• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2017, Vol. 53 ›› Issue (11): 104-112.doi: 10.3901/JME.2017.11.104

• 绿色制造技术 • 上一篇    下一篇

基于分层结构的空间机械臂多约束任务规划

王一帆, 孙汉旭, 陈钢, 贾庆轩   

  1. 北京邮电大学自动化学院 北京 100876
  • 出版日期:2017-06-05 发布日期:2017-06-05
  • 作者简介:

    王一帆(通信作者),男,1987年出生,博士研究生。主要研究方向为空间机械臂任务规划与控制技术。

    E-mail:wangyifan@bupt.edu.cn

  • 基金资助:
    * 国家重点基础研究发展计划(973 计划,2013CB733000)和国家自然科学基金(61403038)资助项目; 20160619收到初稿,20170313收到修改稿;

Hierarchical Task Planning for Space Manipulator with Multi-constraint

WANG Yifan, SUN Hanxu, CHEN Gang, JIA Qingxuan   

  1. School of Automation, Beijing University of Posts and Telecommunications, Beijing 100876
  • Online:2017-06-05 Published:2017-06-05

摘要:

针对空间机械臂在轨环境复杂、约束繁多、任务多样的特点,提出了一种基于分层结构的多约束任务规划方法。将任务规划过程划分为任务剖面分析与任务中间点规划两个层次分别进行。根据空间机械臂的工作特点,提出一种采用状态矩阵结合状态掩码的工作环境表征方式。在此基础上,定义了三种空间机械臂原任务及其算子。基于分层任务网络(Hierarchical task network, HTN)进行任务剖面分析,将复杂任务拆解为原任务序列。系统梳理空间机械臂规划参数集,针对因各类约束条件导致无法直接执行的原任务,在考虑资源优化的同时利用改进的A*算法对原任务进一步划分为多段简单路径,并给出代价最小的任务中间点序列。以物体转移任务为例通过仿真试验验证了基于分层结构的空间机械臂任务规划方法的可行性与有效性。

关键词: A*算法, 分层任务网络, 任务规划, 空间机械臂

Abstract:

A hierarchical task planning method with multi-constraint is presented for space manipulator. The process of task planning is separated into two hierarchies:mission profile analysis and task node planning. A representation method for working environment using state matrix and state mask is proposed, according to the space manipulator’s characteristic. Then, three kinds of original tasks and operators are defined based on those matrixes. A sequence of original tasks can be derived by applying hierarchical task network to a complex task. A sequence of task node of the original task which cannot be executed directly because of some constraints can be calculated by an improved A* algorithm, and optimization problem is introduced into the calculation. The feasibility and effectiveness of the proposed task planning method are verified by simulation.

Key words: A* algorithm, hierarchical task network, task plan, space manipulator