A neuro-muscular elasto-dynamic approach to modeling biomechanical systems: the human upper extremity

In this thesis the author intends to lay the groundwork for the development of a neuromuscular elasto-dynamic model of the upper extremity which will provide a way to relate neurological control to stress development in the skeletal structure. Several benefits will result from such a model, such as realizing inertial effects on stress from dynamic movements, and immediate predictions of neurological and modeling effects on skeletal structures. Using a variational approach to dynamic and elastic modeling, the method used offers a straightforward, systematic, approach which is applicable in other areas of skeletal research as well as being suitable to finite element methods. The work presented herein is simplified to the extent that it is intended as an example of the next logical step in the study of neurological and musculo-skeletal research.

As an example of the method, two elasto-dynamic models have been created. The models differ only in approach to how muscle force application is modeled, one using direct muscle-to-bone insertion and the other the method of joint torque reduction. Two control strategies are then implemented for each model; feedforward using a step input and position feedback. Both control strategies are designed to produce a curling action in the arm. The sensitivity of the system to the changes in control strategies and muscle force application techniques is shown.