Predicting times of waiting on red signals using BERT

TL;DR

This paper introduces a BERT-based model to predict red signal waiting times in traffic, outperforming other machine learning models on simulated traffic data, with potential applications in traffic signal optimization.

Contribution

The paper presents a novel application of BERT for traffic signal timing prediction, demonstrating superior performance over traditional ML models on realistic simulation data.

Findings

BERT-based models outperform other ML models in predicting waiting times.

The approach is effective on realistic traffic simulation data.

Potential for optimizing traffic signals with autonomous vehicles.

Abstract

We present a method for approximating outcomes of road traffic simulations using BERT-based models, which may find applications in, e.g., optimizing traffic signal settings, especially with the presence of autonomous and connected vehicles. The experiments were conducted on a dataset generated using the Traffic Simulation Framework software runs on a realistic road network. The BERT-based models were compared with 4 other types of machine learning models (LightGBM, fully connected neural networks and 2 types of graph neural networks) and gave the best results in terms of all the considered metrics.