Guided Optimization for Balanced LocomotionReportar como inadecuado

Guided Optimization for Balanced Locomotion - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

1 iMAGIS - Models, Algorithms and Geometry for Computer Generated Image Graphics GRAVIR - IMAG - Graphisme, Vision et Robotique, Inria Grenoble - Rhône-Alpes

Abstract : Teaching simulated creatures how to walk and run is a challenging problem. As with a baby learning to walk, however, the task of synthesizing the necessary muscle control is simplified if an external hand to assist in maintaining balance is provided. A method if using guiding forces to allow progressive learning of control actions for balanced locomotion is presented. The process has three stages. Stage one involves using a -hand of God- to facilitate balance while the basic actions of a desired motion are learned. Stage two reduces the dependence on external guidance, yielding a more balanced motion. The method is applied to obtain walking motions for a simple biped and a bird-like mechanical creature, as well as walking, running, and skipping motions for a human model of realistic proportions.

Autor: Michiel Van de Panne Alexis Lamouret -



Documentos relacionados