欠驱动船舶RBF神经网络路径自适应跟踪控制

Path flowing adaptive control for underactuated surface vessels based on RBF neural network

针对模型参数未知和存在外界干扰的三自由度欠驱动水面船舶路径跟踪控制问题,提出一种RBF神经网络控制器.该算法利用神经网络的函数逼近特性对船舶模型未知的非线性部分在线逼近并与反步法相结合进行设计,同时实现前进速度在线可调.通过Lyapunov稳定性方法分析验证了闭环系统的稳定性.仿真计算验证了该控制策略的有效性.

For path flowing control problem of three degrees of freedom underactuated surface vessel model with unknown pa- rameters and external disturbances, a RBF adaptive neural network controller is proposed. The controller was designed by using neural network function approximation properties combi- ning with backstepping method to approach the unknown non- linear part of the ship model online, and realized the forward speed adjustable. Based on Lyapunov stability analysis it is proved that the final closed-loop system is stable. Simulation results show the effectiveness of the adaptive control strategy used.