arXiv：cs.LG（机器学习，全量分类）· Morgan Byrd, Donghoon Baek, Kartik Garg, Hyunyoung Jung, Daesol Cho, Maks Sorokin, Robert Wright, Sehoon Ha·· 15 小时前AI 评分41
AdaptManip：用在线循环状态估计实现人形机器人自适应全身物体举升与配送
AdaptManip: Learning Adaptive Whole-Body Object Lifting and Delivery with Online Recurrent State Estimation
AI 导读
人形机器人框架 AdaptManip 通过强化学习训练全身 loco-manipulation 策略，无需人类演示或遥操作数据。该框架由循环物体状态估计器、全身基础策略加残差操作控制、以及基于 LiDAR 的全局定位估计器三部分组成，全部在仿真中训练并零样本部署到真实硬件。实验显示其在适应性和整体成功率上显著优于模仿学习基线，并在真实人形机器人上完成全自主导航、举升与配送。
正文
Abstract:This paper presents Adaptive Whole-body Loco-Manipulation, AdaptManip, a fully autonomous framework for humanoid robots to perform integrated navigation, object lifting, and delivery. Unlike prior imitation learning-based approaches that rely on human demonstrations and are often brittle to disturbances, AdaptManip aims to train a robust loco-manipulation policy via reinforcement learning without human demonstrations or teleoperation data. The proposed framework consists of three coupled components: (1) a recurrent object state estimator that tracks the manipulated object in real time under limited field-of-view and occlusions; (2) a whole-body base policy for robust locomotion with residual manipulation control for stable object lifting and delivery; and (3) a LiDAR-based robot global position estimator that provides drift-robust localization. All components are trained in simulation using reinforcement learning and deployed on real hardware in a zero-shot manner. Experimental results show that AdaptManip significantly outperforms baseline methods, including imitation learning-based approaches, in adaptability and overall success rate, while the learned estimator keeps tracking the object when visual observations are intermittent. We further demonstrate fully autonomous real-world navigation, object lifting, and delivery on a humanoid robot.
| Comments: | Website: this https URL |
| Subjects: | Robotics (cs.RO); Machine Learning (cs.LG) |
| Cite as: | arXiv:2602.14363 [cs.RO] |
| (or arXiv:2602.14363v2 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2602.14363 arXiv-issued DOI via DataCite |
Submission history
From: Morgan Byrd [view email]
[v1]
Mon, 16 Feb 2026 00:29:53 UTC (2,966 KB)
[v2]
Wed, 30 Sep 2026 21:04:17 UTC (2,968 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org