计算机应用

• 图形图像处理 • 上一篇    下一篇

一个虚拟人手臂操控的运动规划框架

王维 李焱   

  1. 国防科学技术大学机电工程与自动化学院自动化研究所 国防科学技术大学机电工程与自动化学院自动化研究所
  • 收稿日期:2008-10-22 修回日期:2008-12-07 发布日期:2009-04-01 出版日期:2009-04-01
  • 通讯作者: 王维

Motion planning framework for virtual human arm manipulation

Wei WANG Yan LI   

  • Received:2008-10-22 Revised:2008-12-07 Online:2009-04-01 Published:2009-04-01
  • Contact: Wei WANG

摘要: 基于双向扩展的启发式快速扩展随机树(RRT)算法,提出了一种虚拟人手臂操控的运动规划框架。该框架根据是否抓握操控对象,将虚拟人的手臂操控划分为接触和搬运两个阶段,分别采用手臂的前向和逆向运动学两种策略进行规划,保证规划结果快速、可靠。通过实验验证了该方法的有效性。

关键词: 虚拟人, 操控规划, 快速扩展随机树, 启发式策略

Abstract: A motion planning framework for virtual human arm manipulation was proposed based on bidirectional heuristic Rapidly-exploring Random Tree (RRT). According to whether a manipulated object was hold by hand, the framework proceeded in two phases: a reaching phase and a moving phase. Forward as well as inverse kinematic algorithm was used in these two phases respectively in order to make planning result rapid and credible. The effectiveness of our method was demonstrated by experiments.

Key words: virtual human, manipulation planning, Rapidly-exploring Random Tree (RRT), heuristic strategy

中图分类号: