An algorithm that recognizes and reproduces distinct types of humanoid motion based on periodically-constrained nonlinear PCA

Authors:
Rawichote Chalodhorn;Karl MacDorman;Minoru Asada
Affiliations:
Department of Adaptive Machine Systems and Frontier Research Center, Graduate School of Engineering, Osaka University, Osaka, Japan;Department of Adaptive Machine Systems and Frontier Research Center, Graduate School of Engineering, Osaka University, Osaka, Japan;Department of Adaptive Machine Systems and Frontier Research Center, Graduate School of Engineering, Osaka University, Osaka, Japan
Venue:
RoboCup 2004
Year:
2005

Citing 6
Cited 1

Introduction to Reinforcement Learning

Introduction to Reinforcement Learning
Effiicient BackProp

Neural Networks: Tricks of the Trade, this book is an outgrowth of a 1996 NIPS workshop
Periodic Nonlinear Principal Component Neural Networks for Humanoid Motion Segmentation, Generalization, and Generation

ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 4 - Volume 04
Circular nodes in neural networks

Neural Computation
Circular backpropagation networks for classification

IEEE Transactions on Neural Networks
Limitations of nonlinear PCA as performed with generic neural networks

IEEE Transactions on Neural Networks

Using eigenposes for lossless periodic human motion imitation

IROS'09 Proceedings of the 2009 IEEE/RSJ international conference on Intelligent robots and systems

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper proposes a new algorithm for the automatic segmentation of motion data from a humanoid soccer playing robot that allows feed-forward neural networks to generalize and reproduce various kinematic patterns, including walking, turning, and sidestepping. Data from a 20 degree-of-freedom Fujitsu hoap-1 robot is reduced to its intrinsic dimensionality, as determined by the isomap procedure, by means of nonlinear principal component analysis (nlpca). The proposed algorithm then automatically segments motion patterns by incrementally generating periodic temporally-constrained nonlinear pca neural networks and assigning data points to these networks in a conquer-and-divide fashion, that is, each network's ability to learn the data influences the data's division among the networks. The learned networks abstract five out of six types of motion without any prior information about the number or type of motion patterns. The multiple decoding subnetworks that result can serve to generate abstract actions for playing soccer and other complex tasks.