Robust local search for spacecraft operations using adaptive noise
Alex S. Fukunaga, Gregg Rabideau, Steve Chien · 2004
Randomization is a standard technique for improving the per-formance of local search algorithms for constraint satisfac-tion. However, it is well-known that local search algorithms are sessitive to the noise values selected. We investigate the use of an adaptive noise mechanism in an iterative repair-based planner/scheduler for spacecraft operations. Prelimi-nary results indicate that adaptive noise makes the use of ran-domized repair moves safe and robust; that is, using adaptive noise makes it possible to consistently achieve performance comparable with the best tuned noise setting without the need for manually tuning the noise parameter., 1