输入受限的非仿射无人帆船航向系统自适应动态面控制
Adaptive dynamic surface control for nonaffine unmanned sailboat course system with input constraint

针对输入受限和控制方向未知的无人帆船航向控制问题,考虑系统模型存在动态不确定和未知外界扰动的情况,本文提出一种基于非仿射航向运动数学模型的最小参数自适应递归滑模动态面控制策略。该策略通过Taylor展开方法将非仿射模型转化为具有线性结构的仿射时变系统,采用最小参数学习神经网络逼近无人帆船模型不确定部分,并利用双曲正切函数处理控制输入饱和现象,引入Nussbaum函数处理系统中未知控制方向问题,同时综合考虑帆船艏摇角速度误差和航向误差之间关系设计递归滑模动态面舵角控制律,并设计参数自适应律对神经网络逼近误差与复合干扰总和的界进行估计。选取李雅普诺夫函数证明了所设计控制器能够保证航向闭环系统内所有信号的一致最终有界性。最后,基于一艘12m无人帆船进行仿真验证,结果表明无人帆船航向控制响应速度快,所设计的控制器能有效地处理模型不确定项和风浪等外界扰动,具有较强的鲁棒性。
To solve the course control problem with input saturation and unknown control direction, a minimal learning parameter(MLP) based adaptive recursive sliding-mode dynamic surface course control method is proposed in the presence of the model uncertainties and unknown external disturbances for the unmanned sailboat non-affine course motion model. The non-affine system is first transformed into a time-varying system with a linear structure using the Taylor expansion method. The hyperbolic tangent function is used to handle the input constraint, and the MLP is adopted to approximate the model’s uncertain part. The problem of unknown control direction is properly solved by using Nussbaum gain function. Then a recursive sliding-mode dynamic surface rudder control law is designed based on the relationship between yaw angular velocity and course errors. Moreover, the adaptive law is introduced to estimate the boundary value of neural network approximation error and compound disturbances. The application of Lyapunov function proves that all signals of the resulting closed-loop system can be guaranteed the uniformly ultimate boundedness by the proposed controller. The simulation results based on a 12m unmanned sailboat show that the unmanned sailboat course control response speed is fast, and the design controller has strong robustness against the system model uncertainty, wind, flow, as well as other external disturbances