Open AccessImproved SVM algorithm for upper limb rehabilitation mechanical arm Study on the Prediction of Track Tracking Control
Author(s) -
Zhaobi Chu,
Wei Wang,
Di Wang
Publication year - 2020
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1486/7/072003
Subject(s) - support vector machine , control theory (sociology) , decoupling (probability) , robustness (evolution) , trajectory , nonlinear system , computer science , algorithm , artificial intelligence , engineering , control engineering , control (management) , biochemistry , chemistry , physics , quantum mechanics , astronomy , gene
Aiming at the strong coupling, nonlinear and time-varying characteristics of upper limb rehabilitation training manipulator, a trajectory tracking predictive controller based on SVM (support Vector Machine) is designed. The input and output data of the manipulator system are collected, and the generalized inverse system is obtained by SVM identification, which is decoupling from the original system in series. For the decoupling system, the improved SVM algorithm is used to predict the trajectory tracking control, and the SVM algorithm is improved by combining the predictive function control method of PSO optimized rolling control sequence. The improved SVM algorithm can predict the trajectory of upper limb rehabilitation manipulator with high precision, and the experimental results show that the improved algorithm has good adaptability to the stability and robustness of the system.