• DocumentCode
    3208513
  • Title

    Artificial neural network controller for multi-link structures

  • Author

    Styer, Daniel L. ; Vemuri, V.

  • Author_Institution
    Biomed. Eng. Graduate Group, California Univ., Davis, CA, USA
  • fYear
    1995
  • fDate
    5-7Jan 1995
  • Firstpage
    37
  • Lastpage
    42
  • Abstract
    Adaptive critic and direct search methods are used to learn control of a two-link structure. The performance differences between algorithms is more pronounced using the two-link structure than using the cart-pole system. It is shown how the direct search methods can be treated as reinforcement learning methods
  • Keywords
    genetic algorithms; learning (artificial intelligence); neurocontrollers; search problems; adaptive critic method; artificial neural network controller; direct search methods; multi-link structures; reinforcement learning methods; two-link structure; Adaptive control; Artificial neural networks; Biomedical engineering; Control systems; Feedback loop; Learning; Neurofeedback; Programmable control; Search methods; State feedback;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Industrial Automation and Control, 1995 (I A & C'95), IEEE/IAS International Conference on (Cat. No.95TH8005)
  • Conference_Location
    Hyderabad
  • Print_ISBN
    0-7803-2081-6
  • Type

    conf

  • DOI
    10.1109/IACC.1995.465871
  • Filename
    465871