Multi-Agent Systems: Beyond the Warehouse
A full-day workshop at IROS2026: September 27th (Sunday)
This workshop is CONFIRMED
While large-scale multi-agent systems (MAS) thrive in structured spaces like automated warehouses, current control and planning frameworks rely heavily on predictability, grid worlds, and robot homogeneity. To step into fields like autonomous agriculture or heavy construction, massive robotic collectives must leave the lab and handle unstructured, dynamic environments.
This workshop serves as a forum to define this transition. We aim to discuss gaps between foundational MAS theory and real-world deployment by bringing together top minds in planning, control, and learning alongside field-robotics practitioners to pioneer the deployment of robotic collectives into the wild.
This workshop operates as an interactive and collaborative engine built around applied talks, interactive sessions, and breakout rooms. The insights generated throughout the day will be synthesized into a post-workshop roadmap published right here to help steer the future of field-ready multi-robot systems.
To contact the organizers, email gmgutow@mtu.edu and include MASBW26 in the subject line.
Extended Abstract Submission
Opens: July 1st
Closes: August 15th, AOE Time
Acceptance Notification
August 29th (tentative)
"Demos from the Field" videos
Due: September 14th, AOE Time (tentative)
Let's meet at IROS 2026!
September 27-October 1
The Workshop
September 27th, 8:30AM-5:30PM
Talk: Scalable Planning and Asynchronous Execution for Long-Horizon Multi-Robot Manipulation
Jiaoyang Li is an Assistant Professor in the Robotics Institute at Carnegie Mellon University. Her research develops heuristic search algorithms for multi-agent motion planning and coordination, including multi-agent path finding (MAPF), and combines search with machine learning. Her group's work on scalable imitation learning for lifelong MAPF won the ICRA 2025 Best Student Paper Award and Best Paper Award on Multi-Robot Systems, and she has recently worked on AI for collaborative autonomy and multi-robot systems in manufacturing. She completed her PhD at the University of Southern California.
Website:
Talk: High-Dimensional Multi-Agent Robot Learning
Guillaume Sartoretti joined the Mechanical Engineering Department at the National University of Singapore (NUS) as an Assistant Professor in 2019, where he founded the Multi-Agent Robotic Motion (MARMot) lab. Before that, he was a Postdoctoral Fellow in the Robotics Institute at Carnegie Mellon University (USA), where he worked with Prof. Howie Choset. He received his Ph.D. in robotics from EPFL (Switzerland) in 2016 for his dissertation on "Control of Agent Swarms in Random Environments," under the supervision of Prof. Max-Olivier Hongler. His passion and research lie in understanding and eliciting emergent coordination/cooperation in large multi-agent systems, by identifying what information and mechanisms can help agents reason about their individual role/contribution to each other and to the team. Guillaume was a Manufacturing Futures Initiative (MFI) postdoctoral fellow at CMU in 2018-2019, was awarded an Amazon Research Awards in 2022, as well as an Outstanding Early Career Award from NUS' College of Design and Engineering in 2023. His group's work received best paper awards at DARS in 2021, 2022, and two best paper awards at ICRA 2025.
Website:
Talk: Deployable Aerial Swarms: From Safe and Efficient Coordination to Language-Based Interaction
SiQi Zhou is an Assistant Professor in the School of Computing Science at Simon Fraser University, where her research lies at the intersection of robotics, machine learning, and control, with the goal of enabling robots to operate safely in unstructured environments. She completed her PhD in robotics at the University of Toronto Institute for Aerospace Studies and was previously a senior scientist in the Learning Systems and Robotics Lab at the Technical University of Munich. Some of her work in multi-robot systems, such as SwarmGPT and AMSwarm, explores using both control theory and large language models (LLM) to coordinate a swarm of aerial robots.
