicles, Congress . . . )? There are a number of ways to reduce cost risk without increasing cost. It is a profitable opportunity to exploit. One powerful heuristic is: Simplify! If possible, have an alternate, simpler, competitive mission ready in the wings, preferably one with a similar cost-to-benefit ratio. That is, one of less cost and of acceptably less benefit. It is surprising what a bit of competition will do to keep costs within bounds. One common alternate to a new system is a somewhat less capable upgrade of the old one. Another is a completely different system that essentially accomplishes the same purpose. The longer a competitive alternative is available, the lesser is the cost risk for the mission as a whole.

OCR for page 22
--> The Nature Of The Effort A critical value judgment is the level of commitment to the project that follows the current one. Is the current project a single, one-time project or part of a long-range program? Parenthetically, in the early 1960s Congress committed to a one-of-a-kind Apollo lunar mission, not to a continuing manned flight program, to the long-lasting frustration of NASA and considerable expense to the country. As an example, planetary flights in general tend to be one of a kind if for no other reason than that specific mission opportunities are so far apart in time (and often so different) that to do the next mission at a minimal cost requires major changes in system architecture and technology. On the other hand, Earth satellites, for both national security and scientific purposes, tend to be multidecade programs exploiting a common architecture. Unfortunately, perhaps, but for understandable reasons, Congress very seldom commits to long-range programs, only to projects. An unstated assumption to the contrary can be very costly. There are a number of ways to "hedge" those costs: design in options such as the "scars" for airliner stretching, or partition the system into relatively autonomous components. Define success in reachable, intermediate steps. Life-Cycle Costs Life-cycle costs, while theoretically worthwhile, all too often run afoul of the governmental reality of annual appropriations. More than one project has had cost overrun badly because its year-to-year funding did not match efficient implementation. The government does not run on life-cycle costs. It runs on cash flow. No cash, no flow. Designing on the basis that minimal life-cycle costs will "automatically" generate annual appropriations almost never works. Annual appropriations come from continuing needs and go to "programs" that satisfy them—communications, weather, surveillance and, most recently, navigation. The principal value of fife-cycle costs is that it enforces a consideration of operations costs, particularly the costs of testing and failure. It is unfortunate but true that most cost estimates in proposals are based on everything going according to a (life-cycle) plan. This is a near impossibility. It is not yet common practice to try to reduce cost by improving quality, though life-cycle cost strategies mandate it. In short, quality makes money. Improving Schedules To Reduce Costs From time to time assertions have been made that project costs can be saved by optimum scheduling. For example, NASA in its early years maintained that a specific Apollo schedule would result in the least cost, with greater costs on either side of it. In another case, an Air Force manager decided that the way to

OCR for page 22
--> reduce cost was to shorten the schedule and enforce it, regardless. Another insisted that tightening schedules always increased costs. Unfortunately, data from many projects do not support the one-to-one relationship assumed by any of these strategies. There are too many other factors involved, among them erratic cash flow, changes in procurement regulations, technological mismatches, unexpected events in lengthy projects, responsive management, and so on. A more profitable area for cost reduction through scheduling is likely to be the timing of projects relative to each other. Permitting every project or program to have its own funding profile over time results in peaks and valleys in the sum of all of them. Timing them relative to each other, for steady cash flow for example, means that not every project can be scheduled for the same time, much less when it wishes. There have to be years of the planetaries, years of the orbiting telescope, years of Earth observation, years of manned flight. If all parties understand and agree, more or less, to such macro-scheduling, and can plan for it, the frustrations, tragedies, and costs of "no new starts" and the miseries of missed opportunities might be considerably reduced. Galileo is only one example of many in the terrible 1980s for planetaries. To work, of course, each project must stay within its cost boundaries and, if not, take the consequences. And then there is cost allocation. For example, to achieve a reasonable balance between mission effectiveness and launch success rate, the ratio of spacecraft to launcher costs should be between 4 and 5 to 1. Higher than that risks the loss of a very high-value spacecraft due to a relatively cheap, relatively unreliable launcher. Lower than that invests too much in a support function (launching) that has little to do with the final mission of exploration, communication, intelligence, etc. One thing for sure, changing launch vehicles in midstream is very expensive. Some common causes are spacecraft weight growth or misestimation, launch vehicle fleet grounding, failure to meet a deadline, etc. Upgrading Or Block Changing? The effectiveness of upgrading as a way of cost reduction—as opposed to block changes—depends on where in the upgrading S curve the item is. Continually upgrading relatively mature hardware is likely to be less cost-effective than is continually upgrading rapidly evolving software. Note too that software upgrading can be, and has been, done remotely in flight—a major advantage for long-duration flights. On the other hand (all new) block changing of hardware is more straightforward than block changing software because the latter is more dependent on backward compatibility. The system-level value judgment here is the value to the client of being able to upgrade (or downsize) the mission, both before and after launch, and to exercise built-in options depending on the situation at the time. It might seem that such flexibility would automatically be welcomed. But although the natural inclination of design engineers is for plenty of options, the

