A Spherical 3-DOF Parallel Mechanism Driven By Pneumatic Muscle Actuator

Pneumatic muscle actuator(PMA) is mainly made of rubber tube(the bladder) attached to end caps and surrounded by braided fibrous shell. When the bladder is pressurized, PMA shortens in axis and expand in diameter, which generates contractile force simultaneously. Compared with electric and hydraulic actuators, PMA provides superior advantages of quick response, high power/weight and power/volume ratios and natural flexibility similar to biological muscles. As these benefits shown above, PMA has various applications especially in rehabilitation therapy and robotic motion control, so the research of PMA is important to it’s application.The purpose of this paper is to study the characterization of PMA and drive parallel mechanism actuated by PMAs, The main work is as follow:First of all, the working principle of PMA is studied, two static models of PMA based on energy conservation and BP neural network are established and analyzed, and experimental results of the two models are compared after building static test station.Secondly, based on a three-element phenomenological model(contractile, spring and damping coefficient), the dymamic characterization of a DMSP-type PMA manufactured by FESTO is studied in this paper. Relationship between the three coefficients with pressure and external load are expressed. A nearly linear relationship between contractile coefficient and pressure is found, spring coefficient is linear with external load and damping coefficient is nearly a constant in high pressure.Finally, a spherical 3-DOF parallel mechanism driven by PMAs is designed. This thesis studies the inverse kinematic equations, expresses balance requirements in rotation and deduces velocity equations and acceleration equations in this mechanism. The dynamic model based on Kane equations is established. Control model based on energy conservation is constructed. Open-loop and close-loop experiments are conducted and compared aftering establishing experimental station.Static model, dynamic characterization and parallel mechanism driven by PMAs are discussed in this paper, all of those works enable us to understand PMA deeper, improve control precison and accelerate the applications driven by PMAs.