A Specialized Genetic Algorithm for Optimum Path Planning of Mobile Robots

摘要

A specialized genetic algorithm for optimum path planning of mobile robots is presented in this paper. The proposed algorithm consists of an obstacle avoidance scheme that is introduced to generate efficient initial population and domain heuristic knowledge based crossover, mutation, refinement and deletion operators that are tailored to fit path planning for mobile robots. Furthermore, a fuzzy logic control approach is also incorporated into the proposed genetic algorithm to self-adaptively adjust the probabilities of crossover and mutation. Simulations on both off-line and on-line planning with static obstacles are carried out, and the proposed method is studied in comparison with another recently reported method. The results have demonstrated superior performance of the proposed method such as rapid search speed and high search quality.