基于有限时间指令滤波反步法的船舶动力定位自适应控制  

Adaptive dynamic positioning control for ships based on finitetime command filtered backstepping

作　　者：陈兴华[1] 韩旭 刘彩云 夏国清 李娟[1] CHEN Xing−hua;HAN Xu;LIU Cai−yun;XIA Guo−qing;LI Juan(College of Intelligent Systems Science and Engineering,Harbin Engineering University,Harbin 150001,China;Guangdong Technology College,Zhaoqing 526100,China;Shandong Technology and Business University,Yantai 264005,China)

机构地区：[1]哈尔滨工程大学智能科学与工程学院,哈尔滨150001 [2]广东理工学院,广东肇庆526100 [3]山东工商学院,山东烟台264005

出　　处：《船舶力学》2025年第4期547-559,共13页Journal of Ship Mechanics

基　　金：工信部“第七代超深水钻井平台(船)创新专项”。

摘　　要：本文研究船舶动力定位(DP)系统中存在的推进器动态特性、模型参数不确定、推进器输入饱和及外部扰动未知等问题,提出一种基于有限时间指令滤波反步法的鲁棒自适应动力定位控制方法,该方法不仅具有滤波反步法的优点还能保证控制系统在有限时间内收敛。首先,采用神经网络技术对系统中的非线性项进行逼近;其次,设计有限时间辅助动态系统(FTADS)解决推进器输入饱和问题;最后,结合设计的神经网络估计值和FTADS,采用有限时间指令滤波反步法对控制器进行设计,利用Lyapunov稳定性理论证明系统跟踪误差和参数估计误差是有限时间收敛的。并对所提控制方法进行仿真实验,仿真结果表明所提出的控制器是有效性的。This paper focuses on proposing a finite-time command filtered backstepping robust adaptive dynamic positioning control method to deal with the thruster dynamics,parameters uncertainty,input saturation and unknown external disturbance problems.The proposed method not only has the advantages of the command filtered backstepping,but also can guarantee the control system convergence in finite time.Firstly,adaptive neural network was used to estimate uncertain functions in the system.Secondly,the thruster input saturation issue was addressed with a finite-time auxiliary dynamic system.Finally,based on the uncertain estimation and the FTADS,a finite-time command filtered backstepping(FTCFB)control law was introduced,and the system tracking errors and parameter estimating errors were proved to be convergent in finite time by using Lyapunov stability theory.Additionally,the effectiveness of the proposed positioning control method was verified by numerical simulations.

关 键 词：动力定位系统 推进器动态特性 神经网络估计器 有限时间指令滤波反步法 有限时间收敛 

分 类 号：U664.3[交通运输工程—船舶及航道工程] U664.82[交通运输工程—船舶与海洋工程]