基于改进A-star与DWA相融合的移动机器人动态路径规划算法

Dynamic path planning algorithm for mobile robots based on the fusion of improved A-star and DWA

A-star算法常用于移动机器人的全局路径规划,但在复杂场景中A-star算法存在耗时长、搜索节点过多、路径不平滑、不能避开环境中未知的障碍物等问题。针对于此,本文提出一种融合路径规划算法。首先,在A-star算法的基础上引入环境中的障碍物信息和搜索节点到起始位置的距离信息动态调节启发函数的权重,减少搜索节点数,提升A-star算法的性能;然后,利用自适应分段步长的高阶贝塞尔曲线对路径进行优化,减少转折点提升路径的平滑性;最后,将改进A-star算法规划的全局路径作为引导,将路径节点作为DWA算法的中间目标,实现全局路径规划和局部规划的融合,使移动机器人在找到全局最优路径的同时,能够避开环境中的未知障碍物,实现移动机器人的动态路径规划。仿真结果验证了该算法的有效性。

The A-star algorithm is commonly used for global path planning in mobile robotics.However,in complex environments,the A-star algorithm faces issues such as long computation times,excessive search nodes,non-smooth paths,and an inability to avoid unknown obstacles in the environment.To address these challenges,this paper introduces a fusion path planning algorithm.Firstly,it enhances the A-star algorithm by dynamically adjusting the weights of the heuristic function using information about obstacles in the environment and the distance of search nodes from the starting position,thereby reducing the number of search nodes and improving the A-star algorithm’s performance.Next,it optimizes the path using a high-order Bézier curve with adaptive segmented step lengths,reducing turning points to enhance path smoothness.Finally,it combines the globally planned path from the improved A-star algorithm as guidance,treating the path nodes as intermediate goals for the Dynamic Window Approach(DWA)algorithm.This fusion of global and local path planning enables the mobile robot to find the global optimal path while avoiding unknown obstacles in the environment,facilitating dynamic path planning for mobile robots.Simulation experiments validate the algorithm’s effectiveness.