TY - JOUR
T1 - Embodied Scene Rearrangement Planning
AU - Chen, Canzhi
AU - Wang, Zan
AU - Zhu, Siqi
AU - Wu, Qi
AU - Li, Yixuan
AU - Liang, Wei
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2026/11/1
Y1 - 2026/11/1
N2 - This paper introduces Embodied Scene Rearrangement Planning (ESRP), a novel task requiring embodied agents to rearrange furniture in 3D scenes to match a target configuration using only egocentric observations and a top-down target layout. Unlike prior rearrangement tasks, Embodied Scene Rearrangement Planning (ESRP) precludes global state access and introduces mutual object occlusions, reflecting the practical constraints of real-world robotic deployment. These factors make aligning partial egocentric observations with the global target layout particularly challenging for long-horizon planning. To facilitate research, we present ESRP-Bench, a comprehensive benchmark built on OmniGibson featuring over 5,400 scene pairs and 8,200 objects. We define three multi-level metrics to evaluate rearrangement quality and provide four baselines: a hierarchical task-and-motion planning method, a vision-language-model-based method, and two learning-based approaches (IL and RL). Experimental results demonstrate that current methods struggle to complete the task efficiently, highlighting ESRP as a challenging frontier for embodied agents in scene understanding and long-horizon task planning. This work serves as a stepping stone toward deploying intelligent agents in real-world scenarios.
AB - This paper introduces Embodied Scene Rearrangement Planning (ESRP), a novel task requiring embodied agents to rearrange furniture in 3D scenes to match a target configuration using only egocentric observations and a top-down target layout. Unlike prior rearrangement tasks, Embodied Scene Rearrangement Planning (ESRP) precludes global state access and introduces mutual object occlusions, reflecting the practical constraints of real-world robotic deployment. These factors make aligning partial egocentric observations with the global target layout particularly challenging for long-horizon planning. To facilitate research, we present ESRP-Bench, a comprehensive benchmark built on OmniGibson featuring over 5,400 scene pairs and 8,200 objects. We define three multi-level metrics to evaluate rearrangement quality and provide four baselines: a hierarchical task-and-motion planning method, a vision-language-model-based method, and two learning-based approaches (IL and RL). Experimental results demonstrate that current methods struggle to complete the task efficiently, highlighting ESRP as a challenging frontier for embodied agents in scene understanding and long-horizon task planning. This work serves as a stepping stone toward deploying intelligent agents in real-world scenarios.
KW - Embodied AI
KW - mobile manipulation
KW - scene rearrangement planning
KW - semantic scene understanding
UR - https://www.scopus.com/pages/publications/105050161698
U2 - 10.1109/LRA.2026.3728329
DO - 10.1109/LRA.2026.3728329
M3 - Article
AN - SCOPUS:105050161698
SN - 2377-3766
VL - 11
SP - 12839
EP - 12846
JO - IEEE Robotics and Automation Letters
JF - IEEE Robotics and Automation Letters
IS - 11
ER -