Aiming at the path tracking problem of underactuated ships, a neural network sliding mode control method considering the saturation characteristics of control input is proposed in this paper. In order to better integrate with navigation practice, the influence of non diagonal elements of mass matrix and damping matrix is considered in the establishment of underactuated ship model. The input saturation problem is considered in the design of the controller, and the additional virtual control rate can be designed to simultaneously deal with the input saturation and underdrive problems. At the same time, the dynamic surface control (DSC) is introduced to estimate the derivative of the virtual control law to avoid the computational burden of direct derivation. The neural network minimum parameter learning method is used to deal with the uncertainty of the model and approximate the nonlinear term of the system. The adaptive law and practical control law of the system are designed with sliding mode control. The stability of the closed-loop system is proved by Lyapunov direct method. The simulation experiment results show that the control strategy proposed in this paper is more in line with actual engineering requirements while ensuring fast and efficient path tracking.
Path tracking is one of the typical scenarios for the research of ship controlling, which usually includes three parts, i.e., path planning, guidance strategy and control strategy. Among these, the control strategy is the most challenging problem, which requires the comprehensive consideration of underactuated characteristics, the interference of external environments and the uncertainty of ship models (Li, Tong et al. 2016). With assistance of modern advanced control theory, great research achievements have been obtained for the above problems. However, some research results are not compatible with the actual ships due to the imbalance of information between the complex algorithms and the practical applications.
