• CN:11-2187/TH
  • ISSN:0577-6686

机械工程学报 ›› 2022, Vol. 58 ›› Issue (7): 32-43.doi: 10.3901/JME.2022.07.032

• 机器人及机构学 • 上一篇    下一篇

扫码分享

下肢康复机器人人机耦合动力学建模和主动柔顺控制

卢浩1,2,3,4, 王洪波1,2,4, 冯永飞5   

  1. 1. 复旦大学工程与应用技术研究院 上海 200433;2. 上海市老年疾病临床医学研究中心 上海 200040;3. 复旦大学智能机器人教育部工程研究中心 上海 200433;4. 上海智能机器人工程技术研究中心 上海 200433;5. 宁波大学机械工程与力学学院 宁波 315211
  • 收稿日期:2021-08-04 修回日期:2021-11-05 出版日期:2022-04-05 发布日期:2022-05-20
  • 通讯作者: 王洪波(通信作者),男,1956年出生,教授,博士研究生导师。主要研究方向为康复机器人、手术机器人、移乘转运机器人。E-mail:Wanghongbo@fudan.edu.cn
  • 作者简介:卢浩,男,1990年出生,博士研究生。主要研究方向为康复机器人、护理机器人。E-mail:luh20@fudan.edu.cn
  • 基金资助:
    国家重点研发计划(2019YFB1312500)和国家自然科学基金联合基金(U1913216)资助项目。

Human-machine Coupling Dynamics Modeling and Active Compliance Control of Lower Limb Rehabilitation Robot

LU Hao1,2,3,4, WANG Hongbo1,2,4, FENG Yongfei5   

  1. 1. Academy for Engineering & Technology, Fudan University, Shanghai 200433;2. Shanghai Clinical Research Center for Aging and Medicine, Shanghai 200040;3. Institute of AI and Robotics, Fudan University, Shanghai 200433;4. Shanghai Engineering Research Center of AI & Robotics, Shanghai 200433;5. Faculty of Mechanical Engineering & Mechanics, Ningbo University, Ningbo 315211
  • Received:2021-08-04 Revised:2021-11-05 Online:2022-04-05 Published:2022-05-20

摘要: 针对多关节下肢康复机器人非线性人机交互作用力影响训练过程中舒适度和安全性的问题,提出一种人机耦合动力学建模方法和主动柔顺控制策略。结合脑卒中发病初-中期康复训练特点和降低髋关节输出功率的要求,确定出坐卧式下肢康复机器人自由度配置、自平衡结构原理,并完成具体模型设计。基于简化的人体肌骨模型,考虑人机耦合方式,确定出人机交互作用力模型,进一步给出考虑自平衡结构影响的人机耦合动力学模型。引入环境刚度模拟患者踩踏在不同路面的接触力效果,利用动力学前馈进行关节期望力矩补偿控制,并通过李雅普诺夫定理判定控制系统的稳定性。通过实验验证了轨迹跟踪、力跟踪的准确性,进一步通过临床试验,92.2%的脑卒中患者病情得到缓解,验证了研究内容的可行性和安全性。

关键词: 下肢康复机器人, 人机耦合动力学, 柔顺控制, 临床试验

Abstract: Aiming at the problem that the nonlinear human-computer interaction force of lower limb rehabilitation robot affects the comfort and safety of training process, a human-computer coupling dynamic modeling method and a dynamic feedforward control strategy are proposed. Combined with the characteristics of early rehabilitation training of stroke and the requirements of reducing hip output power, the degree of freedom configuration, self-balancing structure principle and specific model design scheme are determined. Based on the simplified human muscle bone model and considering the man-machine coupling mode, the man-machine interaction force model is determined, and the man-machine coupling dynamic model considering the influence of self-balancing structure is further given. The environmental stiffness is introduced to simulate the contact force effect of different soft and hard roads, the dynamic feedforward is used to carry out the feedforward compensation control of joint expected torque, and the stability of the control system is determined by Lyapunov method. The accuracy of trajectory tracking and force tracking were verified by experiments. Further, through clinical trials, 92.2% of stroke patients are relieved, which verifies the feasibility and safety of this research.

Key words: lower limb rehabilitation robot, human-machine coupling dynamics, compliance control, clinical trial

中图分类号: