Predictive Dynamics - Human Motion Simulation: Predictive Dynamics - page 123

FIGURE 5.23H
Jumping.

where the dynamic equations of motion must be integrated, this method provides
a viable and rigorous approach to solving for the motion.
The equations of motion are typically characterized by a set of differential
algebraic equations and contain a high degree of freedom system

they are quite
complex, and highly nonlinear. As a result, these DAEs are difficult to integrate
if not impossible, depending on the complexity.
Predictive dynamics offers a unique method for characterizing human behav-
ior as represented by cost functions (one or multiple objectives) formulated as a
multi-objective optimization problem. A digital human climbing a wall will most
likely have various cost functions driving that motion, including but not limited
to minimizing energy, maximizing stability, maximizing vision, and several
others. The predictive dynamics approach provides for a direct method to enable
“human” or more “naturalistic” predictions of human motion than the typical
direct integration of the equations of motion.
We list below the major important aspects of predictive dynamics:
a. No integration of the equations of motion.
b. Uses human cost functions to drive the motion.
c. Computationally efficient.

...