Road-Side Unit Anomaly Detection

Abstract

Actors of the Cooperative Intelligent Transport Systems (C-ITS) generate various amounts of data. Useful information on various issues such as anomalies, failures, road profiles, etc., could be revealed from the analysis of these data. The analysis, could be managed by operators and vehicles, and its output could be very helpful for future decision making. In this study, we collected real data extracted from road operators. We analyzed these streams in order to verify whether abnormal behaviors could be observed in the data. Our main target was a very sensitive C-ITS failure, which is when a road-side unit (RSU) experiences transmission failure. The detection of such failure is to be achieved by end users (vehicles), which in turn would inform road operators which would then recover the failure. The data we analyzed were collected from various roads in Europe (France, Germany, and Italy) with the aim of studying the RSUs’ behavior. Our mechanism offers compelling results regarding the early detection of RSU failures. We also proposed a new C-ITS message dedicated to raise alerts to road operators when required.

Keywords

• C-ITS
• failure detection
• road-side unit
• data analysis
• machine learning