关键词: 半挂汽车列车, 自动紧急制动控制系统, BP神经网络算法, 模糊控制, PID控制, 滑模控制

Abstract:

The promulgation of the standard on the performance requirements and test method of autonomous emergency baking （AEB） for commercial vehicles in China promotes the development and application of AEB in the field of commercial vehicles. In view of the long braking distance and high mass center height of tractor-semitrailer combination， an AEB control strategy based on time to collision （TTC） predicted by BP neural network is proposed in this paper， drawing on the driver’s experience of emergency braking. First of all， the upper controller is designed. Based on the data of TTC in different emergency braking scenes with different drivers and by using BP neural network algorithm， the prediction model is obtained to calculate the time thresholds of early warning and emergency braking， and with the relative distance and relative speed between front vehicle and ego vehicle and the deceleration of front vehicle as inputs， the desired deceleration of ego vehicle is obtained by fuzzy control rule. Then， the lower controller is designed. By combining the forward control of desired deceleration and the PID backward control of deceleration deviation， the braking pressure in wheel cylinder needed for each wheel is obtained， the wheel lock is prevented by sliding mode control， and the safety， comfort and yaw stability in emergency braking are enhanced. Finally， CCRb， CCRm and CCRs three test scenes are set up in TruckSim to verify the control strategy. The results show that the control strategy proposed can effectively avoid the occurrence of crash and this study provides a theoretical basis for the design and research of the AEB system in tractor-semitrailer combination.

Key words: tractor?semitrailer combination, autonomous emergency braking control system, BP neural network algorithm, fuzzy control, PID control, sliding mode control