实时上肢参数辨识的康复机器人力控制研究

摘要

When rehabilitating training for patients, the auxiliary force calculation errors are caused by inaccurate estima-tion and change of upper extremity kinetics parameters, which affect accurate and stable training control. In order to reduce the auxiliary force calculation error and achieve accurate and stable control performance, an impedance control algorithm with the real-time upper limb dynamic parameter identification is proposed based on impedance control algorithm and using multiple linear regression method to identify the upper extremity kinetics parameters in this paper. The dynamic model of the rehabilitation robot is modeled and the proposed impedance control algorithm is researched by simulation. The simula-tion results show that the impedance control algorithm can identify the upper extremity kinetics parameters and eliminate the auxiliary force calculation errors in the training process, and realize the precise trajectory control.%针对病人进行康复训练时，上肢动力学参数估计不准确和训练过程发生上肢动力学参数变化，所导致康复机器人系统辅助力计算不准确，影响精确和稳定的控制练训。为减小辅助力计算误差，实现精确和稳定的训练控制，基于阻抗控制算法，使用多元线性回归方法对上肢动力学参数进行辨识，提出了一种实时上肢动力学参数辨识的阻抗控制算法，建立了康复机器人动力学模型，同时对控制算法进行仿真研究。仿真结果表明该算法能够准确地对上肢动力学参数进行辨识，有效地消除了辅助力计算误差，实现训练过程中训练轨迹精确控制。