Background
This research project is focused around generating connected vehicle basic safety messages using cameras. Adoption of connected vehicles have been slow. This is due to many reasons, but a primary one is the lack of connected vehicle messages being generated from vehicles. The aim of this research is to generate most of a basic safety message using information from a traffic camera observing the road. By using cameras to generate basic safety messages, we hope to provide a much greater coverage of connected vehicle data, helping accelerate the adoption and use of connected vehicle data.
Approach
This project is leveraging the Active-Vision software, which previously provided basic traffic camera analytics, and are adapting it to be able to generate basic safety message data. Basic safety message requires less than 1 meter of accuracy for position and highly accurate measurements for speed and acceleration. To accomplish these in a manner that is scalable, new techniques for precise calibration, faster processing to enable running at native 30 frames per second, and 3D localization of vehicle of centers from a 2D image needed to be developed.
Accomplishments
This research is currently still in progress but has made great progress on addressing each of the technical goals to achieve basic safety message level accuracy. These results will be demonstrated at the end of the project on the newly build ITS corridor, which are a set of 5 camera poles located along the length of Tom Slick Avenue with a total of 10 cameras. This will enable precise localization of all vehicles travelling along this roadway.