A Methodology of Spatial Risk Estimation for Lane Change Conflicts Based on Real-Time Vehicle Trajectory Data

Drivers’ visual perception is subject to various constraints and heterogeneities, so lane-changing behavior often presents collision risk. To identify potential collisions more accurately, a new method of spatial risk estimation (SRE) considering vehicle trajectory prediction is proposed. Existing research mainly focuses on speed and accuracy in identifying instantaneous or overall risk during lane change, but this paper presents a mobile coordinate system to show relative positions of vehicles clearly. Then, 1,444 sample groups are established by using trajectory data provided by NGSIM. Finally, based on real-time trajectories of the surrounding vehicles, potential conflict events can be classified efficiently by the proposed model, while locations of potential conflict points are visualized by a dynamic coordinate system. Results show the proposed model can effectively identify potential conflict events during lane change. The accuracy of this model shows improvement in recognizing lane-changing conflict events. The findings can provide support for the design of an intelligent lane-changing warning system.