Multi-target Vehicle Identifying And Tracking Of Adaptive Cruise Control System

The ACC system effectively reduces the fatigue strength of drivers and reduces the incidence of traffic accidents to a certain extent.However,the traditional ACC system has some problems,such as:the limitation of the region of interest,false recognition of the tracking on the curve and the inability to respond to lane-change vehicles in the adjacent lanes.In order to solve these problems,the key technologies involved in the system are studied in this paper:(1)Using the Kalman algorithm and the target motion`s characteristics to eliminate the false object in radar data,and effectively identify the static object.On this basis,the judgment stage of candidate object is added,and the new control rules of the existence of the object are set up to ensure the stability of object tracking and reduce the delay time of object replacement.(2)Aiming at recognizing the curve-entry/-exit and lane-change of the forward vehicle,When the speed of the forward vehicle is different from the subject vehicle,the relationship between the speed of the forward vehicle,the azimuth and the relative lateral velocity of two vehicles are analyzed.And a new criterion is proposed to identify the curve-entry/-exit of the forward vehicle.Based on the data of the subject vehicle,a method of fitting road curvature formula is put forward to obtain the change rate of road curvature,so as to realize the calculation of lateral compensation distance of target vehicle on the curve.(3)Using the fuzzy logic control algorithm and the sliding window method to identify the lane change behavior of vehicles in adjacent lane.Considering the relation between the three types of vehicles:the subject vehicle,the adjacent vehicle and the forward vehicle,the lane chane behavior is classified.And the tracking model of the subject vehicle under different lane change behaviors is developed.(4)Aiming at the straight driving and lane change of the forward vehicle,the quartile and boxplot are used to analyze the difference and distributive range of the data of three parameters :the course angle of the forward vehicle,the relative lateral distance and the relative lateral velocity of two following vehicles,so as to provide a reference for the classification of the universe of the fuzzy controller parameters selected.Finally,Carsim and Matlab/Simulink are used to perform simulation to verify the rationality of the identification algorithm of the lane change behavior of adjcacent vehicles and the effectiveness of the tracking model of the subject vehicle under different lane change behaviors.