Grenoble Traffic Lab (GTL) is an experimental platform for real-time collection of traffic data coming from a dense network of wireless sensors installed in the south ring of Grenoble, direction east-west from A41 to A480. The main components of GTL are: 130 magneto-resistive sensors over the 10.5 km two-lanes highway (26 collection points), a database and a show-room. Sensors provide macroscopic traffic measurements such as flow [veh./h], speed [km/h], occupancy [%]. They were put in place in collaboration with local traffic authorities (DIR-CE) and Karrus-ITS. The objective of the GTL is to collect dense data at high frequency. Data are intended for testing new traffic prediction algorithms, validating traffic mathematical models, and evaluating on-line traffic indexes.
Current version: 1.1.2
- Research engineers: V. Bertrand, R. Piotaix, I. Bellicot, P. Bellemain, A. Andreev
- PhD students: A. Ladino-Lopez, L. León-Ojeda
- Post-docs: E. Lovisari, F. Morbidi
- NeCS staff: C. Canudas-de-Wit, H. Fourati, F. Garin, A. Kibangou
- General description of GTL is published in: Grenoble Traffic Lab: an experimental platform for advanced traffic monitoring and control, C. Canudas de Wit, F. Morbidi, L. León Ojeda, A. Y. Kibangou, I. Bellicot, P. Bellemain, IEEE Control Systems Magazine, June 2015.
- Prediction algorithms are published in: Travel time forecasting from clustered time series via optimal fusion strategy, A. Ladino, A. Kibangou, H. Fourati, C. Canudas-de-Wit, European Control Conference 2016, Jun 2016, Aalborg, Denmark.
- An overview of how prediction algorithms worked and the obtained results can be found at NeCS website.
- In the scope of the ERC project Scale-Freeback, the team has also created a version of the GTL covering the whole city of Grenoble (and a bit more). The website is accessible here: GTL-Ville.
- Equipment and human resources of the project have received funding from several sources: UGA, HYCON2, MoCoPo, SPEEDD, ERC-AdG Scale-FreeBack, and the Inria DTI program (ADT).