Programmer. Hiker. Cook. Always looking for interesting problems to solve.
01 Mar 2013 |
Just 10 days left until I have to submit my draft report. This describes how I feel right now:
After performing some experiments with Q-Learning algorithm, reading tons of research papers and finally realizing that evolving walking in Aibo is a research project on its own, I decided to change the simulation to e-puck.
A few thoughts about the evolution. In a population of 50 neural networks, how do I select which one will decide on next motor action?