AIRoA MoMa Dataset: A Large-Scale Hierarchical Dataset for Mobile Manipulation
Abstract
As robots transition from controlled settings to unstructured human environments, building generalist agents that can reliably follow natural language instructions remains a central challenge. Progress in robust mobile manipulation requires large-scale multimodal datasets that capture contact-rich and long-horizon tasks, yet existing resources lack synchronized force-torque sensing, hierarchical annotations, and explicit failure cases. We address this gap with the AIRoA MoMa Dataset, a large-scale real-world multimodal dataset for mobile manipulation. It includes synchronized RGB images, joint states, six-axis wrist force-torque signals, and internal robot states, together with a novel two-layer annotation schema of sub-goals and primitive actions for hierarchical learning and error analysis. The initial dataset comprises 25,469 episodes (approx. 94 hours) collected with the Human Support Robot (HSR) and is fully standardized in the LeRobot v2.1 format. By uniquely integrating mobile manipulation, contact-rich interaction, and long-horizon structure, AIRoA MoMa provides a critical benchmark for advancing the next generation of Vision-Language-Action models. The first version of our dataset is now available at https://huggingface.co/datasets/airoa-org/airoa-moma .
Keywords
Cite
@article{arxiv.2509.25032,
title = {AIRoA MoMa Dataset: A Large-Scale Hierarchical Dataset for Mobile Manipulation},
author = {Ryosuke Takanami and Petr Khrapchenkov and Shu Morikuni and Jumpei Arima and Yuta Takaba and Shunsuke Maeda and Takuya Okubo and Genki Sano and Satoshi Sekioka and Aoi Kadoya and Motonari Kambara and Naoya Nishiura and Haruto Suzuki and Takanori Yoshimoto and Koya Sakamoto and Shinnosuke Ono and Hu Yang and Daichi Yashima and Aoi Horo and Tomohiro Motoda and Kensuke Chiyoma and Hiroshi Ito and Koki Fukuda and Akihito Goto and Kazumi Morinaga and Yuya Ikeda and Riko Kawada and Masaki Yoshikawa and Norio Kosuge and Yuki Noguchi and Kei Ota and Tatsuya Matsushima and Yusuke Iwasawa and Yutaka Matsuo and Tetsuya Ogata},
journal= {arXiv preprint arXiv:2509.25032},
year = {2025}
}