Publications
Peer-reviewed articles, preprints, and submitted manuscripts across four research directions.
Complete publication record on Google ScholarTransportation Research Part C: Emerging Technologies, 190, 105784
Models post-crash traffic speed during freeway all-lane-closure incidents with a dual-encoder graph-attention mixture-of-experts framework.
Accident Analysis & Prevention, 236, 108673
Organizes vision-language traffic-safety research around event understanding, grounded evidence, and verification, with an agenda for generative and counterfactual analysis.
Accident Analysis & Prevention, 232, 108519
Applies causal machine learning to the relationship between in-vehicle warnings, driver workload, and risk perception.
Transportation Research Record, 03611981261424236
Connects crash analysis and conflict assessment approaches in research on vehicle-pedestrian interactions.
Transportation Research Part F: Traffic Psychology and Behaviour, 119, 103585
Combines conflict and crash datasets to examine interactions and risks between right-turning vehicles and pedestrians.
Transportation Research Part C: Emerging Technologies, 183, 105479
Pairs connected-vehicle crash-risk prediction with causal mediation to investigate the pathways linking roadway conditions, driving behavior, and risk.
IEEE Open Journal of Intelligent Transportation Systems, 7, 1293-1310
Uses connected-vehicle data and spatiotemporal sequence learning to study both freeway crash occurrence and crash type.
Analytic Methods in Accident Research, 48, 100406
Jointly models pedestrian injury severity across left-turn, straight, and right-turn vehicle movements while accounting for temporal instability and spatial correlations.
Journal of Safety Research, 95, 345-361
Evaluates ADAS safety effectiveness using real-world two-vehicle crash records, examining how safety effects vary across traffic conditions.
Analytic Methods in Accident Research, 47, 100387
Integrates visual environment features and spatial effects to study intersection crash frequency and variation across locations and periods.
Accident Analysis and Prevention, 213, 107942
A safety-mobility coupling paper that uses cross-stitch networks to jointly model tunnel crash severity and congestion duration.
IEEE Transactions on Intelligent Transportation Systems, 25(11), 18383-18395
A real-time crash-risk modeling paper combining traffic flow and risky-driving behavior with sequence learning.
Accident Analysis and Prevention, 200, 107563
A field-study paper on how mobile-phone distraction changes pedestrian behavior and crossing safety.
ASCE-ASME Journal of Risk and Uncertainty in Engineering Systems, Part A: Civil Engineering, 10(2), 04024009
Brings psychophysiological evidence into minimum horizontal curve radius design for plateau road environments.
Transportation Research Part F: Traffic Psychology and Behaviour, 98, 318-336
A human-factors paper comparing driver perception-reaction time at expressway exit ramps in plain and plateau environments.
Preprints are listed separately from peer-reviewed publications.
arXiv, 2512.17724
A multimodal dataset project for first-person ADAS near-miss and crash event understanding.
Work submitted for review; these manuscripts are not listed as accepted or peer-reviewed publications.
Examines multimodal risk recognition and graded prevention in unsignalized, occluded pedestrian crossing scenarios.
Investigates a local-action safety-boundary benchmark and model for pre-crash safety recovery.
Studies bus safety-risk prediction through driver-group heterogeneity, perception, and road environment-driving behavior interactions.
Examines controller-state recovery for resilient merging between connected and automated vehicles and human-driven vehicles.
Studies CACC platoon safety using tire-force risk envelopes under cyberattacks and reduced-friction roadway conditions.
Examines a deployment-oriented protocol for ADAS candidate-event warning, combining video, structured signals, and timing-sensitive evaluation.
Investigates a mixed-traffic simulation platform for multiple cyberattack classes, vehicle dynamics, and lane changes.
Investigates multi-source freeway safety assessment through deep ensembles and causal learning.
Connects microscopic vehicle dynamics and macroscopic rear-end collision risk in cyberattack impact assessment for connected platoons.
No research records match these filters.