Mobile Manipulation: Emerging Opportunities & Contemporary Challenges

Introduction

Recent advances in large-scale robot data have opened up possibilities for developing general-purpose policies for tabletop manipulation, while progress in universal, cross-embodiment end-to-end navigation has significantly advanced mobility. This convergence naturally motivates research into mobile manipulation (MoMA), an interdisciplinary field that integrates mobility with dexterity to enable robots to operate in large, dynamic, and human-centric environments. Compared to tabletop setups, mobile manipulation presents distinct challenges, including hardware design trade-offs (e.g., bimanual arms, torso structures, wheeled or legged bases), low-level control complexities for whole-body coordination, and scalability challenges in real-world data collection and skill learning. The field also raises fundamental questions about sim-to-real transfer, egocentric vision, and long-horizon reasoning. Furthermore, mobile manipulators introduce new dimensions to human-robot interaction, requiring safety and adaptability in extended workspaces. Addressing these challenges will unlock the full potential of mobile manipulators across unstructured environments such as homes, hospitality, logistics, gastronomy, retail, and agriculture, driving progress toward general-purpose, intelligent robots. Key topics include:

• Tasks and New Challenges: What are the critical tasks and application domains for mobile manipulation that should be prioritized? What additional challenges does mobile manipulation present compared to a tabletop setup?
• Hardware Design: What is the optimal form factor for a mobile manipulator? Should we prioritize anthropomorphic designs or appliance-like robots? Specifically, how should we decide between bimanual arms, torso structures, wheeled bases, or legged systems?
• Low-level Control: Controlling a robot with many DoFs is a significant challenge. How can we effectively integrate planning and control for executing low-level motions and whole-body coordination?
• Egocentric Perception and Scene Representation: What new challenges arise in mobile manipulation with moving cameras? How does it impact the formulation of action space and visual representations in the 3D scenes?
• Real-world Data Scaling: Can advanced human-robot interfaces enhance teleoperation and motion capture systems for demonstration collection? How can action-free videos facilitate skill learning in the mobile manipulation domain?
• Sim-to-Real Transfer: How can simulation accelerate learning for mobile manipulation while ensuring effective real-world deployment? Can sim-to-real techniques developed for navigation be adapted to mobile manipulation?
• Long-horizon Planning and Reasoning: How can mobile manipulators efficiently plan and execute long-horizon tasks that require multiple sequential interactions and reasoning? How can large pre-trained models (e.g., LLM, VLM, VLA, World Models) facilitate new capabilities of mobile robots?
• Human-Robot Interaction: Mobile manipulators work in the extended workspace. What new challenges and opportunities does it pose and what safety considerations should be addressed in human-centric environments?

Call for Papers

We are excited to announce the Call for Papers for the RSS MoMA workshop. We invite original contributions presenting novel ideas, research, and applications relevant to the workshop’s theme.

Important Dates

Submission Guidelines

• Page Limit: There are no page length requirements, but we suggest a length around 4-9 pages long. There is no limit on the number of pages for references or appendices.
• Formatting: Submissions are encouraged to use the RSS 2025, IROS 2025, CoRL 2025, NeurIPS 2025 templates.
• Anonymity: All submissions must be anonymized. Please remove any author names, affiliations, or identifying information.
• Relevant Work: We welcome references to recently published, relevant work (e.g., RSS, CoRL, ICRA, and CVPR).
• Archival Status: All accepted papers are non-archival. Submitting to multiple non-archival workshops is permitted.
• Link: openreview submission

Accepted papers will be presented in the form of posters at the workshop. In addition, selected papers may be invited to deliver spotlight talks.

Paper topics

A non-exhaustive list of relevant topics:

• Benchmarks for Mobile Manipulation
• Novel Designs of Mobile Manipulator Hardware
• Low-Level, Whole-Body Control for Mobile Manipulation
• Representations for Mobile Manipulation
• Data Collection for Learning Mobile Manipulation
• Mobile Manipulation Simulations and Sim-to-Real Transfer
• Reasoning and Planning Algorithms for Mobile Manipulation
• Human-Robot Interaction in Mobile Manipulation

Invited Speakers

listed alphabetically

Jeannette Bohg

Stanford University

Richard Cheng

Toyota Research Institute

Roberto Martín-Martín

The University of Texas at Austin

Lerrel Pinto

New York University

Organizers

Jiafei Duan

University of Washington

Yunfan Jiang

Stanford University

Arjun Gupta

University of Illinois at Urbana-Champaign

Ainaz Eftekhar

University of Washington

Huihan Liu

The University of Texas at Austin

Jingyun Yang

Stanford University

Nur Muhammad "Mahi" Shafiullah

New York University

Ruohan Zhang

Stanford University

Rutav Shah

The University of Texas at Austin

Haoyu Xiong

Stanford University

Kuo-Hao Zeng

Allen Institute for AI

Advisory Board

Yuchen Cui

University of California, Los Angeles

Yuke Zhu

The University of Texas at Austin

Fei-Fei Li

Stanford University