计算机工程 ›› 2010, Vol. 36 ›› Issue (23): 16-18.doi: 10.3969/j.issn.1000-3428.2010.23.006

• 博士论文 • 上一篇    下一篇

一种自适应加权快速探索随机树算法

朱金辉a,梁明杰a,梁颖驹a,闵华清b,张梅c   

  1. (华南理工大学 a. 计算机科学与工程学院; b. 软件学院; c. 自动化科学与工程学院,广州 510006)
  • 出版日期:2010-12-05 发布日期:2010-12-14
  • 作者简介:朱金辉(1977-),男,讲师、博士研究生,主研方向:自适应算法,软件体系结构,智能机器人;梁明杰,博士研究生;梁颖驹,硕士研究生;闵华清,教授、博士;张梅,副教授、博士
  • 基金项目:
    国家自然科学基金资助项目(60873078); 中央高校基本科研业务费专项基金资助项目(2009ZM0297);广东省科技计划基金资助项目(2009A040300008)

Adaptive Weighted Rapidlyexploring Random Tree Algorithm

ZHU Jinhuia,LIANG Mingjiea,LIANG Yingjua,MIN Huaqingb,ZHANG Meic   

  1. (a.School of Computer Science and Engineering; b. School of Software; c. School of Automation Science and Engineering, South China University of Technology, Gaungzhou 510006, China)
  • Online:2010-12-05 Published:2010-12-14

摘要: 针对快速探索随机树算法在局部极小区域做大量失败探索的问题,提出一种自适应加权快速探索随机树算法。分析影响快速探索随机树生长的关键因素,提出在树探索的动态过程中应充分利用探索过程的反馈信息,为树节点赋予自适应权重。根据树节点的自适应权重大小,选择树的生长点。仿真结果表明,该方法能有效地提高树探索效率,缩短规划路径长度。

关键词: 运动规划, 随机采样, 快速探索随机树, 自适应权重

Abstract: Rapidlyexploring Random Tree(RRT) algorithm is a practical and promising solution to motion planning problem. The algorithm easily falls into local minima which leads to massive failure exploring. To overcome the shortcoming, key factors affecting the exploring of the tree are analyzed and an adaptive weight method is proposed. Nodes are weighted according to the heuristic information collected from the dynamic exploring process, tree extension is guided by the weight. Simulation results show that the method can improve the quality of the tree and shorten the length of planning path.

Key words: motion plan, randomized sampling, Rapidlyexploring Random Tree(RRT), adaptive weight

中图分类号: