Periodic Nonlinear Principal Component Neural Networks for Humanoid Motion Segmentation, Generalization, and Generation

In an experiment with a soccer playing robot, periodic temporally-constrained nonlinear principal component neural networks (NLPCNNs) are shown to characterize humanoid motion effectively by exploiting fundamental sensorimotor relationships. Each network learns a periodic or transitional trajectory in a phase space of possible acnd thus abstracts a kind of protosymbol. NLPCNNs can play a key role in a system that learns to imitate people, enabling a robot to recognize the behavior of others because it has grounded that behavior in terms of its own bodily movements.

• Type: Conference
• Year: 2004
• Where: ICPR
• Authors: Karl F. MacDorman, Rawichote Chalodhorn, Minoru Asada
• Keywords: Computer Vision, Fundamental Sensorimotor Relationships, ICPR 2004, Nonlinear Principal Component, Soccer Playing Robot