arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2405.03809cs.AI
收藏

SocialFormer:利用边增强异构图 Transformer 进行社会交互建模以实现轨迹预测

SocialFormer: Social Interaction Modeling with Edge-enhanced Heterogeneous Graph Transformers for Trajectory Prediction

Robert Bosch GmbH(罗伯特·博世公司);Bosch Center for Artificial Intelligence(博世人工智能中心);Technical University of Munich(慕尼黑工业大学)

Zixu Wang, Zhigang Sun, Juergen Luettin, Lavdim Halilaj

AI总结:

本文提出 SocialFormer 轨迹预测方法,利用道路拓扑、EHGT 异构图网络和 GRU 时间编码器建模车辆间社会交互,并融合智能体、车道与交互信息,在 nuScenes 上达到最先进性能。

摘要:

准确的轨迹预测对于确保安全、高效的自动驾驶至关重要。然而,大多数现有方法忽视了交通参与者之间往往决定其未来轨迹的复杂交互。本文提出 SocialFormer,这是一种智能体交互感知的轨迹预测方法,通过利用道路拓扑来挖掘目标车辆与周围车辆之间的语义关系。我们还引入边增强异构图 Transformer(EHGT)作为图神经网络(GNN)中的聚合器,以编码语义和空间智能体交互信息。此外,我们提出一种基于门控循环单元(GRU)的时间编码器,用于建模智能体运动的时间社会行为。最后,我们提出一个信息融合框架,整合智能体编码、车道编码和智能体交互编码,以形成交通场景的整体表示。我们在流行的 nuScenes 基准上评估 SocialFormer 的轨迹预测任务,并取得了最先进的性能。

英文摘要:

Accurate trajectory prediction is crucial for ensuring safe and efficient autonomous driving. However, most existing methods overlook complex interactions between traffic participants that often govern their future trajectories. In this paper, we propose SocialFormer, an agent interaction-aware trajectory prediction method that leverages the semantic relationship between the target vehicle and surrounding vehicles by making use of the road topology. We also introduce an edge-enhanced heterogeneous graph transformer (EHGT) as the aggregator in a graph neural network (GNN) to encode the semantic and spatial agent interaction information. Additionally, we introduce a temporal encoder based on gated recurrent units (GRU) to model the temporal social behavior of agent movements. Finally, we present an information fusion framework that integrates agent encoding, lane encoding, and agent interaction encoding for a holistic representation of the traffic scene. We evaluate SocialFormer for the trajectory prediction task on the popular nuScenes benchmark and achieve state-of-the-art performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