Alan K. Mackworth's Publications

Sorted by DateClassified by Publication TypeSorted by First Author Last NameClassified by Author Last Name

Synthesis of Hybrid Constraint-Based Controllers

Y. Zhang and Alan K. Mackworth. Synthesis of Hybrid Constraint-Based Controllers. In Hybrid Systems II, pp. 552–567, Springer-Verlag, Berlin, February 1995.

Download

[PDF]217.6kB  

Abstract

A robot is an integrated system, with a controller embedded in its plant. We take a robotic system to be the coupling of a robot to its environment. Robotic systems are, in general, hybrid dynamic systems, consisting of continuous, discrete and event-driven components. We call the dynamic relationship of a robot and its environment the behavior of the robotic system. The problem of control synthesis is: given a requirements specification for the behavior, and given dynamic models of the plant and the environment, generate a controller so that the behavior of the robotic system satisfies the specification. We have developed a formal language, Timed Linear Temporal Logic (TLTL) [17], for requirements specification. We have also developed a semantic model, Constraint Nets [19], for modeling hybrid dynamic systems. In this paper, we study the problem of control synthesis using these representations. Control synthesis in general is difficult. We first focus on a special class of requirements specification, called constraint-based specification, in which constraints are associated with properties such as safety, reachability and persistence. Then we develop a systematic approach to synthesizing controllers using constraint methods, in which controllers are embedded constraint solvers that solve constraints in real-time. Finally, we consider hierarchical control structures, in which the higher levels embody digital/symbolic event-driven control derived from discrete constraint methods and the lower levels incorporate analog control based on continuous constraint methods. We illustrate these techniques using a robot soccer player as a running example.

BibTeX

@InProceedings{HSII95,
  author =	 {Y. Zhang and Alan K. Mackworth},
  title =	 {Synthesis of Hybrid Constraint-Based Controllers},
  year =	 {1995}, 
  month =        {February},
  booktitle =	 {Hybrid Systems II},
  Editor =       {P. Antsaklis et al.},
  publisher =    {Springer-Verlag}, 
  series =       {Lecture Notes in Computer Science},
  volume =       {999},
  address =      {Berlin},
  pages =         {552--567},
  abstract =	 {A robot is an integrated system, with a controller embedded in its plant. We take a robotic system to be the coupling of a robot to its environment. Robotic systems are, in general, hybrid dynamic systems, consisting of continuous, discrete and event-driven components. We call the dynamic relationship of a robot and its environment the behavior of the robotic system. The problem of control synthesis is: given a requirements specification for the behavior, and given dynamic models of the plant and the environment, generate a controller so that the behavior of the robotic system satisfies the specification. We have developed a formal language, Timed Linear Temporal Logic (TLTL) [17], for requirements specification. We have also developed a semantic model, Constraint Nets [19], for modeling hybrid dynamic systems. In this paper, we study the problem of control synthesis using these representations. Control synthesis in general is difficult. We first focus on a special class of requirements specification, called constraint-based specification, in which constraints are associated with properties such as safety, reachability and persistence. Then we develop a systematic approach to synthesizing controllers using constraint methods, in which controllers are embedded constraint solvers that solve constraints in real-time. Finally, we consider hierarchical control structures, in which the higher levels embody digital/symbolic event-driven control derived from discrete constraint methods and the lower levels incorporate analog control based on continuous constraint methods. We illustrate these techniques using a robot soccer player as a running example.},
  bib2html_pubtype ={Refereed Conference Proceeding},
  bib2html_rescat ={},
}

Generated by bib2html.pl (written by Patrick Riley ) on Wed Apr 23, 2014 19:08:34