Autonomous software holds the promise of new operation possibilities, easier design and development, and lower operating costs. However, as those system close control loops and arbitrate resources on-board with specialized reasoning, the range of possible situations becomes very large and uncontrollable from the outside, making conventional scenario-based testing very inefficient. Analytic verification and validation (V&V) techniques, and model checking in particular, can provide significant help for designing autonomous systems in a more efficient and reliable manner, by providing a better coverage and allowing early error detection. This article discusses the general issue of V&V of autonomy software, with an emphasis towards model-based autonomy, model-checking techniques, and concrete experiments at NASA.
Verification and Validation of Autonomy Software at NASA
2000-01-01
Preprint
Keine Angabe
Englisch
Verification and Validation of Autonomy Software at NASA
NTRS | 2000
|NTRS | 2019
|NTIS | 2019
|Verification and Validation of NASA LEWICE 2.2 Icing Software Code
Online Contents | 2006
|