Data-Driven Transfer Learning Framework for Estimating Turning Movement Counts

TL;DR

This paper introduces a transfer learning framework that improves the estimation of turning movement counts at intersections using traffic data, outperforming existing models in accuracy and reducing reliance on costly physical sensors.

Contribution

The study presents a novel transfer learning approach that enhances TMC estimation accuracy across diverse intersections by integrating multiple data sources, addressing domain discrepancy issues.

Findings

The proposed TL model achieved the lowest MAE and RMSE among tested models.

Transfer learning improved generalization of TMC estimates to new intersections.

Model outperformed eight state-of-the-art regression models.

Abstract

Urban transportation networks are vital for the efficient movement of people and goods, necessitating effective traffic management and planning. An integral part of traffic management is understanding the turning movement counts (TMCs) at intersections, Accurate TMCs at intersections are crucial for traffic signal control, congestion mitigation, and road safety. In general, TMCs are obtained using physical sensors installed at intersections, but this approach can be cost-prohibitive and technically challenging, especially for cities with extensive road networks. Recent advancements in machine learning and data-driven approaches have offered promising alternatives for estimating TMCs. Traffic patterns can vary significantly across different intersections due to factors such as road geometry, traffic signal settings, and local driver behaviors. This domain discrepancy limits the generalizability and accuracy of machine learning models when applied to new or unseen intersections. In response to these limitations, this research proposes a novel framework leveraging transfer learning (TL) to estimate TMCs at intersections by using traffic controller event-based data, road infrastructure data, and point-of-interest (POI) data. Evaluated on 30 intersections in Tucson, Arizona, the performance of the proposed TL model was compared with eight state-of-the-art regression models and achieved the lowest values in terms of Mean Absolute Error and Root Mean Square Error.