摘要
设计了一种基于无模型自适应的外骨骼式上肢康复机器人主动交互训练控制方法.在机器人与人体上肢接触面安装力传感器采集人机交互力矩信息作为量化的主动运动意图,设计了一种无模型自适应滤波算法使交互力矩变得平滑而连贯;以人机交互力矩为输入,综合考虑机器人末端点与参考轨迹的相对位置和补偿力的信息,设计了人机交互阻抗控制器,用于调节各关节的给定目标速度;设计了将无模型自适应与离散滑模趋近律相结合的速度控制器完成机器人各关节对目标速度的跟踪.仿真结果表明,该控制方法可以实现外骨骼式上肢康复机器人辅助患者完成主动交互训练的功能.通过调节人机交互阻抗控制器的相应参数,机器人可以按照患者的运动意图完成不同的主动交互训练任务,并在运动出现偏差时予以矫正.控制器在设计实现过程中不要求复杂准确的动力学建模和参数识别,并有一定的抗干扰性和通用性.
This paper proposes an active interaction exercise control method for the exoskeleton upper limb rehabilitation robot based on model-free adaptive algorithm. Force sensors are mounted on the contact surface of the robot and the human upper limb to collect human-robot interaction torque information which is used to quantize active movement intention. A model-free adaptive filtering algorithm is designed to make the interaction torque smooth and continuous. A human-robot interaction impedance controller is designed for adjusting the given target velocity of each joint, according to the input of interaction torque and considering the relative position of the robot end-point to the reference trajectory and the compensation force information. By combining the model-free adaptive algorithm and the sliding mode exponential reaching law, a speed controller for all the joints of the robot is developed to implement the target speed tracking. Simulation results show that this control method can achieve the function of the exoskeleton for upper limb rehabilitation, which is to assist patients to complete the active interaction training exercise. By adjusting the corresponding parameters of the human-robot interaction impedance controller, the robot can perform different tasks of the active interaction training exercise in the light of patients＇ active movement intention, and give correction in motion deviation. In the designing and implementing the controller, complicated and accurate dynamic modeling and parameter identification are not required, and the control method has a certain degree of immunity and versatility.
作者
王晓峰
李醒
王建辉
WANG Xiao-Feng LI Xing WANG Jian-Hui(State Key Laboratory of Synthetical Automation for Process Industries, Northeastern University, Shenyang 110819 Institute of Automation, College of Information Science and Engineering, Northeastern University, Shenyang 110819)
出处
《自动化学报》
EI
CSCD
北大核心
2016年第12期1899-1914,共16页
Acta Automatica Sinica
基金
国家自然科学基金(61503070)
中央高校基本科研业务费专项资金(N150804001)
2015辽宁省博士点专项基金(201501142)资助~~
关键词
无模型自适应
康复机器人
主动交互
力传感器
阻抗控制
Model free adaptive, rehabilitation robot, active interaction, force sensor, impedance control
