Research On Longitudinal Finite Time Control Of Intelligent Vehicle Fleet Considering Disturbance And Time Delay

Intelligent Vehicle Fleet(IVF),as one of the important technologies of intelligent traffic system,is of great significance for establishing environmentally friendly transportation,improving road traffic efficiency,and reducing traffic accidents.During intelligent vehicle fleet driving,vehicle spacing needs to be quickly adjusted to safety car spacing to improve driving safety and traffic efficiency.However,in the actual driving process,vehicles are susceptible to interference and communication delays,leading to a decrease in the following efficiency of the fleet and even affecting the safety of the fleet’s driving.This paper proposes to carry out research on intelligent fleet longitudinal following control based on finite-time control,design a global fast integral terminal sliding mode control strategy based on finite-time extend state observer for the disturbance factors in the vehicle travel process,and construct a modified Smith prognosticator to compensate the system time lag by prediction.The main research contents include:Firstly,this paper analyzes the longitudinal following process of the intelligent vehicle fleet and establishes the following distance model of the fleet and the vehicle dynamics model.On this basis,the finite-time stability criterion is introduced,and the finite-time criterion based on the terminal sliding mode theory is discussed.The comprehensive evaluation index of the longitudinal control stability of the intelligent vehicle fleet is analyzed,which lays a theoretical foundation for the subsequent research on the longitudinal control of the intelligent vehicle fleet.Secondly,this paper carries out the study of longitudinal finite-time control strategies for intelligent fleets considering matched and unmatched disturbances.The impact of matching interference and mismatch interference on the longitudinal following behavior of the vehicle fleet during the driving process is analyzed.Aiming at the matching disturbance during driving,a global fast integration terminal sliding mode controller(GFITSMC)based on non-smooth reaching law(NSRL)is designed to improve the stability of the motorcade system.Further considering the effect of mismatch disturbance,a finite-time extended state observer(FTESO)is constructed to observe the mismatch disturbance,and based on the results,the FTESO-GFITSMC is designed to improve the robustness of the fleet system.Then,this paper analyzes the influence of the distribution position of the system time delay on the longitudinal steady-state performance of the intelligent vehicle fleet when the system disturbance and time delay exist simultaneously.Considering the influence of the system disturbance and time-delay,an improved Smith predictor is introduced to synchronize the input of the extended state observer,and a predictive FTESO-GFITSMC is designed to improve the robustness of the fleet system under disturbances and time delay.Finally,simulation experiments are conducted on two different communication topologies for the longitudinal finite time control strategy of intelligent vehicle queues.The results show that compared with sliding mode controller(SMC)and dual power sliding mode controller(DPSMC),the designed GFITSMC can effectively suppress the influence of matching interference.Under the lead-follow(LF)communication topology,the convergence time of vehicle spacing error is reduced by 56.3% and 50%,and the peak error is reduced by72% and 58%,respectively.Under the preceding-follow(PF)communication topology,the convergence time of vehicle spacing error has been reduced by 51.4% and 48.6%,respectively.For the impact of mismatched interference,although GFITSMC can still maintain good convergence performance,it is possible to further improve the performance of acceleration buffeting.And the FTESO-GFITSMC considered can well restrain acceleration buffeting while maintaining the performance of GFITSMC.In response to the phenomenon of time delay in the system,the proposed improved Smith time delay prediction strategy can still achieve good stability under the influence of time delay.Under LF structure and PF structure,the convergence time of vehicle spacing is only 13 s and 17 s,respectively.Under two different communication topologies,there is no significant difference in the convergence process except for the convergence time,indicating that the proposed strategy can adapt to two different communication topologies.