Thesis supervisor: Katalin Hangos
co-supervisor: Lőrinc Márton
Location of studies (in Hungarian): University of Pannonia, Faculty of Information Technology, Department of Computer Science and Systems Technology Abbreviation of location of studies: PE
Description of the research topic:
Description of the research topic:
To solve the control of complex systems, it is beneficial to apply such modeling strategies that treat the system as a set of subsystems which are physically interconnected. Typical examples are cooperative mobile robot teams, smart-grids or industrial process networks.
In interconnected systems the failure of a single subsystem can induce the faulty behavior of the entire system: due to the fault propagation through the physical interconnections, the fault event also influences the dynamic behavior of the other subsystems.
To achieve a prescribed control goal in complex systems, networked control methods can be applied. In this approach to each subsystem a local controller is attached. These local controllers are able interchange information among each other to guarantee the desired behavior of the entire system.
The fault-tolerant control strategies are meant to assure acceptable performances in the control systems in the presence of a fault event. First the implementation of a fault diagnosis module is necessary, which solves the fault detection, isolation and identification problem. Based on the information provided by the diagnostic module, the nominal controller is reconfigured to assure fault tolerance. In the case of physically interconnected systems, the fault isolation and identification is difficult due to the fault propagation through the physical couplings.
The goal of the research work is the development of new control strategies that assure the fault tolerant behavior of networked control systems. Both the fault diagnosis and fault tolerant control problems in networked systems will be treated. The developed control methods should stand for low communication costs by applying distributed control strategies. During the research both the achievable control performances (theoretical investigation) and the applicability (e.g. fault-tolerant control of multi-robot systems in a virtual-reality environment) of the developed strategies will be examined.
To solve the formulated research problem, Lyapunov methods, the theory of passive systems, estimation methods developed for large-scale systems, synchronization methods for multi-agent systems, distributed optimization methods can be applied.
Preliminary results can be found in the following publications:
F. L. Lewis, H. Zhang, K. Hengster-Movric, A. Das, Cooperative Control of Multi-Agent Systems, Optimal and Adaptive Design Approaches, Communications and Control Engineering, Springer, 2014.
Mogens Blanke, Michel Kinnaert, Jan Lunze, Marcel Staroswiecki, Diagnosis and Fault-Tolerant Control, 2006, Springer-Verlag Berlin Heidelberg.
Further requirements: Expectations of applicants: English language – upper intermediate; basic knowledge in system and control theory. It is an advantage if the applicant has participated at Scientific Students’ Conferences.
Login
