The Research Of Improved Variable Structure Controller Of Autonomous Underwater Vehicle

An Autonomous Underwater Vehicle (AUV) is a seriously nonlinear system with strong coupling in every displacement. In general, it is difficult to get its precise moving equations. As an integrate control method, the merit of sliding mode control (SMC) method lies in that it can use imprecise mathematics models to design the controller. It has quite strong robustness, so it is suitable to control AUV, but the chattering exists in the sliding mode .In this dissertation, The basic theory of the sliding mode control and controller's design steps are introduced. For the controllable regular multi-input nonlinear system, two methods of designing the switching function are presented: the pole setting and the quadratic form optimization. In addition, For the deficiencies of traditional method—chattering, some improved methods are developed, which are the exponent reaching law solution based on fuzzy control and neural network respectively.This dissertation applies SMC method to AUV. Based on the need of practice and the convenience of designing the controller, the model is properly simplified from the dynamic model with 6DOF into 5 DOF (without rolling). Tendency ration strategy is presented to acquire the equation of the sliding mode control. The pole setting method and the quadratic form optimization are presented to design the switching function respectively. And then two improved sliding-mode controllers are presented to eliminate chattering.AUV's position control and velocity control are researched. In this dissertation, simulation experiment is carried out at two situations: in deep ocean with no current and in ocean with current. Simulation results show that the controller presented in this dissertation can not only effectively weaken the effect of chattering to system, but also have strong adaptability and high control precision.