Stochastic Traffic

Stochastic traffic modeling aims to understand and predict the unpredictable nature of traffic flow, focusing on improving traffic management and autonomous vehicle navigation. Current research emphasizes developing data-driven models, particularly employing graph neural networks (GNNs) and recurrent neural networks (RNNs) like GRUs, to capture complex spatiotemporal dependencies in traffic data and address challenges like incomplete sensor coverage and non-equilibrium flows. These advancements improve traffic volume estimation, congestion pattern retrieval, and uncertainty quantification in traffic state prediction, ultimately leading to more efficient traffic systems and safer autonomous driving.

Papers

November 28, 2023

Pattern retrieval of traffic congestion using graph-based associations of traffic domain-specific features
Tin T. Nguyen, Simeon C. Calvert, Guopeng Li, Hans van Lint
Traffic Data Pattern Recognition Traffic Efficiency Domain Specific Feature Traffic Dynamic Spatiotemporal Pattern Traffic Congestion Association Graph Stochastic Traffic

March 10, 2023

Towards better traffic volume estimation: Jointly addressing the underdetermination and nonequilibrium problems with correlation-adaptive GNNs
Tong Nie, Guoyang Qin, Yunpeng Wang, Jian Sun
Traffic Datasets Non Equilibrium Network Wide Traffic Traffic Estimation Transmission Determinacy Stochastic Traffic

November 13, 2022

GC-GRU-N for Traffic Prediction using Loop Detector Data
Maged Shoman, Armstrong Aboah, Abdulateef Daud, Yaw Adu-Gyamfi
Long Short Term Memory Temporal Dependency Graph Convolution Traffic Prediction Loop Detection Stochastic Traffic

June 19, 2022

Quantifying Uncertainty In Traffic State Estimation Using Generative Adversarial Networks
Zhaobin Mo, Yongjie Fu, Xuan Di
Generative Adversarial Network Uncertainty Quantification Physic Informed Deep Learning Traffic Model Traffic State Estimation Stochastic Traffic

February 21, 2022

RTGNN: A Novel Approach to Model Stochastic Traffic Dynamics
Ke Sun, Stephen Chaves, Paul Martin, Vijay Kumar
Graph Neural Network Novel Approach Traffic Dynamic Future Traffic Stochastic Traffic

Stochastic Traffic

Papers

Pattern retrieval of traffic congestion using graph-based associations of traffic domain-specific features

Towards better traffic volume estimation: Jointly addressing the underdetermination and nonequilibrium problems with correlation-adaptive GNNs

GC-GRU-N for Traffic Prediction using Loop Detector Data

Quantifying Uncertainty In Traffic State Estimation Using Generative Adversarial Networks

RTGNN: A Novel Approach to Model Stochastic Traffic Dynamics