Tri-HGNN: Learning triple policies fused hierarchical graph neural networks for pedestrian trajectory prediction

doi:10.1016/j.patcog.2023.109772

	Tri-HGNN: Learning triple policies fused hierarchical graph neural networks for pedestrian trajectory prediction
	Zhu, Wenjun1 ; Liu, Yanghong 1; Wang, Peng 2; Zhang, Mengyi 1; Wang, Tian 3; Yi, Yang 1
发表期刊	PATTERN RECOGNITION
ISSN	0031-3203
	2023-11-01
卷号	143 页码:11
通讯作者	Yi, Yang(yiyang@njtech.edu.cn)
摘要	In complex and dynamic urban traffic scenarios, the accurate trajectory prediction of surrounding pedes-trians with interactive behaviors plays a vital role in the self-driving system. Intrinsic factors and extrinsic factors will inevitably influence the pedestrians trajectory. Intrinsic factors such as pedestrians diversified intentions bring rich and diverse multi-modal future possibilities. Besides, extrinsic factors affecting the future trajectory are accompanied by context semantics such as interactions among pedestrians. However, most of the existing methods discuss two problems (interaction and intention) separately. Considering both two factors impact the trajectory of pedestrians, a Triple Policies Fused Hierarchical Graph Neural Networks (Tri-HGNN) is proposed to model spatial and temporal interactions and intentions among the whole scene of pedestrians at each time step and predict the multiple future trajectories. Tri-HGNN con-tains three different policies: (i) Extrinsic-level policy is used to extract spatial nodes embedding from the interaction graph of pedestrian trajectories by using the Graph Attention Network. (ii) Intrinsic-level policy adopts the Graph Convolutional Network to infer the human intention for more accurate predic-tion. Moreover, human intention is influenced by the intrinsic interaction generated among pedestrians, so we fuse the interaction features to grasp the influence of the extrinsic interaction. (iii) Basic-level pol-icy then integrates the heuristic information obtained from other two policies and concatenates it with historical trajectories to make multiple predictions through Temporal Convolution Network. Experimen-tal results show that our model improves performance compared with state-of-the-art methods on the ETH/UCY and SDD benchmarks.& COPY; 2023 Elsevier Ltd. All rights reserved.
关键词	Trajectory prediction Hierarchical policy Graph neural networks
DOI	10.1016/j.patcog.2023.109772
关键词[WOS]	ATTENTION
收录类别	SCI
语种	英语
资助项目	National Key Research and Development Program of China[2021YFB3301300] ; Natural Science Foundation of the Higher Education Institutions of Jiangsu Province of China ; Postgraduate Research amp; Practice Innovation Program of Jiangsu Province[21KJB520007] ; [KYCX23_1450]
项目资助者	National Key Research and Development Program of China ; Natural Science Foundation of the Higher Education Institutions of Jiangsu Province of China ; Postgraduate Research amp; Practice Innovation Program of Jiangsu Province
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:001039043900001
出版者	ELSEVIER SCI LTD
引用统计	被引频次：1[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/53817
专题	多模态人工智能系统全国重点实验室
通讯作者	Yi, Yang
作者单位	1.Nanjing Tech Univ, Inst Elect Engn & Control Sci, Nanjing 211816, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 3.Beihang Univ, Inst Artificial Intelligence, Beijing 100083, Peoples R China
推荐引用方式 GB/T 7714	Zhu, Wenjun,Liu, Yanghong,Wang, Peng,et al. Tri-HGNN: Learning triple policies fused hierarchical graph neural networks for pedestrian trajectory prediction[J]. PATTERN RECOGNITION,2023,143:11.
APA	Zhu, Wenjun,Liu, Yanghong,Wang, Peng,Zhang, Mengyi,Wang, Tian,&Yi, Yang.(2023).Tri-HGNN: Learning triple policies fused hierarchical graph neural networks for pedestrian trajectory prediction.PATTERN RECOGNITION,143,11.
MLA	Zhu, Wenjun,et al."Tri-HGNN: Learning triple policies fused hierarchical graph neural networks for pedestrian trajectory prediction".PATTERN RECOGNITION 143(2023):11.