Adaptive AEB control strategy for driverless vehicles in campus scenario

It’s difficult for the traditional autonomous emergency braking (AEB) system on the driverless vehicle to work well in the campus environment with number of sudden and complex scenarios. To address this problem, firstly, one two-layer adaptive AEB control strategy is proposed in this paper, in the upper layer, an adaptive risk assessment model based on lateral and longitudinal Time-to-Collision (TTC) with adaptive TTC threshold algorithm is used to make a comprehensive assessment of the danger degree of the dangerous target, the desired deceleration based on the danger degree of the target in the three-stage braking strategy is obtained. The lower layer controller is based on an inverse longitudinal dynamics model and employs a BP neural network PID control algorithm to track and control the conversion of desired deceleration into braking pressure. Second, an estimation method based on Lagrange interpolation formula is designed to make the driverless vehicle adaptive to the road friction coefficients changes, and the peak road friction coefficient is estimated in real-time. Finally, the adaptive AEB control strategy is validated on a Carsim-Matlab/Simulink joint simulation platform, the results show that it has good adaptability to both the pavement and the complex targets on the campus.