Simulation-Based Design Space Exploration of Swarm Cyber-Physical Systems Using Contracts
John B. Finn, Pierluigi Nuzzo, Alberto Sangiovanni-Vincentelli

Citation
John B. Finn, Pierluigi Nuzzo, Alberto Sangiovanni-Vincentelli. "Simulation-Based Design Space Exploration of Swarm Cyber-Physical Systems Using Contracts". Talk or presentation, 29, October, 2014; Poster presented at the 2014 TerraSwarm Annual Meeting.

Abstract
Swarm cyber-physical systems are distributed control systems characterized by complex and nonlinear plant dynamics that are tightly coupled with the behavior of a control algorithm. In the design and optimization of such systems, the impact of continuous dynamics and transient behaviors can no longer be neglected. A scalable design methodology that can efficiently deal with these interactions is missing. The goal of this project is to develop algorithms for efficient and optimal co-design of plant architecture and distributed control algorithms for swarm cyber-physical systems modeled as hybrid systems. We define a library of hybrid component models using Modelica to accurately capture the behaviors of the physical plant. Then, we leverage assume-guarantee contracts to formalize system requirements and enable the generation of monitors to analyze simulation traces during optimization. Our framework allows efficient exploration of design tradeoffs between architecture sizing and control parameters, while ensuring that the specifications are satisfied. We demonstrate our approach on the sizing of the heat exchanger and the low-level proportional control of an aircraft air management system.

Electronic downloads

Internal. This publication has been marked by the author for TerraSwarm-only distribution, so electronic downloads are not available without logging in.

Citation formats

• HTML

  John B. Finn, Pierluigi Nuzzo, Alberto
  Sangiovanni-Vincentelli. <a
  href="http://www.terraswarm.org/pubs/442.html"
  ><i>Simulation-Based Design Space Exploration of
  Swarm Cyber-Physical Systems Using
  Contracts</i></a>, Talk or presentation,  29,
  October, 2014; Poster presented at the 2014 TerraSwarm
  Annual Meeting.

• Plain text

  John B. Finn, Pierluigi Nuzzo, Alberto
  Sangiovanni-Vincentelli. "Simulation-Based Design Space
  Exploration of Swarm Cyber-Physical Systems Using
  Contracts". Talk or presentation,  29, October, 2014;
  Poster presented at the 2014 TerraSwarm Annual Meeting.

• BibTeX

  @presentation{FinnNuzzoSangiovanniVincentelli14_SimulationBasedDesignSpaceExplorationOfSwarmCyberPhysical,
      author = {John B. Finn and Pierluigi Nuzzo and Alberto
                Sangiovanni-Vincentelli},
      title = {Simulation-Based Design Space Exploration of Swarm
                Cyber-Physical Systems Using Contracts},
      day = {29},
      month = {October},
      year = {2014},
      note = {Poster presented at the 2014 TerraSwarm Annual
                Meeting},
      abstract = {Swarm cyber-physical systems are distributed
                control systems characterized by complex and
                nonlinear plant dynamics that are tightly coupled
                with the behavior of a control algorithm. In the
                design and optimization of such systems, the
                impact of continuous dynamics and transient
                behaviors can no longer be neglected. A scalable
                design methodology that can efficiently deal with
                these interactions is missing. The goal of this
                project is to develop algorithms for efficient and
                optimal co-design of plant architecture and
                distributed control algorithms for swarm
                cyber-physical systems modeled as hybrid systems.
                We define a library of hybrid component models
                using Modelica to accurately capture the behaviors
                of the physical plant. Then, we leverage
                assume-guarantee contracts to formalize system
                requirements and enable the generation of monitors
                to analyze simulation traces during optimization.
                Our framework allows efficient exploration of
                design tradeoffs between architecture sizing and
                control parameters, while ensuring that the
                specifications are satisfied. We demonstrate our
                approach on the sizing of the heat exchanger and
                the low-level proportional control of an aircraft
                air management system. },
      URL = {http://terraswarm.org/pubs/442.html}
  }
  

Posted by John B. Finn, IV on 5 Nov 2014.
Groups: tools