TY  - JOUR
PY  - 2022//
TI  - Short-term traffic prediction using physics-aware neural networks
JO  - Transportation research part C: emerging technologies
A1  - Pereira, Mike
A1  - Lang, Annika
A1  - Kulcsár, Balázs
SP  - e103772
EP  - e103772
VL  - 142
IS  - 
N2  - In this work, we propose an algorithm performing short-term predictions of the flow and speed of vehicles on a stretch of road, using past measurements of these quantities. 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 traffic state estimations and predictions for the flow and speed of vehicles, which are physically-constrained by the macroscopic traffic flow model and based on estimated and predicted space-time dependent traffic parameters. These parameters are themselves obtained using a succession of LSTM recurrent neural networks. The algorithm is tested on raw flow measurements obtained from loop detectors.<p />  <p>Language: en</p>
LA  - en
SN  - 0968-090X
UR  - http://dx.doi.org/10.1016/j.trc.2022.103772
ID  - ref1
ER  -