Stable adaptive neural network control of path following for underactuated ships
欠驱动船舶路径跟踪的神经网络稳定自适应控制

We propose a stable adaptive-neural-network control for the path following in a 3 degrees-of-freedom (3DOF) underactuated ship with parameters uncertainties and disturbances. Based on the diffeomorphism transformation and Lyapunov direct method, controllers for the reference yaw angle and the reference speed are designed. Signal errors in the closed-loop path following system are proved to be uniformly ultimately bounded in a small neighborhood of zero. Numerical simulation results are given to illustrate the effectiveness of the stable adaptive-neural-network algorithm.