Skip to main content
QUICK REVIEW

[论文解读] V2X-Seq: A Large-Scale Sequential Dataset for Vehicle-Infrastructure Cooperative Perception and Forecasting

Haibao Yu, Wenxian Yang|arXiv (Cornell University)|May 10, 2023
Autonomous Vehicle Technology and Safety被引用 4
一句话总结

V2X-Seq 提供了首个大规模真实世界序列数据集,用于车辆-基础设施协同(VIC)感知与预测,包含15,000帧用于感知任务,210,000个场景用于轨迹预测,数据覆盖672小时。本文提出了三项新任务——VIC3D追踪、在线-VIC预测与离线-VIC预测,并建立了基准。实验表明,融合基础设施数据可提升追踪与预测性能,其中FF-Tracking在300ms延迟下相较早期融合方法实现最高4%的MOTA提升。

ABSTRACT

Utilizing infrastructure and vehicle-side information to track and forecast the behaviors of surrounding traffic participants can significantly improve decision-making and safety in autonomous driving. However, the lack of real-world sequential datasets limits research in this area. To address this issue, we introduce V2X-Seq, the first large-scale sequential V2X dataset, which includes data frames, trajectories, vector maps, and traffic lights captured from natural scenery. V2X-Seq comprises two parts: the sequential perception dataset, which includes more than 15,000 frames captured from 95 scenarios, and the trajectory forecasting dataset, which contains about 80,000 infrastructure-view scenarios, 80,000 vehicle-view scenarios, and 50,000 cooperative-view scenarios captured from 28 intersections' areas, covering 672 hours of data. Based on V2X-Seq, we introduce three new tasks for vehicle-infrastructure cooperative (VIC) autonomous driving: VIC3D Tracking, Online-VIC Forecasting, and Offline-VIC Forecasting. We also provide benchmarks for the introduced tasks. Find data, code, and more up-to-date information at \href{https://github.com/AIR-THU/DAIR-V2X-Seq}{https://github.com/AIR-THU/DAIR-V2X-Seq}.

研究动机与目标

  • 解决自动驾驶领域中车辆-基础设施协同(VIC)感知与预测缺乏真实世界序列数据集的问题。
  • 通过提供大规模真实世界数据集并同步多模态数据,推动协同追踪与行为预测研究。
  • 提出三项新任务——VIC3D追踪、在线-VIC预测与离线-VIC预测,以推进基于VIC的自动驾驶系统发展。
  • 基于来自28个路口的672小时真实世界数据,建立公平的基准以评估VIC感知与预测模型。
  • 开发并验证一种新型FF-Tracking框架,提升在可变延迟条件下的追踪鲁棒性。

提出的方法

  • 从95个真实世界场景中采集并标注大规模序列化V2X数据集(V2X-Seq),包含基础设施与车辆侧图像、点云、3D检测结果、矢量地图及交通灯信号。
  • 设计并实现三项新任务:VIC3D追踪(协同3D目标追踪)、在线-VIC预测(利用实时基础设施与本车数据进行行为预测)与离线-VIC预测(从历史基础设施数据中蒸馏知识)。
  • 提出FF-Tracking,一种基于晚期融合的框架,通过在特征层面融合车辆与基础设施输入,提升追踪精度与延迟鲁棒性。
  • 开发PP-VIC,一种分层感知-预测框架,在将轨迹输入TNT与HiVT等轨迹模型前,融合基础设施与本车轨迹。
  • 在80,000条基础设施视角轨迹上预训练轨迹预测模型(TNT、HiVT),以提取适用于离线预测的通用知识。
  • 在轨迹预测数据集上采用5:2:3的训练/验证/测试划分,确保所有基线方法与所提方法的公平评估。
Figure 1 : Autonomous driving datasets. V2X-Seq is the first large-scale, real-world, and sequential V2X dataset. The green circle denotes the real-world dataset, and the pink triangle denotes the simulated dataset. The abscissa represents the number of sequences.
Figure 1 : Autonomous driving datasets. V2X-Seq is the first large-scale, real-world, and sequential V2X dataset. The green circle denotes the real-world dataset, and the pink triangle denotes the simulated dataset. The abscissa represents the number of sequences.

实验结果

研究问题

  • RQ1基础设施数据能否显著提升协同感知系统中3D目标追踪的准确率与鲁棒性?
  • RQ2相较于仅依赖本车数据的模型,实时访问基础设施轨迹在多大程度上能提升在线轨迹预测性能?
  • RQ3从历史基础设施轨迹中提取的知识能否改善离线轨迹预测性能?
  • RQ4在不同延迟条件下,所提出的FF-Tracking框架与早期融合基线相比,追踪性能如何?
  • RQ5在仅基于基础设施的轨迹上进行预训练,对提升下游预测模型泛化能力有何影响?

主要发现

  • 在300ms延迟下,FF-Tracking相较早期融合方法在MOTA上最高提升4%,证明其对通信延迟具有更强鲁棒性。
  • 在200ms延迟下,FF-Tracking实现最佳追踪性能,证实其在实时协同感知中的有效性。
  • 使用在线基础设施轨迹可使minADE降低3.74(TNT)与0.28(HiVT),相较于仅本车数据基线,证明其在实时预测中的价值。
  • 在80,000条基础设施视角轨迹上进行预训练,可使TNT的minADE降低7.65,HiVT降低0.03,表明离线知识提取具有显著优势。
  • PP-VIC框架的minADE与minFDE均低于仅本车数据模型,证实融合基础设施与车辆轨迹可提升预测精度。
  • V2X-Seq数据集实现了首次大规模真实世界VIC感知与预测评估,涵盖50,000个协同视角场景与672小时数据,为该领域树立了新基准。
Figure 2 : Total number and average tracking length of 3D tracked objects per category for the sequential perception dataset (SPD). The distribution of tracked objects is relatively balanced.
Figure 2 : Total number and average tracking length of 3D tracked objects per category for the sequential perception dataset (SPD). The distribution of tracked objects is relatively balanced.

更好的研究,从现在开始

从阅读论文到最终审阅,大幅缩短您的研究时间。

无需绑定信用卡

本解读由 AI 生成,并经人工编辑审核。