Verification and Validation of AI Systems . . .

  • Published 1997

Abstract

NASA is developing technology for the next generation of deep-space robotic spacecraft, with the aim of enabling new types of missions and radically reducing costs. One technology under development is Autonomy: highly capable spacecraft that perform significant scientific missions with little or no commanding and monitoring from Earth. Artificial Intelligence provides a basis for autonomy technology, but raises issues of verification and validation outside the scope of empirical testing technology for conventionally commanded spacecraft. This paper describes research towards extending formal methods verification techniques for the mathematical verification of AI systems controlling deep-space spacecraft. This paper first overviews a planned space mission called DS-1 which includes an AI-based autonomy experiment. It then describes part of this AI system called the executive, which includes an ‘intelligent’ operating system based on goal-oriented constructs. The paper then describes focused research on applying and extending model-checking technology for verifying both the core services of the executive and the concurrent task programs run by the executive. 1 NASA’s New Millennium Program The successful landing of Mars Pathfinder on Independence day (July 4, 1997) signalled a new era in man’s exploration of the solar system: faster, better, and cheaper. The Mars Pathfinder project was completed in four years, delivered widely sampled geological data from a mobile rover and cost just $250 million (1997) dollars. In contrast, the two Viking missions of twenty years ago took over eight years to develop, delivered data from fixed landers and orbiters, and cost over $3 billion (1997) dollars. The Mars pathfinder project took advantage of off-the-shelf technology to reduce development costs. NASA is preparing for an order-of-magnitude expanded space exploration program in the next decade within the constraints of a flat-lined budget. One key aspect of this plan is the New Millennium program: a series of technology validation flights whose objective is to accelerate the flight-qualification of new spacecraft technology. For example, new generations of radiation-hardened microprocessors, based on commercial designs, will be flight-qualified in New Millennium missions. Up to now the functional performance of space-qualified hardware has often lagged a decade or more behind commercial hardware. New Millennium will greatly accelerate the space-hardening and space qualification of new technology. This will reduce development costs for subsequent science-oriented missions and enhance the technology base for these missions. The New Millennium program is also aimed towards decreasing operations costs while enhancing science return. Operations costs are largely determined by two factors: launch weight and personnel. Microelectronics and other miniaturization technology can greatly reduce the Slightly revised version of paper that Foundations of Intelligent Systems, (Eds. Z.W. Ras, A. Skowron), TenthInternational Symposium on Methodologies for Intelligent Systems, Charlotte,North Carolina, October 15-18, 1997, Lecture Notes in ArtificialIntelligence, Springer-Verlag, Vol. 1325. appeared in:

02040'98'00'02'04'06'08'10'12'14'16
Citations per Year

140 Citations

Semantic Scholar estimates that this publication has 140 citations based on the available data.

See our FAQ for additional information.

Cite this paper

@inproceedings{1997VerificationAV, title={Verification and Validation of AI Systems . . .}, author={}, year={1997} }