一种智能膝关节假肢及其控制算法研究  

Research on an intelligent above-knee prosthesis and its control algorithm

作　　者：张意彬[1,2] 吕杰 喻洪流 Zhang Yibin;Lü Jie;Yu Hongliu(School of Health Science and Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China;College of Medical Instrumentation,Shanghai University of Medicine&Health Sciences,Shanghai201318,China;College of Rehabilitation Sciences,Shanghai University of Medicine&Health Sciences,Shanghai201318,China)

机构地区：[1]上海理工大学健康科学与工程学院,上海200093 [2]上海健康医学院医疗器械学院,上海201318 [3]上海健康医学院康复学院,上海201318

出　　处：《现代仪器与医疗》2022年第6期19-27,共9页Modern Instruments & Medical Treatment

基　　金：国家自然科学基金资助项目(61473193)。

摘　　要：为了使下肢假肢的功能更趋近于人体下肢,提出了一种智能膝关节假肢的设计方法及其相应的控制算法.首先,根据人体膝关节的结构特点,设计仿生膝关节的机械组件.并在此基础上,为其添加分体式的膝关节阻尼控制模块.然后,为阻尼控制模块添加控制算法[传统PD控制算法、微粒群(PSO)优化的BP神经网络PD控制算法].最后通过步态测试对比两种不同算法的实际控制效果.最终的实验结果表明:在行走的摆动相,PSO-BP-PD控制相较于传统PD控制,膝关节摆动角度的控制准确度提升5.1°,稳定度下降0.7°.In order to make the function of lower limb prosthesis more similar to that of human lower limb, a design method of intelligent above-knee prosthesis and its related control algorithm were proposed. Firstly, according to the structural characteristics of human knee joint, the mechanical components of bionic knee joint are designed. On this basis, a separated damping control module of knee joint is added. Then, add control algorithms [traditional PD control algorithm, particle swarm optimization(PSO) BP neural network PD control algorithm] to the damping control module. Finally, the actual control effects of two different algorithms are compared through gait test. The final experimental results show that the swing angle of above-knee prosthesis increased 5.1° in accuracy and decreased 0.7° in stability.

关 键 词：微粒群优化 BP神经网络 智能 膝关节假肢 比例微分控制 

分 类 号：R197.39[医药卫生—卫生事业管理]