Optimized vision-directed deployment of UAVs for rapid traffic monitoring

Christos Kyrkou, Stelios Timotheou, Panayiotis Kolios, Theocharis Theocharides, Christos G. Panayiotou

Abstract: The flexibility and cost efficiency of traffic monitoring using Unmanned Aerial Vehicles (UAVs) has made such a proposition an attractive topic of research. To date, the main focus was placed on the types of sensors used to capture the data, and the alternative data processing options to achieve good monitoring performance. In this work we move a step further, and explore the deployment strategies that can be realized for rapid traffic monitoring over particular regions of the transportation network by considering a monitoring scheme that captures data from a visual sensor on-board the UAV, and subsequently analyzes it through a specific vision processing pipeline to extract network state information. These innovative deployment strategies can be used in real-time to assess traffic conditions, while for longer periods, to validate the underlying mobility models that characterise traffic patterns.