OCR for page 22
--> natural inclination of operators and clients is almost the opposite—no changes at all unless components have checked out ahead of time. What is it worth to the client? Better not prejudge. Component-Level Possibilities Modified List II Component-Level Opportunities For Cost Reduction The use of software to reduce hardware and mission costs. The use of ultraquality parts to reduce testing costs. New technology versus space-qualified technology. Military specification versus screened commercial parts (COTS, pro and con). Piggyback payloads, etc. I will address here only the first component. Using Software To Reduce Hardware And Mission Costs Where are the opportunities for using software to reduce costs? The most apparent opportunity is to use "smart" systems, both on the ground and in flight. For many missions, software can increase dramatically the mission's worth per dollar spent, even to the point where previously anticipated flights are not needed. And that, if planned in time, can be a real cost reduction. Another possibility is to replace hardware with software in the interests of greater reliability and lower cost. Clearly the ratio of hardware to software, expressed as costs, has changed dramatically in the last decade from a cost ratio of 10:1 to 1:2. Part of this change is justified on mission effectiveness grounds; "smart" systems are not only more valuable, they cost less at the margin. Very soon software will be at the center of all real-time, software-intensive systems. Earth satellites and planetary spacecraft are certainly examples. However, as software becomes a larger and larger part of mission costs, it becomes, and should become, a primary target for cost reduction. The question is how. It is a complex and controversial subject, perhaps better for a subsequent workshop. But from my point of view, the answer lies in new software architectures, reusable mission-certified modules, and progressive, integrated modeling. For example, proponents of reusing software modules claim great savings. However, industry experience with COTS equipment and software has been mixed. On the one hand, the lack of knowledge of the details of a COTS component increases system risk, just as it did in the case of space lubricants whose composition was proprietary and unreleaseable. COTS "standards" can create otherwise unnecessary constraints on systems design. Supplier support for

OCR for page 22
--> specialized hardware and consumer software may evaporate with time or if the COTS is modified in any way. As the heuristic states, "COTS is COTS," period. On the positive side, the use of COTS ought to reduce development time (and costs), and in a world of "time to market" economics, time is money. All things considered, it would seem the better part of wisdom to use only COTS coming from the same or similar programs and subject to the same acceptance tests. Heuristic Guidelines For Cost Reduction1 Simplify! Keep it simple, stupid! (KISS). Efficiency is inversely proportional to universality. Scope! Scope! Scope! Success and risk are defined by the beholder, not the architect. In architecting a new program, all the serious mistakes are made in the first day. In architecting a new system, by the time of the first design review, performance, cost, and schedule have been predetermined. One might not know what they are yet, but to first order, all the critical assumptions and choices have been made that will determine those key parameters. Do not assume that the original statement of the problem is necessarily the best, or even the right one. The most dangerous assumptions are the unstated ones. Any extreme requirement must be intrinsic to the system's design philosophy and must validate its selection. Everything must pay its way onto the airplane. In partitioning, choose the elements so that they are as independent as possible, that is, elements with low external complexity and high internal complexity. Choose a configuration with minimal communications between the subsystems. "Proven" and "state of the art" are mutually exclusive qualities. Complex systems will develop and evolve within an overall architecture much more rapidly if there are stable intermediate forms than if there are not. In any resource-limited situation, the true value given a service or product is determined by what one is willing to give up to obtain it. The bitterness of poor performance remains long after the sweetness of low prices and prompt delivery are forgotten. Quality makes money. 1    Rechtin, E., and M. Maier. 1997. The Art of Systems Architecting. Boca Raton, Fla.: CRC Press.

