Evolutionary algorithms are powerful search algorithms which can be used for design exploration, parameter optimization, and adaptation to changing conditions. Evolutionary robotics is the field of applying evolutionary algorithms to robotics, and has the potential to improve robots in a number of ways - from the design of robotic bodies and controllers, to adaptation when the robot suffers from damage or encounters a new environment.
There are several different themes which could be relevant, and the specific project should be worked out with the student and supervisor(s) according to interests. Here are some example themes:
- Morphology evolution (E.g. Exploring various ways of co-evolving robot shapes and control systems, how to explore the vast design space efficiently, new morphology encodings). In-depth description.
Example MSc thesis, Example paper, another paper - Evolution of locomotion patterns for robots using various algorithms and control systems (E.g. evolving a repertoire of different locomotion patterns, overcoming obstacles, evolving robust gaits, using neural networks for control, comparing with other learning methods). In-depth description.
- Reality gap research (Testing various approaches for a smooth transfer from simulator to reality).
Example paper based on an MSc project - Adaptivity and learning (E.g. detecting changes in the environment, adapting accordingly, running learning algorithms on robot in simulation and/or the real world)
Example MSc thesis
A Master's project in evolutionary robotics normally involves the following components:
- Literature survey and writing essay
- Implementation of some experimental system, e.g. programming a variant of an evolutionary algorithm, designing test scenarios, or fitness functions
- Running experiments on the implemented approach. This may be using a physics simulation system, and/or on a real robot
- Analyzing the results and re-iterating on the steps above
- Writing the MSc thesis document