虚假数据注入式攻击下无人水面船舶自适应神经输出反馈轨迹跟踪控制  

Adaptive Neural Output Feedback Trajectory Tracking Control for USVs Under False-data-injection Attacks

作　　者：祝贵兵 吴晨 马勇[2,3,4,5] ZHU Gui-Bing;WU Chen;MA Yong(School of Naval Architecture and Maritime,Zhejiang Ocean University,Zhoushan 316022;State Key Laboratory of Maritime Technology and Safety,Wuhan University of Technology,Wuhan 430063;School of Navigation,Wuhan University of Technology,Wuhan 430063;National Engineering Research Center for Water Transport Safety,Wuhan University of Technology,Wuhan 430063;Academician Workstation of COSCO SHIPPING Group,Shanghai 200135)

机构地区：[1]浙江海洋大学船舶与海运学院,舟山316022 [2]武汉理工大学水路交通控制全国重点实验室,武汉430063 [3]武汉理工大学航运学院,武汉430063 [4]武汉理工大学国家水运安全工程技术研究中心,武汉430063 [5]中国远洋海运集团院士工作站,上海200135

出　　处：《自动化学报》2024年第7期1472-1484,共13页Acta Automatica Sinica

基　　金：国家自然科学基金(52261160383,52022073,62073251);装备预研教育部联合基金(8091B022239);海南自然科学基金创新研究团队项目(722CXTD518);武汉基础研究知识创新计划(2022010801010181);舟山科技局项目(2022C41006);武汉理工大学重庆研究院研究项目(YF2021-12)资助。

摘　　要：本文主要研究网络环境下无人水面船舶(Unmanned surface vessels,USVs)遭受虚假数据注入式(False-data-injection,FDI)攻击的跟踪控制问题.其中,内部和外部不确定以及输入饱和约束等实际因素均考虑在设计中.在控制设计过程中,为避免将船舶速度的攻击信号引入闭环系统,采用分类重构思想,构造一种新的神经网络(Neural network,NN)状态观测器,同时重构船舶速度和攻击信号.进一步,在backstepping设计框架下,利用重构的攻击信号补偿USVs运动学通道因虚假数据注入式攻击引起的非匹配不确定项.在动力学设计通道中,利用自适应神经技术和单参数学习法,重构由内部和外部不确定组成的复合不确定部分,进而提出自适应神经输出反馈控制方案.理论分析表明,即便在FDI攻击、内外不确定以及执行器饱和约束的情况下,所提控制方案仍能迫使USVs跟踪给定的参考轨迹.同时,仿真和比较结果证实了所提控制方案的有效性和优越性.This paper investigates the tracking control issue of unmanned surface vessels(USVs)under the attack of false-data-injection(FDI)in the network environment,and these actual factors such as internal and external uncertainties and input saturation constraints are also considered in the design.In the control design,to avoid FDI attack signals from the velocity channel being introduced into the closed-loop system,the idea of classification reconstruction is developed.Based on this idea,a novel neural network(NN)state observer is constructed to reconstruct vessels velocity and FDI attack signals.Furthermore,under the backstepping design framework,utilizing the reconstructed attack signals to compensate the mismatched uncertainties in USVs kinematic channel,which is caused by false-data-injection attacks.In the dynamic design channel,adaptive neural technology and single parameter learning method are used to reconstruct the lumped uncertain parts,which consist of internal and external uncertainties,and then the adaptive neural output feedback control scheme is proposed.The theoretical analysis shows that the proposed control scheme can make USVs track a given reference trajectory,even in the presence of FDI attacks,internal and external uncertainties,and actuator saturation constraints.At the same time,the simulation and comparison results illustrate the effectiveness and superiority of the proposed control scheme.

关 键 词：无人水面船舶 虚假数据注入式攻击 跟踪控制 单参数学习法 自适应神经控制 输出反馈 

分 类 号：U664.82[交通运输工程—船舶及航道工程] TP273[交通运输工程—船舶与海洋工程]