OCR for page 22
--> Perspectives On Small Spacecraft: Results Of A Recent RAND Workshop1 Liam P. Sarsfield RAND Critical Technologies Institute Overview Small spacecraft are becoming an increasingly important element of civil space policy. RAND is currently studying small spacecraft missions and their future in space science missions. This briefing provides a review of this on-going study and highlights from a recent RAND workshop covering trends in the development of small spacecraft. This material is presented as an in-process briefing. Results discussed here are preliminary and are provided in support of ongoing intellectual discourse on methods to reduce the cost of space science missions. Background Of The Study The purpose of RAND' s current work on small spacecraft is to review current and future programs to assess the efficacy of various development approaches and to gain insight into the performance of small spacecraft. In preparing the course of study, three key questions were established to guide the review of programs: What role are small spacecraft playing in civil space programs? What strategies have proven especially effective in reducing cost and increasing performance of small spacecraft? What role does advanced technology play in the process of building small spacecraft? At the conclusion of the study, policy options will be synthesized and recommendations prepared regarding future directions for small spacecraft programs. The study focuses on NASA science spacecraft with a dry mass of under 500 kg. Some analysis of DOD unclassified programs is also included. Although the definition of what constitutes "small" is arbitrary, the 500 kg. limit captures the programs that are generally considered small within both NASA and DOD. Occasionally larger programs are referred to for comparative purposes. The Role Of Small Spacecraft Small spacecraft have become important policy tools. We already rely on small spacecraft to meet many of our national objectives in space and this trend is likely to continue. Small programs offer opportunities for international cooperation, in civil space most notably, but also possible future military cooperation in 1    From Cosmos on a Shoestring. Washington, D.C.: RAND Critical Technologies Institute. MR-864-OSTP (in press). Reprinted with permission.

OCR for page 22
--> communications and remote sensing. As missions become smaller these programs will also carry an increasing burden of helping the U.S. maintain a skilled aerospace workforce. On the economic side, small spacecraft currently represent a sizable national investment. Figure C-1 provides an estimate of NASA spending in FY96 for (1) all spacecraft research programs, and (2) for small spacecraft programs within the scope of this study.2 Cost breakdowns are provided in five areas; hardware (development of flight systems), launch systems, operations, research and analysis (R&A), and personnel. Approximately $4B is spent on spacecraft research programs within NASA, with $1B devoted to small spacecraft missions. Small spacecraft are also of economic importance because they provide a means of conducting science at lower cost. They can be built faster requiring less engineering and testing; as a result they are cheaper. They also provide platforms for testing cost reduction strategies that have implications for programs of all sizes. Competitiveness is another economic advantage inherent in small spacecraft. Built on faster timelines, small spacecraft are better able to approach the state-of-the-art. Since scientific objectives have remained ambitious spacecraft developers have been forced to aggressively pursue technology to increase performance. One of the most important characteristics of small spacecraft, however, centers around a set of technical advantages. Firstly they are responsive to scientific needs. The ability to return results in 24 to 36 months is especially important in the realm of space science. Before budgets shrank spacecraft had already started to become leaner, a trend initiated by the demands of the science community: "Rapid, elegant response is imperative . . . science is not best served by exclusive emphasis on major missions." Crisis in Space and Earth Science NASA Space and Earth Advisory Committee, 1986 "Efficient conduct of science and applications missions cannot be based solely upon intermittent, very large missions that require 10-20 years to complete. Mission time constants must be commensurate with the time constants of scientific understanding, competitive technological advances, and inherent changes in the systems under study. . .. NASA's new initiative for smaller, less expensive, and more frequent missions is not simply a response to budget pressures; it is a scientific and technical imperative." Improving NASA's Technology for Space Science NRC, 1993 2    This is a crosscut of the NASA budget prepared using RAND's RaDiUS (Research and Development in the United States) data base, which contains detailed budget information on the federal budget. Spacecraft costs are a summation of development costs for flight segments within the respective crosscuts. Launch costs include vehicle procurement and spacecraft-to-vehicle integration costs. Operational costs represent an aggregate of individual mission operations and data analysis (MO&DA) costs, construction of facilities (COF), and ground segment line items. Research and analysis (R&A) is a simple accumulation of these identified line items. Personnel costs are estimated as a fixed percentage of overall research and program management (R&PM) accounts.

