arXiv:2605.27884v1 Announce Type: new 
Abstract: City-wide traffic forecasting is important for congestion management, route guidance, and intelligent transportation systems, but accurate prediction remains challenging when future traffic must be generated as spatial maps over an entire urban network. Existing traffic movie prediction methods have improved frame-level accuracy, yet many still treat forecasting mainly as image reconstruction. This can produce traffic maps that are numerically close to the ground truth but weakly constrained by road layout, connectivity, travel direction, and congestion propagation, especially in cross-city settings where both traffic behavior and road structure change. To address this limitation, this study proposes RCSNet, a road-conditioned spatiotemporal network that reformulates traffic movie prediction as topology-guided future-state generation. RCSNet extracts road-aware representations from static road maps, models multi-horizon traffic dynamics from historical observations, aligns directional traffic features with local road structure, and progressively generates future traffic maps for improved temporal consistency. A structure-consistent learning objective further encourages predictions to remain accurate, road-aligned, and spatially stable. Experiments across multiple cities show that RCSNet improves both forecasting accuracy and structural consistency. In same-city forecasting on Berlin, Antwerp, and Moscow, RCSNet reduces average MAE, MSE, and RMSE by 11.5%, 10.0%, and 5.1%, respectively, compared with the closest baseline. In cross-city testing on unseen Chicago and Bangkok, it reduces RMSE by 10.6% and 10.5% without target-city fine-tuning. Additional horizon-wise, road-structure, explainability, statistical, and efficiency analyses show that RCSNet produces more accurate, transferable, road-aligned, and computationally efficient traffic forecasts.

تم اقتراح نموذج جديد لتوقع حركة المرور يسمى RCSNet، والذي يستخدم التعلم المكاني الزمني المشروط بالطريق لتحسين دقة توقعات حركة المرور على مستوى المدينة. يعالج هذا النموذج قيود الطرق الحالية من خلال دمج تخطيط الطرق والاتصال في عملية التوقع، مما يجعله أكثر فعالية في البيئات الحضرية المتنوعة مثل برلين وأنتويرب وموسكو وشيكاغو وبانكوك.

Se ha propuesto un nuevo modelo de predicción del tráfico llamado RCSNet, que utiliza el aprendizaje espaciotemporal condicionado por la carretera para mejorar la precisión de las predicciones de tráfico a nivel de ciudad. Este modelo aborda las limitaciones de los métodos existentes al incorporar el diseño de las carreteras y la conectividad en el proceso de predicción, haciéndolo más efectivo en diversos entornos urbanos como Berlín, Amberes, Moscú, Chicago y Bangkok.

Un nouveau modèle de prévision du trafic, nommé RCSNet, a été proposé, utilisant un apprentissage spatiotemporel conditionné par la route pour améliorer la précision des prévisions de trafic à l'échelle d'une ville. Ce modèle répond aux limites des méthodes existantes en intégrant la configuration routière et la connectivité dans le processus de prévision, le rendant plus efficace dans divers environnements urbains tels que Berlin, Anvers, Moscou, Chicago et Bangkok.

A new traffic forecasting model named RCSNet has been proposed, which utilizes road-conditioned spatiotemporal learning to enhance the accuracy of city-wide traffic predictions. This model addresses the limitations of existing methods by incorporating road layout and connectivity into the forecasting process, making it more effective in diverse urban environments like Berlin, Antwerp, Moscow, Chicago, and Bangkok.