Modelling the neural controller of a swimming lamprey - a comparison of naturally and artificially evolved networks

Lines connect the coordinator with its partners. CORDIS does not always give exact coordinates for projects before 2014. These points are placed at city or country level.

Project objective

Research objectives and content This project is interested in the neural locomotion controller of lampregs. It has as objective to develop artificial neural control- lers and compare them with the biological model. This research has a double motivation : firstly to study the alternative possibilities Nature could have choses to obtain adapted controllers and secondly to evaluate how good Genetic Algorithms are as tool for creating neural controllers for a specific task. The project will be achieved in three stages : 1) simulation the biological model (reproduction of ekeberg's work) 2) development of encodings and fitness functions in order to evolve artificial controllers 3) Comparison of the artificial and biological networks The thesis will also make a review of dynamic neural networks Training content (objective, benefit and expected impact) applied to control theory. The main benefit of this project is to use a new tool (Evolutionary Algorithms) in a recent field (Dynamic Neural Networks for robot control) I kope this PHD will provide me a deep knowledge of both promising fields.

Original text from CORDIS.

Participants

• UNIVERSITY OF EDINBURGH · EDINBURGHCoordinatorUnited Kingdom
• Not availableCity levelNetherlands

Links

• View on CORDIS