Human2LocoMan: Learning Versatile Quadrupedal Manipulation with Human Pretraining
Yaru Niu, Yunzhe Zhang, Mingyang Yu, Changyi Lin, Chenhao Li, Yikai Wang, Yuxiang Yang, Wenhao Yu, Tingnan Zhang, Zhenzhen Li, Jonathan Francis, Bingqing Chen, Jie Tan, Ding Xuan Zhao · 2025
Robot Data Finetuning Human Data Pretraining Policy RolloutFig. 1: Human2LocoMan provides a unified framework for collecting human demonstrations and teleoperated robot wholebody motions, enabling flexible and scalable data collection.Human data is used for cross-embodiment model pretraining, while robot data is leveraged for policy finetuning.Human2LocoMan achieves positive transfer from human to quadrupedal embodiments, facilitating versatile quadrupedal manipulation.Abstract-Quadrupedal robots have demonstrated impressive locomotion capabilities in complex environments, but equipping them with autonomous versatile manipulation skills in a scalable way remains a significant challenge.In this work, we introduce a system that integrates data collection and imitation learning from both humans and LocoMan, a quadrupedal robot with multiple manipulation modes.Specifically, we introduce a teleoperation and data collection pipeline, supported by dedicated hardware, which unifies and modularizes the observation and action spaces of the human and the robot.To effectively leverage the collected data, we propose an efficient learning architecture that supports co-training and pretraining with multimodal data across different embodiments.Additionally, we construct the first manipulation dataset for the LocoMan robot, covering various household tasks in both unimanual and bimanual modes, supplemented by a corresponding human dataset.Experimental results demonstrate that our data collection and training framework significantly improves the efficiency and effectiveness of imitation learn-*Authors contributed equally to this work.ing, enabling more versatile quadrupedal manipulation capabilities.Our hardware,