I will present how evolutionary robotics can be used for designing and studying collective (robotic) systems. This talk will cover both evolutionary robotics as an off-line design method (search for a solution, then use it), and as an on-line design method (search for a solution while the robots are already deployed). I will also describe our recent work on environment-driven evolutionary adaptation, i.e. on-line distributed algorithms that can be used to deploy a swarm of robots in open environments, where the tasks to achieve self-sustainability are unknown to the experimenter prior to deployment. Finally, I will discuss how evolutionary robotics for collective systems can be used both for engineering (as a design method) and biology (as a modelling and simulation method).
Authors
Nicolas Bredeche
Nicolas Bredeche is a professor of computer science at UPMC - Sorbonne Universités in Paris, France. He works at the Institute of Intelligent Systems and Robotics (ISIR, UPMC/CNRS). His research topics are at the crossroad of computer science, biology and robotics. He is interested in adaptive collective systems, whether these systems are natural (collective behaviours in animals) or artificial (collective robotics). His research topic is on evolutionary adaptation, to understand the mechanisms at work in nature (ex.: the role of coordination in the evolution of cooperation), and to design evolutionary algorithm for artificial systems (ex.: embodied evolutionary robotics in open environment).
Doncieux S., Bredeche N., Mouret J.-B., Eiben A.E. Evolutionary Robotics: What, Why, and Where to. Frontiers in Robotics and AI, Volume 2, number 4, 2015. doi: 10.3389/frobt.2015.00004.
E. Haasdijk, N. Bredeche, A.E. Eiben. Combining Environment-Driven Adaptation and Task-Driven Optimisation in Evolutionary Robotics. PLoS ONE 9(6): e98466. doi:10.1371/journal.pone.0098466. 2014.
S. Anderson, N. Bredeche, A.E. Eiben, G. Kampis, M. van Steen. Adaptive Collective Systems. Publisher: Focas. ISBN: 978-2-7466-6930-7 (print), 978-2-9548251-0-6 (electronic).
N. Bredeche, J.-M. Montanier, W. Liu, A. FT. Winfield. Environment-driven Distributed Evolutionary Adaptation in a Population of Autonomous Robotic Agents. Mathematical and Computer Modelling of Dynamical Systems, Special Issue: Modelling the swarm - analysing biological and engineered swarm systems. Taylor & Francis Eds. Volume 18, Issue 1, p.101-129, 2012.
A. Devert, N. Bredeche, M. Schoenauer. Robustness and the Halting Problem for Multi-Cellular Artificial Ontogeny. IEEE Transactions on Evolutionary Computation 15(3), pages 387-404, 2011.
Invited Talk e-session
Keywords
Tags: collective systems, design for engineering, evolutionary adaptation, evolutionary robotics, modelling and simulation for biology, on-line learning, swarm robotics
Photos by : Anonymous