Website: https://siqizhou.com/
Talk: Towards High Performance AI-Enabled Robot Manipulation
Stefan Schaal is Director of Science, Research, Robotics, and AI at Intrinsic, an Alphabet company, where his work spans robotics and machine learning for industrial automation, including recent work with Google DeepMind on learning multi-robot coordination. He previously served as Intrinsic's Chief Science Officer and as Director of Robotics at X, the moonshot factory. He was the founding director of the Autonomous Motion Department at the Max Planck Institute for Intelligent Systems and a professor of computer science, neuroscience, and biomedical engineering at the University of Southern California for over two decades
Website: https://stefan-schaal.net/
Talk: Scaling Multi-Agent Systems for the dynamic Outdoors
Dr. Bae is an assistant professor in Mechanical and Aerospace Engineering and the College of Computing at Michigan Technological University. Prior to this appointment in 2019, she was a research professor for Intelligent Systems and Robotics Laboratory at Korea University in Seoul, South Korea from 2014 to 2019. She received her BS and MS in Mechanical Engineering at Hongik University in Seoul, South Korea. She was a research assistant for the Division for Applied Robot Technology in Korea Institute of Industrial Technology (KITECH) in Ansan, South Korea in 2007 for a year. She earned her PhD in Mechanical Engineering at Texas A&M University in College Station, Texas. Dr. Bae is interested in Robotics, especially with Multi-robot systems. Her research interests includes Coordination of Heterogeneous Robot Teams, Vehicle Routing Problems, Multi-robot System Control and Optimization, and Autonomous Navigation. Her main goal of research is development of operational strategy for multi-agent autonomous vehicle systems.
Talk: Multi-Agent Space Systems: Challenges and Opportunities
Dr. Manoranjan Majji earned his Ph. D. in Aerospace Engineering from Texas A&M University in 2009. He teaches graduate and undergraduate courses in mechanics, control, systems analysis, estimation of dynamical systems and astrodynamics. He developed an undergraduate aerospace engineering mechanics course, developed or co-developed five new graduate classes in the college of engineering. He received the Milton Plesur award for excellence in teaching from the University at Buffalo, State University of New York. Dr. Majji’s research focuses on computational vision, alternative positioning and navigation systems, tensegrity systems and aerospace robotics. His fundamental contributions are documented in over 180 publications (including 49 journal articles). He has chaired or co- chaired twelve doctoral committees, eighteen masters committees and served on twenty eight other graduate committees. His graduates are leading research in major corporations, startups, national laboratories, and universities. Dr. Majji received the New Investigator Award from the National Geospatial Intelligence Agency in 2014, Dean of Engineering Excellence award (Assistant Professor) in 2021, and the TEES Faculty Fellow award in 2023. He is a Fellow of the American Astronautical Society (AAS).
Website: https://lasr.tamu.edu/
Talk: From Beaver Canals to Robot Earthworks: Environment-Mediated Coordination in Multi-Agent Systems
Dr. Degani is an Associate Professor at the Faculty of Civil and Environmental Engineering at the Technion – Israel Institute of Technology. He heads the CEAR Lab – Civil, Environmental, and Agricultural Robotics Lab which focuses on dynamic locomotion, minimalism and autonomous systems in unstructured environments with applications such as search and rescue and agriculture. Before coming to the Technion Dr. Degani did his MSc, PhD and Postdoc at the Robotics Institute at Carnegie Mellon University, Pittsburgh PA. At the Robotics Institute he worked in the Manipulation Lab and the Bio-Robotics lab, supervised by Howie Choset and Matt Mason. Dr. Degani's main interests are Locomotion and Manipulation (mainly control and planning of dynamic moving mechanisms), Kinematics (singularity analysis of parallel mechanisms), Mechanical design (medical mechanisms), other medical applications (knee pathology classifications) and more…
Website: https://cear.net.technion.ac.il/
Our schedule pairs expert invited talks and poster sessions with two collaborative breakout rooms blocks. The morning breakout focuses on mapping out the open technical challenges of deploying MAS in unstructured environments (e.g., capability heterogeneity, communication degradation). The afternoon breakout translates these challenges into concrete community needs (e.g., datasets, benchmarks), directly seeding our concluding expert synthesis panel.
