Robots Need Some Education: On the complexity of learning in evolutionary robotics

TL;DR

This paper explores the complexities of integrating learning algorithms into evolutionary robotics, highlighting the challenges and presenting new algorithms tailored for this combined approach.

Contribution

It investigates the effects of learning within evolutionary robotics and introduces several novel learning algorithms suited for this integration.

Findings

Introducing learning affects evolutionary processes in complex ways

New algorithms improve robot controller optimization

Understanding learning effects enhances evolutionary robotics design

Abstract

Evolutionary Robotics and Robot Learning are two fields in robotics that aim to automatically optimize robot designs. The key difference between them lies in what is being optimized and the time scale involved. Evolutionary Robotics is a field that applies evolutionary computation techniques to evolve the morphologies or controllers, or both. Robot Learning, on the other hand, involves any learning technique aimed at optimizing a robot's controller in a given morphology. In terms of time scales, evolution occurs across multiple generations, whereas learning takes place within the `lifespan' of an individual robot. Integrating Robot Learning with Evolutionary Robotics requires the careful design of suitable learning algorithms in the context of evolutionary robotics. The effects of introducing learning into the evolutionary process are not well-understood and can thus be tricky. This thesis investigates these intricacies and presents several learning algorithms developed for an Evolutionary Robotics context.