Research On The Motion Control Method For High-speed And Large-load AUV In Deep Sea

The ocean is a vast treasure trove of resources,and Autonomous Underwater Vehicle(AUV)is one of the powerful tools to explore the ocean,however,AUV is a typical dynamic system with strong coupling,nonlinearity and under-actuated.Compared with the traditional AUV,the lift type high-speed AUV has designed a lift wing,which will produce different lift at different speeds.Compared with the traditional AUV,the lift and high-speed AUV has the characteristics of wide speed and large load,which makes the precise motion control of the lift and high-speed AUV very difficult.Therefore,this thesis studies the characteristics of the lift wing,establishes the motion model of the lift AUV,and designs the sliding mode control algorithm based on the radial basis function neural network on the basis of ensuring the stability of the AUV,which effectively improves the steady-state accuracy and suppresses the chattering of the system.The specific research contents and work are as follows:Firstly,the differences between the lift and high-speed AUV and the traditional AUV are studied,the characteristics of the lift and high-speed AUV lift wing are analyzed,the kinematics and dynamics models of the lift and high-speed AUV are established,and the system model parameters are identified,which provides the basis for the design of the lift and high-speed AUV motion control algorithm and the simulation verification of the algorithm.Secondly,the basic principle of sliding mode control is studied,and the sliding mode control algorithm is designed based on Lyapunov stability using AUV system model.Through the further analysis of the sliding mode control theory aiming at the defect of its chattering which is hard to be applied in practice,the depth and heading controller of AUV is designed by improving the characteristics of the boundary layer.The simulation results show that the improved sliding mode control algorithm can reduce the chattering of the system,make the output of the controller smooth and improve its engineering practicability on the premise of ensuring the stability and control accuracy.Thirdly,the basic principle of RBF neural network is studied,and its online modeling ability is analyzed.In order to improve the accuracy of the model,the imprecise part of the model is compensated online by radial basis function neural network.The depth and heading controller of AUV is designed by the sliding mode control algorithm based on RBF neural network compensation.The simulation results show that the compensated sliding mode control algorithm can achieve better control accuracy under the premise of wreakening chattering compared with the former.Finally,the thesis analyzes the application of sliding mode control based on RBF neural network in AUV motion control,summarizes the research content and achievements of this thesis,and puts forward further prospects for the thesis.