📄 中文摘要
准确的交通速度预测对于主动交通管理、提升交通效率和安全性至关重要。现有研究主要依赖于聚合的宏观交通流数据来预测未来交通趋势，而道路交通动态也受到个体微观人类驾驶行为的影响。近期的连接车辆（CV）数据提供了丰富的驾驶行为特征，为将这些行为洞察纳入速度预测提供了新机会。为此，提出了宏观-微观交叉注意力变换器（MMCAformer），旨在将基于CV数据的微观驾驶行为特征与宏观交通特征相结合进行速度预测。MMCAformer特别采用自注意力机制来学习内部特征，从而提升预测的准确性。
MMCAformer：用于交通速度预测的宏观-微观交叉注意力变换器，结合微观连接车辆驾驶行为
📄 English Summary
MMCAformer: Macro-Micro Cross-Attention Transformer for Traffic Speed Prediction with Microscopic Connected Vehicle Driving Behavior
Accurate speed prediction is essential for proactive traffic management to improve traffic efficiency and safety. Existing studies have primarily relied on aggregated macroscopic traffic flow data to forecast future traffic trends, while road traffic dynamics are also influenced by individual microscopic human driving behaviors. Recent Connected Vehicle (CV) data provide rich driving behavior features, creating new opportunities to incorporate these behavioral insights into speed prediction. To address this, the Macro-Micro Cross-Attention Transformer (MMCAformer) is proposed to integrate CV data-based micro driving behavior features with macro traffic features for speed prediction. Specifically, MMCAformer employs self-attention to learn intrinsic features, enhancing the accuracy of predictions.