Time Event
08:30-08:45 Welcome and Introduction-Organizing Committee
08:45-09:10 Invited Talk: Jiaoyang Li
Scalable Planning and Asynchronous Execution for Long-Horizon
Multi-Robot Manipulation
09:10-09:35 Invited Talk: Guillaume Sartoretti
High-Dimensional Multi-Agent Robot Learning
09:35-10:00 Invited Talk: Siqi Zhou
Deployable Aerial Swarms: From Safe and Efficient Coordination
to Language-Based Interaction
10:00-10:30 Poster Lightning Talks
10:30-11:00 Morning coffee break & Poster Session
11:00-11:40 Breakout Rooms:
Brainstorming Guiding Challenges for Beyond the Warehouse
11:40-12:05 Invited Talk: Stefan Schaal
Towards High Performance AI-Enabled Robot Manipulation
12:05-12:30 Invited Talk: Jungyun Bae
Scaling Multi-Agent Systems for the dynamic Outdoors
12:30-13:30 Lunch break
13:30-13:55 Invited Talk: Manoranjan Majji
Multi-Agent Space Systems: Challenges and Opportunities
13:55-14:20 Invited Talk: Amir Degani
From Beaver Canals to Robot Earthworks:
Environment-Mediated Coordination in Multi-Agent Systems
14:20-14:45 Invited Talk: To be announced
14:45-15:30 Breakout Rooms:
Community Needs to Tackle Guiding Challenges
15:30-16:00 Afternoon Coffee break & Poster Session
16:00-16:30 Conclusions and insights from Breakouts
16:30-17:00 Invited Speaker Panel
Response to Breakout rooms
17:00-17:30 Best Poster Award, Feedback, and Open Forum
17:30 Adjourn
We invite the community to submit 2-4 page extended abstracts relevant to multi-agent systems, broadly interpreted. The focus should be on capabilities (algorithmic or otherwise) necessary to move multi-agent systems beyond highly structured environments such as warehouse work, and/or on application domains that demand multi-agent systems. We invite early stage work, position papers, recent successes, and new ideas in need of stress testing. Students and early-career researchers are strongly encouraged to submit. Abstracts will be lightly reviewed by the workshop organizers for relevance and quality. Selected abstracts will be archived on this site and be invited to give a lightning talk and present a poster during the workshop. A Best Poster Award will be made.
In addition to their talk and poster, for accepted abstracts we invite submission of 30 second "Demos from the Field" videos which will be publicized on social media and the workshop website prior to the event to draw interest.
Submission Format: A single PDF file (no more than 4 pages including references) formatted following the IEEE conference template; available at https://www.ieee.org/conferences/publishing/templates
Submission Link: https://openreview.net/group?id=IEEE.org/IROS/2026/Workshop/MAS-BW
Submission Opens: 1st July 2026
Submission Closes: 15th August 2026, AOE time
Acceptance Notification: 29th August 2026 (tentative)
Demos from the Field videos due: 14th September 2026, AOE time (tentative)
Dr. Geordan Gutow
Assistant Professor, Mechanical and Aerospace Engineering
Michigan Technological University
Dr. Gutow joined Michigan Tech in Fall 2025. His research focuses on developing uncertainty-aware optimization-based planners, estimators, and controllers that allow robots to reason about the properties and capabilities of their embodiments. Such problems are made tractable by finding ways to exploit structure in the underlying mathematics, particularly precomputation, parallelism, upper and lower bounds, and convexity.
Website: https://www.mtu.edu/mechanical-aerospace/people/faculty/gutow/
Email: gmgutow@mtu.edu
PhD Student in Robotics
Carnegie Mellon University
Philip Huang is a PhD student in the Robotics Institute at Carnegie Mellon University, advised by Jiaoyang Li. His research centers on multi-robot-arm coordination and task-and-motion planning (TAMP), with a focus on applications in robotic assembly. He has previously worked as a research intern at NVIDIA's Seattle Robotics Lab. Philip holds a Master's degree from the University of Toronto.
Website: https://philip-huang.com
PhD Student in Robotics
Carnegie Mellon University
Anoop's research deals with traveling salesman problems (TSP) and vehicle routing problems (VRP) involving moving targets and obstacles.
Postdoctoral researcher
Civil, Environmental, and Agricultural Robotics Lab
Technion – Israel Institute of Technology
Federico Oliva is a postdoctoral researcher at the Civil, Environmental, and Agricultural Robotics Lab (CEAR) at the Technion – Israel Institute of Technology, focusing on multi-agent active localization and robotic aggregate formation techniques. He holds a Master's in Automation Engineering from Alma Mater Studiorum – Università di Bologna (2017-2020), where he graduated with honors. He completed his Ph.D. at Università di Roma Tor Vergata (2020-2023) in Computer Science, Control, and Geoinformation, focusing on nonlinear observers and system identification.
Github: https://github.com/fedeoli