Short-term traffic prediction using physics-aware neural networks

Abstract

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.

17 pages, 11 figures, 2 tables

References in corpus (3)

• Macroscopic Traffic Flow Modeling with Physics Regularized Gaussian Process: A New Insight into Machine Learning Applications
• Solving inverse-PDE problems with physics-aware neural networks
• Parameter and density estimation from real-world traffic data: A kinetic compartmental approach

Cited by in corpus (4)

• Fourier neural operator for learning solutions to macroscopic traffic flow models: Application to the forward and inverse problems
• The Traffic Reaction Model: A kinetic compartmental approach to road traffic modeling
• Reconstructing Physics-Informed Machine Learning for Traffic Flow Modeling: a Multi-Gradient Descent and Pareto Learning Approach
• Proxy Reconstruction Pre-training for Ramp Flow Prediction at Highway Interchanges