Proactive Crash Risk Prediction

Sun, 26 Apr 2026 00:00:00 +0000

This project line develops proactive crash risk and crash type prediction models using connected vehicle trajectories, traffic states, risky-driving behavior, and video-derived context.

Representative methods include BiLSTM-Transformer sequence models, deep ensembles, causal feature interpretation, and macro-micro exposure fusion for behavior-aware freeway safety assessment.

From prediction to explanation: A machine learning and causal mediation framework for roadway crash risk with connected vehicle data

Mon, 01 Dec 2025 00:00:00 +0000

This paper represents the connected-vehicle and causal-explainability line of work: predictive safety models are paired with causal mediation analysis so the model output can be translated into interpretable crash-risk mechanisms.

Crash Risk Prediction | Chenzhu Wang

Proactive Crash Risk Prediction

From prediction to explanation: A machine learning and causal mediation framework for roadway crash risk with connected vehicle data