Engineering Mechanics

Proceedings Vol. 10 (2004)

ENGINEERING MECHANICS 2004

Copyright © 2004 Institute of Thermomechanics, Academy of Sciences of the Czech Republic, Prague

 

Krejsa J., Grepl R., Věchet S.
pages 159 - +8p., full text

The paper compares global and local approximation methods used for walking robot stability model. Global approximators are represented by feedforward multilayer neural network (ffNN) trained by gradient method; local approximators are represented by Locally Weighted Regression (LWR) and Receptive Field Weighted Regression (RFWR) methods. Global approximators try to learn global non-linear function which fits all the training data (minimizes training error), while local approximators use spatially limited data in query point neighborhood to generate appropriate response. Various aspects of used approximation methods are discussed (precision, robustness, computational and memory requirements).

Text and facts may be copied and used freely, but credit should be given to these Proceedings.

All papers were reviewed by members of the scientific committee.