Research On Event-based Predictive Control Of Resilient For Mobile Robot

The mobile robot system in the framework of cyber-physical technology mainly connects the actuator,controller,sensor and other physical units through the network layer to form a closed-loop control.However,the openness of the network layer makes the mobile robot system vulnerable to network attack during information transmission.To resist network attack such as false data injection(FDI),an resilient predictive control strategy needs to be developed on the basis of the traditional model predictive control(MPC).But the existing resilient control strategy only aims at ensuring the stability of the system,and lacks the consideration of the consumption of network resources such as online computing burden and communication bandwidth of the robot system.Therefore,it is difficult to directly apply to the actual mobile robot system with limited resources.To solve the above problems,this paper focuses on the research of resilient predictive control method based on event triggering mechanism,which mainly includes the following contents:Firstly,a predictive control method of resilient event-triggered for mobile robot based on sample-and-hold fashion is proposed.Two signal reconstruction strategies based on single sample and multiple samples sample-and-hold are constructed,and corresponding resilient event-triggered predictive control algorithms are designed.Then,the feasibility of the algorithms and the stability of the closed-loop system are proved by theoretical analysis.Combined with the mobile robot system,the simulation results show that the proposed algorithms has better resilient control performance than the traditional model predictive control algorithm,and with the increase of the number of samples,the control performance and resilient performance of the mobile robot gradually improved,which verifies the effectiveness and superiority of the algorithms.Secondly,in order to further reduce the state monitoring resource consumption of the robot system,a resilient self-triggered predictive control method based on energy balance transmission is proposed.A resilient control mechanism of energy balance is constructed which only provides extra protection for two key control samples,and an resilient self-triggered predictive control algorithm is proposed.Then,the theoretical characteristics of the proposed algorithm are analyzed,and the iterative feasibility and closed-loop stability of the resilient self-triggered predictive control system are proved.Combined with the mobile robot system,the simulation results show that the proposed algorithm has not only good triggering performance,but also better resilient control performance than the classical self-triggered model predictive control(STMPC)algorithm,which verifies the effectiveness and practicability of the algorithm.Finally,the proposed resilient self-triggered predictive control algorithm was verified experimentally on the SCOUT MINI mobile robot experimental platform.The mobile robot hardware experiment platform and robot operating system were configured,the local area network was established and the software algorithm was designed.Rviz was used to visualize the real-time state of the mobile robot,and the point stabilization motion control experiment of mobile robot was carried out.Through the analysis of the experimental data,it can be concluded that the resilient self-triggered predictive control strategy proposed in this paper can not only make the system quickly recover to the stable operating state when it encounters false data injection attack,but also can effectively balance the system resources.