面向边缘设备的低延迟轨迹预测模型，兼顾精度与实时性。
EdgeVTP: Exploration of Latency-efficient Trajectory Prediction for Edge-based Embedded Vision Applications
- 用轻量Transformer+图建模捕捉车辆交互，预测曲线参数而非逐点生成。
- 在三组高速数据集上实现最低端到端延迟，两个达当前最佳精度。
- 适合部署于嵌入式边缘计算平台，尤其适用于高密度交通场景。
车辆轨迹预测是高速公路感知的核心，但部署于路侧边缘设备需保证确定性的端到端延迟。我们提出EdgeVTP，一种以嵌入式系统为先的设计方案，结合考虑交互的图建模、轻量级Transformer主干和一次性曲线解码器。通过将未来运动预测为以最后观测位置为锚点的紧凑曲线参数，而非时序展开的自回归点序列，显著降低解码开销并生成平滑轨迹。为在复杂场景中保持运行时可预测性，采用带硬邻居数限制的局部图结构显式控制交互复杂度。在三个高速公路基准测试和两个Jetson类平台上的实验表明，EdgeVTP在包含图构建与后处理的完整流程下实现了最低测量端到端延迟，且在其中两个数据集上达到当前最佳（SotA）预测精度，在其余基准上表现具有竞争力。代码已公开于https://github.com/SeungjinStevenKim/EdgeVTP。
原文摘要 · Abstract (English)
Vehicle trajectory prediction is central to highway perception, but deployment on roadside edge devices necessitates bounded, deterministic end-to-end latency. We present EdgeVTP, an embedded-first trajectory predictor that combines interaction-aware graph modeling with a lightweight transformer backbone and a one-shot curve decoder. By predicting future motion as compact curve parameters (anchored at the last observed position) rather than horizon-scaled autoregressive waypoints, EdgeVTP reduces decoding overhead while producing smooth trajectories. To keep runtime predictable in crowded scenes, we explicitly bound interaction complexity via a locality graph with a hard neighbor cap. Across three highway benchmarks and two Jetson-class platforms, EdgeVTP achieves the lowest measured end-to-end latency under a protocol that includes graph construction and post-processing, while attaining state-of-the-art (SotA) prediction accuracy on two of the three datasets and competitive error on other benchmarks. Our code is available at https://github.com/SeungjinStevenKim/EdgeVTP.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。