Title :
Test and evaluation by genetic algorithms
Author :
Schultz, Alan C. ; Grefenstette, John J. ; De Jong, Kenneth A.
Author_Institution :
US Naval Res. Lab., Washington, DC, USA
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 :
computerised control; digital simulation; genetic algorithms; learning (artificial intelligence); program testing; vehicles; adaptively chosen set; autonomous-vehicle software controllers; controller tests process; fault combinations; fault scenarios; fault tolerance; genetic algorithm; machine learning technique; vehicle performance; vehicle simulator; Artificial intelligence; Automatic control; Automotive engineering; Genetic algorithms; Machine learning; Performance loss; Robust control; Software algorithms; Software testing; Vehicles;
Journal_Title :
IEEE Expert
