Design and Testing of a Robust Control System for Industrial Manipulators using Matlab and Simulink

The Thesis project has been developed in COMAU S.p.A, one of the most important leading companies in the industrial automation field. The purpose of the activity is the development and simulation of a robust control system for an industrial six axes manipulator, the most common mechanical structure. This is a very important goal to reach, since robustness to disturbances and uncertainties has always been the central focus in feedback control theory. To fulfill this objective, the $H_\infty$ approach has been used. Moreover, the development of the controller has given the chance to investigate the advantages and disadvantages of this method when it is applied to a more realistic and complex field, such as the robotics one. Matlab has been adopted as computing environment for the development of the design. The analysis of the obtained results has been carried out through Simulink toolbox. After the derivation of the model describing the dynamics of the robot under analysis, a simple feedback control system has been implemented to control the position of a single axis of the manipulator. To enhance the robustness of the system stability, the previous structure has been subsequently improved by introducing an inner loop aimed at controlling the velocity. Once the controllers have been derived, their robustness with respect to variations, in a given range of uncertainty, of the parameters characterizing the system has been studied. Finally, a parametric controller adapting to the changes of the link inertia has been built.