Anomaly Stop Detection by Smartphone

Abstract

This paper proposes a method for detecting anomaly stop events from GPS logs of a vehicle recorded by smartphone. Despite of many researches of strong braking event detection are proposed, they cannot discriminate between a normal and an abnormal braking event. Our proposal includes an IMAC model for GPS based map generation and an algorithm for anomaly stops detection based on the acquired map. The real world experiment shows our map generation method and its anomaly stop detection rule works well.

Evaluation of a safety map generated from a collection of difference of Individuals

“…We refer to the proposed traffic safety map as a TimeDiff map. The map integrates a collection of traffic-incident maps that are generated from users' smartphone log data (proposed in [2][3]). In this section, we explain the method for constructing a user's traffic-incident map which is the left side of Fig.1.…”