改进粒子群优化算法的移动机器人路径规划

针对单一智能优化算法求解机器人路径规划时易陷入局部误区的问题,提出改进粒子群优化算法(GB_PSO)用于机器人路径规划.该算法以粒子群优化算法(particle swarm optimization,PSO)为主体,由于遗传算法(genetic algorithm,GA)和细菌觅食算法(bacterial foraging optimization algorithm,BFO)更新策略所受环境影响的不同,拟合两种环境参数;然后计算粒子与不同环境参数之间的相关性将粒子群划分为两类,分别通过GA的选择、交叉、变异算子和BFO的趋化操作并行加强局部优化;最后通过改进的粒子群更新公式对粒子进行更新,实现机器人全局和局部路径的优化.实验结果表明,改进粒子群优化算法进行路径规划提高了局部和整体的搜索能力,路径规划速度快且路径距离短,同时具备更强的鲁棒性.

Improved particle swarm optimization algorithm for mobile robot path planning

Improved particle swarm optimization algorithm(GB_PSO) proposed for robot path planning to address the problems of easy to fall into local misconceptions when solving robot path planning by a single intelligent optimization algorithm. The algorithm took on the PSO as the main body, due to the GA and the BFO algorithm update strategies was subject to different environmental influences, fitting two environmental parameters. Then the correlation between particles and different environmental parameters calculated to classify the particle swarm into two classes, and local optimization enhanced in parallel by selection, crossover, and variation operators of GA and convergence operations of BFO, respectively. Finally the improved particle swarm update formulation used to achieve the optimization of global and local paths of the robot. The experimental results show that the GB_PSO algorithm for path planning improves the local and overall search capability. The path planning is fast and short distance, while having stronger robustness.