HOZ plus plus : Versatile Hierarchical Object-to-Zone Graph for Object Navigation

doi:10.1109/TPAMI.2025.3552987

	HOZ plus plus : Versatile Hierarchical Object-to-Zone Graph for Object Navigation
	Zhang, Sixian 1,2; Song, Xinhang 1,2; Yu, Xinyao 1,2; Bai, Yubing 1,2; Guo, Xinlong 1,2; Li, Weijie 1,2; Jiang, Shuqiang 1,2
	2025-07-01
发表期刊	IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
ISSN	0162-8828
卷号	47 期号:7 页码:5958-5975
摘要	The goal of object navigation task is to reach the expected objects using visual information in unseen environments. Previous works typically implement deep models as agents that are trained to predict actions based on visual observations. Despite extensive training, agents often fail to make wise decisions when navigating in unseen environments toward invisible targets. In contrast, humans demonstrate a remarkable talent to navigate toward targets even in unseen environments. This superior capability is attributed to the cognitive map in the hippocampus, which enables humans to recall past experiences in similar situations and anticipate future occurrences during navigation. It is also dynamically updated with new observations from unseen environments. The cognitive map equips humans with a wealth of prior knowledge, significantly enhancing their navigation capabilities. Inspired by human navigation mechanisms, we propose the Hierarchical Object-to-Zone (HOZ++) graph, which encapsulates the regularities among objects, zones, and scenes. The HOZ++ graph helps the agent to identify the current zone and the target zone, and computes an optimal path between them, then selects the next zone along the path as the guidance for the agent. Moreover, the HOZ++ graph continuously updates based on real-time observations in new environments, thereby enhancing its adaptability to new environments. Our HOZ++ graph is versatile and can be integrated into existing methods, including end-to-end RL and modular methods. Our method is evaluated across four simulators, including AI2-THOR, RoboTHOR, Gibson, and Matterport 3D. Additionally, we build a realistic environment to evaluate our method in the real world. Experimental results demonstrate the effectiveness and efficiency of our proposed method.
关键词	Navigation Training Visualization Layout Artificial intelligence Semantics Planning Periodic structures Location awareness Three-dimensional displays Embodied AI visual navigation object goal navigation hierarchical knowledge graph
DOI	10.1109/TPAMI.2025.3552987
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[62125207] ; National Natural Science Foundation of China[62032022] ; National Natural Science Foundation of China[62272443] ; National Natural Science Foundation of China[U23B2012] ; Beijing Natural Science Foundation[JQ22012] ; Beijing Natural Science Foundation[L242020]
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:001504146900038
出版者	IEEE COMPUTER SOC
引用统计
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/42344
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Jiang, Shuqiang
作者单位	1.Chinese Acad Sci, Inst Comp Technol, Key Lab Intelligent Informat Proc, Beijing 100190, Peoples R China 2.Univ Chinese Acad Sci, Beijing 100049, Peoples R China
推荐引用方式 GB/T 7714	Zhang, Sixian,Song, Xinhang,Yu, Xinyao,et al. HOZ plus plus : Versatile Hierarchical Object-to-Zone Graph for Object Navigation[J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,2025,47(7):5958-5975.
APA	Zhang, Sixian.,Song, Xinhang.,Yu, Xinyao.,Bai, Yubing.,Guo, Xinlong.,...&Jiang, Shuqiang.(2025).HOZ plus plus : Versatile Hierarchical Object-to-Zone Graph for Object Navigation.IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE,47(7),5958-5975.
MLA	Zhang, Sixian,et al."HOZ plus plus : Versatile Hierarchical Object-to-Zone Graph for Object Navigation".IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 47.7(2025):5958-5975.