基于RRT^(*)改进的移动机器人路径规划算法

Improved path planning algorithm for mobile robot based on RRT^(*)

针对RRT^(*)算法在复杂环境路径规划中存在的盲目搜索、冗余节点及路径较长等问题,提出一种融合树扩展策略和采样策略的改进RRT^(*)算法(AF-RRT^(*))。通过创造父节点改进RRT^(*)扩展树的结构,缩小路径长度;引入自适应探索,增加采样导向的选择性,减少路径搜索时间,同时不会陷入局部最优陷阱;通过动态步长,减少冗余节点。仿真结果表明,AF-RRT^(*)算法在多种环境下,路径获取效率和路径质量均优于RRT^(*)和F-RRT^(*)。消融实验验证了AF-RRT^(*)算法和算法各功能模块的有效性。

The optimal rapidly-exploring random tree(RRT^(*))algorithm is usually used in complex environments for robot path planning.However,it has issues of blind search,redundant nodes and long path length.An improved RRT^(*)algorithm(AF-RRT^(*))that combined tree expansion strategy with sampling strategy was proposed.To reduce the path length,parent node creation strategy was used to improve the structure of the RRT^(*)extension tree.The adaptive exploration strategy was introduced to increase the sampling-oriented selectivity and reduce the path search time without falling into the local optimum trap.The dynamic step size was utilized to decrease the redundant nodes.Simulation results show that AF-RRT^(*)algorithm is better than RRT^(*)and F-RRT^(*)in path acquisition efficiency and path quality.The ablation experiment verifies the effectiveness of AF-RRT^(*)algorithm and its function modules.