跳到主要导航 跳到搜索 跳到主要内容

Multi-modal Trajectory Prediction Network that Integrates Historical Motion and Spatio-Temporal Interaction

  • Chenlong Li
  • , Mingxing Li*
  • , Jian Zhao
  • *此作品的通讯作者
  • Beihang University
  • Innovation Centor for Control Actuators
  • Beijing Institute of Precise Mechatronics and Controls

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Multi-modal trajectory prediction (MTP) has become the research trend in the field of autonomous driving, as it provides multiple plausible trajectories. However, related works lack attention to the temporal dependence of historical features and inherent association between multiple trajectory modes, which may lead to large deviations. To address these critical limitations, we propose a multi-modal trajectory prediction network that integrates historical motion and spatio-temporal interaction (MTPN-IMI). In MTPN-IMI, a local spatio-temporal graph (LSTG) is constructed to model local agent-agent interaction. Furthermore, a Causal Convolution Module (CCM) and a Causal Self-Attention Module (CSAM) are introduced to focus on local and global temporal dependence in historical motion feature and local agent-agent interaction feature. Moreover, a Cross Attention Module (CAM) is utilized to capture the inherent association between multiple modes. Experiments show that our model outperforms related models on Argoverse1.1 validation set, achieving superior prediction accuracy.

源语言英语
主期刊名Proceedings of 2025 Chinese Intelligent Systems Conference
编辑Yingmin Jia, Yang Liu, Weicun Zhang, Yongling Fu
出版商Springer Science and Business Media Deutschland GmbH
433-441
页数9
ISBN(印刷版)9789819565528
DOI
出版状态已出版 - 2026
活动21st Chinese Intelligent Systems Conference, CISC 2025 - Beijing, 中国
期限: 25 10月 202526 10月 2025

出版系列

姓名Lecture Notes in Electrical Engineering
1549 LNEE
ISSN(印刷版)1876-1100
ISSN(电子版)1876-1119

会议

会议21st Chinese Intelligent Systems Conference, CISC 2025
国家/地区中国
Beijing
时期25/10/2526/10/25

学术指纹

探究 'Multi-modal Trajectory Prediction Network that Integrates Historical Motion and Spatio-Temporal Interaction' 的科研主题。它们共同构成独一无二的学术指纹。

引用此