Issue No. 05 - October (1993 vol. 8)
DOI Bookmark: http://doi.ieeecomputersociety.org/10.1109/64.236476
<p>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.</p>
J. J. Grefenstette, K. A. De Jong and A. C. Schultz, "Test and Evaluation by Genetic Algorithms," in IEEE Intelligent Systems, vol. 8, no. , pp. 9-14, 1993.