Parameter determination for a genetic algorithm applied to robot control

Abstract

Automatic path planning is an important aspect in the development of autonomous robots. This paper shows how a genetic algorithm (GA) may be used to find feasible minimum distance paths between an initial and final configuration without colliding with physical objects present in its workspace. The paper emphasises the importance of determining appropriate GA parameters, i.e. the population size and probabilities of crossover and mutation, and the method used for selection generations, which has a dominant influence on the effectiveness of the algorithm. >