OCR for page 22
--> Figure C-1 NASA spending on small spacecraft.

OCR for page 22
--> Eliminate any unnecessary military standards and specifications, and facilitate the use of commercial best practices. Increase the dollar limit for noncompetitive contracts, and make users and operators accountable for funding and system cost. Then provide incentives for government and contractors to reduce cost—don't penalize your program team for taking calculated risks. Manage The Requirements Process To Facilitate Reducing Cost Identify and implement a process for managing requirements that provides the user or customer what they need; however, generate fiscally responsible requirements, and make a concerted effort to identify the truly difficult and costly requirements. Bundle program requirements to facilitate affordable systems. Establish a timely process to trade on the requirements and negotiate acceptable compromises by motivating key players in government and industry to identify tough requirements and provide options to change or meet them. Develop better integrated mission concepts, document them, and be willing and able to negotiate by integrating space into the everyday lives of users and operators. State your program requirements in a more constructive fashion by describing what is needed, not how to provide it, by including ranges of performance and by stating the rationale or reason for requirements—the goal is to communicate. Develop Cost-Effective Mission Concepts Recognize and facilitate different classes of missions and payloads, select the proper class of mission and payload, then implement accordingly. Perform up-front space mission engineering to develop innovative mission concepts using air, space, land, and sea resources, but implement them conservatively. The most important cost savings occur while deciding how to meet operational requirements, not how to implement a set of technical specifications. Do trades among mission elements early, as they provide the best opportunity for reducing cost. Examples include data processing, orbit insertion, propulsion, and autonomy. Make maximum use of cost-effective commercial products—look for commercial capabilities first, then remember that large missions can be done by using several large or many small spacecraft—many trade-offs exist. Technical risk and cost may increase when you put all of your eggs in one basket, for example, dollar per bit may be lower for one large integrated spacecraft, and the consequence of failure is also larger. Use a "design-to-cost" approach and adjust the mission concept, requirements, and design to meet a life-cycle cost goal. Emphasize Managing Programs Carefully select an experienced program leader and give him sole responsibility and accountability for development, test, and operations; then support him or her. Use committees to gather sage advice, wisdom, and good ideas but don't

OCR for page 22
--> make decisions by committee—when things go wrong, it turns out no one can remember being at the meeting! Select your program leadership based on the type of program and desired attributes for a specific program (large or small, technical or not, hands on, motivation, and skill mix). Build competitive hardware; do not focus on paper studies. Actively use prototyping and simulation where appropriate. Minimize documentation and reviews. Only develop and maintain necessary documentation. Zero-base documentation works, but it is painful. Explore using existing contractor documentation and augment, if necessary, and reuse generic documentation. Track cost and schedule in near real time—work the problems in real time and don't accept schedule slips. Encourage mutual trust between the government and contractor team—integrated product teams can work well. Facilitate easy communication among all players. Determine the appropriate approach for government interaction with contractors (small government and contractor program office—separate or joint). Use concurrent design and fabrication judiciously and avoid jointly funded programs. Manage the requirements and design change process with an iron fist once you have selected the proper philosophy (in any event do not allow changes on changes to go unchallenged). Government program managers can save time and money by using contractor/contracting officers to procure hardware, software, and services as well as some facilities. Be wary of reducing contract funding by 10 percent each year because it causes a very destructive phenomenon. Consider canceling lower-priority activities and leave others unchanged. A compressed schedule can reduce overhead of a "standing army." It forces your program to move rapidly and can reduce cost. A tight schedule can be a wonderful excuse to expedite procurement. Reducing the budget should result in reduced capability. If not, you are reducing program margin and increasing risk. Use increased spacecraft margins to reduce cost because it provides more flexibility and makes the system more robust during the development and operations phases. It can also reduce operations, engineering, and manufacturing costs. Make maximum use of cost-effective commercial products by looking for commercial capabilities first. Use of commercial off-the-shelf items should be strongly encouraged. Share cost among nations, organizations, and companies. This may reduce the cost of one piece, but be sure that the overall cost will be higher. More interfaces usually imply more complexity and higher cost. Incorporate Spacecraft Design, Development, And Test Strategies Develop and use standard interfaces where possible. Automate appropriate spacecraft functions to reduce life-cycle cost. Automating the wrong ones will drive costs up. We know that automating things such as anomaly detection can

