The invention discloses a
pedestrian collision risk assessment method applied to a
signal-free road, and the method comprises the steps: firstly, detecting the relative distance and relative angle information of a
pedestrian through a
millimeter-
wave radar; and then estimating the current speed and course angle of the
pedestrian through extended Kalman filtering (EKF). And according to the current speed of the pedestrian, a
Markov chain is used to predict the pedestrian trajectory, and the future speed of the pedestrian is predicted. The driving information of the vehicle is detected by a vehicle sensor (IMU and the like). A
collision domain is obtained based on predicted future trajectory information of pedestrians and vehicles to judge whether the pedestrians and the vehicles collide, finally
collision risk levels are evaluated based on TTC, and different collision avoidance measures are taken according to different risk levels.