On oscillation free controller changes

Abstract Control systems are typically subject to strong, and often conflicting, constraints. One of the techniques that have been extensively investigated consists in using several controllers, designed to have different resource needs, and consequently, to exhibit different performance levels. These controllers are switched dynamically, to obtain the desired quality of control while using the lowest possible amount of resources. However, the research made so far has been essentially focused on the rules for triggering the controller switching, neglecting the full extent that such changes have in the system. In particular, switching controllers often causes output oscillations that may negate the potential performance gains. In this paper, firstly, the cause for oscillations in the presence of period changes is investigated. Then, it is presented a solution, based on a change of basis matrix. The experimental evaluation shows that important performance gains are achieved in key control performance indicators such as overshoot, settling time and error.

[1]  Anton Cervin,et al.  Feedback scheduling of control tasks , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[2]  Enrico Bini,et al.  The Optimal Boundary and Regulator Design Problem for Event-Driven Controllers , 2009, HSCC.

[3]  Anton Cervin,et al.  Optimal Online Sampling Period Assignment: Theory and Experiments , 2011, IEEE Transactions on Control Systems Technology.

[4]  Michael D. Lemmon,et al.  Generalized Elastic Scheduling , 2006, 2006 27th IEEE International Real-Time Systems Symposium (RTSS'06).

[5]  Karl-Erik Årzén,et al.  Feedback–Feedforward Scheduling of Control Tasks , 2002, Real-Time Systems.

[6]  Enrico Bini,et al.  Control-Driven Tasks: Modeling and Analysis , 2008, 2008 Real-Time Systems Symposium.

[7]  Richard H. Bartels,et al.  Algorithm 432 [C2]: Solution of the matrix equation AX + XB = C [F4] , 1972, Commun. ACM.

[8]  Scott A. Brandt,et al.  Experimental evaluation of slack management in real-time control systems: Coordinated vs. self-triggered approach , 2010, J. Syst. Archit..

[9]  Anton Cervin,et al.  Resource management for control tasks based on the transient dynamics of closed-loop systems , 2006, 18th Euromicro Conference on Real-Time Systems (ECRTS'06).

[10]  Paulo Pedreiras,et al.  Adapting the sampling period of a real-time adaptive distributed controller to the bus load , 2005, 2005 IEEE Conference on Emerging Technologies and Factory Automation.

[11]  G. Golub,et al.  A Hessenberg-Schur method for the problem AX + XB= C , 1979 .

[12]  Giorgio C. Buttazzo,et al.  Quality-of-Control Management in Overloaded Real-Time Systems , 2007, IEEE Transactions on Computers.

[13]  Paulo Tabuada,et al.  On the Benefits of Relaxing the Periodicity Assumption for Networked Control Systems over CAN , 2009, 2009 30th IEEE Real-Time Systems Symposium.

[14]  A. Varga.  Robust pole assignment techniques via state feedback , 2000, Proceedings of the 39th IEEE Conference on Decision and Control (Cat. No.00CH37187).

[15]  Pau Marti,et al.  Control Performance Evaluation of Selected Methods of Feedback Scheduling of Real-time Control Tasks , 2008 .

[16]  Giuseppe Lipari,et al.  Elastic Scheduling for Flexible Workload Management , 2002, IEEE Trans. Computers.