Real-Time Visualization of Postural Stability in a Dynamic Environment – A Hybrid Artificial Intelligence Approach

PDF

Authors
  1. Nickerson, J.
Corporate Authors
Defence R&D Canada - Atlantic, Dartmouth NS (CAN)
Abstract
Modelling realistic 3-D human motions in a dynamic environment is a challenging task. When fully modelled the human body has over 200 segments, 600 muscles and approximately 200 degrees of freedom. Human motion is not computationally well defined and there is no one definitive motion that can be categorized as humanlike. This paper presents an approach for real-time modelling of realistic human motion in a dynamic environment. It focuses on movements of the limbs and uses genetic algorithms and a knowledge based system with fuzzy logic rules to animate statically stable poses.

Il y a un résumé en français ici.

Report Number
DRDC-ATLANTIC-TM-2010-024 — Technical Memorandum
Date of publication
01 Jul 2010
Number of Pages
42
DSTKIM No
CA035401
CANDIS No
534934
Format(s):
Electronic Document(PDF)

Permanent link

Document 1 of 1

Date modified: