[robotics-worldwide] Deadline Extension for Abstract Submission - ICRA 2013 Workshop on "Towards Fully Decentralized Multi-Robot Systems: Hardware, Software and Integration"

Antonio Franchi antonio.franchi at tuebingen.mpg.de
Mon Mar 18 15:54:21 PDT 2013

*Full-Day Workshop at ICRA 2013*
Towards Fully Decentralized Multi-Robot Systems: Hardware, Software and Integration

*Open Call for an EXTENDED ABSTRACT*
- Acceptance Notification: April 12, 2013

*Author Instructions*
- Original works as well as reviews of previous consolidated contributions are both welcomed
- Minimum length: 2 pages, maximum length: 6 pages.
- Authors must send their pdf file to icra2013mrs at tuebingen.mpg.de
- Authors are recommended to use the the IEEE RAS Manuscript Preparation Guidelines (http://ras.papercept.net/conferences/support/support.php)
- An accompanying video is optional and can also be provided as a weblink (e.g., youtube).

*Presentation of Accepted Contribution*
- Accepted contributions will be presented either in the oral or interactive form depending on the total number of accepted submissions and on the overall schedule
- The accepted abstracts  will be included in the workshop proceedings and available to download in the proceedings webpage.

- Antonio Franchi (Max Planck Institute for Biological Cybernetics, Germany)
- Paolo Robuffo Giordano (CNRS, France)
- Martin Riedel (Max Planck Institute for Biological Cybernetics, Germany)
- Johannes Lächele (Max Planck Institute for Biological Cybernetics, Germany)

*Date and Time*
Monday, May 6th, 2013, 8:30 am--6:15pm

ICRA 2013 in Karlsrhue, Germany,


The program will include both invited talks and contributions from an open call for extended abstracts. Please, refer to the workshop website http://icra2013mrs.tuebingen.mpg.de/  for updates on the program.

*List of Invited Speakers*

- Vijay Kumar (University of Pennsylvania )
- Volkan Isler (University of Minnesota )
- Alcherio Martinoli (École Polytechnique Fédérale de Lausanne EPFL)
- Daniela Rus (Massachusetts Institute of Technology)
- Oskar von Stryk and Karen Petersen (Technische Universität Darmstadt)
- Gaurav Sukhatme (University of Southern California)

*List of Topics*

This workshop is supported by the IEEE RAS Networked Robots Technical Committee.

Contributions from both new theoretical advancements and/or successful experimental implementations with possible live demos are welcomed.

The workshop topics cover, but are not limited to:
- multi-robot systems
- software architectures for many robots
- mobile and aerial robotics
- robotics middleware and testbeds
- distributed multi-agent control
- mobile sensor swarms
- robustness to hardware/system failures
- scalability
- heterogeneous vs. homogeneous platforms
- communication reliability
- decentralized communication infrastructures
- mobile ad-hoc networks
- code deployment and debugging for multiple robots
- human/multi-robot interaction and cooperation
- real-time multi-robot simulation
- hardware robotic frameworks
- modular robotics

*Motivation and objectives*

Multi-robot systems are a major research topic in robotics due to their intrinsic robustness, versatility, and potential pervasiveness. Designing, testing and deploying in the real world a large number of robots is nowadays a concrete possibility thanks to the recent technological advances in the production of miniaturized sensors, actuators, and small- scale/embedded processors. However, actual realizations of fully decentralized multi-robot systems (i.e., combining state- of-the-art decentralized methods for localization, communication, planning, control, and computation) are rarely found in the community. Nevertheless, building a decentralized and scalable multi-robot testbed is a fundamental step for a full development and validation, and in order to actually reduce the gap between theoretical results and real applications.

Testbeds for multi-robot systems substantially differ from the ones for single robots. The effort (in terms of time and resources) spent to conceive, design, and experimentally validate a multi-robot application is considerably larger than in a single robot scenario. The same holds for assembling, deploying, and maintaining a large team of robots. Overall, research groups working in this field are required to invest a considerable amount of resources in order to obtain a successful scientific output. The main challenges related to multi-robot testbeds can be summarized as follows: 1) hardware assembly, customization, and maintenance require a more efficient approach in the management of the equipment than for a single-robot setup; 2) current robotics software does not provide a friendly way to deal with a large number of robots (e.g., for code deployment and debugging); 3) real distributed (ad-hoc) communication infrastructures are current research topics and not yet available on the market (off-the-shelf); 4) general lack of automatized solutions for the repetitive phases of a multi-robot experiment, e.g., startup, physical deployment (homing), synchronization, and shutdown; 5) standardized tools are missing for online supervision of multiple robots and for keeping track of large amounts of data coming from concurrent processes; 6) need of transparent integration of simulation software into the overall development process for enabling an easy transition to the experimental validation.

Although many different aspects of multi-robot systems have been discussed in the past, description of specific testbeds is rarely found in the literature. Research groups generally do not provide many details of their testbed setup, and their software frameworks are typically based on custom solutions that are hard to port to other setups.

In this workshop we want to provide a round table for presenting and discussing different working solutions for the aforementioned challenges. This will give the opportunity to active researches in the field to disseminate an in-depth illustration of their multi-robot hardware/software testbeds. We hope that highlighting different needs and solutions will yield to ideas for new concepts that are very general, yet powerful enough to perform a vast range of complex multi-robot experiments. Another goal of the workshop is to plant the seeds for the future development of a universal infrastructure for multi-robot systems including, e.g., an SDK, a suitable set libraries, tools, testing utilities, as well as hardware layouts, sensors, etc. A fruitful dialog between the robotics distributors and academic researchers on the topic will be also encouraged.

More information about the robotics-worldwide mailing list