Active Spatial Inspection for Effective and Efficient Embodied Exploration

📅 2026-09-17
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
本文针对具身探索中任务成功率和效率问题,提出了一种基于空间检查的ACE框架,通过证据感知与暴露信息移动相结合的方法来提高导航任务的成功率和探索效率。
📝 Abstract
Achieving high task success and efficiency remains a central pursuit in embodied exploration. Existing frameworks typically guide agent behavior through a spatially coarse and indirect assessment of suggestive cues and directions, yet such designs may struggle to judge cue sufficiency and the need for further inspection, leading to early termination or excessive continuation and ultimately reducing task success and efficiency. This work rethinks embodied exploration from a spatially explicit standpoint and introduces the spatial-inspection-guided ACtive Exploration (ACE) framework coupling evidence-grounded perception with exposure-informed movement and establishing a spatially resolved decision paradigm. Evidence-grounded perception strengthens inferential validity by integrating granular localization with focused verification, turning suggestive cues into precise decisive visual support. Exposure-informed movement promotes directional judgment through prospective prioritization and retrospective suppression, selecting promising directions for efficient spatial progress. ACE alleviates the tension between preventing early termination and avoiding unnecessary continuation. Extensive experiments demonstrate that ACE achieves 18.0% higher navigation task success and 10.3% higher exploration efficiency for question answering than prior state-of-the-art baselines, advancing effective and efficient embodied exploration.
Problem

Research questions and friction points this paper is trying to address.

embodied exploration
spatial inspection
task success
efficiency
cue sufficiency
Innovation

Methods, ideas, or system contributions that make the work stand out.

spatial-inspection-guided
evidence-grounded perception
exposure-informed movement
embodied exploration
🔎 Similar Papers
No similar papers found.
💼 Related Jobs
No related jobs found.
W
Wenbin Wang
The Hong Kong Polytechnic University and China Three Gorges University
Xiang Bai
Xiang Bai
Huazhong University of Science and Technology (HUST)
Computer VisionOCR
Y
Yizhao Wang
The Hong Kong Polytechnic University
H
Hang Sun
China Three Gorges University
D
Dong Ren
China Three Gorges University
Jie Qin
Jie Qin
Professor, Nanjing University of Aeronautics and Astronautics
Computer VisionMachine LearningPattern Recognition
Q
Qingquan Li
Shenzhen University
Bing Wang
Bing Wang
Professor of Optics, Huazhong University of Science and Technology
PlasmonicsNanophotonics