Towards Next-Gen Mobile Manipulation: Modularity, Reconfiguration, and Generalisation

Workshop Day: 20th/October/2025 | Monday | Time: 2 pm to 6 pm | Room 401

Abstract

In modern industries and service sectors, there is a growing demand for general purpose
mobile manipulators that can perform diverse tasks in multi-stage workflows, handle a
wide range of goods and materials in various shapes, sizes and weights, and operate in
dynamic and unstructured environments. However, the current mobile manipulation
systems are often task specific, limited in generalization and constrained by fixed
embodiments.
To elaborate, embodiment such as humanoids shown advances in generalization but are
limited by payload capacity, efficiency and speed, whereas task specific robots are
efficient but lack flexibility to handle diverse tasks, objects and environments. However,
neither of them fully meets the demand for adaptable, scalable, and workflow-driven
automation.
This workshop addresses and explores an emerging paradigm of modular and
reconfigurable mobile manipulation, capable of adapting physically and behaviorally (e.g.
swapping end-effectors, attaching mobile bases, changing embodiment, reconfiguring
arms) and generalizing across multiple tasks (e.g., multistage workflows in warehouses
such as unloading, palletizing, sorting, shelf stocking and retrieval, depalletizing,
unloading). A paradigm that integrates mobility, manipulation, and adaptive
hardware/software architectures to achieve task-variability and generalization in diverse
workflows.
By discussing robot design, control architectures, multi-modal sensing, multi-task
adaptation, skill learning, AI-driven generalization and real-world scalability, this
workshop aims to bring out the latest advancements, draw inspiration and lay out the
future of mobile manipulation beyond humanoids and task specific robots.
This workshop brings together researchers, experts in mobile manipulation, robot learning,
humanoids and industry practitioners to discuss the latest advancements, challenges,
evolving industry needs and future directions for next-gen mobile manipulation systems
and solutions that could become mainstream and transform automation across industries.

Scope

• Modular Hardware Architectures for Mobile Manipulation
  Design and integration of reconfigurable arms, end-effectors, mobile bases, and embodiment systems for task adaptability.
• AI-Driven Task and Workflow Generalization
  Learning and planning approaches that enable mobile manipulators to perform multiple tasks across dynamic, multi-stage workflows.
• Multi-Modal Sensing for Robust Perception and Interaction
  Fusion of RGBD, tactile, event-based, and proprioceptive sensing for reliable operation in unstructured environments.
• Skill Learning and Transfer in Mobile Manipulation
  Approaches to acquiring, transferring, and generalizing manipulation skills using deep learning, reinforcement learning, and imitation learning.
• Control and Planning for Reconfigurable Systems
  Adaptive control architectures, behavior switching, and real-time motion planning for changing robot configurations and tasks.
• Human-Robot Collaboration and Shared Autonomy
  Interactive systems where humans guide, supervise, or collaborate with mobile manipulators in semi-structured settings.
• Scalability and Deployment in Real-World Environments
  Case studies, challenges, and solutions for deploying modular mobile manipulators in logistics, warehouses, and service applications.
• Benchmarking and Evaluation of General-Purpose Manipulation
  Metrics, testbeds, and standardized tasks to evaluate generalization, adaptability, robustness, and performance of mobile manipulation systems.

Time	Session	Duration
1:00 – 2:00 PM	DFL Sponsored Lunch + Networking	60 min
2:00 – 2:10 PM	Dr. Rajkumar Muthusamy — Introduction to the Workshop	10 min
2:10 – 2:35 PM	Speaker 1: Dr. Maximo Roa — Novel End-Effector Designs for Generalizable Manipulation	20 + 5 min Q&A
2:35 – 3:00 PM	Speaker 2: Prof. Kensuke Harada — Robotic Gripper Design for Reconfiguration, and Generalization	20 + 5 min Q&A
3:00 – 3:25 PM	Speaker 3: Prof. Chao Chen — Enabling Efficient Mobile Manipulation with Base Placement and Operational Part Detection	20 + 5 min Q&A
3:25 – 4:10 PM	Poster Session & Coffee Break	45 min
4:10 – 4:35 PM	Speaker 4: Prof. Guyue Zhou — Fast Deployment for Mobile Manipulation in General Scenarios	20 + 5 min Q&A
4:35 – 5:00 PM	Speaker 5: Prof. Roberto Martin-Martin — Learning to Safely Imitate a Mobile Manipulation Task from a Single Human Video	20 + 5 min Q&A
5:00 – 5:25 PM	Speaker 6: Dr. Suraj Nair — Speedcargo: Digitalizing Airfreight for an Automated Future – Robotics for Air Cargo	20 + 5 min Q&A
5:25 – 6:00 PM	Panel Discussion & Best Poster Award	35 min

