Abstract
This research extends the three-joint arm's optimal control model with a freezing-like mechanism in its hand joint to a model more analogous to the human three-joint arm's motion control mechanism by changing the model's desired value of its state variable at the final point from joint angle information to visual one (i.e., x and y coordinates of the hand point), and it discusses its effectiveness. Consequently, the following results are obtained: (1) the extended model reproduces human arm's reaching movement characteristics including the freezing characteristic where the hand joint hardly moves during movements; (2) for the extended model's success in reproducing reaching movements, the model requires a larger weight value of the energy term in its criterion function than the previous model so that it can enhance the degree of consumed energy minimization of the extended model; (3) the extended model's reproduction error of the hand-joint angle is magnified with the movement time. These results suggest that the extended model can become a mathematical model more analogous to the human three-joint arm's motion control mechanism, that it requires another constrained condition to further enhance the degree of reproducibility of the hand joint angle, and that there is a stronger possibility that human reaching movements can be generated based on consumed energy minimization.
