基于一种混合遗传算法的移动机器人路径规划

Mobile robot path planning based on a hybrid genetic algorithm

移动机器人路径规划问题一直是机器人学研究的核心内容之一,而遗传算法作为智能仿生学算法在路径规划中得到了广泛的应用。针对传统遗传算法存在局部搜索能力差的问题,文中研究在已知环境下运用一种基于遗传算法和模拟退火算法相结合的技术对移动机器人进行最优路径的规划方法。算法采用栅格法对环境建立模型,同时在遗传算子中添加插入算子和删除算子以优化路径。Matlab仿真实验结果表明,该算法相对于基本遗传算法的收敛速度,搜索质量等有了明显的提高。

The path planning problem of the mobile robot is always one of the core contents of robotics research,and the genetic algorithm,as an intelligent bionics algorithm,has been widely used in path planning. In allusion to the poor local search capability problem of the traditional genetic algorithm,the mobile robot optimal path planning method is researched in this paper by using the genetic algorithm and simulated annealing algorithm combined technology in a known environment. In the algorithm,the grid method is used to construct the environmental model. The insert operator and delete operator are added in genetic operators to optimize the path. The experimental results of the Matlab simulation show that in comparison with the basic genetic algorithm,the algorithm has an obvious improvement in convergence speed and search quality.