OCR for page 22
--> work, but anomaly resolution has been less successful. Automating some functions on the spacecraft to ease operations cost has, in some cases, had the opposite effect. Consider weight-optimized, smaller spacecraft versus fewer larger spacecraft to reduce launch cost. This may drive spacecraft cost up. Develop designs that are robust to known or anticipated changes (historically based). Shoehorning software into a computer and increasing the speed at which a code operates is one of the most expensive things we can do. Mass, power, and throughput must be robust. Make considered, maximum use of existing capabilities and infrastructures if they are cost effective. Use the 80 percent rule in developing multiuser systems; but trying to be all things to all users drives the cost through the roof. Don't sacrifice on integration and test; however, it is possible to eliminate some development and performance tests if qualification tests are used. Emphasize validation and testing from day one. Include Mission Operations And Ground Infrastructure Concepts Develop and use standard interfaces, protocols, and procedures. Software and procedure reuse coupled with up-front participation can reduce cost significantly. Automate appropriate functions to reduce cost and enhance reliability. Eliminating one low-cost operator and replacing him, or her, with a high-cost software maintainer is not cost effective in many cases. Use automation to enhance reliability and reduce life-cycle cost. Carefully consider data flow to minimize organizations and steps. Review data push and data pull approaches. The data pull approach has the potential to reduce the cost of data. Implement an accounting system for mission operations and make the developer and operator responsible for the cost of operations. Make as much of operations as possible direct costs. Allow adequate spacecraft margin needed for expensive analysis during development and operations. This makes the spacecraft more operable and less expensive to operate. Check mission and spacecraft operability prior to committing to mission and spacecraft design. Periodically revisit the design throughout development. Identify Technology Focus on technologies that provide savings in the "-ilities"—producibility, testability, reliability, and operability—are prime. Fly operational demos in addition to tech demos—the philosophy and approach may be different. Technologies to reduce mission operations cost include autonomous orbit determination and maintenance, on-board data processing and health monitoring, standardized communication interfaces, use of spacecraft command language, and on-board

OCR for page 22
--> solid-state memory. Technologies to reduce the cost of space missions include those that improve up-front development of mission concepts, operations planning, and systems engineering. For example, miniaturization of electronics, solar electric power generation, and electric propulsion, autonomous navigation of spacecraft, the Global Positioning System, or other technologies. Summary We share many of the same problems; however, we can work together to solve many of them. We've seen many useful approaches and many examples of how to reduce cost, but the strongest approach is to select a combination of approaches that suit a program's particular needs.

OCR for page 22
--> Mars Exploration Program Strategy: 1995-20201 Donna L. Shirley and Daniel J. McCleese NASA Jet Propulsion Laboratory Abstract In the wake of the failure of the Mars Observer mission in 1993 a long-term program of robotic exploration of Mars was established. The themes of the Mars Exploration Program are to understand Life, Climate and Resources on Mars, with these themes tied together by the common thread of Water. The Mars Exploration Program comprises at least one Discovery mission (Mars Pathfinder), the Mars Surveyor Program, plus sample return missions and other missions to prepare for possible human expeditions to Mars. The program will launch (on average) two missions every 26 months. The missions launched between 1996 and 2001 will include a lander and an orbiter at each opportunity, launched on the Delta family of launch vehicles. International participation is an important factor in the program, and relationships are being established with Russia, Europe and Japan. The program is severely cost constrained, with missions costing about $150M apiece or less, including launch and operations. The "Water Strategy" The Mars Exploration Program will continue the exploration of the red planet which has fascinated humankind for thousands of years. Robotic spacecraft began visiting Mars in 1965, and landed on the surface in 1976. Mars was found to be a planet of stark contrasts. The surface features range from ancient, cratered terrain like Earth's moon, to giant volcanoes and a canyon as long as the United States is wide. The atmosphere is less than 1 percent as dense as Earth's, but there are constant polar caps with reservoirs of water ice. Close-up, Mars resembles an earthly desert like California's Mojave, but there is evidence that water once flowed and cut channels on the surface. 1    Copyright © 1997 by the American Institute of Aeronautics and Astronautics. Reprinted with permission.

