Path Planning for Mobile Robot Based on Improved Simulated Annealing Artificial Neural Network

摘要

A mobile robot path planning method based on improved simulated annealing algorithm and artificial neural network is proposed. First the simulated annealing algorithm with the best reserve mechanism is introduced and it is combined with Powell algorithm to form improved simulated annealing mixed optimize algorithm which not only add the good solution protective measures but also improve the convergence rate of simulated annealing algorithm. Then we take the obstacle collision penalty function which expressed using neural network and the path length as the energy function of improved simulated annealing mixed optimize algorithm, thereby the solution which obtained by the improved simulated annealing mixed optimize algorithm can not only satisfy the path shortest but also effective avoid the collision with obstacle. The simulation result shows that the proposed method is feasible and valid.