Thursday, July 18, 2024 | 12:00 – 1:00 PM (ET)
The research initiative, conducted jointly by the University of South Florida (USF) and the University of Puerto Rico at Mayagüez (UPRM), focuses on exploring the feasibility of using Unmanned Aerial Vehicles (UAVs) for incident detection in high-speed multi-lane and freeway corridors. This initiative stems from the National Institute for Congestion Reduction’s (NICR) mission to enhance transportation systems through technology and data-driven strategies. Previous phases of this research have laid the groundwork for developing and validating UAV-based incident detection methodologies, with the overarching goal of improving traffic management efficiency and roadway safety. The research methodology involved developing an incident detection methodology using real-time video data from single and multiple flying UAVs, UAV path planning for corridor incident detection, and designing experiments to establish protocols, standards, and guidance for using UAVs in accordance with FAA regulations. Key findings indicate that traditional methods of incident detection, such as manual patrolling, are not efficient in high-speed and multi-lane environments, highlighting the need for alternative solutions like UAVs. The study successfully validated the object detection and incident detection algorithms developed in previous phases, emphasizing the importance of proper site selection and data collection methodologies. The study’s conclusions underscore the significant potential of UAVs in enhancing traffic monitoring and incident detection capabilities. It highlights key insights into the performance of the YOLO7 algorithm, emphasizing the impact of factors such as drone elevation and type on its efficiency. The findings indicate that UAVs offer a promising alternative to traditional manual patrolling methods. The validation of object detection and incident detection algorithms further validates the feasibility of using UAVs for real-time traffic monitoring, emphasizing the importance of meticulous site selection and data collection processes.
About the Presenter:
Dr. Ivette Cruzado serves as a professor within the Department of Civil Engineering and Surveying at the University of Puerto Rico at Mayaguez, with experience in both academia and administration. Previously, she served as the Associate Dean of Research in the Engineering faculty and currently holds the role of Associate Director of the Department of Civil Engineering and Surveying. Her background includes a bachelor’s degree in Civil Engineering from the University of Puerto Rico at Mayaguez, a Master’s degree from Michigan State University, and a PhD from Penn State University. While her doctoral thesis focused on the geometric design of highways, Dr. Cruzado has diversified her research interests to encompass areas such as public transportation, big data, incident management using aerial unmanned systems, highway safety, and engineering education. Currently, she contributes as a co-investigator in the UTC Social and Economic Mobility for People and Communities Through Transportation (SEMPACT), a research project that focuses simulation platforms are used to model driving behavior along dynamic toll lanes. Dr. Cruzado has served as a member and friend of TRB committees and is a reviewer for several transportation-focused journals. She is currently serving as the Coordinator of the Research Academy, an entity dedicated to provide support to new faculty members in their research endeavors. She is also the president of UPRM’s Institutional Committee for Research.
