arXiv cs.AI· Ran Tu, Zihao Li, Arash Mirzaei·· 12 小时前AI 评分26
GCRN+：用探针监督图卷积循环网络从稀疏固定传感器预测全网高速公路速度
Network-Wide Freeway Speed Nowcasting and Forecasting from Sparse Fixed Sensors with a Probe-Supervised Graph Convolutional Recurrent Network
AI 导读
GCRN+ 在达拉斯-沃斯堡地区 4127 条方向性高速路段上，仅用 643 条路段（15.6%）的传感器数据实现全路段当前小时与下一小时速度预测，无传感器路段 RMSE 分别为 2.21 mph 和 2.39 mph，优于历史均值基线的 3.23 mph。
正文 · 原文
Abstract:Traffic agencies need current and near-future speeds on every link of their network, but their fixed sensors cover only part of it. We study joint nowcasting and next-hour forecasting of hourly speeds on all 4{,}127 directional freeway links of the Dallas--Fort Worth region from sensors on 643 links (15.6%). Historical probe-vehicle data supervise every link in training; only sensor readings are used at inference. We call this setting probe-supervised sparse sensing. We propose GCRN+, a graph convolutional recurrent network that predicts deviations from a historical speed profile. It combines a parameter-free, direction-aware kernel interpolation of sensor anomalies, used as input features and as an output baseline, with a diffusion-convolutional GRU on the directed road graph, a low-rank adaptive graph and learnable link embeddings. It scales linearly with network size and trains on a CPU. Sensor readings are emulated from probe speeds, which isolates spatio-temporal inference from detector calibration. On four held-out test days, GCRN+ attains an RMSE of 2.21~mph for the current hour and 2.39~mph for the next hour on unsensored links, against 3.23~mph for the historical average. The best baselines that use the same interpolation features reach 2.53 and 2.75~mph. Adapted recent models without these features (STID, STAEformer, ImputeFormer, MoGERNN) reach at best 2.54 and 2.53~mph; this comparison is between complete pipelines rather than architectures. GCRN+ is best in every distance band except links not connected to any sensor through the road graph. Ablations within one recurrent backbone attribute most of the gain to the link embeddings and the adaptive graph, which can be trained because probe labels exist for every link. For congestion detection on unsensored links, recall rises from 77\% (historical average) and 86% (best competitor) to 89% at aprecision of 0.94.
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.22630 [cs.LG] |
| (or arXiv:2610.22630v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.22630 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ran Tu [view email]
[v1]
Thu, 15 Oct 2026 20:35:45 UTC (819 KB)
来源：arXiv cs.AI · arxiv.org