NsBM-GAT: A Non-stationary Block Maximum and Graph Attention Framework for General Traffic Crash Risk Prediction

2025-03-06

AI Summary

Current traffic collision risk prediction methods face two key bottlenecks: (1) scarcity of real-world pre-collision ego-vehicle data, forcing reliance on manually constructed hazardous scenarios; and (2) dashcam videos capturing only ego-vehicle behavior while lacking critical motion information of neighboring vehicles. To address these, we propose a real-time risk prediction framework integrating non-stationary Extreme Value Theory (EVT) with Graph Attention Networks (GAT). Our method introduces the first non-stationary block maxima model coupled with a nonlinear covariate-aware GAT architecture, unifying EVT’s interpretability with expressive nonlinear modeling of multi-vehicle dynamic interactions. It enables joint prediction of diverse collision types—including rear-end and sideswipe collisions. Evaluated on 100 real-world forward-collision trajectories collected via drone in merging/interweaving road segments, our approach significantly outperforms state-of-the-art methods, achieving superior micro-causal distribution fitting and全面提升 in risk prediction accuracy.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Computer Vision: Motion & Tracking
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems

Abstract

Accurate prediction of traffic crash risks for individual vehicles is essential for enhancing vehicle safety. While significant attention has been given to traffic crash risk prediction, existing studies face two main challenges: First, due to the scarcity of individual vehicle data before crashes, most models rely on hypothetical scenarios deemed dangerous by researchers. This raises doubts about their applicability to actual pre-crash conditions. Second, some crash risk prediction frameworks were learned from dashcam videos. Although such videos capture the pre-crash behavior of individual vehicles, they often lack critical information about the movements of surrounding vehicles. However, the interaction between a vehicle and its surrounding vehicles is highly influential in crash occurrences. To overcome these challenges, we propose a novel non-stationary extreme value theory (EVT), where the covariate function is optimized in a nonlinear fashion using a graph attention network. The EVT component incorporates the stochastic nature of crashes through probability distribution, which enhances model interpretability. Notably, the nonlinear covariate function enables the model to capture the interactive behavior between the target vehicle and its multiple surrounding vehicles, facilitating crash risk prediction across different driving tasks. We train and test our model using 100 sets of vehicle trajectory data before real crashes, collected via drones over three years from merging and weaving segments. We demonstrate that our model successfully learns micro-level precursors of crashes and fits a more accurate distribution with the aid of the nonlinear covariate function. Our experiments on the testing dataset show that the proposed model outperforms existing models by providing more accurate predictions for both rear-end and sideswipe crashes simultaneously.

Problem

• Predicts individual vehicle crash risks accurately.
• Addresses lack of pre-crash individual vehicle data.
• Captures interactions between target and surrounding vehicles.

Innovation

• Non-stationary EVT with nonlinear covariate optimization
• Graph attention network for vehicle interaction modeling
• Drone-collected trajectory data for real crash analysis