Advanced

Merging Real-Time and Control Theory for Improving the Performance of Embedded Control Systems

Cervin, Anton LU (2004)
Abstract
This report describes the work carried out within the research project ``Merging Real-Time and Control Theory for Improving the Performance of Embedded Control Systems''. The overall objective of the work has been to develop integrated control and scheduling methods for improving the performance of real-time control systems with limited resources. The work has fallen into three categories. First, overrun methods for control tasks has been investigated. Specifically, a reservation-based scheduling concept called the Control Server has been further developed, and control experiments on a ball-and-place process have been performed. Second, the issue of jitter in real-time control systems has been explored. The concept of Jitter Margin has... (More)
This report describes the work carried out within the research project ``Merging Real-Time and Control Theory for Improving the Performance of Embedded Control Systems''. The overall objective of the work has been to develop integrated control and scheduling methods for improving the performance of real-time control systems with limited resources. The work has fallen into three categories. First, overrun methods for control tasks has been investigated. Specifically, a reservation-based scheduling concept called the Control Server has been further developed, and control experiments on a ball-and-place process have been performed. Second, the issue of jitter in real-time control systems has been explored. The concept of Jitter Margin has been introduced as a link between control stability theory and scheduling theory. In this context, best-case response-time analysis under earliest-deadline-first scheduling has been researched. Third, some development work on the S.Ha.R.K. real-time kernel has been performed. The rate-monotonic and earliest-deadline-first scheduling modules have been extended, and new modules for the elastic task model and the control server model have been implemented. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Book/Report
publication status
published
subject
publisher
Department of Computer Engineering and Systems Science, University of Pavia, Italy
language
English
LU publication?
yes
id
59c890ad-f8ef-4ba4-9050-423064d7eca3 (old id 8602633)
alternative location
http://www.control.lth.se/documents/2004/cer04rep.pdf
date added to LUP
2016-02-15 18:34:00
date last changed
2016-04-16 09:12:05
@techreport{59c890ad-f8ef-4ba4-9050-423064d7eca3,
  abstract     = {This report describes the work carried out within the research project ``Merging Real-Time and Control Theory for Improving the Performance of Embedded Control Systems''. The overall objective of the work has been to develop integrated control and scheduling methods for improving the performance of real-time control systems with limited resources. The work has fallen into three categories. First, overrun methods for control tasks has been investigated. Specifically, a reservation-based scheduling concept called the Control Server has been further developed, and control experiments on a ball-and-place process have been performed. Second, the issue of jitter in real-time control systems has been explored. The concept of Jitter Margin has been introduced as a link between control stability theory and scheduling theory. In this context, best-case response-time analysis under earliest-deadline-first scheduling has been researched. Third, some development work on the S.Ha.R.K. real-time kernel has been performed. The rate-monotonic and earliest-deadline-first scheduling modules have been extended, and new modules for the elastic task model and the control server model have been implemented.},
  author       = {Cervin, Anton},
  institution  = {Department of Computer Engineering and Systems Science, University of Pavia, Italy},
  language     = {eng},
  title        = {Merging Real-Time and Control Theory for Improving the Performance of Embedded Control Systems},
  year         = {2004},
}