arXiv cs.AI· Qishen Zhou, Yifan Zhang, Michail A. Makridis, Anastasios Kouvelas, Yibing Wang, Simon Hu·· 13 小时前AI 评分29
TA-ANP：融合多源数据的大都市级弹性可信交通流推断框架
Metropolis-Scale Resilient and Trustworthy Traffic Flow Inference Using Multi-Source Data
AI 导读
研究者提出任务感知注意力神经过程 TA-ANP，将浮动车数据与稀疏固定检测器测量融合，实现全网交通状态的概率化推断，无需重训练即可适应传感配置变化。团队构建覆盖 26,166 条路段（含 2,371 条装传感器路段）的大都市多源交通数据集 MMTD，TA-ANP 在实时估计与观测/未观测位置预测等子任务上均取得最佳性能。
正文 · 原文
Abstract:Inferring network-wide traffic states from sparse observations with high accuracy and trustworthy uncertainty quantification is essential for intelligent transportation systems, yet it remains challenging due to the underdetermined nature of the problem, multifaceted disturbances in sensing networks, and the inherent conflicts among multiple inference subtasks when modeled jointly. We propose the Task-Aware Attentive Neural Process (TA-ANP), a unified probabilistic framework for resilient and trustworthy global traffic state inference (GTSI) by fusing floating car data (FCD) with sparse fixed-detector measurements. By casting GTSI as a stochastic process, TA-ANP leverages the meta-learning properties of neural processes to adapt rapidly to changes in sensing configurations without retraining. A task-aware multi-query attention module with distinct spatiotemporal inductive biases is introduced to handle three GTSI subtasks jointly ---real-time estimation at unobserved locations, forecasting at both observed and unobserved locations---while mitigating cross-task interference. For uncertainty quantification, we combine neural processes with Monte Carlo Dropout to capture both aleatoric and epistemic uncertainty. To support metropolis-scale evaluation, we construct the Metropolitan Multi-Source Traffic Dataset (MMTD), integrating fixed-loop sensor measurements, FCD statistics, and OpenStreetMap road-network data over an urban network of 26,166 road segments, including 2,371 fixed-sensor-equipped segments used for quantitative evaluation. Experiments across multiple independently generated sensor configurations show that TA-ANP achieves best performance across all subtasks under deterministic and probabilistic metrics. The resulting well-calibrated uncertainties enable more efficient fixed-sensor placement with fewer sensor deployments.
| Comments: | The paper has been accepted by Transportation Research Part C: Emerging Technologies |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.25004 [cs.LG] |
| (or arXiv:2605.25004v2 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2605.25004 arXiv-issued DOI via DataCite |
|
| Journal reference: | Transportation Research Part C: Emerging Technologies, Volume 194, Part B, January 2027, 106083 |
| Related DOI: | https://doi.org/10.1016/j.trc.2026.106083
DOI(s) linking to related resources |
Submission history
From: Qishen Zhou [view email]
[v1]
Sun, 24 May 2026 11:12:02 UTC (5,967 KB)
[v2]
Wed, 21 Oct 2026 05:45:57 UTC (18,150 KB)
来源：arXiv cs.AI · arxiv.org