CycleSense: Detecting Near Miss Incidents in Bicycle Traffic from Mobile Motion Sensors

TL;DR

CycleSense leverages signal processing and machine learning to automatically detect near miss incidents in bicycle traffic, enhancing safety insights for city planning using data from a crowdsourcing platform.

Contribution

We introduce CycleSense, a novel system combining signal processing and machine learning to automate near miss incident detection from mobile sensors.

Findings

CycleSense significantly outperforms the baseline method in incident detection accuracy.

The system demonstrates effective use of crowdsourced bicycle ride data.

Improves safety insights for urban cycling infrastructure planning.

Abstract

In cities worldwide, cars cause health and traffic problems whichcould be partly mitigated through an increased modal share of bicycles. Many people, however, avoid cycling due to a lack of perceived safety. For city planners, addressing this is hard as they lack insights intowhere cyclists feel safe and where they do not. To gain such insights,we have in previous work proposed the crowdsourcing platform SimRa,which allows cyclists to record their rides and report near miss incidentsvia a smartphone app. In this paper, we present CycleSense, a combination of signal pro-cessing and Machine Learning techniques, which partially automatesthe detection of near miss incidents, thus making the reporting of nearmiss incidents easier. Using the SimRa data set, we evaluate CycleSenseby comparing it to a baseline method used by SimRa and show that itsignificantly improves incident detection.