基于改进蚁群算法的移动机器人路径规划

Improved ant colony algorithm for mobile robot path planning

针对大多数路径规划方法所忽视的路径尖峰,以及传统蚁群算法(ACA)易出现的早熟、陷入局部最优等问题,提出一种改进ACA以用于路径规划.首先,在ACA中融入遗传算子,利用交叉与变异操作来扩大解的搜索空间,提升解的全局性.然后,引入简化与平滑操作优化算子,对所寻路径做进一步处理,消除路径中不必要的尖峰,提高其平滑性.栅格环境下的机器人路径规划仿真结果表明,与A*以及传统ACA相比,所提算法能够得到更为平滑的最短路径.

In order to remove peak points obtained by most path planning methods and to decrease the risks of trapping in premature convergence and occurrence of local optimization in the traditional ant colony algorithm （ACA）, an improved ACA was proposed for a mobile robot path planning. At first, genetic operators were introduced into the traditional ACA using crossover and mutation operators to expand the search space and enhance the global property of solution. Then the optimization operators, such as simplification and smoothness operators, were applied to remove the redundant nodes and to increase the smoothness of the solved path. The simulation results concerning path planning for mobile robot in two grid environments illustrate that, compared with algorithm A^＊ and traditional ACA, the proposed algorithm can obtain a much shorter and smoother path.