发表机构
Tennessee State University; Nashville Department of Transportation & Multimodal Infrastructure（田纳西州立大学; 纳什维尔交通与多式联运基础设施部）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
提出LTV-CTDNet框架，通过组合式分解分别预测非负进口道总量与转向比例，在纳什维尔激光雷达数据上实现无负值且精度具竞争力的短期转向流量预测。
AI 中文摘要
短期转向流量预测可支持信号控制与走廊运营，但无约束的神经网络可能产生物理上不可能的负计数，或输出未明确关联到进口道需求总量的结果。本研究提出了线性时变组合式转向分解网络（LTV-CTDNet），这是一种旨在兼顾竞争性精度与结构上可接受输出的预测框架。LTV-CTDNet 使用来自田纳西州纳什维尔八个受监控走廊位置、为期七个月的 15 分钟激光雷达观测数据进行评估。其轻量级编码器结合了近期转向历史、周时段嵌入和位置嵌入。组合式转向分解框架分别预测非负的进口道总量和进口道内转向比例，然后从这些分量重建转向流量预测。在评估的预定义配置中，LTV-CTDNet 实现了转向级平均绝对误差（MAE）为 1.8189，均方根误差（RMSE）为 3.8072。其相对于最强序列模型的精度提升较为温和，但它未产生任何负预测，而无约束的学习模型在约 10.6% 至 29.2% 的原始预测单元中产生了负值。该框架通过构造确保输出非负，并确保每个模型预测的进口道总量与其各转向分量之和精确一致，从而无需裁剪或一致性校正即可提供直接可解释的预测。
英文摘要
Short-term turning-movement forecasts can support signal control and corridor operations, but unconstrained neural networks may produce physically impossible negative counts or outputs that are not explicitly tied to an approach-demand total. This study introduces the Linear Temporal-Variable Compositional Turning Decomposition Network (LTV-CTDNet), a forecasting framework designed to combine competitive accuracy with structurally admissible outputs. LTV-CTDNet was evaluated using seven months of 15-minute LiDAR observations from eight monitored corridor locations in Nashville, Tennessee. Its lightweight encoder combines recent turning-movement history, weekly time-slot embeddings, and location embeddings. The Compositional Turning Decomposition framework separately predicts nonnegative approach totals and within-approach turning proportions, then reconstructs movement forecasts from these components. Among the evaluated predefined configurations, LTV-CTDNet achieved a movement-level MAE of 1.8189 and RMSE of 3.8072. Its accuracy gains over the strongest sequence models were modest, but it produced no negative forecasts, while unconstrained learned models generated negative values in approximately 10.6% to 29.2% of raw forecast cells. The framework enforces nonnegative outputs and exact agreement between each model-predicted approach total and the sum of its component movements by construction, providing directly interpretable forecasts without clipping or coherence correction.
CommentsAccepted: September 26, 2026, for presentation at the 2027 TRB Annual Meeting, Washington, D.C. (Paper No. TRBAM-27-06161)
