Research On Event-triggered Safety Control Of Autonomous Unmanned Vehicles Under Cyber Attack

Information space and physical system interlace with networked control system(NCS).NCS has been widely used in the traffic and transportation.With the development of emerging tech-nologies such as industrial Internet,cloud computing and big data,the security problems in NCS have gradually becoming more pressing.In this paper,several event-triggered control strategies with unmanned ground vehicle(UGV)system,which are proposed to solve the matters in NCS,such as limited communication bandwith,cyber-attack.The main contributions of this paper are listed as follows:1.State-sensitive based event-triggered_∞control for path following of UGVA state-sensitive event-triggered_∞control strategy to solve the problem of UGV path fol-lowing control under limited communication network.Firstly,the path following control model is established according to the dynamics of UGV.Secondly,a novel state-sensitive event-triggered communication strategy is proposed,its main advantage is that the event-triggered threshold can be dynamically adjusted refer to the state of control system.Then,we combine with time-delay system modeling method and Lyapunov stability theory,an event-triggered_∞controller is de-signed under the proposed event-triggered communication scheme.Finally,the effectiveness of the proposed control strategy is verified by some simulation experiments.2.Security control of UGV subject to Sigmoid-like event-triggered communication schemeA Sigmoid-like event-triggered scheme for UGV security cruise control systems is proposed to cope with the stochastic false data injection(FDI)attacks in communication network.To im-prove the perception ability of event-triggered scheme to system state,an event-triggered commu-nication scheme based on Sigmoid-like function is proposed,and the event-triggered threshold can be dynamically adjusted according to real-time system state.Then,we assume a kind of stochastic FDI attacks which obey Bernoulli distribution exists in cruise control systems,the co-design of se-curity cruise control and Sigmoid-like event-triggered communication scheme is proposed in this paper.Next,a sufficient stability and stabilization criterion of an_∞control is well derived ac-cording to control objectives.Finally,some simulation experiments are used to verify the validity of Sigmoid-like event-triggered security cruise control scheme for networked UGVs.3.Output-sensitive event-triggered security control of UGV under stochastic FDI at-tacksAn event-triggered control scheme based on output-sensitive state is proposed for the prob-lems of UGV security control under cyber-attacks.In the case of controlled system state cannot be fully measured,the proposed communication scheme dynamically adjusts the event-triggered threshold according to available output measurements.Considering a class of stochastic FDI at-tacks and output-sensitive event-triggered communication scheme,the discrete sampling control system is transformed into a time-delay system model by input-delay method.By virtue of Lya-punov stability theory and LMI techniques,sufficient criterions on both stability analysis and con-troller synthesis are well derived.Finally,the effectiveness of the proposed security control strategy is verified by simulation results.4.Neural-network-based event-triggered security control of UGV under attacked actu-atorTo cope with stochastic cyber-attack on actuator of UGV path-following control system,an event-triggered adaptive security control scheme based on RBF neural network is proposed.First-ly,RBF neural-network is used to adaptively estimate and compensate such malicious cyber-attack signal exist in actuator.Then,modelling issues on the co-design of Sigmoid-like event-triggered communication scheme and UGV path-following control are shown when actuators are suffering from abnormal signal.Based on the proposed cooperative control model and combine with Lya-punov stability theory and LMI techniques,the stability criteria and stabilization controller are well proposed in this work.Finally,some simulation experiments are used to verify the effectiveness of neural-network-based event-triggered path following security control scheme.