Friedrich-Alexander-Universität UnivisDeutsch FAU-Logo
Techn. Fakultät Willkommen am Institut für Informatik FAU-Logo
Logo LSS
Chair for System Simulation (Department of Computer Science 10)
Klaus Iglberger
Dept. of Computer Science  >  Computer Science 10  >  People  >  Klaus Iglberger

Diplomarbeit / Master thesis

Rigid Body Dynamics: Triangle Meshes

Supervision:

Background:

The pe rigid body physics engine is a C++ framework for the physically correct simulation of rigid bodies of arbitrary shape. Currently, the arbitrary shape of the rigid bodies is achieved by unions of geometric primitives, like for instance spheres, boxes, capsules and planes. For the simulation of truely arbitrary geometries, the physics engine has to be extended by triangle meshes.



The integration of triangle meshes involves a couple of necessary steps. The first task in realizing triangle meshes is the formulation of a suitable data structure for the triangle meshes. This data structure has to support large numbers of triangles and an efficient update of the involved vertices in every time step. A second step is the import of a triangle mesh from a file (STL, 3DS, ...). Depending on this input file format, several problems may occur. For example, it may be possible that a triangle mesh is not closed and therefore offers no possibility to calculate a center of mass and the moment of inertia. The third step is the treatment of collisions between rigid bodies built from triangles and rigid body primitives. This step requires the implementation of several special purpose functions to handle for instance sphere-triangle, box-triangle, ... collisions. The last step is the visualization of the triangle meshes in both the real-time visualization system and the POV-Ray visualization.

The task of this thesis is the integration of triangle meshes into the pe framework. This integration involves all the afore mentioned steps. Helpful for this task is a preceeding thesis about a possible implementation of triangle meshes [1] and several books about collision detection and treatment [2].



Tasks:

  • Integration of joint motion constraints into the pe framework
  • Development and implementation of a general scheme to enable the easy integration of other joint types
  • Implementation of several example joints to demonstrate the suitability of the extension
  • Development of suitable test scenarios to prove the correctness of the implementation
  • Development of demonstration examples

Recommended knowledge:

  • Advanced C++ programming
  • Basic background in engineering and mechanics

Status:

Reserved

References:

[1] Martin Ketzer: Geometrieprimitive und Dreiecksgitter für die pe Physikengine, Lehrstuhl für Systemsimulation, März 2007
[2] Christer Ericson: Real-Time Collision Detection, Morgan Kaufmann Series, ISBN: 1558607323

  Contact Last modified: 2008-11-07 08:37   cf