摘要
下肢假肢的膝关节是一种具有高度非线性、时变、强耦合的阻尼系统,传统控制方法很难达到良好控制效果.针对这一问题,提出将神经网络(NN)应用于下肢假肢控制.以学习矢量量化(LVQ)神经网络为基础,提出神经网络模型参考自适应控制方法.该方法通过选择适当的参考模型和自适应算法,利用参考模型输出与实际系统输出之间的误差信号,由自适应算法计算当前的控制量以控制智能下肢假肢,达到自适应控制的目的.该方法不需要进行性能指标的变换,容易实现且自适应速度快,仿真结果表明了该方法的有效性.
The knee joint of lower limb prosthesis is a damp system with high nonlinearity,time-varying and strong coupling,and the traditional control method can hardly achieve good performance.Aimed at the problem,a neural network(NN)-based model reference adaptive control method was proposed based on the learning vector quantization(LVQ) neural network.Based on an appropriate reference model and an adaptive algorithm,the current control variable was calculated by using the error between the reference model output and the actual system output in order to control the intelligent lower-limb prosthesis and achieve the adaptive control.The method does not require the transformation of performance criteria,and is fast and easy to implement.Simulation results showed the validity of the method.
出处
《浙江大学学报(工学版)》
EI
CAS
CSCD
北大核心
2010年第7期1373-1376,1381,共5页
Journal of Zhejiang University:Engineering Science
基金
国家"863"高技术研究发展计划资助项目(2008AA04Z212)
国家自然科学基金资助项目(60874102
60705010)
浙江省自然科学基金资助项目(Y1080854
Y1090761)
浙江省教育厅基金资助项目(Y200907366
Y200805157)