Research on AEB-P control strategy based on adaptive weight MPC

Abstract

To further optimize the automotive AEB-P (autonomous emergency braking for pedestrian) control algorithm, an AEB-P hierarchical control strategy that integrates driving comfort and pedestrian injury risk was proposed. A braking safety distance model considering the driver′s comfort during braking was designed for the AEB-P evaluation standard of C-NCAP. The weight coefficient adjustment strategy was obtained by introducing fuzzy rules to comprehensively consider the risk of pedestrian injury and scenario working conditions. Based on this, the adaptive weight coefficient MPC (model predictive control) upper controller was designed, and the PID (proportion integration differentiation) lower controller was used to correct the actual deceleration of the auto-vehicle. The vehicle longitudinal dynamics model was established and the test scenario and control algorithm were constructed through CarSim and Matlab/Simulink, then the proposed method and the fixed TTC (time to collision) threshold algorithm were compared through hardware-in-the-loop experiments. The results show that the proposed control algorithm can effectively avoid collision in 93.4% of the test conditions, while the fixed TTC threshold algorithm has a successful rate of only 43.75% in obstacle avoidance. Compared with the traditional control strategy, the proposed method can maintain a more stable minimum distance between the vehicle and the pedestrian in front of it, with better robustness, which provides a reference basis for the AEB-P control theory.

Keywords

• vehicle engineering; automatic emergency braking; model predicative control; pedestrian injury; pedestrian collision avoidance; hardware-in-the-loop (hil)