Project Details
Projekt Print View

Optimiertes Codesign von Regler und Scheduling für eingebettete Regelungssysteme mit beschränkten Ressourcen

Subject Area Automation, Mechatronics, Control Systems, Intelligent Technical Systems, Robotics
Term from 2010 to 2013
Project identifier Deutsche Forschungsgemeinschaft (DFG) - Project number 169488262
 
Final Report Year 2013

Final Report Abstract

Various robust control and scheduling codesign strategies with different favors have been proposed in this project. The first strategy was based on the receding-horizon concept in which a finite- or infinite-horizon optimization problem is solved at each sampling instant. It has been shown that the resulting performance using the receding-horizon concept is always no larger than the resulting performance under offline robust control and scheduling codesign without utilization of the receding-horizon concept. The associated computational complexity has been also tackled using specific methods which can be seen as a tradeoff between performance and complexity. The second strategy was based on partitioning the state-space into a number of regions and then searching for Lyapunov-like functions associated with each region which in combination form an overall multiple Lyapunov function. This idea was inspired from the switched systems theory since networked and /or embedded control systems are modeled as discrete-time switched systems. The associated complexity with such state-space partitioning codesign strategy is not an issue as in the first strategy. The third strategy was based on the event-triggering concept in which transmissions of new measurements and computations of control signals are performed only when necessary from a stability or performance point of view. All proposed strategies have been extended to be robust against uncertain communication and computation delays. Furthermore, the closed-loop stability conditions required under each strategy are formulated as linear matrix inequalities which can be solved efficiently. Finally, the proposed strategies in this project have been implemented in MATLAB and encapsulated as toolboxes in which users can use them by simply running a computer program without any knowledge of the theory needed to derive them.

Publications

  • "Optimal control of switched systems with application to networked embedded control systems," Dissertation, Institute of Control Systems, Department of Electrical and Computer Engineering, University of Kaiserslautern, 2011
    D. Görges
 
 

Additional Information

Textvergrößerung und Kontrastanpassung