Abstract

Heterogeneous multi-robot teams have the potential to perform complex tasks that require considerable coordination in unstructured environments. To fully realize this potential, however, we need to address an array of significant challenges in enabling the coordinated behavior of inherently different kinds of agents. Particularly interesting are the questions what, who, and when of multi-robot coordination. The answers to “what needs to be done?” provide insights into the nature of complex multi-task missions and how to accomplish them. The question of who, whether in team composition or in task allocation, is particularly impactful when dealing with teams of heterogeneous agents. Finally, the answer to “when to perform the tasks?” deals with realizing a conflict-free plan for the agents to perform a number of tasks, both sequentially and simultaneously.

The aim of this workshop is to bring together perspectives on how to tackle these three fundamental questions of multi-robot coordination from different communities within robotics. We wish to have the perspectives of renowned expert researchers in the field, and offer young researchers the opportunity to share their new ideas and preliminary results of ongoing research. We hope this workshop will help the community discover exciting collaborations and effective directions of future research.

Topics of Interest

• Task assignment

• Team Composition

• Task planning and scheduling for multi-robot systems

• Modeling multi-agent teams

• Abstraction and analysis of heterogeneity

• Verifiable coordination and task performance

• Resilience and adaptive behavior

• Learning for coordination

• Adversarial games

Invited Speakers

• Hadas Kress-Gazit, Cornell University

• Johnathan How, Massachusetts Institute of Technology

• Gal Kaminka, Bar Ilan University

• Kiril Solovey, Stanford University

• Vaibhav Srivastava, Michigan State University

• Dimitra Panagou, University of Michigan - tentative

Format

• Invited talks, representing different perspectives on multi-robot coordination

• Poster sessions (for all accepted contributions)

• Spotlight paper presentations (for a subset of the accepted contributions)

• A moderated panel consisting of the invited speakers and senior organizers

Call for Contributions

In order to increase the participation of young and senior researchers in the topic, along with the quality of the contributions, we invite submissions of the following types:

1. Traditional workshop submissions presenting preliminary results of ongoing research.

2. Submissions linked to already published papers coming from:

   • Recent journal papers, which have not had the chance to be discussed in a conference, (imitating the strategy of RA-L and TR-O).

   • Recent conference papers, which were presented at machine learning conferences (e.g., NIPS, ICLR, ICML).

Note that the authors must explicitly indicate the type of their submissions.

We encourage participants to submit their research in the form of a paper (6 pages maximum including references) in the IEEE conference proceedings format. Contributors are encouraged to submit their manuscripts with videos (max. duration: 3 minutes) illustrating their work.

The submitted contributions will go through a single blind review process. All the accepted contributions will be presented as posters during the interactive sessions. A subset of the accepted papers/abstracts will be presented during spotlight presentations, before the interactive sessions.

All submissions should be sent in PDF format to mrta.rss2020@gmail.com

Important Dates

• Submission deadline: TBD

• Notification of acceptance: TBD

• Workshop date: July 12, 2020

Organizers

• Harish Ravichandar, Georgia Institute of Technology

• Ragesh Ramachandran, University of Southern California

• Sonia Chernova, Georgia Institute of Technology

• Seth Hutchinson, Georgia Institute of Technology

• Gaurav Sukhatme, University of Southern California

• Vijay Kumar, University of Pennsylvania