Assessing the effect of pedestrian deceleration on pedestrian autonomous emergency braking systems

The aim of the manuscript is to access whether the Rate of Collision Avoidance (RCA) will be improved by considering the pedestrian deceleration in the Pedestrian Autonomous Emergency Braking Systems (AEB-P). A vehicle braking method by considering the pedestrian deceleration was proposed firstly, and then through the Monte Carlo optimization algorithm optimal combinations of 7 input parameters were obtained by applying the method to 124 actual collisions, finally the single factor analysis method was used to explore the effects of each parameter on the RCA. Through observation from actual collisions, the interval of pedestrian deceleration was [0.1, 8.2]m/s2 with a median value of 4.2 m/s2 in an emergency. The optimal average value of the detection angle, the minimum detection distance, the lateral trigger distance, the pedestrian warning trigger distance, the lag time, the pedestrian deceleration and its reaction time is 60°, 25 m, 3.5 m, 0 m, 0 s, −8 m/s2 and 0 s, respectively. The maximum RCA = 84.68% was obtained based on optimal combinations. All analyses consistently found that the RCA increases as pedestrian deceleration increases. Results shown that we can consider pedestrian deceleration in the AEB-P to improve the performance of the system and subsequently improve the RCA.