Citation
Andreas Thuy. "Towards flexible and robust cyber-physical-systems through self organization". Talk or presentation, 16, February, 2011; Poster presented at the Ninth Biennial Ptolemy Miniconference, Berkeley, CA.

Abstract
Today's real-time systems always run in the worst case scenario. A lot of built-in knowledge, such as sensor sampling rates and operating system task priorities, determined at design time to work in the worst case are included in the embedded software. Some of these systems have the ability to switch between a few operating modes. However, these modes are designed (additional work) for specific situations envisioned at design time. Aiming for cyber-physical-systems (CPSs), it must be possible to make decisions at run-time to react online to wanted system changes or failures. PtolemyII supports designers in the development process of CPSs by allowing the combination of different models of computation from different domains as well as through the hierarchical actor-oriented design paradigm. Regarding timing behavior, it is usually hard to guess from the model how timing behavior might change, when application goals coupled with physics change. For example, when reduced requirements allow slower sensor sampling rates, traffic on busses could be reduced. This can be of advantage in emergency situations with broken communication links, where messages can be routed differently than what was originally planned. The ability to adapt to new situations at run-time is necessary on the way towards the next generation of flexible, robust hard real-time systems, the CPSs. Self-organization is presented as a means to achieve robustness. To allow system components to self-organize behavior according to given goals, the needed information must be made available, which has impact on the models of computation and the communication.

Electronic downloads

• p14ThuyPtolemyMiniconferencePosterThuy.pdf · application/pdf · 248 kbytes

• HTML

  Andreas Thuy. <a
  href="http://chess.eecs.berkeley.edu/pubs/829.html"><i>Towards
  flexible and robust cyber-physical-systems through self
  organization</i></a>, Talk or presentation,  16,
  February, 2011; Poster presented at the <a
  href="http://ptolemy.eecs.berkeley.edu/conferences/11"
  >Ninth Biennial Ptolemy Miniconference</a>,
  Berkeley, CA.

• Plain text

  Andreas Thuy. "Towards flexible and robust
  cyber-physical-systems through self organization". Talk
  or presentation,  16, February, 2011; Poster presented at
  the <a
  href="http://ptolemy.eecs.berkeley.edu/conferences/11"
  >Ninth Biennial Ptolemy Miniconference</a>,
  Berkeley, CA.

• BibTeX

  @presentation{Thuy11_TowardsFlexibleRobustCyberphysicalsystemsThroughSelf,
      author = {Andreas Thuy},
      title = {Towards flexible and robust cyber-physical-systems
                through self organization},
      day = {16},
      month = {February},
      year = {2011},
      note = {Poster presented at the <a
                href="http://ptolemy.eecs.berkeley.edu/conferences/11"
                >Ninth Biennial Ptolemy Miniconference</a>,
                Berkeley, CA.},
      abstract = {Today's real-time systems always run in the worst
                case scenario. A lot of built-in knowledge, such
                as sensor sampling rates and operating system task
                priorities, determined at design time to work in
                the worst case are included in the embedded
                software. Some of these systems have the ability
                to switch between a few operating modes. However,
                these modes are designed (additional work) for
                specific situations envisioned at design time.
                Aiming for cyber-physical-systems (CPSs), it must
                be possible to make decisions at run-time to react
                online to wanted system changes or failures.
                PtolemyII supports designers in the development
                process of CPSs by allowing the combination of
                different models of computation from different
                domains as well as through the hierarchical
                actor-oriented design paradigm. Regarding timing
                behavior, it is usually hard to guess from the
                model how timing behavior might change, when
                application goals coupled with physics change. For
                example, when reduced requirements allow slower
                sensor sampling rates, traffic on busses could be
                reduced. This can be of advantage in emergency
                situations with broken communication links, where
                messages can be routed differently than what was
                originally planned. The ability to adapt to new
                situations at run-time is necessary on the way
                towards the next generation of flexible, robust
                hard real-time systems, the CPSs.
                Self-organization is presented as a means to
                achieve robustness. To allow system components to
                self-organize behavior according to given goals,
                the needed information must be made available,
                which has impact on the models of computation and
                the communication.},
      URL = {http://chess.eecs.berkeley.edu/pubs/829.html}
  }
  

Posted by Christopher Brooks on 18 Feb 2011.
Groups: ptolemy
For additional information, see the Publications FAQ or contact webmaster at chess eecs berkeley edu.