Accepted Papers

The Following Papers have been accepted for Poster Presentation

1. “Human-Robot Co-Transportation using Disturbance-Aware MPC with Pose Optimisation” Al Jaber Mahmud, Amir Hossain Raj, Duc M. Nguyen, Weizi Li, Xuesu Xiao, and Xuan Wang.
2. “CubiX: Portable Wire-Driven Parallel Robot Overcoming Physical Limitations by Utilizing the Environment” Shintaro Inoue, Kento Kawaharazuka, Sota Yuzaki, Temma Suzuki, and Kei Okada.
3. “Mutual Adaptation in Human-Robot Co-Transportation with Human Preference Uncertainty” Al Jaber Mahmud, Weizi Li, and Xuan Wang.
4. “Equivariant Triplane Representation for Volumetric Grasping“ Pinhao Song, Yutong Hu, Pengteng Li, Renaud Detry.
5. “A Probabilistic Shared Control Framework for Navigation” Pinhao Song, Yurui Du, Ophelie Saussus, Sofie De Schrijver, Irene Caprara, Peter Janssen, Renaud Detry.
6. “Learning Lyapunov Functions from Data with Automatic Maximum Decay Rate Estimation” Zhe Shen.
7. “A Generalized LLM-Based Architecture for Robotic Manipulation: Self-Directing Tool Use Without Task-Specific Coding“ Xiaoqian Huang, Hussain Sajwanj, Abdulla Ayyad, Asim Khan, Rajkumar Muthusamy ,Tarek Taha, Yahya Zweiri.
8. “Learning Mobile Robot Pushing for Unwanted Object Removal in Dense Clutter” Tatsuki Nishimura, Weiwei Wan, Takuya Kiyokawa and Kensuke Harada
9. “Scalable Whole-body Trajectory Generation for Coordinated Mobile Manipulation” Zeyu Zhang, Yida Niu, Ziyuan Jiao.
10. “Offline World Model based Reinforcement Learning for Mobile Manipulation Behaviours” Xiaoxu Feng and Takato Horii.
11. “Towards Generalization: Reconfigurable End-Tool for Multiple Warehouse Tasks” Rajkumar Muthusamy, Tanzim Ahmed, Abhinav Pathak and Tarek Taha.
12. “LLM-Powered Multimodal Framework for Safe and Efficient Human-Robot Collaboration in Warehouse Shelf-Picking” Abhinav Pathak, Tarek Taha and Rajkumar Muthusamy.

Call for Papers

We welcome authors to submit extended abstracts (2-4 pages excluding references) showcasing relevant work on workshop topics, including field-collected dataset analysis, robot technologies and platforms, interactive demonstrations, and innovative, creative, and futuristic solutions within the workshop context.

• Format: Submissions should be in IROS conference standard PDF format.
• Page Limit: Maximum of 2-4 pages (excluding references).
• Review Process: All submissions will undergo peer review based on relevance, technical quality, and ethics.
• Presentation Opportunities: Accepted papers will be presented as posters, with some selected for spotlight talks during the workshop.
• Publication: Accepted papers will be available on the workshop website.

Submission Timelines

**Submission deadline: 20th September 2025 28th/September/2025**

**Notification of acceptance: 22nd September 2025 1st/October/2025**

Workshop Day: 20th/October/2025 | Monday | Time: 2 pm to 6 pm | Room 401

**Please send your contribution to: rajkumar.muthusamy@dubaifuture.gov.ae (Subject : IROS2025-MoMA4 Workshop Contribution | Corresponding Author Name | University/Industry Name)**

Awards

• Best Poster Award: 500$ USD

Sponsors