Adaptive Control for Evolutionary Robotics

Adaptive Control for Evolutionary Robotics: And its effect on learning directed locomotion

van Diggelen, Fuda (TU Delft Mechanical, Maritime and Materials Engineering)

Babuska, R. (mentor)
Eiben, Guszti (graduation committee)
Dodou, D. (graduation committee)
Della Santina, C. (graduation committee)

Delft University of Technology

2020-10-28

This thesis is motivated by evolutionary robot systems where robot bodies and brains evolve simultaneously. In such a robot system, `birth' must be followed by `infant learning' by a learning method that works for various morphologies evolution may produce. Here we address the task of directed locomotion in modular robots with controllers based on Central Pattern Generators. We present a bio-inspired adaptive feedback mechanism that uses a forward model and an inverse model that can be learned on-the-fly. We compare two versions (a simple and a sophisticated one) of this concept to a traditional (open-loop) controller using Bayesian Optimization as a learning algorithm.
The experimental results show that the sophisticated version outperforms the simple one and the traditional controller. It leads to improvement in performance and more robust controllers that cope better with noise.

Adaptive control
Directed Locomotion
Reality Gap
Evolutionary Robotics

http://resolver.tudelft.nl/uuid:63fdb4b7-14a5-4fa2-9c65-87e0323fd115

Student theses

master thesis

© 2020 Fuda van Diggelen