| Principal Investigator | Ivette Cruzado, Ph.D. |
| Final Report (DOI) | View Final Report |
| TRID | Available Soon |
| Policy Brief | Available Soon |
| RIP | View RIP entry |
Abstract
Automatic freeway incident detection is expected to reduce the response time and improve the efficiency of incident management. UAV platforms have been used to obtain video data and develop methodologies for incident detection. Traffic Management Centers also use video data to monitor freeway traffic operations and incident detection. Liability concerns regarding the privacy of road users are always present. The FHWA report on TMC’s video recording and archiving identifies technologies that can lower image resolution to remove private information (i.e., license plate). Sensing technologies (e.g., thermal sensors) can collect images for extracting traffic information without capturing private information. Limited studies report on the benefits of automatic incident detection with thermal images. UAVs have been used to monitor traffic operations at multiple locations, to cover longer freeway segments and interchanges at different angles and heights while collecting traffic information, making it cost-effective as compared to conventional sensing technologies. Furthermore, it has the potential of reducing secondary traffic incidents thus improving the overall safety of the freeway corridor being studied.
This research requires three phases. Phase I focused on the design and testing of the operations of multiple UAVs for collecting traffic information and the development of incident detection methodology (see NICR Project 4-3: Corridor-Wide Surveillance Using Unmanned Aircraft Systems). Phase II will involve two separate but related research efforts by the University of Puerto Rico, Mayaguez (Part A), and the University of South Florida (Part B). Based on the research outcomes of Phase I of this study the University of Puerto Rico, Mayaguez research team will continue collecting field data from PR-52 Highway in Phase II Part A using both RGB camera and thermal sensors on the UAV platform and using probe vehicles. The probe vehicle will be used to further compare the performance of traffic information extraction (e.g., speed) from analyzing thermal images. Also, the team will emphasize developing automatic incident detection algorithms for thermal images collected from fixed (mounted) and moving stations (UAV platforms). The team will also work with stakeholders from Puerto Rico DTPW, Metric Engineering, and Florida DOT District 7 to identify barriers and challenges of implementing emerging technologies in automatic incident detection and provide future research directions. In Phase III of this project, the research team will focus on the validation of the algorithms developed in the previous phases and implementation matters of Phase II.
