Dealing with the Ill-Conditioned Equations of Motion for Articulated
Figures (Maciejewski)

In this paper, the author discusses the simulation of motion for
articulated figures, why the forumulations are ill-conditioned, and
what to do about them. He proposes the use of the damped least-squares
formulation in order to provide continuous, good looking motion near
singularities and to provide meaningful bounds for the numerical
integration involved.

Key ideas/Contributions:

1. An analysis of current equations for motion and pointing out that
they are "inherent in the figures' physical structure and not due to a
choice in mathematical formulation."

2. The use of a damped least-squares formulation to provide continuity
in pseudoinverse solutions near a singularity.

Questions:

Just to better understand the math. I gave this paper a terse read and
will have to read through it more thoroughly over the weekend.