Autonomous vehicles require sophisticated
software controllers to maintain vehicle performance in the presence of
vehicle faults. The test and evaluation of comples software controllers
is a challenging task. The goal of this effort is to apply machine learning
techniques to the general problem of evaluating an intelligent controller
for an autonomous vehicle. The approach involves subjecting a controller
to an adaptively chosen set of fault scenarios within a vehicle simulator,
and searching for combinations of faults that produce noteworthy performance
by the vehicle controller. The search employs a genetic algorithm. The
evidence suggests that this approach is an effective supplement to manual
and other forms of automated testing of sophisticated software controllers.
Several intelligent controllers were tested in this project using several
different genetic algorithm-based learning programs.
to break things: adaptive testing of intelligent controllers,
In Handbook on Evolutionary Computation , G3.5, IOP Publishing
Ltd. and Oxford Press, 1997.
Full publication list for Adaptive Testing
Alan C. Schultz, Principal Investigator (Click for contact information)