Comparison of robust optimal QFT controller with TFC and MFC controller in a multi-input multi-output system

Abstract

This paper suggests a practical approach for the development of a stable robot controller using the Quantitative Feedback Principle (QFT). Robot manipulators have a multivariable nonlinear transfer function, the implementation of the QFT method includes, first the conversion of their nonlinear plant into a group of linear and uncertain plant set, and then an ideal robust controller for each set has been designed. To demonstrate the effectiveness of our algorithm, we show the implementation of the two degrees of freedom manipulator. In the approach provided, the controller has been designed directly by specifying and optimizing the transfer function coefficients using a genetic algorithm. The consistency and limitations of the method are considered to be the restrictions of the problem in the optimization process. System stability and tracking problem are perceived to be the limitations of the system in the optimization process. Non-linear simulations on the tracking problem are carried out and the results illustrate the performance of the controllers. Finally, the controller constructed based on the QFT approach is compared with the TFC and MFC (Fuzzy) controllers and it is shown that the QFT methodology indicates a controller that has increased control efficiency.