移动机器人导航路径的自主学习粒子群规划方法

Mobile Robot Navigation Path Planning Method Autonomous Learning Particle Swarm Algorithm

为了减小移动机器人行驶路径长度,提出了基于自主学习粒子群算法的导航路径规划方法。以减小路径长度为目标建立了路径规划模型;为了防止机器人发生碰撞,给出了障碍物膨化处理方法。在粒子群算法中引入了由多种粒子学习策略组成的学习策略池,并给出了粒子对学习策略进行选择的自主学习策略,从而提出了具有较强进化能力的自主学习粒子群算法。经算法性能测试,自主学习粒子群算法的优化能力强于传统粒子群算法和文献[11]改进粒子群算法;将自主学习粒子群算法应用于简单场景和复杂场景的路径规划,该算法规划的路径均值和标准差均小于传统粒子群算法,验证了自主学习粒子群算法在机器人路径规划中的优越性。

In order to reduce the path length of mobile robot,a navigation path planning method based on autonomous learning particle swarm algorithm is proposed.A path planning model is established aiming at reducing the path length;In order to prevent collision of robot,the method of obstacle expansion is given.In particle swarm optimization,a learning strategy pool composed of multiple particle-learning strategies is introduced,and the autonomous learning strategy of selecting learning strategies by particles is given,thus an autonomous learning particle swarm optimization algorithm with strong evolutionary ability is proposed.Through the algorithm performance test,the optimization ability of the autonomous learning particle swarm optimization algorithm is better than the traditional particle swarm algorithm and the improved particle swarm algorithm in reference[11];The autonomous learning particle swarm algorithm is applied to the path planning of simple scene and complex scene.The mean and standard deviation of the path planning of the algorithm are both less than those of the traditional particle swarm algorithm,which verifies the superiority of the autonomous learning particle swarm optimization algorithm in robot path planning.