Research On Active Safety Control Method For Automated Heterogeneous Platoons

With the continuous development of economy and society,the improvement of intercity highway provide guarantee for economic development.However,with the increasing traffic capacity of highway,traffic congestion is becoming increasingly serious,resulting in increasingly serious traffic safety problems.Traffic inefficiency caused by traditional single-vehicle driving mode is the main reason for the above problems.With the development of intelligent vehicle infrastructure cooperative technology and automatic drive technology,intelligent platoon can organize different intelligent vehicles by improving the information interaction ability between vehicles.By reducing the intra platoon spacing,the highway efficiency can be improved and the energy cost can be decreased.With the improvement of traffic efficiency brought by platoon driving mode,the traffic safety risk also increased.Therefore,on the basis of the existing research on platooning,it is an urgent problem for the application of intelligent platoon to put forward the active safety method,optimize the platoon formation mode,predict and resolve different types of traffic risks in advance by means of collision waring and vehicle control.From the point of view of ensuring the safety of heterogeneous platoon,the dissertation research on the active safety method of automated heterogeneous platoon in different traffic scenarios.The heterogeneous platoon in the dissertation refers to the platoon composed of different types of vehicles with different characteristics.Based on modeling uncertain and diversity vehicle and V2V(Vehicle to Vehicle)communication parameters,a heterogeneous vehicle platoon formation method is proposed to meet the safety requirement through risk assessment based on typical scenarios and cruise control stability analysis.In views of inter platoon risk,the cruise control stability and active steering mode control are adopted to ensure the safety of platoon in normal and emergency driving mode respectively.In views of external platoon risk,the end of queue collision warning guarantee platoon safety based on V2V communication.Based on the research of this paper,the following innovation points have been achieved:(1)A dynamic model and a quantitative evaluation method for the optimization of safety spacing in heterogeneous platoon are proposed.For the random and uncertain factors affecting traffic safety.The communication distance and maximum braking capacity model based on truncated Gauss distribution,the communication delay model based on Rayleigh distribution and the non-linear vehicle braking model are proposed.By setting typical traffic risk scenarios and risk assessment indicators,the traffic risk under different spacing corresponding to different speed is quantified.A quadratic optimal spacing model is proposed to ensure driving safety and traffic flow stability through data fitting.(2)The cruise control method of heterogeneous fleet based on sliding mode control in normal driving mode is proposed.Based on Lyapunov stability theory and stability analysis method based on frequency domain analysis,a sliding mode controller is designed to satisfy the single vehicle stability and string stability under the condition that the delay parameters are unknown and bounded,aiming at cruise control stability in quadratic safety spacing model.The maximum allowable number of vehicles and the maximum allowable delay parameters of V2V for heterogeneous platoon are given by stability analysis and robustness analysis.(3)The active steering collision avoidance strategy in emergencies mode is proposed,Based on the model of vehicle safety inter platoon spacing,considering the dynamic adjustment of vehicle motion state,dynamic change of lane changing time with motion state,the lane changing opportunity decision for collision avoidance is presented.Based on the constraints of lateral acceleration saturation,nonholonomic constraint,lane changing time and boundary constraints in the process of emergency lane changing,the high degree lane changing trajectory based on decoupling model is given.Based on the simplified longitudinal and lateral coupled dynamic model,an adaptive non-singular fast terminal sliding mode controller is designed under the condition that some parameters of the model are unknown,which satisfies the stability and real-time requirements of the controller in the process of emergency lane changing.(4)A collision avoidance method for low-speed traffic flow based on V2V is proposed.Based on kinematics analysis of early warning scenario and Brill deceleration braking model,an early warning strategy based on velocity difference perception information is designed.Combined with data analysis of 1-880 highway in San Francisco,early warning thresholds for different weather environments in this section are given.The simulation of early warning evaluation in different traffic scenarios is carried out based on the safety measurement method and the success rate indicator,which improves the early warning effect on the basis of the existing variable speed limit early warning method based on loop detector.This dissertation designs and implements the control and early warning simulation under the influence of uncertain random parameters such as communication delay.actuator delay and maximum braking capacity of vehicles in heterogeneous platoon by collecting data from highway section and simulating experiments under multi-level and multi-scenarios.By comparing the simulation results with the traditional algorithm,it is proved that the proposed active safety method is superior to the traditional control and early warning methods.