Predictive Dynamics: Lifting - Human Motion Simulation: Predictive Dynamics - page 197

(A)

(B)

(C)

FIGURE 8.10
Snapshots of simulated box-lifting motion: (A) stability criterion, (B) dynamic effort
criterion, (C) a weighted sum of w 1 5

0.1 and w 2 5 0.9.

The total error in Equation (8.27) is used to quantify accuracy of the predicted
lifting motion and to determine weighting coefficients corresponding to each
objective function for the multi-objective optimization problem.

8.7 Predictive dynamics simulation
For the results below, a weighted sum approach to multi-objective optimization
can be used for stability and effort. The box lifting is simulated by solving
the MOO problem, and the simulation results are validated with the experimental
data in this section. The effects of different objective functions are also illustrated.
Results of the predictive dynamics simulation for the lifting task using stabil-
ity, dynamic effort, and the weighted sum performance measures are depicted in
Figure 8.10 . The ZMP trajectories for the box-lifting simulations using different
performance measures are illustrated in Figure 8.11 . In addition, GRF and joint
torque profiles obtained from the motion prediction using the predictive dynamics
approach are shown in Figures 8.12 and 8.13 , respectively.
A detailed reference on using predictive dynamics in simulating lifting was
reported by Xiang et al. (2009b) . As observed in Figure 8.10 ,

the stability