Tutorial on Evolutionary Robotics

In the same way as evolution shapes animals, is it possible to use artificial evolution to automatically design robots? This attractive vision gave birth to Evolutionary Robotics (ER), an interdisciplinary field that incorporates ideas from Biology, Robotics and Artificial Intelligence. Within the last 20 years, Evolutionary Robotics yielded to a large variety of scientific questions, from understanding Natural Evolution thanks to ER models, to addressing the autonomous design of robots' bodies and brains. This tutorial will first give an overview of ER and its motivations. It will then present the main features of ER to introduce the main topics of the tutorials, which are: (1) how to make it work? Fitness function design and the influence of selection pressure; (2) ER for real robots: the reality gap and how to cross it; (3) ER and collective robotic systems: how to build such systems? How to use these tools to understand collective behaviors in nature? The questions related to the representation and the principles of neuro-evolution will be only briefly mentioned, as they are described in another tutorial.