• DocumentCode
    488010
  • Title

    Analysis of Two Step Learning Systems

  • Author

    Judd, R.P. ; Hideg, L.M.

  • Author_Institution
    Oakland University, Department of Electrical and Systems Engineering, School of Engineering and Computer Science, Rochester, Michigan 48309-4401
  • fYear
    1989
  • fDate
    21-23 June 1989
  • Firstpage
    2327
  • Lastpage
    2333
  • Abstract
    Learning systems are evolving into a significant method to allow physical systems to track a desired trajectory. Frequency domain analysis has contributed to a better understanding of the behavior of learning systems. This paper proposes a method which uses the errors from both the previous trial and the second trial back. This law yields better convergence and model error tolerance than single step laws.
  • Keywords
    Computer errors; Computer science; Convergence; Filters; Frequency domain analysis; Learning systems; Manufacturing; Robots; Systems engineering and theory; Trajectory;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    American Control Conference, 1989
  • Conference_Location
    Pittsburgh, PA, USA
  • Type

    conf

  • Filename
    4790577