Consensus and cooperation in networked multi-agent systems is a topic that is starting to receive significant attention in control theory and distributed computing owing to numerous possible engineering applications. For instance, the power grid, urban traffic, arrays of distributed sensors, multi-robot systems, and social networks are various examples of collective systems requiring an effective response to local perturbations. Using a number of theoretical models, we present some unexpected features arising in the collective dynamics of networked multi-agent systems. With the help of two swarm robotics testbeds, we experimentally analyse these features, and discuss their far-reaching practical implications for the design and understanding of distributed systems.
Biography
Prof Bouffanais and his group focus on both fundamental and applied interdisciplinary problems rooted in the field of Complexity. Depending on the nature of the problem, experiments, analytical theory and computation are considered and practiced. Specifically, we investigate complexity in engineering, biological and physical systems with a particular emphasis on the study of dynamical processes in relation with emergent collective behaviors: cell aggregation, fish schooling, cellular sensory networks, quantum emergent behaviors, turbulence in fluids, etc. We aim at devising some commonalities in those awe-inspiring collective behaviors by studying the system dynamics of a set of parameters borrowed from information theory, control theory and network science.