利用改进的粒子群算法的机器人全局路径规划

Global path planning for mobile robot based on improved particle swarm optimization

针对常规路径规划单纯追求路径最短、路径规划不灵活和实现复杂的缺点,提出了一种改进的移动机器人全局路径规划方法.该方法综合人工势场(APF)与粒子群优化算法(PSO)的优点,利用障碍物的排斥力生成路径的危险度地图(DDM),将路径长度与危险度的加权和作为PSO的适应度函数,获得了一条全局最优路径.该方法具有3个优点:粒子初始化及更新过程中会自动避开有障碍物的危险区域,规划出一条既安全相对长度又较短的路径;通过调整加权因子平衡长度与危险度在适应度函数中的比重,路径规划灵活;算法实现简单,收敛速度快,能满足移动机器人实时路径规划的要求.仿真结果证明了该算法的可行性和有效性.

In conventional path planning methods,the length of the path is the principal factor,so the path we get is the shortest but is not flexible and is complex in realization.In order to overcome the above defects,a new path planning approach based on artificial potential field(APF) and particle swarm optimization(PSO) is presented in the paper.The first step is to make a danger degree map(DDM) based on the repulsive force of obstacles in the environment.Then the PSO whose fitness function is the weighted sum of...