The animation of articulated bodies presents interest in the areas of biomechanics, sports, medicine and the entertainment industry. Traditional motion control methods for these bodies, such as kinematics and rotoscoping are either expensive to use or very laborious. The motion of articulated bodies is complex mostly because of their number of articulations and the diversity of possible motions. This thesis investigates the possibility of using dynamic analysis in order to define the motion of articulated bodies. Dynamic analysis uses physical quantities such as forces, torques and accelerations, to calculate the motion of the body. The method used in this thesis is based upon the inverse Lagrangian dynamics formulation, which, given the accelerations, velocities and positions of each of the articulations of the body, finds the forces or torques that are necessary to generate such motion. Dynamic analysis offers the possibility of generating more realistic motion and also of automating the process of motion control. The Lagrangian formulation was used first in robotics and thus the necessary adaptations for using it in computer animation are presented. An analytical method for the calculation of ground reaction forces is also derived, as these are the most important external forces in the case of humans and the other animals that are of special interest in computer animation. The application of dynamic analysis in bipedal walking is investigated. Two models of increasing complexity are discussed. The issue of motion specification for articulated bodies is also examined. A software environment, Solaris, is described which includes the facility of dynamic and kinematic motion control for articulated bodies. Finally, the advantages and problematics of dynamic analysis with respect to kinematics and other methods are discussed.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:314493 |
| Date | January 1991 |
| Creators | Vasilonikolidakis, Nickos A. |
| Publisher | London Metropolitan University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://repository.londonmet.ac.uk/3385/ |
Page generated in 0.0018 seconds