Research On AEB Control Strategy Of Passenger Car Based On Multi-sensor Data Fusion

With the continuous improvement of people’s living standards in China,more and more people choose cars as their common travel tools,however,the phenomenon increases the frequency of traffic rear-end accidents.Reducing traffic rear-end accidents has become a common problem for the traffic development and automobile industry development.In order to solve this social problem,the active safety technology for vehicle driving safety has also been developed rapidly.As an important vehicle active safety technology,automatic emergency braking system can prevent the collision by early warning and active braking before the collision occurs,which has been favored by major domestic and foreign automobile enterprises.In this paper,a data fusion algorithm based on millimeter-wave radar and camera vision is used to study the control strategy of automatic emergency braking system(AEB)for passenger cars.Firstly,a vehicle detection method based on data fusion of millimeter-wave radar and camera is proposed for target recognition module of automatic emergency braking system(AEB).By the way of threshold filter,this paper selects the primary target based on millimeter-wave radar and screens out the potential dangerous target.Then,a vehicle target detection algorithm based on Haar-like feature and Addaboost has been constructed.On this basis,a fusion model of millimeter-wave radar and camera has been built and the multi-sensor data fusion strategy has been studied.In addition,vehicle ROI partition and target information correction have been completed,and the integration of wave radar and camera in space and time has been realized.Then,the safety model of the automatic emergency braking system is studied.The requirement of safety model is discussed,and the typical safety models based on collision time and safety distance are analyzed and compared.On this basis,a modified safety model based on collision time and safety distance is proposed.According to the requirements of working conditions,the collision time is revised,and the safety distance models under three states are determined,including complete braking,partial braking and safety early warning.At the same time,the key parameters of safety model such as TC threshold,braking deceleration,final workshop distance and driver reaction time are analyzed.Afterwards,the control strategy of automatic emergency braking system(AEB)is studied.The idea of hierarchical control is put forward,and the hierarchical control structure of the control system is determined.The dangerous targets are distinguished and classified according to the collision time and the safe distance between vehicles.What’s more,the corresponding three-level safety strategy and the solution method of the expected deceleration under different working conditions are formulated.In order to realize the braking control of vehicles,a throttle/brake switching control model is proposed,and the expected deceleration following controller based on PID controller is constructed.Finally,the simulation model of AEB system and the vehicle test platform are built to carry out the simulation test and the vehicle test under typical working conditions respectively.The simulation analysis and vehicle test results show that the safety control strategy of the automatic emergency braking system(AEB)constructed in this paper is effective and reliable.The data fusion strategy based on millimeter-wave radar and camera vision,the revised safety model based on collision time and safe vehicle distance,the safety strategy based on dangerous target grade and the method of test verification proposed in this paper have important theoretical guidance and practical value for the key technical research of automatic emergency braking system(AEB)for passenger cars,and it is also of practical significance to promote the development of intelligent vehicle and improve traffic safety.