Ship Course And Cooperative Optimal Control Based On Adaptive Dynamic Programming

The ocean is a treasure house of human resources and an important strategic space for sustainable development.In recent years,more and more attention has been paid to the exploration and development of marine resources all over the world and the research on ship motion control has been paid more and more attention by experts and scholars.However,with the rapid growth of the number of vessels and the task of offshore operations becomes more complicated,the single vessel can no longer meet the actual engineering needs.The multiple vessels cooperative control has the advantages of good maneuverability and anti-interference,which has been widely used in the marine practical engineering,such as maritime search and rescue,submarine operation and marine survey.This paper focuses on ship course and multiple ships cooperative optimal control problem.The main work of this paper are as follows:(1)In order to solve the problem that it is difficult to establish an accurate mathematical model due to the complexity of ship mathematical model and the fact that the ship is affected by the external disturbance during the voyage,this paper adopts the offline adaptive dynamic programming algorithm to establish the optimal course controller for model-free vessel.This method does not need the accurate model parameter information of the ship,uses the input-output data of the ship,combines the neural network technology to design the controller,and proves the stability of the control system by the Lyapunov stability theorem.Finally,the Yupeng ship of Dalian maritime university is used as the simulation object,and the simulation experiment is carried out with MATLAB software to verify the effectiveness of the algorithm.(2)To solve the problem of multiple ships cooperative control with unknown model,an online adaptive dynamic programming algorithm based on time is adopted in this paper.The critic and actor neural network are used to approximate the performance index function and the optimal control policies of multi-agent systems,respectively.The controller is designed based on the current and historical data of the ship.and the state of the follower and the leader tends to be consistent.The stability analysis of Lyapunov shows that the error system is uniform and ultimately bounded.Finally,numerical simulation is carried out with MATLAB software to verify the effectiveness of the algorithm.