DMCA
Chaos (1987)
Cached
Download Links
| Venue: | ACM Transactions on Graphics (SIGGRAPH |
| Citations: | 57 - 10 self |
BibTeX
@ARTICLE{Yin87chaos,
author = {Kangkang Yin and Kevin Loken and Michiel Panne},
title = {Chaos},
journal = {ACM Transactions on Graphics (SIGGRAPH},
year = {1987},
volume = {2007}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Figure 1: Real-time physics-based character simulation with our framework. (a) A single controller for a planar biped responds to unanticipated changes in terrain. (b) A walk controller reconstructed from motion capture data responds to a 350N,0.2s diagonal push to the torso. Physics-based simulation and control of biped locomotion is difficult because bipeds are unstable, underactuated, high-dimensional dynamical systems. We develop a simple control strategy that can be used to generate a large variety of gaits and styles in real-time, including walking in all directions (forwards, backwards, sideways, turning), running, skipping, and hopping. Controllers can be authored using a small number of parameters, or their construction can be informed by motion capture data. The controllers are applied to 2D and 3D physically-simulated character models. Their robustness is demonstrated with respect to pushes in all directions, unexpected steps and slopes, and unexpected variations in kinematic and dynamic parameters. Direct transitions between controllers are demonstrated as well as parameterized control of changes in direction and speed. Feedback-error learning is applied to learn predictive torque models, which allows for the low-gain control that typifies many natural motions as well as producing smoother simulated motion. 1
Keyphrases
physically-simulated character model direct transition diagonal push unexpected variation high-dimensional dynamical system real-time physics-based character simulation motion capture data responds biped locomotion low-gain control predictive torque model physics-based simulation large variety feedback-error learning walk controller many natural motion single controller simple control strategy unanticipated change motion capture data small number dynamic parameter unexpected step
