Design of Fuzzy Controller of the Cycle-to-Cycle Control for Swing Phase of Hemiplegic Gait Induced by FES
Design of Fuzzy Controller of the Cycle-to-Cycle Control for Swing Phase of Hemiplegic Gait Induced by FES
复制标题
DOI：
10.1093/ietisy/e89-d.4.1525
复制
发表时间：
2006-04
期刊：
影响因子：
--
通讯作者：
A. Arifin;Takashi Watanabe;N. Hoshimiya
中科院分区：
文献类型：
--
作者：
A. Arifin;Takashi Watanabe;N. Hoshimiya
本研究的目的是设计一种实用的模糊控制器，用于功能电刺激（FES）诱导步态摆动阶段的多关节运动的周期到周期控制。首先，我们设计了三个模糊控制器（固定模糊控制器、基于梯度下降法的参数调整模糊控制器和基于模糊模型的参数调整模糊控制器）和两个PID控制器（固定PID控制器和自适应PID控制器）来控制双关节（膝盖和脚踝）的运动。通过肌肉-骨骼模型的计算机模拟，测试了所设计的控制器在自动生成刺激爆发持续时间和补偿肌肉疲劳方面的控制能力。模糊控制器在两方面的控制性能均优于PID控制器。结果表明，基于模糊模型的参数调整在主体间可变性引起振荡响应时是有效的。在此基础上，设计了模糊控制器，并利用模糊模型实现对三关节（髋关节、膝关节和踝关节）运动的模糊调节。计算机仿真得到的受控步态模式与正常步态模式无明显差异，可定性地用于临床FES步态控制。针对FES步态摆动阶段多关节运动的周期对周期控制所设计的模糊控制器有望得到临床检验。
The goal of this study was to design a practical fuzzy controller of the cycle-to-cycle control for multi-joint movements of swing phase of functional electrical stimulation (FES) induced gait. First, we designed three fuzzy controllers (a fixed fuzzy controller, a fuzzy controller with parameter adjustment based on the gradient descent method, and a fuzzy controller with parameter adjustment based on a fuzzy model) and two PID controllers (a fixed PID and an adaptive PID controllers) for controlling two-joint (knee and ankle) movements. Control capabilities of the designed controllers were tested in automatic generation of stimulation burst duration and in compensation of muscle fatigue through computer simulations using a musculo-skeletal model. The fuzzy controllers showed better responses than the PID controllers in the both control capabilities. The parameter adjustment based on the fuzzy model was shown to be effective when oscillating response was caused due to the inter-subject variability. Based on these results, we designed the fuzzy controller with the parameter adjustment realized using the fuzzy model for controlling three-joint (hip, knee, and ankle) movements. The controlled gait pattern obtained by computer simulation was not significantly different from the normal gait pattern and it could be qualitatively accepted in clinical FES gait control. The fuzzy controller designed for the cycle-to-cycle control for multi-joint movements during the swing phase of the FES gait was expected to be examined clinically.