Research On Adaptive Iterative Sliding Mode Control For Underactuated Ship Motion

In recent years,with the rapid development of underactuated system control technology and the increasing requirements of ship automation,it is of great practical significance to study the motion control of underactuated ships.The nonlinear control method of underactuated ship can improve the maneuverability,safety and economy of ship.At the same time,with the development of marine economy,it is necessary to complete such complicated tasks as laying of submarine pipelines,positioning of offshore drilling platforms,exploration of marine resources and replenishment of ocean.The requirements for precise control of ships should be increasing.Therefore,it is of great practical and theoretical value to study the motion control of underactuated ships.Firstly,an adaptive iterative sliding mode control method based on reinforcement learning is proposed for the path tracking problem of the three degrees underactuated ship.The method introduces a hyperbolic tangent function to design a iterative sliding mode for system state,and uses the neural network to optimize control parameters and enhance adaptive performance of the controller.The structure and parameters of the neural network can be adjusted online by defining a control amount chattering measured variable and reinforcement learning signal,it can further inhibit the chattering of control amount.Secondly,for the trajectory tracking problem of the underactuated ship,the iterative sliding mode controller based on reinforcement learning is extended to two-way control.And the iterative sliding mode controller is designed for the lateral and longitudinal deviations of the tracking trajectory respectively.The controller output is rudder angle and diesel engine speed.The parameters are adjusted according to the chatting of rudder angle and speed.The controller control structure of trajectory tracking is similar with path tracking control.Then,the iterative sliding mode controller based on reinforcement learning is applied to the automatic berthing control of underactuated ships.And an improved particle swarm optimization(PSO)algorithm is introduced to optimize the initial values of controller's parameters,which make the controller has a relatively good control effect of the ship in the berthing starting point.It's more conducive to the accuracy of the relatively high demand for automatic berthing control.Finally,The simulation of 5446 TEU container ship mathematical model is carried out in Matlab.The results of simulations show that the designed controller can successfully complete the control task and track the target trajectory quickly under the disturbance of the wind and wave.Compared with the iterative sliding mode controller,the control time and dynamic error are smaller,the chattering is reduced.The controller also successfully is carried out in the automatic berthing control.The control effect added the particle swarm adjustment parameter is better,the overshoot can't occur,and the safety of the ship is improved,which is more in line with the actual operational requirements of the ship.