Study on evolution of the artificial flying creature controlled by neuro-evolution

Authors:
Ryosuke Ooe;Ikuo Suzuki;Masahito Yamamoto;Masashi Furukawa
Affiliations:
Department of Information Science and Technology, Hokkaido University, Hokkaido, Japan 060-0814;Department of Information Science and Technology, Hokkaido University, Hokkaido, Japan 060-0814;Department of Information Science and Technology, Hokkaido University, Hokkaido, Japan 060-0814;Department of Information Science and Technology, Hokkaido University, Hokkaido, Japan 060-0814
Venue:
Artificial Life and Robotics
Year:
2013

Citing 9
Cited 0

Analytical methods for dynamic simulation of non-penetrating rigid bodies

SIGGRAPH '89 Proceedings of the 16th annual conference on Computer graphics and interactive techniques
Evolving virtual creatures

SIGGRAPH '94 Proceedings of the 21st annual conference on Computer graphics and interactive techniques
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Realistic modeling of bird flight animations

ACM SIGGRAPH 2003 Papers
Evolving soft robotic locomotion in PhysX

Proceedings of the 11th Annual Conference Companion on Genetic and Evolutionary Computation Conference: Late Breaking Papers
Complex networks of simple neurons for bipedal locomotion

IROS'09 Proceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems
Feathered flyer: integrating morphological computation and sensory reflexes into a physically simulated flapping-wing robot for robust flight manoeuvre

ECAL'07 Proceedings of the 9th European conference on Advances in artificial life
Incremental evolution of target-following neuro-controllers for flapping-wing animats

SAB'06 Proceedings of the 9th international conference on From Animals to Animats: simulation of Adaptive Behavior
Evolution of central pattern generators for bipedal walking in areal-time physics environment

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

The objective of this study is to realize efficient learning for the high generalization ability of evolutionary artificial neural network (EANN). In order to achieve this objective, the evolutionary process of behavior acquisition is analyzed, and then an efficient evaluation function is led by the analysis. An artificial flying creature (AFC) is controlled to fly towards a given target point by EANN. The three-dimensional motion of the AFC is calculated by the physical engine PhysX and a numerical expression of the simple drag force. To evolve ANNs and to have the AFC flight suitably for given target points, particle swarm optimization optimizes parameters of ANNs. The results of evolutionary simulation show that generalization ability of ANNs does not always increase as evolution progresses, and it depends on given tasks of the AFC. It is also shown that diversity of input signals about target points, which the AFC goes through in flight, has positive correlation with generalization ability.