Parallelization of Dynamics Algorithms

Motivation

Multi-body dynamics plays an important role in simulation and control. As the dynamic systems become more complex, the time complexity of dynamics algorithm increases.
This causes the degradation of the system performance and reliability. For this reason, efficient dynamics algorithms have been developed.

Contribution of our research

  • Derivation of explicit formulations for computing the system inertia and bias terms of multi-body
    systems by generalizing the principle of dynamical balance.
  • Development of parallel dynamics algorithms using explicit formulations of the system inertia and bias parameters
  • Results
    캡처
  • Computation of the system inertia and bias matrix with respect to the degrees-of-freedom(DOF) of the serial type robots (Left Column), 20-DOF Robots with Different Tree Typologies (Right Column).