动态环境下移动机器人路径规划的改进蚁群算法

研究动态环境下移动机器人路径规划问题,采用栅格法对机器人工作空间进行建模,在使用蚁群算法进行全局路径搜索过程中引入人工势场的概念,使蚂蚁对最优路径更加敏感;机器人针对动态环境中可能出现的不同类型障碍物分别执行不同的避障策略;同时提出一种最优路径预测模型用于预测在避障过程中是否出现新的最优路径。算法结合人工势场法和蚁群算法的特点,将全局路径规划与局部路径规划相融合以提高路径搜索的效率。仿真结果验证了该算法的有效性。

An Improved Ant Colony Algorithm for Mobile Robot Path Planning Under Dynamic Environment

The paper studied the problem of mobile robot path planning under dynamic environment.Grid method was used to establish workspace model of the mobile robot.The artificial potential field concept which makes ants more sensitive to the optimal path was introduced into the ant colony algorithm during the global searching process.The mobile robot carries out different obstacle avoidance strategies respectively when different types of obstacle appear in the dynamic environment.Meanwhile an optimal path prediction model was proposed to predict if there is a better path in the process of obstacle avoidance.This algorithm combined the characteristics of artificial potential field method with ant colony algorithm,integrated the global path planning and the local path planning so as to improve path searching efficiency.The simulation results indicate the validity of the algorithm.