• Computer Science > Machine Learning [Submitted on 17 Feb 2026] Title:MMCAformer: Macro-Micro Cross-Attention Transformer for Traffic Speed Prediction with Microscopic Connected Vehicle Driving Behavior View PDFAbstract: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.
Article Summaries:
- A new study introduces the Macro‑Micro Cross‑Attention Transformer (MMCAformer), a neural network that fuses aggregated traffic‑flow data with fine‑grained driving‑behavior signals from Connected Vehicles (CVs). Using self‑attention to model macro‑level dependencies and cross‑attention to link them with micro‑level features such as hard braking and acceleration frequencies, the model predicts future traffic speeds while estimating uncertainty via a Student‑t negative log‑likelihood loss. Experiments on four Florida freeways show that adding micro‑behavior data reduces RMSE, MAE, and MAPE by roughly 9-10 % and shrinks predictive intervals by 10-24 %. The gains are most pronounced in congested, low‑speed conditions.
Sources: