Title :
Piecewise Affine Modeling and Constrained Optimal Control for a Pneumatic Artificial Muscle
Author :
Andrikopoulos, G. ; Nikolakopoulos, George ; Arvanitakis, Ioannis ; Manesis, Stamatis
Author_Institution :
Dept. of Electr. & Comput. Eng., Univ. of Patras, Rio, Greece
Abstract :
In this paper, the modeling and control problem of a pneumatic artificial muscle (PAM) is being considered. The PAM is an actuator characterized by a decrease in the actuating length when pressurized. Its nonlinear nature and time-varying parameters cause difficulties in modeling their characteristics, as well as in designing controllers for high-performance positioning systems. A constrained linear and piecewise affine system model approximation is formulated, and a control scheme composed of the following is being synthesized: 1) a feedforward term regulating the control input at specific set points and 2) a constrained finite-time optimal controller handling any deviations from the system´s equilibrium points. Extended experimental studies are utilized to prove the efficacy of the suggested controller.
Keywords :
approximation theory; control system synthesis; feedforward; optimal control; pneumatic actuators; PAM; actuator length; constrained finite-time optimal controller; constrained linear system; constrained optimal control; controller design; feedforward term; high-performance positioning system; piecewise affine modeling; piecewise affine system; pneumatic artificial muscle; system model approximation; Actuators; Approximation methods; Feedforward neural networks; Mathematical model; Muscles; Switches; Constrained control; feedforward control; fluidic muscle; optimal control; piecewise affine (PWA) system; pneumatic muscle actuator;
Journal_Title :
Industrial Electronics, IEEE Transactions on
DOI :
10.1109/TIE.2013.2254094