Robot Path Planning Based on Adaptive Isolation Niche Genetic Algorithm
Xianyi Cheng, Changyu Yang · 2008
To solve two problems of niche genetic algorithm: premature convergence and lower convergence,a adaptive isolation niche genetic algorithm is proposed. The algrorithm forms niches based on isolation which well keeps diversities of the population,and designs adaptive operator according to generation. Crossover probability and mutation probability adjust to the evolution, which will accelerate convergence and avoid premature convergence. Through Shubert function test, stability on keeping diversities of the population and high efficiency on searching are proved. Finally, the simulation shows that the proposed algorithm is reliable and efficient in robots' path planning.