北京航空航天大学学报 ›› 2018, Vol. 44 ›› Issue (9): 1918-1925.doi: 10.13700/j.bh.1001-5965.2017.0753

• 论文 • 上一篇    下一篇

水下机械手不确定遥操作自适应双边控制

张建军1, 刘卫东1,2, 高立娥1,2, 李乐1, 李泽宇1   

  1. 1. 西北工业大学 航海学院, 西安 710072;
    2. 西北工业大学 水下信息与控制重点实验室, 西安 710072
  • 收稿日期:2017-12-05 出版日期:2018-09-20 发布日期:2018-09-21
  • 通讯作者: 刘卫东.E-mail:liuwd@nwpu.edu.cn E-mail:liuwd@nwpu.edu.cn
  • 作者简介:张建军 男,博士研究生。主要研究方向:基于力觉感知的水下机械手遥操作;刘卫东 男,博士,教授,博士生导师。主要研究方向:水下航行器控制与仿真。
  • 基金资助:
    国家自然科学基金(61473224);国家重点研发计划(2016YFC0301700);中央高校基本科研业务费专项资金(3102017OQD069)

Adaptive bilateral control for underwater manipulator in uncertainty teleoperation

ZHANG Jianjun1, LIU Weidong1,2, GAO Li'e1,2, LI Le1, LI Zeyu1   

  1. 1. School of Marine Science and Technology, Northwestern Polytechnical University, Xi'an 710072;
    2. Science and Technology on Underwater Information and Control Laboratory, Northwestern Polytechnical University, Xi'an 710072, China
  • Received:2017-12-05 Online:2018-09-20 Published:2018-09-21
  • Supported by:
    National Natural Science Foundation of China (61473224); National Key R&D Program of China (2016YFC0301700); the Fundamental Research Funds for the Central Universities (3102017OQD069)

摘要: 针对水下机械手遥操作过程中数学模型及外部干扰引起不确定问题提出了自适应双边控制策略。对主机械手模型参数与外部干扰引起的不确定,设计了基于名义模型的参考自适应阻抗控制律,根据主手力与从手力误差来调节期望模型的参考位置,利用自适应控制律补偿模型不确定性。针对从机械手的不确定性采用径向基函数(RBF)神经网络进行自适应补偿,通过设计滑模变结构控制器与鲁棒自适应控制器消除逼近误差,满足了从机械手对主机械手位置跟踪。设计了李雅普诺夫函数证明跟踪性能与全局稳定性,保证力-位置跟踪的渐进收敛性能。结果表明:整体控制在模型不确定及外部干扰条件下具有很好的力-位置跟踪能力,整体系统具有稳定性和可靠性,并且具有鲁棒性及自适应控制能力。

关键词: 遥操作, 阻抗控制, 神经滑模控制, 力-位置跟踪, 鲁棒性, 自适应

Abstract: An adaptive bilateral control strategy is proposed for the uncertainty of the mathematical model and external disturbances during the teleoperation of underwater manipulator. A reference adaptive impedance control law based on the nominal model is designed for the uncertainty of the parameters of the master manipulator model and the external disturbance. The reference position of the expected model is adjusted by the deviation between the force of operator and the slave manipulator, and the model uncertainty is compensated by the adaptive control law. Aimed at the uncertainty of slave manipulator, the adaptive compensation is achieved by the radial basis function (RBF) neural network, and the approximation deviation is eliminated by the design of the sliding mode variable structure controller and the robust adaptive controller, which satisfies the position tracking of the slave manipulator to the master manipulator. The tracking performance and global stability are proved by Lyapunov function, and the asymptotic convergence of force-position tracking is guaranteed. The results show that the overall controller has good force-position tracking ability under the conditions of model uncertainty and external disturbance. The whole system is stable and feasible, and has robustness and adaptive control ability.

Key words: teleoperation, impedance control, neural sliding mode control, force-position tracking, robustness, adapation

中图分类号: 


版权所有 © 《北京航空航天大学学报》编辑部
通讯地址:北京市海淀区学院路37号 北京航空航天大学学报编辑部 邮编:100191 E-mail:jbuaa@buaa.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发