OCR for page 22
--> NASA Administrator Dan Goldin and NASA Associate Administrator for Space Science Wes Huntress have agreed on a strategy for the exploration of Mars for the next 10 years. The strategy is to explore and study Mars in three areas: evidence of past or present life climate (weather, processes and history) resources (environment and utilization) Mars will be our first footfall in the search for life beyond Earth. We began our search for life on Mars using the tools of astronomy. As we extend our reach with new astronomical tools in search of planets and life in other solar systems, Mars remains our touchstone for understanding planetary evolution different from Earth's. Robots and humans will go to Mars to explore intensively. We seek the markers of life from which we will learn how to find and study hospitable worlds. Using both remote presence and physical contact, our skills will be honed and our reach lengthened sufficiently to understand whether we occupy a unique place in the universe or one of many such places scattered throughout it. If life ever arose on Mars it would almost surely have been connected with water. And understanding the water-connected processes which led (or didn't lead!) to life on Mars will help us understand the potential for life elsewhere in the Universe. The climate and resources themes are also connected with the search for water on Mars. When and where was water present in the past, and what is its current form and amount? We know from previous missions that the Martian polar caps include water ice as well as frozen carbon dioxide. The Viking and Mariner 9 orbiter images show evidence of past great floods (the Pathfinder lander is planning to land in such an area) and of dry rivers and lake beds. Where did the all the water go? Water is key to climate both on Earth and Mars, and understanding the history of the Martian climate will help us understand better the Earth's climate change processes. Water will also be a major resource for future human exploration of Mars, and if we understand how Mars evolved (including discovering the sources and sinks of water, past and present) we may be able to locate reservoirs of water for human use. A Series Of Missions To Build Up "Water" Knowledge Our exploration of Mars for greater understanding of life, climate and resources will focus, in large part, on the study of water and its role in the history of the planet. How do we go about finding out about water on Mars? Dr. Daniel McCleese of JPL, the Mars Exploration Program Scientist, and Dr. Steven Squyres of Cornell, the head of the Mars Science Working Group, led that group to define a strategy for the "water search." They looked at how small Mars

OCR for page 22
--> orbiters, landers, ''networks'' of landers, and sample returns could be combined in a logical progression of missions that will build up an understanding of how water has existed and is existing on Mars today. Small orbiter missions will search for accessible water (we know that ice is accessible at the poles, but are there reservoirs underground or in the soil?). They'll search for ancient sediments and hydrothermal deposits (dry lake beds and hydrothermal vents). They'll provide data needed to understand the present Mars climate and study how water escapes from the atmosphere into space. The orbiters will also study the surface of Mars and identify good landing sites for the landers, and provide radio links between the landers and the Earth. Small lander missions will search for carbonates and evaporites that could only have formed in the presence of water. Landers can investigate water reservoirs in detail: for example, they can measure the amount of water which is in the soil, or examine the polar ice caps (using drilled core samples and electro-magnetic sounding) to see how, when, and how much ice was laid down. Investigation of surface chemistry and how the rocks and soil have "weathered" will tell us about the past climates. And the landers may find organic compounds or even evidence that tells us whether life was ever present on the surface of Mars; and if not, why not. "Networks" of more than a dozen very small landers scattered over the planet could be used as weather stations to study Martian weather and the circulation of its atmosphere. If the network landers also have seismometers on board, and if they detect "Marsquakes," that information will tell us about what Mars is like deep in its interior, and how the interior has evolved over time. Mobility will be important for understanding the Martian surface and accessing features of particular interest, so missions involving long range rovers and balloons are being studied. Finally, sample return missions can bring rocks and soil to laboratories on Earth for analysis by our most sophisticated instruments (too large and massive to send to Mars) which can tell us about the chronology of the planet's evolution, and may even allow us to detect compounds which could have led to life, or which are evidence of past life. (The odds of being able to select a rock with a fossil, however, are very low, even if fossils exist on Mars.) Baseline Mission Set All of these missions must be done within the very fight cost constraints of the Mars Exploration Program. The entire program over the next 10 years will be conducted for about one-third the cost of the Viking missions which orbited and landed on Mars twenty years ago. Each mission will cost about the same as a major motion picture, and the total cost of the first 10 missions to Mars will be about that of a single major military aircraft. The Mars Science Working Group laid out a "strawman" strategy for fitting

