• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2020, Vol. 56 ›› Issue (5): 39-46.doi: 10.3901/JME.2020.05.039

• 航天重器——空间大型可展机构与装备专刊 • 上一篇    下一篇

桁架式可展开抓取机构展开动力学建模及其自适应鲁棒控制

李国涛1,2, 李兵1,2, 郭宏伟1, 黄海林1,2   

  1. 1. 机器人技术与系统国家重点实验室(哈尔滨工业大学)哈尔滨 150001;
    2. 哈尔滨工业大学(深圳)机电工程与自动化学院 深圳 518055
  • 收稿日期:2019-01-15 修回日期:2019-05-24 出版日期:2020-03-05 发布日期:2020-04-23
  • 通讯作者: 李兵(通信作者),男,1971年出生,博士,教授,博士研究生导师。主要研究方向为机器人学、并联机床和并联隔振平台、空间可展开抓取机构、护理机器人等。E-mail:libing.sgs@hit.edu.cn
  • 作者简介:李国涛,男,1988年出生,博士研究生。主要研究方向为可展开抓取机构设计、动力学建模与分析、自适应鲁棒控制和自抗扰控制等运动控制方法。E-mail:lghitsz@163.com
  • 基金资助:
    国家自然科学基金联合基金重点支持项目(U1613201)。

Deployment Dynamic Modelling and Adaptive Robust Control of Truss-shaped Deployable Grasping Mechanism

LI Guotao1,2, LI Bing1,2, GUO Hongwei1, HUANG Hailin1,2   

  1. 1. State Key Laboratory of Robotics and System(HIT), Harbin 150001;
    2. Department of Mechanical Engineering and Automation, Harbin Institute of Technology, Shenzhen 518055
  • Received:2019-01-15 Revised:2019-05-24 Online:2020-03-05 Published:2020-04-23

摘要: 针对一种新型桁架式可展开抓捕机构的展开运动,提出该机构的动力学建模方法并设计考虑过约束等非线性因素的控制策略以提高其运动性能。对该桁架式可展开抓取机构进行简要描述,利用拉格朗日力学并通过递归的方法建立该机构展开运动的动力学模型。利用动力学模型的已知信息,设计扩张状态观测器以估计机构中的过约束等非线性因素引起的动力学项。使用扩张状态观测器作为补偿项,提出改进型的自适应鲁棒控制器,其中,通过设计非连续映射自适应控制律来估计机构的惯性参数,通过设计自适应滑模控制律来克服传统自适应鲁棒控制器中的问题。搭建可展开抓取机构的试验平台,通过与现有的不同控制器的比较试验来验证提出的控制器的有效性。

关键词: 展开机构, 过约束, 动力学建模, 扩张状态观测器, 自适应鲁棒控制

Abstract: A dynamic modeling method for the deployment motion of a new type of truss-shaped deployable grasping mechanism is proposed, and a control strategy considering nonlinear factors, such as over-constraints, is designed to improve the motion performance. The truss-shaped deployable grasping mechanism is briefly introduced, and the dynamic model of the deployment motion of the mechanism is established by using Lagrangian mechanics and the recursive method. Using the known information of the dynamic model, an extended state observer is designed to estimate the dynamic part caused by the over-constraints in the mechanism. An improved adaptive robust controller based on extended state observer is proposed, where the inertia parameter of the mechanism is estimated by a discontinuous-projection-based adaptive control law, and the problem of the traditional adaptive robust controller is overcome by proposing an adaptive sliding mode control law. The experimental platform of deployable grasping mechanism is built, and the effectiveness of the proposed controller is verified in comparison with the existing state-of-the-art controllers.

Key words: deployable mechanism, over-constraint, dynamic modelling, extended state observer, adaptive robust control

中图分类号: