Technological solutions for automated traffic monitoring with UAVs

ronda tool 700 x 450

The KIOS Research and Innovation team has developed a computer vision software system for the road traffic monitoring with the use of Unmanned Aerial Vehicles (UAVs). The tool was developed within the framework of the research project “RONDA” (Roadway Network Distress Assessment).

The RONDA project aims to develop a data-driven framework, pattern recognition techniques and low-cost sensor technologies for the detection, classification and georeferencing of roadway pavement surface anomalies, as well as to develop technologies for adaptive-location traffic monitoring with the use of drones. For this reason, the project utilizes a real-life transport network (in the area of Lakatamia Municipality) to study the roadway infrastructure, including several vulnerability aspects.

In this context, KIOS researchers have developed a computer vision software system, using deep learning algorithms, capable of analyzing video streamed from a UAV in real-time, detecting and classifying the types of vehicles, counting them, and estimating individual vehicle speeds and trajectories at selected network junctions.

This software has the potential to be deployed in a swarm of UAVs flying above a traffic network in a fully autonomous and coordinated manner to collect traffic data at scale, process the captured videos on the edge, and communicate the traffic data to a traffic management center in real-time.

It is worth mentioning that the technology developed by KIOS CoE will be part of the integrated RONDA platform. Traffic data made available through this platform can assist relevant stakeholders (transport engineers, policy makers) to better assess the condition of the roadway network, leading to more informed decision making in an effort to reduce congestion and subsequently greenhouse emissions and noise pollution, as well as prevent accidents.

Find out more about the project:

The project is co-financed by the European Regional Development Fund and the Republic of Cyprus through the Cyprus Research and Innovation Foundation (‘RESTART 2016-2020’ Program) (Grant No. INTEGRATED/0918/0056).