OCR for page 22
--> the science goals into a set of missions which can gradually build up our knowledge of Mars over the next 10 years. This set of missions has evolved over the past year to that shown in Figure C-6. Figure C-6 shows the Mars launch opportunities from 1996 through 2005. The bottom half of the chart is the "U.S.-Only" component of the program, while the top half is actual or potential augmentation by international partners. The "baseline" missions are clear, with possible alternative missions or augmentation shaded. The numbers in parentheses are, respectively, the development cost, operations cost, and approximate launch costs of each year's mission set. Mars Pathfinder will be the second mission in the series of NASA' s Discovery program of planetary exploration missions. It was launched in December 1996 on a McDonnell Douglas Delta II 7925 rocket (capable of throwing about 1000 kg to Mars). Mars Pathfinder will fly directly to Mars and plunge into the atmosphere at 17,000 mph without going into orbit. Using a combination of a heat shield, parachute, rockets and airbags, Pathfinder will land on the surface in an ancient flood plain which is expected to be littered with a wide variety of rocks. Pathfinder will image the Martian terrain in 13 different colors, monitor the weather, and deploy a small rover to explore the region around the lander and measure the composition of the surface. Figure C-6 Mars Exploration Program strategy.

OCR for page 22
--> Mars Global Surveyor, which was launched in November 1996 (also on a Delta II 7925), will go into orbit around Mars in September 1997. It will use "aerobraking," skimming through the upper part of the thin Martian atmosphere, to go from a long, looping orbit into a circular polar orbit. Mars Global Surveyor will scan the surface of Mars for a full Martian year (about two Earth years) using 6 of the 8 instruments that were originally flown on Mars Observer (which was lost in 1993—the first planetary spacecraft failure in 27 years). Mars Global Surveyor is the first of a series of missions called the Mars Surveyor Program. This program will fly two missions to Mars every opportunity (about every 26 months), and, with Pathfinder, is pioneering the "better, faster, cheaper" approach to planetary missions. Through competitive procurements, Lockheed Martin Astronautics of Denver, Colorado, has been selected as JPL's industrial partner for Mars Global Surveyor, and for at least the subsequent set of Surveyor missions to be flown in 1998. In late 1998, Mars Surveyor '98 will launch an orbiter and a lander on a Delta 7325 "Med-lite" launch vehicle. (The Med-lite will only throw about 565 kg to Mars, but is expected to cost considerably less than a Delta 7925.) The orbiter will carry a Pressure Modulator Infrared Radiometer (PMIRR) to map atmospheric temperature, water vapor, and dust over a full Martian year. This instrument was also previously flown on Mars Observer. The lander launched in 1998 will come to rest near the south pole of Mars and will carry a payload, including a robotic arm, which will excavate Martian history by trenching down through thin layers of dust (and possibly ice) deposited in the layered terrain. The polar lander will also chemically analyze the soil, including a search for organic molecules. The final element of the lost Mars Observer payload (a gamma ray spectrometer) will search for water in 2001 on the Mars Surveyor '01 orbiter. Also to be launched in '01 is a lander which may explore the ancient highlands of Mars in areas where water is thought to have once flowed. The 2001 lander may analyze rocks to determine the ancient history of the climate and geology of Mars. The 2001 orbiter will be launched on a Delta 7325, but the lander may be launched on a new "Delta-lite" configuration which will reduce the lander's mass allocation. The 2001 mission may be conducted in partnership with the Russians, with the orbiter being launched on a Russian Molniya. This "Mars Together '01" launch would also include one or more Russian landers. A Russian lander could include a large rover, or perhaps two ''small stations" of the Russian Mars '96 mission type. In 2003, the Mars Surveyor Program is exploring a partnership with the European Space Agency (ESA) to launch three U.S. landers carrying international payloads. These landers, plus a communication orbiter, provided by ESA, would be launched on a European rocket (Ariane 5). This joint NASA/ESA mission is called InterMarsNet. The landers would explore the interior of the planet using seismology to detect "Marsquakes," study geochemistry at three sites, and act as weather stations. In addition, a separate United States mission may be flown,

