基于AIA优化LS-SVR的参数选取及在非线性系统辨识中的应用(英文)

Parameters Selection of LS-SVR Based on AIA and Its Application in Nonlinear System Modeling

研究了LS-SVR的参数对LS-SVR泛化性能的影响,在此基础上提出了基于自适应免疫优化算法和k-折交叉检验相结合的LS-SVR参数整定方法。并将其应用于两自由度机器人的逆运动学建模中,仿真结果表明了该方法的有效性。

The effect of parameters selection on approximating performance of least squares support vector regression （LS-SVR） was investigated. Then, parameters selection method of LS-SVR was proposed based on adaptive immune algorithm （AIA） plus 5-fold cross validation, which was applied to inverse kinematics modeling of 2DOF robot. Simulation results verify the feasibility and validity of this method.