Research on adaptive AUV tracking control system based on least squares support vector machine

Research on adaptive AUV tracking control system based on least squares support vector machine

Song Xiao-ru1, Chen Hua2, Song Bao-wei3, Wu Jie1

COMPUTER MODELLING & NEW TECHNOLOGIES 2014 18(2) 33-39

1 School of Electronic and Information Engineering, Xi’an Technological University, 710032, Xi’an, P.R.China
2 School of Mechatronic Engineering, Xi’an Technological University, 710032, Xi’an, P.R.China
3 School of Marine Engineering, Northwestern Polytechnical University, 710072, Xi’an, P.R.China


Aimed at the nonlinear, uncertainties and impreciseness in AUV tracking control, an adaptive LSSVM control is highlighted. It is including the approximate dynamic inversion control law, linear controller and the robust adaptive LSSVM controller. The key is designed that adaptive LSSVM controller, which is to decrease or offset the uncertain dynamic inverse model errors. The online adjustment LSSVM parameters rules are deduced by the Lyapunov stability theorem. So the closed-loop tracking control system's stability and asymptotical convergence of tracking error can be guaranteed. It can be seen that the tracking errors converge and stay at a small neighbourhood of zero. At last taken a certain real AUV as an example, the numerical simulation results are given in the presence of the ocean current wave interference. They show the proposed method has good tracking performance and a certain robustness against modelling errors.