To solve the problems of low-speed premature intervention and high-speed collision avoidance failure in Autonomous Emergence Braking (AEB) system, a graded braking control method of AEB system with multi-target dangerous target recognition in bend is proposed. Firstly, the vehicle dynamic model is established. Then, based on the multi-target dangerous state recognition, the AEB system is designed and applied to identify and judge the dangerous state under the multi-target working condition of the bend. Next, a segmented braking force distribution scheme is designed to improve the vehicle braking stability. Furthermore, based on the multi-level Time-to-Collision model, a graded early warning braking model is constructed, which effectively solves the problems of low-speed premature intervention and high-speed collision avoidance failure in AEB system. Finally, under the multi-target test scenario for bends, the simulation is carried out. The simulation results show that the proposed control method not only improves the accuracy of the AEB system in the recognition of multiple targets in bends, and enhances the safety and comfort of the AEB collision avoidance system, but also reduces the occurrence of false triggering in bends.
Wang et al. (2026) studied this question.