Learning internal representations by error propagation
Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Non-Linear Dimensionality Reduction
Advances in Neural Information Processing Systems 5, [NIPS Conference]
Hi-index | 0.00 |
A method of generating new motions associatively from novel trajectories that the robot receives is described. The associative motion generation system is composed of two neural networks: nonlinear principal component analysis (NLPCA) and Jordan recurrent neural network (JRNN). First, these networks learn the relationship between a trajectory and a motion using training data. Second, associative values are extracted for associating a new corresponding motion from a new trajectory using NLPCA. Finally, a new motion is generated through calculation by JRNN using the associative values. Experimental results demonstrated that our method enabled a humanoid robot, KHR-2HV, to associatively generate the new motions corresponding to trajectories that the robot had not learned.