MMCAformer: Macro-Micro Cross-Attention Transformer for Traffic Speed Prediction with Microscopic Connected Vehicle Driving Behavior

2026-02-17

AI Summary

This study addresses the limitation of existing traffic speed prediction methods that predominantly rely on macroscopic traffic flow data while neglecting the influence of microscopic driving behaviors. To bridge this gap, the authors propose a novel macro-micro cross-attention mechanism within a Transformer architecture, which integrates microscopic driving behavior features—such as hard braking and rapid acceleration frequencies—derived from connected vehicles into the macroscopic prediction framework. The model jointly captures spatiotemporal interactions between macro and micro levels and employs a Student-t negative log-likelihood loss to quantify prediction uncertainty. Experiments on four Florida highways demonstrate that incorporating microscopic features reduces RMSE, MAE, and MAPE by 9.0%, 6.9%, and 10.2%, respectively, compared to using only macroscopic features, while narrowing prediction intervals by 10.1%–24.0%, with particularly notable improvements during congested conditions.

Technology Category

• Planning, Routing, and Scheduling: Model-Based Reasoning
• Application Domains: Transportation
• Machine Learning: Feature Construction/Reformulation

Application Category

• Security and Privacy: Large-scale security measurements
• Web Mining and Content Analysis: Large pretrained models with web data
• Social Networks and Social Media: Influence propagation, information diffusion, and the prediction on networks

Abstract

Accurate speed prediction is crucial for proactive traffic management to enhance traffic efficiency and safety. Existing studies have primarily relied on aggregated, macroscopic traffic flow data to predict future traffic trends, whereas road traffic dynamics are also influenced by individual, microscopic human driving behaviors. Recent Connected Vehicle (CV) data provide rich driving behavior features, offering new opportunities to incorporate these behavioral insights into speed prediction. To this end, we propose the Macro-Micro Cross-Attention Transformer (MMCAformer) to integrate CV data-based micro driving behavior features with macro traffic features for speed prediction. Specifically, MMCAformer employs self-attention to learn intrinsic dependencies in macro traffic flow and cross-attention to capture spatiotemporal interplays between macro traffic status and micro driving behavior. MMCAformer is optimized with a Student-t negative log-likelihood loss to provide point-wise speed prediction and estimate uncertainty. Experiments on four Florida freeways demonstrate the superior performance of the proposed MMCAformer compared to baselines. Compared with only using macro features, introducing micro driving behavior features not only enhances prediction accuracy (e.g., overall RMSE, MAE, and MAPE reduced by 9.0%, 6.9%, and 10.2%, respectively) but also shrinks model prediction uncertainty (e.g., mean predictive intervals decreased by 10.1-24.0% across the four freeways). Results reveal that hard braking and acceleration frequencies emerge as the most influential features. Such improvements are more pronounced under congested, low-speed traffic conditions.

Problem

• traffic speed prediction
• connected vehicle
• microscopic driving behavior
• macro-micro integration
• traffic dynamics

Innovation

• Macro-Micro Cross-Attention
• Connected Vehicle
• Traffic Speed Prediction
• Uncertainty Quantification
• Transformer