School of Mechatronic Engineering and Automation, Shanghai University, Shanghai, China.
Department of Aerospace Engineering, Ryerson University, Toronto, ON, Canada.
J Healthc Eng. 2017;2017:6732459. doi: 10.1155/2017/6732459. Epub 2017 Aug 1.
This paper discusses the problem of squatting training of stroke patients. The main idea is to correct the patient's training trajectory through an iterative learning control (ILC) method. To obtain better rehabilitation effect, a patient will typically be required to practice a reference posture for many times, while most of active training methods can hardly keep the patients training with correct posture. Instead of the conventional ILC strategy, an impedance-based iterative learning method is proposed to regulate the impedance value dynamically and smartly which will help patients correct their posture gradually and perform better. To facilitate impedance-based ILC, we propose two objectives. The first objective is to find the suitable values of impedance based on the ILC scheme. The second objective is to search the moderate learning convergence speed and robustness in the iterative domain. The simulation and experimental results demonstrate that the performance of trajectory tracking will be improved greatly via the proposed algorithm.
本文讨论了脑卒中患者蹲起训练的问题。主要思想是通过迭代学习控制（ILC）方法来纠正患者的训练轨迹。为了获得更好的康复效果，患者通常需要多次练习参考姿势，而大多数主动训练方法很难让患者保持正确的姿势。本文提出了一种基于阻抗的迭代学习方法，代替传统的 ILC 策略，通过动态、智能地调节阻抗值，帮助患者逐渐纠正姿势，提高训练效果。为了方便基于阻抗的 ILC，我们提出了两个目标。第一个目标是根据 ILC 方案找到合适的阻抗值。第二个目标是在迭代域中搜索适度的学习收敛速度和鲁棒性。仿真和实验结果表明，通过所提出的算法可以大大提高轨迹跟踪的性能。
