基于改进遗传算法的机器人路径规划

Robot Path Planning Based on Improved Genetic Algorithm

针对传统路径规划算法存在路径不可达、寻优计算规模大等缺陷导致算法的计算量大、收敛精度低等问题,本文提出采用改进遗传算法优化中间节点,结合Dijkstra算法求最短路径算法补齐节点间的路径形成一条完整路径的方式,保证遗传操作中的路径全部为可行路径。与传统遗传算法作对比,试验结果表明改进后的算法在收敛精度和寻优能力上都取得了明显的效果。

In view of the problems of traditional path planning algorithms such as unreachable paths and large-scale optimization calculations,which lead to large calculations and low convergence accuracy,this paper proposes an improved genetic algorithm to optimize intermediate nodes,combined with Dijkstra's shortest path algorithm to fill in the path between nodes to form a complete path,to ensure that all paths in the genetic operation are feasible paths.Compared with the traditional genetic algorithm,the experimental results show that the improved algorithm has achieved obvious results in convergence accuracy and optimization ability.