Abstract: To improve the control accuracy of high power density drive units in supernumerary robotic limbs, a low-speed dead zone adaptive compensation method was proposed. Firstly, the drive unit model was established based on the least square system identification and damping identification methods. Then, a computed torque control method for the drive units was proposed. Finally, an adaptive compensation method for the low-speed dead zone of the drive units was proposed and verified. The experimental results show that compared with the traditional PID control method, the maximum angle tracking errors of the drive units are reduced by about 53% and the average angle tracking errors are reduced by about 38% after the computed torque control method is utilized. Based on the computed torque control method, the maximum angle tracking errors of the drive units are reduced by about 45% and the average angle tracking errors are reduced by about 60% after the low-speed dead zone adaptive compensation method is adopted. The control accuracy of the drive units is significantly improved.

Key words: drive unit, dead zone compensation, control method, supernumerary robotic limbs, system identification

摘要： 为提高外肢体机器人高功率密度驱动单元的控制精度，提出一种低速死区自适应补偿方法。首先基于最小二乘系统辨识与阻尼辨识方法建立驱动单元模型，然后提出驱动单元的计算力矩控制方法，最后提出驱动单元的低速死区自适应补偿方法，并进行了实验验证。研究结果表明，与传统PID控制方法相比，采用计算力矩控制方法后驱动单元最大角度跟踪误差减小了约53%，平均角度跟踪误差减小了约38%；在计算力矩控制方法的基础上，采用低速死区自适应补偿方法后，驱动单元最大角度跟踪误差减小了约45%，平均角度跟踪误差减小了约60%，驱动单元的控制精度得到了显著提高。

关键词: 驱动单元, 死区补偿, 控制方法, 外肢体机器人, 系统辨识