Machine Learning based Algorithms and Software for Transportation and Smart Cities

Funded by NSF Cyber Physical Systems, NSF Smart Communities and Florida Department of Transportation

My group develops machine learning algorithms and software for analyzing traffic and improving intersection safety and operations. We combine video, LiDAR, and ground-sensor data to detect and track vehicles and pedestrians, reconstruct trajectories, and identify potential collisions and near-misses.

The system supports processing on edge computers and in the cloud. Our work with Gainesville and Orlando uses these data to assess intersection safety and support the communication of unsafe conditions to connected vehicles and pedestrians.

Our traffic-modeling research includes recurrent attention networks, graph attention networks, digital twins, and generative models for intersection and corridor dynamics. These methods support traffic-flow prediction, simulation calibration, measures of effectiveness, and signal-policy optimization.

Recent work includes topology-invariant intersection simulation, temporal graph digital twins for urban corridors, generative vehicle-trajectory models, and parallel frameworks for corridor calibration and signal optimization.

Selected Publications

1. N. Yousefzadeh, R. Sengupta, Y. Karnati, A. Rangarajan, and S. Ranka, Graph attention network for lane-wise and topology-invariant intersection traffic simulation. IEEE Transactions on Intelligent Transportation Systems, 26(4), 2025, pp. 5082–5093.
2. Y. Karnati, R. Sengupta, A. Rangarajan, and S. Ranka. Subcycle Waveform Modeling of Traffic Intersections Using Recurrent Attention Networks. IEEE Transactions on Intelligent Transportation Systems, special issue on Artificial Intelligence, 23(3):2538–2548, 2022.
3. R. Sengupta, R. Casburn, T. Patel, J. Dilmore, and S. Ranka. A Practical Pipeline for Urban Arterial Simulation Calibration. Proceedings of 2026 Vehicle Technology and Intelligent Transportation System (VEHITS), 2026, pp. 382–392.
4. Y. Ranjan, R. Sengupta, A. Rangarajan, and S. Ranka. Enactor: From Traffic Simulators to Surrogate World Models. Proceedings of 2026 Vehicle Technology and Intelligent Transportation System (VEHITS), 2026, pp. 430–438.
5. R. Sengupta, R. Casburn, T. Patel, J. Dilmore, and S. Ranka. A Parallel Framework for Corridor Flow Calibration and Signal Optimization. Proceedings of 2026 Vehicle Technology and Intelligent Transportation System (VEHITS), 2026, pp. 464–470.
6. N. Yousefzadeh, R. Sengupta, J. Dilmore, and S. Ranka. TGDT: A Temporal Graph-based Digital Twin for Urban Traffic Corridors. Proceedings of the IEEE International Conference on Intelligent Transportation Systems (ITSC), 2025, pp. 700–707.
7. Y. Ranjan, R. Sengupta, A. Rangarajan, and S. Ranka. IntTrajSim: Trajectory Prediction for Simulating Multi-Vehicle Driving at Signalized Intersections. Proceedings of the IEEE International Conference on Intelligent Transportation Systems (ITSC), 2025, pp. 1603–1609.
8. Y. Ranjan, R. Sengupta, A. Rangarajan, and S. Ranka, Evaluating Generative Vehicle Trajectory Models for Traffic Intersection Dynamics. Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD), 2025, pp. 262-274.