Abstract：A path planning method based on artificial potential field optimization is proposed. The ant algorithm is initialized by the planning result of the artificial potential field method as the prior knowledge, which improves the algorithm's efficiency. On the other hand, the path obtained by the artificial potential field method is optimized by the ant algorithm, which overcomes the local minima problem in the artificial potential field method. Results of computer simulation experiment indicate that the method can implement optimal path planning in complex environments. In order to improve the planning efficiency a feasible idea of combining traditional planning methods with statistic optimization is also presented.
[1] Philip Kilby, Patrick Prosser, Paui Shaw. Guided local search for the vehicle routing problem[A].2nd Inter-national Conference on Metaheuristics-MIC97 [C]. [2] Boronstein J, Koren y. The vector field histogram-fast obstacle avoidance for mobile Robots [J].IEEE Journal of Robotics and Automation, 1991,7(3): 278-288. [3] 庄晓东,孟庆春,殷波等.动态环境中基于模糊概念的机器人路径搜索方法[J],机器人,2001,23(5):397-399. [4] Sato K. Deadlock-free motion planning using the Laplace potential field[J].Advanced Robotics, 1993,7(5): 449-461. [5] 金飞虎,洪炳熔,高庆吉.基于蚁群算法的自由飞行空间机器人路径规划[J].机器人,2002,24(6):526-529. [6] 吴丽娟,徐心和.基于遗传算法的足球机器人比赛中障碍回避策略的设计[J].机器人,2001,23(2):142-145. [7] 庄晓东,孟庆春,王汉萍等.多障碍环境中基于增强式学习的势场优化和机器人路径规划[J].青岛海洋大学学报(自然科学版),2001,31(6):937-942. [8] 唐世浩,朱启疆.遗传算法中初始种群与交叉、变异率对解的影响及其解决方案[J].科技通报,2001,17(3):1-7. [9] 杨伟,倪黔东,吴军基.BP神经网络权值初始值与收敛性问题研究[J].电力系统及其自动化学报,2002,14(1):20-22. [10] Dorigo M, Di Caro G, Gambardella L. Ant colony optimization:New Meta-Heuristic[A].Proceedings of the Congress on Evolutionary Computation[C].1999. 1470-1477. [11] Oussama Khatib. Real-time obstacle avoidance for manipulators and mobile robots[J].The International Journal of Robotics Research,1986,5(1): 90-98.
