Identification and depth determination of potholes based on vibration acceleration of vehicles

The vibration acceleration of vehicles passing through potholes have been only used to identify the location of potholes with high accuracy, while this paper proposed a method for both identifying potholes and determining their depths based on vibration accelerations. The study first preprocessed the collected acceleration data. Then, a Newmark method was established to solve the vibration equations and get the acceleration of the vehicles passing potholes, in which vehicle vibration parameters were identified through drop test. Finally, the particle swarm optimisation algorithm was used to back estimate the pothole depths according to the vibration acceleration, followed by a verification test. Result show that the proposed pothole identification algorithm, based on the vibration acceleration threshold, can accurately identify potholes and mark their positions on the time-domain map. For pothole depth determination, the average error rate based on vibration accelerations is 8.94%, with a minimum error rate of just 0.50%, proving the method's high accuracy.