OCR for page 22
--> perhaps deploying a "network" of complementary and very small weather stations around the planet. Mars Surveyor '05 mission may be the first in a series of missions to return samples from the Martian surface. Another possible sample return target is the Martian moon, Phobos. The Russians are especially interested in a Phobos sample return mission. Sample return missions, in general, will probably require violating some of the constraints of the Mars Surveyor program. They may be too expensive to be completely funded by the Mars Surveyor Program, and/or they may require violating the "two-launch" per opportunity rule. Therefore, these missions may be in partnership with the Russians and/or Europeans. A continuing program of robotic missions, including the return of samples, over ten years or so will pave the way for future human exploration. New Technology Infusion More instruments can be carried, or more landers and orbiters sent, if new technology improvements can be introduced into the U.S. spacecraft to make them smaller, lighter and cheaper. The Mars Pathfinder mission has introduced a new flight computer, based on the commercial IBM/Loral RS6000 computer, which will be the basis for the computers of a number of future planetary missions. This provides an enormous increase in computational power. Pathfinder is also utilizing a commercial operating system for its computer, and has pioneered a concurrently engineered flight/ground data system which has greatly reduced costs. Pathfinder has also pioneered a low cost entry and landing approach, of which all but the final airbag impact system is being baselined for future Mars missions. Mars Global Surveyor is utilizing a composite structure for the spacecraft, although its electronic systems and instruments are inherited from Mars Observer. A program called "New Millennium" is currently being planned to develop and demonstrate the next generation of space technologies to reduce costs and improve performance for both planetary and Earth missions. The Mars Exploration Program will be a "customer" for this new technology, and some of the New Millennium demonstrations may "piggyback" on Mars missions. For 1998 the feasibility of the Mars Surveyor '98 lander carrying one or two New Millennium ''microlanders" to Mars is being studied. Investment strategies are being developed by the Mars Exploration Program at JPL in partnership with Lockheed Martin, the New Millennium Program, and NASA's Office of Space Access and Technology (OSAT). Technology investment is required to shrink the '01 and '03 landers so that they are compatible with the limitations imposed by the Delta-lite launch vehicle, and the even more stringent limitations of the InterMarsNet mission. With the current InterMarsNet concept the U.S. landers must mass no more than 415 kg each, which means that the landers must decrease 150 kg from the 565 kg in 1998. Key technology advances are required to accomplish this mass reduction, while hopefully maintaining or increasing the payload fraction. These advances center around the electronics: an

OCR for page 22
--> advanced flight computer and memory, a small deep space transponder (X-band), light weight batteries, a high efficiency solid state power amplifier, advanced power electronics, and an inertial fiber-optic gyroscope. In addition, because of the harsh environment on the surface of Mars, technology advances in temperature tolerant electronics and light weight insulation are required to enable long-term lander missions. The Pathfinder rover has pioneered an approach to light weight insulation using silica aerogel, however phase change materials are expected to be necessary for future missions. The Pathfinder landing ellipse is about 70 by 150 km and its rover can only travel a few hundred meters. The lander mission in 2001 is expected to require much more accurate landing (<< 50 km landing ellipse) and considerable mobility (10s of km) to enable access to ancient lakebeds which may hold clues to the climate history of the planet. Advances in sample collection and storage, and in sample return technology (such as utilization of the atmosphere to manufacture fuel) will be required to enable low cost sample return by 2005. Preparation For Human Missions Each of the robotic missions in the Mars Exploration Program will be gathering information needed to plan future human missions to Mars. The robots will find and scout safe and interesting human landing sites, characterize the atmosphere and surface environments so that human missions can be designed properly, look for water and other resources needed by humans, and develop technologies (such as very low mass electronics) which will be important for human space flights to Mars. Over the next couple of decades the robotic part of the Mars Exploration Program will result in a detailed understanding of Mars, which is of interest not only to scientists but to understanding more about the Earth's environment, and eventually, for future human exploration. NASA is currently developing a long range "road map" for the human exploration of Mars. The road map builds upon the capability of the international space station to understand how people can live and work in space. Trips to Mars will utilize new launch vehicle technologies currently just beginning development, including re-usable and expendable rockets. The use of commercial technologies such as advanced electronics will greatly reduce the cost of human exploration of Mars. A current goal of this road map is to enable the first human Mars mission in 2018. Humans on Mars, in partnership with robots, will explore the planet in more detail than robots alone can. Human presence may be required to finally answer the question of whether Mars has or once had life, and humans will seek to understand the implications of the answer to that question for the possibility of life elsewhere in the universe. Humans will utilize the resources of Mars to investigate how the planet can be made more easily habitable for future generations. And finally, our grandchildren may become citizens of Mars.