Automatic
The AUTOMATic project aims to develop and to test, in a simulation environment, a content-aware urban traffic management system relying on a swarm of unmanned aerial vehicles (UAVs). Urban traffic refers to the movement of public and private vehicles as well as of pedestrians in an urban road network. With the term content-aware system for traffic management, we refer to a set of artificial devices that, by extracting data describing urban traffic and by operating ground based units, can autonomously and automatically: i) evaluate the traffic current status; ii) predict its future; and iii) mitigate or avoid undesired developments (e.g., congestion).
Members
 |
 |
 |
 |
 |
| Elio Tuci (PI) | Alexandre Mauroy (Co-PI) | Jean-Noël Colin (Co-PI) | Davoud Alahvirdi (PhD) | Julien Pietquin (PhD) |
Publications
- Alahvirdi D., Tuci E., Autonomous Traffic Monitoring and Management by a Simulated Swarm of UAVs. Proceedings of the IEEE International Conference on Robotics and Mechatronics (ICROM), Tehran, Iran, December 2023, https://ieeexplore.ieee.org/xpl/conhome/1804824/all-proceedings
Abstract: Traffic monitoring and management have become a key issue for improving the quality of life and well being of citizens. Among the different technologies used for these tasks, unmanned aerial vehicles (UAVs) or drones promise to be particularly effective in generating solutions for problems such as traffic congestion, security, resource management, waste of time, and pollution. This paper presents a swarm of simulated UAVs continuously monitoring the traffic in a simulated road network where traffic is implemented using the cell transition model (CTM). The UAVs monitor the data about the status of traffic on distinctive portions of the network with traffic units, and they share their traffic information with a smart traffic light units. The later regulate their signal cycle using an optimisation model named simultaneous perturbation stochastic approximation (SPSA), in a way to mitigate or eventually resolve road congestion. The simulation results indicate that the system we describe is particularly effective in reducing traffic jams.
