Test and evaluation by genetic algorithms
- 1 October 1993
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Expert
- Vol. 8 (5) , 9-14
- https://doi.org/10.1109/64.236476
Abstract
A machine learning technique for automating the traditional controller tests process that evaluates autonomous-vehicle software controllers is discussed. In the proposed technique, a controller is subjected to an adaptively chosen set of fault scenarios in a vehicle simulator, and then a genetic algorithm is used to search for fault combinations that produce noteworthy actions in the controller. This approach has been applied to find a minimal set of faults that produces degraded vehicle performance and a maximal set of faults that can be tolerated without significant performance loss.Keywords
This publication has 5 references indexed in Scilit:
- Experiments with the subsumption architecturePublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Robustness analysis method for underwater vehicle control systemsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Using a genetic algorithm to learn behaviors for autonomous vehiclesPublished by American Institute of Aeronautics and Astronautics (AIAA) ,1992
- Learning sequential decision rules using simulation models and competitionMachine Learning, 1990
- Adaptive System Design: A Genetic ApproachIEEE Transactions on Systems, Man, and Cybernetics, 1980