Download
Full Text
8.63 MB PDF file
Download
Full Text
8.63 MB PDF file
Research described in this dissertation expands the use of a biomechanical phenomenological model to a commercially available pneumatic muscle actuator (PMA). Due to the nonlinearities of the device, achieving accurate control is challenging. Experiments have been conducted that define boundaries of operation where linear approximations can be used to describe the dynamics of the PMA. Empirical data presented in this dissertation show that nonlinearities exist more prevalently at higher loads (on average >70% contractile force value). When the PMA is under high loads, low displacement occurs. Therefore, these regions are of less interest to design engineers. Once conditions of nonlinearities were defined, operational areas of interest were characterized.
Open-loop linear systems analysis utilized the characterization profiles for the PMA in combination with a model for the D.C. servo motor to develop a system transfer function describing the dynamics of the overall plant. A Tustin (bilinear) transform was applied to the transfer function to generate a discrete time recursion equation. This equation describes the interaction of the PMA and the D.C. servo motor. It was then used to generate motor voltage profiles to demonstrate various tasks on the system.
Finally, this dissertation describes a new concept of using PMAs as an antagonist in a resistive training device. One such application is in a microgravity environment (prolonged space flight). The characterization analysis presents a method to demonstrate this task on the Dynamic Test Station (DTS). In this demonstration the PMA acts as an antagonist generating a resistive load, which the D.C. servo motor, representing the human operator, works against. A 90 degree isokinetic (constant velocity) rotation of the D.C. servo motor pulley is achieved at eight PMA pressures each of which generates a different resistive load.
Document number: wright1227233038.
Bookmark this page as
<http://rave.ohiolink.edu/etdc/view?acc_num=wright1227233038>.