Abstract
The McKibben style artificial muscle is a type of pneumatic artificial muscle. When the muscle is pressurized a length contraction and contractive force occur. Modeling a McKibben style muscle presents many challenges. The physical dynamics of the muscle are highly nonlinear, which makes accurate modeling difficult. Modeling using experimentally determined coefficients enables the creation of an equation that ignores the nonlinear nature of the system, but it can have a limited range or accuracy and cannot adjust for changing system properties. Modeling using the system properties requires the consideration of the nonlinear dynamics of the muscle; however, it provides the ability to run numeric simulations before constructing muscles. This thesis focused on dynamic modeling of a two muscle system using system properties.
In order to develop a two muscle dynamic system model, the following steps were taken. Models using physical properties for individual static muscles were examined. Two muscles were linked around a pulley to form a two muscle system with an output of angular displacement and torque. Finally, the effect of valves was added to allow for modeling of transient system response.
In order to validate the model, an experimental system was constructed. The muscle system was simulated using MATLAB and outputs were compared to experimental results. Good agreement between theoretical and experiment results was obtained. A PID controller was then implemented on the new model to demonstrate the feasibility of using the model for control of a two muscle system. The controller was run through an optimization routine to determine the gains which gave the least position error.
This work is the first to provide a dynamic model for a system of two of opposed McKibben style muscles based on the physical properties of the muscle system.
Library of Congress Subject Headings
Pneumatic machinery--Mathematical models; Smart structures--Mathematical models
Publication Date
1-2018
Document Type
Thesis
Student Type
Graduate
Degree Name
Mechanical Engineering (MS)
Department, Program, or Center
Mechanical Engineering (KGCOE)
Advisor
Kathleen Lamkin-Kennard
Advisor/Committee Member
Agamemnon Crassidis
Advisor/Committee Member
Jason Kolodziej
Recommended Citation
Arcus, Alexander G., "Modeling of a Dynamic McKibben Style Muscle System Using Material Properties" (2018). Thesis. Rochester Institute of Technology. Accessed from
https://repository.rit.edu/theses/9726
Campus
RIT – Main Campus
Plan Codes
MECE-MS