人机不确定条件下康复步行训练机器人的部分记忆迭代学习限速控制

Partial Memory Iterative Learning Control with Velocity Constraints for Rehabilitative Training Walker under Human-robot Uncertainty

为了提高康复步行训练机器人的跟踪精度及安全性,提出了一种带有运动速度约束和部分记忆信息的自适应迭代学习控制方法,目的是抑制人机不确定性及速度突变对系统跟踪性能的影响.在考虑人机不确定性的基础上,建立了康复步行训练机器人的动力学模型.提出了基于模型预测的速度约束方法,通过限制每个轮子的运动速度,约束了机器人的实际运动速度.进一步,利用受约束的运动速度建立了动力学跟踪误差系统,提出了具有部分记忆信息的自适应迭代学习控制器设计方法,并验证了跟踪误差系统的稳定性.仿真对比分析和实验研究结果表明,文中提出的控制方法能抑制人机不确定性并使康复者在安全速度下完成步行训练.

To improve the tracking accuracy and safety of the rehabilitative training walker, an adaptive iterative learning control method with velocity constraints and partial memory information is proposed to reduce the influence of humanrobot uncertainty and velocity mutation on the tracking performance of the human-robot system. Taking the human-robot uncertainty into consideration, a dynamic model of the rehabilitative training walker is established. A model prediction based velocity constraints method is proposed, which restricts the actual movement velocity of the robot by limiting the movement velocity of each wheel. Furthermore, a dynamic tracking error system is established by using the constrained movement velocity, a design method of adaptive iterative learning controller with partial memory information is proposed, and the stability of the tracking error system is verified. Comparative simulation analysis and experimental research are carried out,and the results show that the proposed control method can restrain the human-robot uncertainty, and enable the recovered person to walk at a safe velocity.