Tianheng Zhu
PhD StudentPurdue University
- Multi-Agent 3D Perception
- Multimodal Learning
- Autonomous Vehicle Simulation
Building toward cooperative intelligence for autonomous systems, where heterogeneous sensing across agents becomes shared understanding and coordinated action.
I’m a PhD student in Transportation and Infrastructure System Engineering at Purdue University, advised by Prof. Yiheng Feng. Before that, I earned my BS in Transportation Engineering at Tongji University in Shanghai. My Chinese name is 朱天恒: 天 sky, 恒 persistence.
News
AGI-Coop and Drones as Annotators accepted at the TRB Annual Meeting 2027.
Our paper “Empowering Microscopic Traffic Simulators with Realistic Perception using Surrogate Sensor Models” is out in Transportation Research Part C.
Invited talk at the Trajectory Data Camp, IEEE ITSC 2026, in Naples, Italy.
Received an IEEE ITSC Student Travel Grant.
Our paper “A Systematic Evaluation of Infrastructure-Based Radar System for Highway Traffic Monitoring” is accepted at IEEE ITSC 2026.
Our paper “Integrated Optimization for Vehicle Trajectory Reconstruction under Cooperative Perception Environment” is out in Transportation Research Part C.
Publications
-
2026
AGI-Coop: A Real World Air-Ground-Infrastructure Cooperative Perception Dataset
Accepted at the TRB Annual Meeting 2027
The first real-world dataset for air-ground-infrastructure (AGI) cooperative perception.
-
Drones as Annotators: Amodal 3D Auto-Labeling for Ground LiDAR with Aerial Priors
Accepted at the TRB Annual Meeting 2027
A drone-assisted, training-free framework that exploits the complementary strengths of ground LiDAR and aerial cameras for amodal 3D auto-labeling.
-
Empowering Microscopic Traffic Simulators with Realistic Perception using Surrogate Sensor Models
Transportation Research Part C: Emerging Technologies
A surrogate sensor model that enables realistic perception simulation for large-scale AV deployments with minimal computational overhead.
-
A Systematic Evaluation of Infrastructure-Based Radar System for Highway Traffic Monitoring
IEEE 29th International Conference on Intelligent Transportation Systems (ITSC 2026)
A dual-modality dataset, DRaT (Drone and Radar Trajectories), for evaluating radar-based highway traffic monitoring against drone-derived ground truth.
-
Integrated Optimization for Vehicle Trajectory Reconstruction under Cooperative Perception Environment
Transportation Research Part C: Emerging Technologies
An integrated MILP model that reconstructs complete vehicle trajectories from cooperative perception data at extremely low CAV penetration rates.
-
2025
Learning Adversarial Attacks on Adaptive Traffic Signal Control Systems Under Cooperative Perception
USENIX VehicleSec 2025
Learns attacks that mislead adaptive traffic signal control by manipulating sensor data sharing messages (SDSMs) sent by CAVs.
-
Vehicle-Group-Based Crash Risk Prediction and Interpretation on Highways
IEEE Transactions on Intelligent Transportation Systems
Interaction-aware and interpretable highway crash risk prediction based on vehicle groups.
-
2024
A Cooperative Perception Based Adaptive Signal Control under Early Deployment of Connected and Automated Vehicles
Transportation Research Part C: Emerging Technologies
An adaptive traffic signal controller that uses cooperative perception data under low CAV penetration rates.
Experience
-
2023 – Now
Graduate Research Assistant · Purdue University
Advised by Prof. Yiheng Feng.
Contact
Whether you have a question or we’re working toward a similar goal, I’d love to hear from you.