On Applying Neuroevolutionary Methods to Complex Robotic Tasks
Yohannes Kassahun, José de Gea Fernández, Malte Langosz, Frank Kirchner
Editors: Nikos Papanikolopoulo, Shigeki Sugano, Stefano Chiaverini, Max Meng
In IEEE IROS Workshops on Exploring new horizons in Evolutionary Design of robots, (IROS-09), 11.10.-15.10.2009, St. Louis, Missouri, o.A., pages 26-30, Oct/2009.
Abstract
:
In this paper, we describe possible methods of solving two problems encountered in evolutionary robotics, while applying neuroevolutionary methods to complex robotic tasks. The first problem is the requirement of a large number of evaluations needed to obtain a solution. We propose that this problem can be addressed by accelerating neuroevolutionary methods using a Kalman filter. The second problem is the difficulty of obtaining a mesirable solution that results from a difficulty of choosing an appropriate fitness function. The solution towards this problem is to apply the principles of behavior based systems to decompose the solution space into smaller subsolutions with lower number of intrinsic dimensions, and incrementally modify the fitness function.