Paper Info
Title
Evolution of vision capabilities in embodied virtual creatures
Abstract
We evolve light following behaviours in virtual creatures through neural network training using an incremental evolution approach. The neural controllers of creatures evolved for movement are augmented with simple visual neurons and neural connections. Using an evolutionary algorithm, the resulting creatures are trained to identify and follow a light source. Through this process, we are able to train the neural controllers to create various light following behaviours. Many of the evolved behaviours show stability and adaptiveness to environmental perturbations of body orientation.
Year
DOI
Venue
2010
10.1145/1830483.1830502
GECCO
Keywords
Field
DocType
neural controller,light following behaviour,neural network training,virtual creature,light source,vision capability,environmental perturbation,neural connection,various light,behaviours show stability,body orientation,evolutionary algorithm,vision,evolution,artificial life,neural network,genetic algorithm,genetic algorithms
Creatures,Artificial life,Incremental evolution,Evolutionary algorithm,Computer science,Embodied cognition,Artificial intelligence,Artificial neural network,Light source,Genetic algorithm,Machine learning
Conference
Citations 
PageRank 
References 
11
0.81
10
Authors
2
Name
Order
Citations
PageRank
Marcin L. Pilat19517.81
Christian Jacob221133.00