Research On Pedestrian Collision Avoidance AEB System Considering Driver Characteristics

The Autonomous Emergency Braking System(AEB)is a type of active safety configuration in the Advanced Driver Assistance System(ADAS),which detects the distance and speed of the vehicle in front to determine whether there is a risk of collision.If it is determined that the vehicle is in a sudden dangerous situation or is in danger of collision,the system will actively intervene to brake and brake to avoid or reduce the occurrence of collisions such as rear-end collisions,thereby improving driving safety and protecting drivers and passengers.The existing AEB system research mainly focuses on the design of the collision avoidance strategy for the preceding vehicle.With the deepening of research,various related scientific research institutions have successively carried out research on the effective collision avoidance of vehicles in the AEB system,but due to different driver characteristics There are problems such as differences,difficulty in identifying pedestrians,and uncertainty in pedestrian movement.To achieve pedestrian collision avoidance,many technical difficulties need to be overcome.In terms of how to meet the needs of different drivers,realize pedestrian avoidance,simplify simulation conditions,etc.,the paper conducts related research,the main contents are:In view of the functional requirements of the pedestrian collision avoidance automatic emergency braking system,the driving information required by the AEB-P system is determined and the relevant calculation is carried out.The existing safety distance model is summarized and the safety distance model based on the braking process is selected as the basis for this paper.The relevant parameters in the safe distance model are analyzed and the range of values of the parameters is selected.The relative motion of cross-portrait man-car is analyzed,and the corresponding collision avoidance and early warning strategies are dedged.Driver characteristics identification research,the vehicle driving scene is divided into three types: free driving,car following and lane changing according to road conditions and vehicle conditions,and the driver characteristic classification index in the corresponding vehicle motion scene is selected,and the driver characteristic classification system is constructed based on this;According to the analytic hierarchy process,the actual vehicle experimental data of each index is calculated to realize the identification of the driver’s characteristics.The cluster analysis of the actual vehicle experimental data is used to verify the driver’s characteristic identification and classification results.The results show that the classification results calculated by the adopted driver characteristics objective classification method are in good agreement with the cluster analysis results.The equation is established by the parameters of the driver characteristic identification result and the safety distance model,so that the early warning and the execution of the action of the AEB system meet the subjective needs of different drivers.The dynamic model of a Class D vehicle is established by CarSim simulation analysis software and Matlab/Simulink software,and the inverse dynamic model of the vehicle braking system is obtained by theoretical deducing of the vehicle braking process.The upper fuzzy neural network controller of AEB-P system and the lower controller of AEB-P system based on PID theory are designed to realize the conversion of expected reduction speed to vehicle brake line pressure,so as to control vehicle speed.Based on the relevant regulations of pedestrian collision avoidance test in C-NCAP in China,the setting parameters of the simulation scene and operating conditions are determined,the three-dimensional pedestrian model is established according to the specified size,the parameter configuration of the simulated vehicle is configured in the CarSim software,and the relevant pedestrian test scene is constructed.The joint simulation model of CarSim and Simulink of AEB-P system is established and the result analysis is carried out to verify the feasibility of the pedestrian collision avoidance strategy proposed in this paper.