In order to improve robotic efficiency of path planning in a complex environment,this paper proposes a global path planning algorithm which makes use of ant colony optimization (ACO) to optimize rapidly exploring random tree algo-rithm.The new algorithm effectively combines the strengths of the two algorithms,taking advantage of the high efficiency of the rapidly exploring ran-dom algorithm to quickly converge a possible path, make it into a ACO initial distribution of phero-mones,reduce the number of iterations and accel-erate the convergence.At the same time,by using the positive feedback technology,the process of finding the optimum path is effectively improved. The simulation experiment shows that the algo-rithm increases the efficiency of robotic path plan-ning greatly and can plan the best path in a com-plex environment quickly.%为提高复杂环境下机器人的路径规划效率,提出了一种用蚁群算法来优化随机树算法的新的全局路径规划算法。该算法有效地结合了蚁群和随机树算法的优点,利用随机树算法的高效性快速收敛到一条可行路径,将该路径转换为蚁群的初始信息素分布,可以减少蚁群算法初期迭代;然后利用蚁群算法的反馈性优化路径,求得最优路径。仿真实验表明,该蚁群随机树算法可以提高机器人路径规划的速度,并且在任何复杂环境下迅速规划出最优路径。
展开▼