Impact Factor:15.5

DOI number:10.1016/j.inffus.2025.103495

Journal:Information Fusion

Abstract:Traffic forecasting is a critical task in intelligent transportation systems, requiring accurate modeling of spatiotemporal dependencies among traffic sensors. Traditional deep-learning methods face two key challenges: (1) decoupled spatial–temporal pipelines that process and fuse spatial and temporal dimensions separately fail to capture their intricate interdependencies; and (2) state-of-the-art (SOTA) models relying on Transformer architectures often struggle to balance computational efficiency with representational capacity. To address these limitations, we propose ST-Camba, a novel decoupled-free spatiotemporal graph fusion state space model that unifies spatial and temporal dimensions within a single framework. ST-Camba is the first to integrate a spatial dimension axis into state space equations, enabling effective coupled spatiotemporal modeling through graph convolutions while inheriting the linear complexity advantage of Mamba series models. Additionally, we design an Adaptive Spatial Structure (ASS) Injector and a Lerp-based Gated Unit (LGU) to facilitate adaptive spatial structure capture and control information flow in spatiotemporal modeling. Extensive experiments on flow and speed prediction tasks across standard datasets demonstrate ST-Camba’s superiority. Specifically, on the PEMS07 dataset, our model achieves a 1.8% reduction in MAE compared to other baselines, while reducing computational costs by up to 14.5%. This work underscores the necessity of coupled spatiotemporal modeling and provides a theoretical foundation for scalable solutions in urban traffic systems.

Co-author:Tianhong Zhao, Bowen Zhang, Guanzhou Chen, Zhenhui Li, Haolin Chen, Wei Tu, Qingquan Li

Indexed by:Journal paper

Correspondence Author:Jinzhou Cao

Document Type:J

Volume:126

Page Number:103495

ISSN No.:1566-2535

Translation or Not:no

Date of Publication:2025-07-24