Non-smooth Finite-time Control Of Autonomous Underwater Vehicles

The 21st century is the century of development of marine, underwater vehicles play a key role in the marine development and will gain greater development. The development of underwater vehicles and control technology advances are inseparable, but the traditional control algorithm can not meet the demand for underwater vehicles. Since finite-time control algorithm can converge in finite time and has strong anti-jamming features, it attracts the attention of a growing number of scholars. Control algorithms continuously improve and the control precision constantly improves with the increasing demands for underwater vehicles control performance. How to design control algorithms with strong anti-disturbance ability, fast convergence and high precision without increasing the complexity of control algorithms has become a research hotspot.Firstly, the paper describes search background, current status at home and abroad and future prospects of autonomous underwater vehicles, the existing problems of autonomous underwater vehicles are summarized in the development process.Secondly, some basic concepts and theories about finite-time control and homogeneity are presented, the models of autonomous underwater vehicles and autonomous underwater vehicles with manipulators are respectively discussed.Thirdly, based on the model of autonomous underwater vehicles, for the rapid dive of autonomous underwater vehicles when the state is limited switching control algorithm is proposed. Then MATLAB is used to simulate the algorithm and the conventional algorithm, then the simulation results are compared.Fourthly, according to the model of autonomous underwater vehicles system with manipulators, finite-time control and bounded finite-time control are studied, MATLAB simulation results compare with the similarities and differences between traditional smooth controller and non-smooth finite-time controller, the simulation results also present the parameter values of finite-time controller affect on the convergence velocity. Moreover, adaptive finite-time control of the system is also discussed, then the adaptive finite-time control law is given. Finally, the results of the thesis are summarized, and the further researches and the possible works are pointed.