多策略蚁群算法求解机器人路径规划

Multi-strategy ant colony algorithm for robot path planning

针对基本蚁群算法的缺点,提出用多策略的蚁群算法求解机器人路径规划问题.采用栅格法建立机器人全局路径规划工作空间模型,进行两次凸化改进处理.提出惩罚策略,并配合使用保健算子策略、治病算子策略,同时引入遗传算子策略、精英蚂蚁策略和最大最小蚂蚁策略.介绍在Matlab环境下编程实现的方法及步骤,求解100个栅格点的路径规划问题,得到最优距离为15.070.仿真结果表明,即使在复杂的地形环境中用本算法也可迅速规划出令人满意的最优路径.

A multi -strategy ant colony algorithm is proposed for robot path planning problem to deal with the default of the basic ant colony algorithm. The grid method is established for work space model of global path planning, and the convex optimization is processed twice. The punishment strategy, nourishing operator and remedying operator are introduced along with the genetic operator, max - min ant system and elite ant strategy. The Matlab programming method and steps are described in details. The simulation that applied the robot path planning about 100 points reached an optimum of 15. 070. The results of simulation show that the best path can be found in short time, and the effect is very satisfying even if the geographic conditions with obstacles are exceedingly complicated.