Short-term traffic prediction using physics-aware neural networks
Preprint, 2021

In this work, we propose an algorithm performing short-term predictions of the flux of vehicles on a stretch of road, using past measurements of the flux. This algorithm is based on a physics-aware recurrent neural network. A discretization of a macroscopic traffic flow model (using the so-called Traffic Reaction Model) is embedded in the architecture of the network and yields flux predictions based on estimated and predicted space-time dependent traffic parameters. These parameters are themselves obtained using a succession of LSTM ans simple recurrent neural networks. Besides, on top of the predictions, the algorithm yields a smoothing of its inputs which is also physically-constrained by the macroscopic traffic flow model. The algorithm is tested on raw flux measurements obtained from loop detectors.

Författare

Mike Pereira

Chalmers, Elektroteknik, System- och reglerteknik, Reglerteknik

Annika Lang

Chalmers, Matematiska vetenskaper, Tillämpad matematik och statistik

Balázs Adam Kulcsár

Chalmers, Elektroteknik, System- och reglerteknik, Reglerteknik

Real-Time Robust and AdaptIve Learning in ElecTric VEhicles (RITE)

Chalmers AI-forskningscentrum (CHAIR), 2020-01-01 -- 2021-12-31.

Chalmers, 2020-01-01 -- 2021-12-31.

IRIS: Inverse förstärkning-lärande och intelligenta svarmalgoritmer för elastiska transportnät

Chalmers, 2020-01-01 -- 2021-12-31.

Efficienta approximeringsmetoder för stokastiska fält på mångfalder

Vetenskapsrådet (VR) (2020-04170), 2021-01-01 -- 2024-12-31.

STOchastic Traffic NEtworks (STONE)

Chalmers AI-forskningscentrum (CHAIR), -- .

Chalmers, 2020-02-01 -- 2022-01-31.

Stochastic Continuous-Depth Neural Networks

Chalmers AI-forskningscentrum (CHAIR), 2020-08-15 -- .

Optimal energihantering för nätverk av elektrifierade bussar (OPNET)

Energimyndigheten (46365-1), 2018-10-01 -- 2021-12-31.

Styrkeområden

Transport

Ämneskategorier

Beräkningsmatematik

Sannolikhetsteori och statistik

Reglerteknik

Mer information

Skapat

2021-09-30