Accurate Control of a Quanser Flexible Joint Manipulator robot based onMulti-objective Uniform-diversity Genetic Algorithm (MUGA)

In this paper, a novel online hybrid adaptive robust controller based on Multi-objective Uniform-diversityGenetic Algorithm (MUGA) is presented for a Quanser Flexible Joint Manipulator. Since, the Quanser FJMoffers various benefits such as flexibility, faster performance, lower weight and power consumption; thesefeatures move towards to design a complex controller. The aim of the proposed controller is to reach stabilityand trajectory tracking of the Quanser FJM link by minimizing the error of angular position and deflectionand achieve an optimal control inputs. For this reason, after transforming the nonlinear equation into lineartypes and applying the adaptive mechanism to update the optimal parameters the chattering appears on thesystem. Since the proposed controller structure needs to be improved, adaptive mechanism based on gradientdescend method and chain rule of the derivation (a novel mathematical relation) and optimization of thedesign parameters based on MUGA was considered. In the following, to evaluate the proposed controllerability in confronting parametric and non-parametric uncertainties, the results was compared with othercontrollers.Finally, the analysis demonstrated the robustness and proper performance of the proposed controller inrespect of stability, minimum tracking error, overshoot and optimal control efforts in comparison with other controllers.

Accurate Control of a Quanser Flexible Joint Manipulator robot based onMulti-objective Uniform-diversity Genetic Algorithm (MUGA) Keywords:

Multi-objective Uniform-diversity Genetic Algorithm (MUGA) , Quanser Flexible JointManipulator , Optimal parameters , Adaptive Mechanism.