Real-Time Lane Change Detection Through Steering Wheel Rotation

Abstract

Lane change is an important traffic information. An abnormally high number of lane changes on a particular road section typically suggests that some lanes are blocked due to traffic incidents such as accidents and vehicle break downs. As a result, the lane change information is useful for automatic traffic incident detection. Currently, the lane change information of vehicles on an urban road is typically obtained from over-roadway fixed sensors such as surveillance cameras. However, using fixed sensors has limitations in terms of cost and coverage. It would be more effective if the lane change information could be collected directly from each individual vehicle. In this paper, we introduce a new mobile sensing approach to automatic real-time lane change detection. The lane change event is detected through the corresponding pattern of steering wheel rotation. The results show that the detection algorithm performs extremely well. Overall, it is able to achieve higher than 95% both in terms of precision and recall.