The control of material transportation systems, especially cranes - due to civil and military applications - has achieved a significant interest in the last decades. Many different types of cranes are used in various industries. In spite of different structures, numerous types of cranes have similar mechanical properties. In particular, all of them are underactuated mechatronical systems showing oscillatory behavior. The automatization of these plants is one of the most interesting and important research area in nonlinear control, even in nowadays.
In general, there are two important issues that have to be concerned. One is the accurate positioning of the payload and the other is the minimum payload swing during the transportation (for safety reasons). So both positioning and anti-sway control are important in the controller design process. Therefore a robust controller, which is able to diminish the undesirable swing phenomena and manage uncertainty problems, needs to be developed.
Measurements on all configuration variables are generally not available (especially as far as the rope angles or the load position are concerned). Therefore the observation of the necessary variables becomes essential.
Recently, in the laboratory of the Department of Control Engineering and Information Technology an overhead crane has evolved. The goal of my MSc Thesis is to design a control methodology to this overhead crane, allowing full automated and efficient load positioning.
Based on the state of the art papers found in the references I improved a system model describing the overhead crane. The linearization of the system was done by dynamic feedback techniques based on differential geometry. In order to be able to reconstruct the immeasurable physical variables an observer was designed on the basis of the dynamic model. Trajectory generation and input computation were performed without integrating differential equations.
The model of the overhead crane and the surrounding closed loop control system was implemented in MATLAB environment. In order to test the effectiveness of the system perturbed by different parameter settings and noise I made several simulations. These simulations proved that the designed closed loop control system is suitable for path tracking applications and to control the real overhead crane.