UA's Multimodal Intersection Radar Project Gains Attention
The multimodal traffic monitoring radar project at Tucson intersections by Cao's lab is highlighted for its contribution to smart city sensing.
In December 2021, a feature article on "New Radar Sensor Technology for Intelligent Multimodal Traffic Monitoring at Intersections" described how Dr. Cao's team deployed mmWave radar to distinguish vehicles, pedestrians, and directions at intersections under varying lighting and weather conditions.
Real-World Testing
This system was tested at an intersection in Tucson, using a combination of radar point cloud segmentation (Gaussian Mixture Models) and machine learning to deliver accurate counts, speeds, and classification of objects.
Technical Approach
- mmWave radar deployment at signalized intersections
- Gaussian Mixture Models for point cloud segmentation
- Machine learning for object classification
- Real-time speed and direction estimation
- Robust performance in all weather and lighting conditions
Project Support
The project is funded by the National Institute for Transportation and Communities (NITC). Key deliverables include hardware prototypes, comprehensive datasets for research, and a demonstration platform showcasing the technology's capabilities in real urban environments.