The identification model for ship motion PID control using wavelet neural network

To identify the dynamic of ship's motion at sea quickly with high accuracy, a wavelet network is proposed for ship motion identification. The wavelet network is implemented as the on-line system identifier, whose parameters are tuned at each step. By combing the advantages of the fast learning speed of wavelet network and the robustness of custom PID control, a wavelet-network-based PID controller is proposed for control applications. The simulation results of ship course control demonstrate that the proposed controller can track the setting course accurately with fast computational speed.