Abstract: To solve the problem on global optimization and real-time obstacle avoidance for a mobile robot’s path planning in complex environment,the Safe-point Guide (SPG)hybrid algorithm with heuristic dynamic path planning was proposed by combining the improved A* algorithm with the artificial potential field.The end-point approximation strategy was designed to solve the problem of too many redundant turning points in the traditional A* algorithm.The plane vector product method was adopted to avoid the path oscillation of the conventional artificial potential field.A heuristic strategy guided by safe points was designed for combining the two improved algorithms to ensure a safe path and escape from the local minimum point.The SPG hybrid algorithm was simulated in the static and dynamic environments.Compared with the traditional hybrid algorithms in the static environment,the path length and running time were shortened by 10% and 25.6% respectively.In the dynamic environment,the path length and running time were shortened by 9.5% and 30.9% respectively.The results showed that the SPG hybrid algorithm had good global path-planning and dynamic obstacle-avoidance abilities.The effectiveness of the proposed algorithm was verified in real scenes.

Key words: robots, dynamic environment, hybrid path planning algorithm, safe-point guide

摘要： 针对机器人在复杂环境下路径规划难以兼顾全局最优和实时避障的问题,将改进的A*算法和人工势场法结合,提出一种具有启发式动态路径规划的安全点引导(SPG)混合算法。设计了终点逼近策略,解决了传统A*算法规划路径转折点过多的问题,同时采用平面向量积法避免传统人工势场法路径振荡。设计安全点引导的启发式策略将两种改进算法结合,既保证安全路径,又逃离局部极小值点。分别在静态和动态环境下对SPG混合算法进行仿真并与传统混合算法相比,静态环境和动态环境下的路径长度与运行时间分别缩短了10%,25.6%和9.5%,30.9%,表明SPG混合算法具有良好的全局路径规划与动态避障能力。最后在真实场景中验证了SPG混合算法的有效性。

关键词: 机器人, 动态环境, 混合路径规划算法, 安全点引导