基于反步自适应神经网络的船舶航迹控制

针对欠驱动船舶在稳定航速条件下轨迹跟踪问题,提出了一种基于自适应神经网络与反步法相结合的控制算法.该算法将实际的欠驱动船舶视为模型完全未知的非线性系统,利用神经网络的函数逼近特性实现控制器中非线性部分的在线估计,采用同时调整输入层-隐层、隐层-输出层间的权值阵的方法进行神经网络权值调整.通过选取积分型Lyapunov函数证明了闭环系统的稳定性.仿真实验表明该控制策略具有良好的跟踪特性,可以实现对期望航迹的精确跟踪.

Trajectory tracking control of ships based on an adaptive backstepping neural network

In this paper,the trajectory tracking problem for an underactuated ship at a stable speed was studied,and a control algorithm was put forward based on an adaptive neural network and backstepping.This algorithm regarded the mathematical model of an underactuated ship as a system with the model uncertainties.The uncertainties of the proposed controller were estimated by approximating the characteristics of a neural network online.Furthermore,the weight matrix of the hidden layer and output layer and the weight matrix of the input layer and hidden layer simultaneously adjusted.The stability of the proposed control strategy was accomplished by an integral-type Lyapunov function.Numerical simulation results indicate that the control strategy possesses favorable tracking characteristics and can track the desired path accurately.