Development Of Low-Cost Radar-Based Sensor For Multi-Modal Traffic Monitoring

Siyang Cao, University of Arizona



Multimodal traffic monitoring is critical for improving the mobility and safety at the intersections with potential conflicts among various modes of transportation. Traditional traffic monitoring approaches utilizing camera cannot work reliably during the night and under hazard weather conditions. We propose to build a new intelligent multimodal traffic monitoring device using the low-cost mmWave radar. The proposed device can reliably distinguish different modes (such as buses, pedestrians, bicyclists, trucks, motorcycles, etc.), and determine the counts, speed, and moving directions of every single target in an urban environment under various lighting and weather conditions. In the study, a low-cost prototype system will also be built and tested for data collection at a major intersection near the University of Arizona. The proposed research can build a foundation for commercializing a new traffic monitoring device through the project.


The intended users of this research include transportation researchers, engineers, planner, and practitioners. The principle involved in our research is to combine the advantages from the mmWave radar sensor and machine learning, and this novel idea provides the transportation researchers a new playground to work out a multimodal traffic monitoring solution as we do not see any researches similar to ours.  And we will also provide a traffic monitoring system demo for demonstrative purpose as an outcome of this research, and this demo system will include many implementation details as a design reference for the engineer's usage in the future.
Specifically, we will provide the mmWave radar specifications for the designer to acquire sufficient radar points for each mode, the deep learning architecture for mode classification, and the real-world data sets for further evaluation.

Project Details

Project Type:
Project Status:
In Progress
End Date:
April 30,2021
UTC Grant Cycle:
NITC 16 Round 3
UTC Funding: