Method of monitoring the condition of road surface based on accelerometer signals and fuzzy logic apparatus

Purpose of research. The purpose of this research is to improve the accuracy of monitoring the condition of the road surface of an urban agglomeration by analyzing accelerometer signals in real time based on the developed integrated method for assessing evenness indicators. Methods. The following methods were used in this research: analysis of existing methods for monitoring road surface conditions; methods and algorithms for filtering accelerometer signal noise (a model for preprocessing accelerometer signals was developed and described, including a Butterworth low-pass filter, a median filter, an exponential smoothing method, and calculation of threshold values); fuzzy logic algorithms (a model for classifying road surface conditions into 5 categories was developed); simulation modeling (test runs were conducted using the author's simulation model developed in the Unity environment). Results. The presented method provides automated monitoring of the road surface condition with an accuracy of at least 93%, and the possibility of integration into the smart city system. The method allows monitoring the road surface condition in real time, and the classification of the road surface condition is of a recommendatory nature for road repairs. The prospects of the research include conducting a full-scale experiment, visualizing data with reference to a city map, and using trajectory clustering algorithms to determine the general trajectories of a vehicle when going around uneven surfaces. Conclusion. Based on the developed method of monitoring the condition of the road surface, an integrated assessment of the compliance of the accuracy of monitoring the condition of the road surface of at least 93% was obtained.