National Aeronautics and Space Administration
Small Business Innovation Research 2000 Program Solicitation

TOPIC 24 Thinking Space Systems

[ back to Solicitation ] [ to Chapter 8 ]

24.01 Automated Reasoning for Autonomous Systems
24.02 Human-Centered Computing

The Thinking Systems Cross Enterprise has the goal of enabling better, faster, cheaper, more reliable space missions by extending the scope of decisions and actions that can be done under computer control. The Cross Enterprise enables unmanned missions to accomplish more by making better autonomous decisions, and better interpretation of the science data that is brought back. It enables manned missions to be cheaper and safer by providing more sophisticated interactions between astronaut and machines. Finally, it enables ground operations to be cheaper and faster, by allowing a reduced ground operations team to send more complex high-level instructions. In all cases, the Cross Enterprise will support both earth orbit and deep space missions. The Cross Enterprise also provides an infrastructure in model-based representation and reasoning, software validation and verification, and human-centered computing that will support a wide range of NASA programs. Technologies are sought that enable curious, self-reliant, self-commanding space systems that plan and conduct measurements based on current or historical observations or inputs; recognize desired phenomenon and concentrate observations or activities accordingly monitor and maintain desired status/configuration for long periods of time without frequent communication with ground.

24.01 Automated Reasoning for Autonomous Systems
Lead Center: ARC
Participating Center(s): GSFC, JPL

NASA is planning to fill space with robotic explorers, carrying our intelligence and our curiosity outward in ways never before possible. To survive decades of operation, these remote agents need to be smart, adaptable, curious, wary, and self-reliant in harsh and unpredictable environments. NASA is soliciting research in automated reasoning for autonomous systems that will enable the design, construction and operation of a new generation of remote agents that perform progressively more exploration at much lower cost than traditional approaches.

NASA also needs automated reasoning to improve its operations closer to home. For instance, software for monitoring shuttle and space station systems and diagnosing faults when they occur or software agents for processing, classifying and archiving the mountains of data from earth orbiting satellites. Specific areas of interest for automated reasoning include the following:

Agent Architectures

Autonomy architectures that support plug and play of automated reasoning components.
Architectures for homogeneous and heterogeneous distributed systems of agents.
Capabilities related to Autonomous Performance.
Planning and scheduling systems that support planning concurrent with execution, plan optimization, resource management and/or distributed plan creation capabilities.
Model-based and statistical methods for monitoring, command confirmation, fault isolation, and diagnosis from sensor information.
Methods for robust recovery & repair.
Algorithms for real-time deduction and search.
Novel environment sensing or mapping capabilities.
Machine learning and adaptive control technologies.
Methods for precisely and dynamically adjusting the level of human control.

Capabilities Related to Design

Declarative specification of software and hardware behaviors, collaborative environments for large scale model building.
Methods for code synthesis and controller generation from declarative specifications.
Automated generation of test sequences from component models and analytic verification methods, including model checking and theorem proving.
Methods for modeling, code synthesis, simulation, testing and validation, as above, that operate from hybrid discrete/continuous models.

[back to top]

24.02 Human-Centered Computing
Lead Center: ARC
Participating Center(s): JSC

NASA is planning to use human explorers, highly trained with scientific and technical skills, to explore our solar system in ways never before possible. To survive years of living and working in space, these astronauts need to be outfitted with life support, data gathering, and spacecraft tools that will enable them to be productive and thrive in harsh and unpredictable environments. Not the least of these astronauts’ concerns will be coping with breakdowns and uncertainties in operating the increasingly complex technologies of spacecraft, rovers, and habitats, which will require ongoing monitoring, control, diagnosis, and repair.

To achieve NASA’s ambitious exploration goals, researchers must develop robust control systems and exploration tools that can be understood by people, easily learned, maintained, and directed. For example, life support systems for either spacecraft or habitat systems must aid people in diagnosis and repair. Operations assistants, integrated into just-in-time training systems, will be necessary to help people understand the state of the system and help them correct errant procedures. The design of computer systems necessarily must take into account not only how people will "interface" with the systems, but fundamental aspects of human perceptual-motor coordination, cognitive operations, and group dynamics. Human-centered computing focuses on the "delta" -- what is the difference between the best computer systems and people? What are the particular contributions of humans and machines? How can we design machines and operational procedures to complement each other?

Human-centered computing is a design approach that integrates computational systems with human performance and capabilities, such that the total system amplifies, corrects, and leverages the capabilities of both people and machines. The architectural requirements of autonomous systems are required, plus fundamental theories of human perceptual, cognitive, and social systems that anticipate the context and contribution of human behavior in which technologies are utilized and maintained. Beyond this, the harsh realities of working in space environments must be thoroughly understood, so tools such as electronic notebooks, alarm systems, and scheduling systems are adapted to the living and work environment of a space habitat or planetary surface rover. To advance along these lines, proposals are sought in the following areas:

Perceptual Performance Enhancers

Visualization tools combining "virtual reality" projection with actual objects in the environment, conveying information about object identity, part relationships, and assembly or operational procedures.
"Cognitive prostheses" that qualitatively change the capabilities of human perception, pattern analysis, scientific domain modeling, reasoning, and collaborative activity. Such tools could incorporate any of a variety of modeling techniques such as knowledge-based systems and neural networks, and fit tool operations to ongoing human physical interaction, judgment, and collaborative activity.
Robust, Mixed-Initiative Information Systems.
Advanced AI systems/architectures for mixed-initiative system planning, monitoring, and control, with provision for crew oversight.
Agent-based tools for information gathering, reminding, and alerting; job performance aids that provide cognitive assistance in the context of daily activities.

Collaborative, Knowledge Amplifiers for Scientists and Engineers

Information technology enabling comprehensive sharing of project-related information and data, which supports intelligent organization, access, and presentation of the information.
"Knowledge management" tools that relate technical models of human knowledge to: a) nonverbal concepts and perceptual skills; b) the daily activities of workers, including especially how databases are actually used in practice; c) informal on-the-job learning; and d) the career trajectories of novices, experts, and retiring employees.
Software systems that provide specialized support for collaborative science and engineering tasks, including design, data collection, experimentation, analysis, and model construction to enable scientists and engineers to collaborate as part of distributed project teams at physically separate sites.

[back to top]

[ back to top ] table of content [ chapter 8 ]