Distributed Intelligent Systems
A number of natural and artificial systems can be considered as intrinsically distributed and consisting of nodes presenting a certain degree of intelligence. Typical examples of distributed intelligent systems include social insect colonies, flocks of vertebrates, multi-agent systems, transportation systems, multi-robot systems, and wireless sensor networks. The goals of this course are two-fold: first, to provide students with a sufficient mathematical and computational background to analyze distributed intelligent systems through appropriate models, and second, to illustrate several coordination strategies and show how to concretely implement and optimize them. The course is a well-balanced mixture of theory and laboratory exercises using simulation and real hardware platforms.
It involves the following topics:
- Introduction to key concepts such as self-organization and software and hardware tools used in the course.
- Examples of natural, artificial, and hybrid distributed intelligent systems.
- Modeling methods: microscopic and macroscopic, multi-level; spatial and non-spatial; mean field and stochastic approaches.
- Machine-learning methods: single- and multi-agent techniques; expensive optimization problems and noise resistance.
- Coordination strategies and distributed control: direct and indirect schemes; communication channels and cost; distributed sensing and action; performance evaluation.
Announcements and Additional Information
Please send questions and concerns to the TA mailing list:
Office hours: by appointment only. Please send your time availability and discussion subject to the TA mailing list. A TA will respond you directly and set-up an appointment for discussion.
Cyrill Baumann (head TA)
Chiara Ercolani (TA)
Faëzeh Rahbar (TA)
Anwar Quraishi (TA)
The support staff will be involved in the mantainance of the course web site, design and testing of specific exercises, and supervision of course projects.
Zeki Erden (help-TA)
First of all, we would like to acknowledge all the students attending the Swarm Intelligence course during the WS 2004-2005, WS 2005-2006, WS 2006-2007, 2007-2008, 2008-2009, and WS 2009-20010: their extremely constructive feedback has been incorporated in this substantially redesigned new edition of the course.
Second, we would like to acknowledge the EPFL-FIFO program (Fonds d’Innovation pour la FOrmation) and the CRAFT team for having supported two major tool development projects for this course: Simulations Interactives en Robotique Mobile and e-puck: robot mobile de table pour une éducation interdisciplinaire.
Third, we would like to acknowledge Prof. Guy Theraulaz (UPS and CNRS Toulouse, France) for sharing with us lecture notes from his course on Collective Intelligence in Biological Societies; Dr. Francesco Mondada (EPFL) and Dr. Olivier Michel (Cyberbotics Ltd.) for all the development and discussions around the real and simulated e-puck robot and, more generally, educational tools for mobile robotics; and several other scientists for sharing with us additional teaching material (in alphabetical order, new names might be added during the course period): Dr. Guillermo Barrenetxea (SensorScope S.a.r.l), Dr. Maurice Clerc (consultant with France Telecom, R&D Division), Prof. Jean-Louis Deneubourg (ULB, Bruxelles, Belgium), Prof. Marco Dorigo (ULB, Bruxelles, Belgium), Prof. Deborah Estrin (UCLA, Los Angeles, U.S.A.), Prof. Dario Floreano (EPFL), Prof. Owen Holland (University of Essex, UK), Dr. Nidhi Kalra (Carnegie Mellon University, U.S.A.), Prof. Chris Melhuish (University of West England, UK), Prof. Roland Siegwart (ETH Zurich, Switzerland), Prof. Alan Winfield (University of West England, UK).
Finally, we would like to thank Cyberbotics Ltd. and Crossbow Inc., for providing the realistic simulator and sensor nodes used in the course.