发表机构
University of Florida（佛罗里达大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
PhaseShift是拓扑感知框架，可协调异构路边轨迹并训练可复用主干，在佛罗里达5个交叉口的评估中，其池化模型等方案在长时预测指标上优于局部训练模型，实现了跨交叉口的模型整合。
AI 中文摘要
学习型交通行为模型通常针对每个交叉口单独训练，形成的模型组合无法在不同站点间共享证据。本文提出PhaseShift，这是一种拓扑感知框架，可将异构路边轨迹协调为共享的以智能体为中心的表示，并训练一个可复用的主干网络。自相对坐标、轨迹诱导运动路径、归一化信号上下文以及可变基数交互标记消除了站点惯例，同时保留了与行为相关的拓扑结构。该主干网络支持池化操作、在保留的交叉口上的零样本迁移以及低数据量自适应。我们在佛罗里达州两个区域的五个交叉口上，使用平衡的现场数据、10万个训练窗口和每个站点等量的测试集，在重放条件下的最优采样轨迹协议下进行评估。在10秒时，一个池化模型相对于训练的局部模型，在所有五个站点均降低了minADE和minFDE，中位数降幅分别为36.8%和22.0%。留一个交叉口的部署（包括一个跨区域折叠）在五个站点中的四个站点的10秒指标上优于局部训练，尽管短时间范围内的性能不太均匀。用1000个目标更新窗口进行微调，在三个站点上优于零样本迁移，并且在一个站点上是最强的方案。在7号站点，在固定的10万个窗口预算下，每个跨站点混合模型都显著降低了长时误差；测试似然增益表明，这并非仅由最优样本分散所解释。在长自回归滚动后，局部模型在两个流量最高的站点落后于校准的IDM；而预训练主干网络方案则不会。在这个五站点评估中，PhaseShift展示了在异构物理控制设置下的整合能力，同时识别出仍需自适应的站点。该协议测量的是重放上下文下的条件单车辆生成，而非闭环交通仿真。
英文摘要
Learned traffic-behavior models are commonly trained separately for each intersection, creating model portfolios that cannot share evidence across sites. We present PhaseShift, a topology-aware framework that harmonizes heterogeneous roadside trajectories into a shared actor-centric representation and trains one reusable backbone. Ego-relative coordinates, trajectory-induced movement paths, normalized signal context, and variable-cardinality interaction tokens remove site conventions while preserving behaviorally relevant topology. The backbone supports pooled operation, zero-shot at a held-out intersection, and low-data adaptation. We evaluate five intersections in two Florida regions on balanced field data, 100k training windows and equal-sized test sets per site under a replay-conditioned, best-of-sampled-trajectory protocol. At 10s, one pooled model lowers both minADE and minFDE relative to trained local models at all five sites, with median reductions of 36.8% and 22.0%. Leave-one-intersection-out deployment, including one cross-region fold, beats local training on both 10-s metrics at four of five sites, although short-horizon performance is less uniform. Fine-tuning with 1,000 target update windows improves on zero-shot at three sites and is the strongest regime at one. At site 7, every cross-site mixture sharply lowers long-horizon error under a fixed 100k-window budget; test-likelihood gains argue against a best-of-sample dispersion-only explanation. Local models fall behind calibrated IDM at the two highest-flow sites after long autoregressive rollouts; pretrained-backbone regimes do not. Within this five-site evaluation, PhaseShift demonstrates consolidation across heterogeneous physical control settings while identifying sites that still require adaptation. The protocol measures conditional single-vehicle generation under replayed context, not closed-loop traffic simulation.
