A New Framework for Traffic Conflict Identification by Incorporating Predicted High-Resolution Trajectory and Vehicle Profiles in a CAV Context

Abstract

Surrogate safety measures (SSMs) have been widely used to identify traffic conflicts in traditional road environments. However, with advances in connected and autonomous vehicles (CAVs), it is crucial to achieve more accurate and timely conflict identification to enhance safety standards. Time-to-collision (TTC), one of the most common SSMs, assumes that the vehicles involved in a conflict maintain uniform motion. However, this assumption overlooks the complex and dynamic nature of vehicle movement. Moreover, TTC is designed to detect one-dimensional conflicts, while real-world collisions occur in two-dimensional space. To address these limitations, this paper proposes a novel trajectory-based TTC (TTTC), which enhances conflict identification in CAV environments by incorporating high-resolution predicted trajectories and vehicle profiles. Unlike traditional SSMs, which rely on simplified motion assumptions, TTTC uses predicted trajectories that more accurately reflect future vehicle movement, thereby improving conflict detection accuracy. Additionally, this paper introduces the Planar Collision Index (PCI), a collision identification method that enables TTTC to detect two-dimensional collisions between rectangular vehicle profiles at various angles. The effectiveness of TTTC is evaluated across multiple scenarios, including car-following, non-conflicting interactions, and dangerous overtaking. The results show that TTTC identifies conflicts earlier and more accurately than traditional SSMs in these scenarios. These findings demonstrate that TTTC is a promising tool for conflict identification, which could contribute to advancing future CAV safety research.