Automatic control of Dynamic Systems

Members : Margarita Domènech, Francesc Palacios, Josep Maria Rossell.

Research contents and social projection :

Control Systems Theory deals with the basic principles underlying the analysis and design of control systems. To control a mechanism means to influence its behaviour so as achieve a desired objective. In order to implement this influence, scientists build devices that incorporate various mathematical techniques. A wide variety of well-known automatic systems can be quoted: the water tank level, airplane autopilots, industrial robots, control of irrigation canals, etc.    

Firstly, it is necessary to dispose of a physical-mathematician model basically described by differential equations. A model has to reproduce with great fidelity the behaviour of the system. Then, it is necessary to design convenient control laws such that the system has some required properties, like as stability, controllability, robustness or reliability, among other important characteristics of any dynamical system.  Sometimes, it will not be possible to achieve all these properties simultaneously. We will have to choose those that are essential in each case and looking for control laws that provide them. With this idea in mind, different kinds of controllers will be designed. Concepts as centralized controller, decentralized control laws, optimal control, predictive control, adaptive control, H-infinity control, etc. appear in a natural form. Obviously, these kinds of controllers can be combined in an appropriate way for each specific model.

In a wide variety of natural and man-made systems, they are composed by a high number of subsystems. For either conceptual or computational reasons, it is advantageous to recognize this fact, and build decentralized controllers. In certain complex control problems it is the only effective way to go. Thus, decentralize control is the control technique most utilized and applied by our research group. Moreover, a considerable effort has been made to known the current control problems which appear in some enterprises with the purpose to adapting and implanting the obtained theoretical results.

CoDAlab, acronym of Control, Dynamics and Applications, is a group of people in the Department of Applied Mathematics III, one of the departments of the Universitat Politècnica de Catalunya (UPC). The group is active in an interdisciplinary intersection of applied mathematics, systems and control theory and engineering. The group covers both theoretical and applied research and has grown through the collaboration with other national and foreign leading research groups. It is composed by people with background in mathematics, physics, control engineering and civil engineering.

Email Address : josep.maria.rossell@upc.edu