计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (36): 1-3.

• 博士论坛 • 上一篇    下一篇

基于评价选择策略的Internet多机器人协作控制研究

李 响1,2,王越超1   

  1. 1.中国科学院 沈阳自动化研究所 机器人学重点实验室,沈阳 100016
    2.中国科学院 研究生院,北京 100049
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-12-21 发布日期:2007-12-21
  • 通讯作者: 李 响

Coordination control of Internet-based multi-robot based on evaluation selective strategy

LI Xiang1,2,WANG Yue-chao1   

  1. 1.Robotics Laboratory of Shenyang Institute of Automation,Chinese Academy of Sciences,Shenyang 100016,China
    2.Graduate School of Chinese Academy of Sciences,Beijing 100049,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-12-21 Published:2007-12-21
  • Contact: LI Xiang

摘要: 针对Internet多机器人系统中存在的操作指令延迟、工作效率低、协作能力差等问题,提出了多机器人神经元群网络控制模型。在学习过程中,来自不同功能区域的多类型神经元连接形成动态神经元群集,来描述各机器人的运动行为与外部条件、内部状态之间复杂的映射关系,通过对内部权值连接的评价选择,以实现最佳的多机器人运动行为协调。以互联网足球机器人系统为实验平台,给出了学习算法描述。仿真结果表明,己方机器人成功实现了配合射门的任务要求,所提模型和方法提高了多机器人的协作能力,并满足系统稳定性和实时性要求。

关键词: Internet, 多机器人, 协作, 神经元群网络, 评价

Abstract: In this paper,we construct a neural group network controller for resolving the problems that exist in the Internet-based multi-robot system,such as:delayed instruction,low efficiency and poor collaboration ability,etc.During the learning process,multi-type neuron coming from the different function regions forms the dynamic neural group through connection.This structure may describe the complex mapping relationship among the external conditions,internal states and behaviors of robot.Through the evaluation to the internal connection weights,it can achieve the best behavior coordination of the multi-robot.We adopt the Internet-based soccer robot system as the experimental platform,and describe the learning algorithm.Simulation result shows that it successfully realizes the shooting mission by one team robots.The proposed model and method improve the coordination ability of multi-robot,and fulfill the stability and real-time requirement of system.

Key words: Internet, multi-robot, coordination, neural group network, evaluation