• DocumentCode
    3860533
  • Title

    Tuning of a nonanalytical hierarchical control system for reaching with FES

  • Author

    D. Popovic;M. Popovic

  • Author_Institution
    Fac. of Electr. Eng., Belgrade Univ., Serbia
  • Volume
    45
  • Issue
    2
  • fYear
    1998
  • Firstpage
    203
  • Lastpage
    212
  • Abstract
    Point-to-point functional movements involve simultaneous shoulder and elbow joint rotations. In able-bodied subjects these movements are fully automatic, and feed-forward control ensures the synergistic activity of many muscles. Synergy between joint rotations was defined and described as a scaling between joint angular velocities (M. Popovic and D. Popovic, J. Electromyog. Kinesiol., vol. 4, p. 242-53, 1994). Similarly, subjects who can control their shoulder movements may be assisted in reaching tasks by functional electrical stimulation (FES) of elbow extensor muscles. The synergistic control paradigm can be implemented in real-time by employing a hierarchically structured production-rules method. The use of production-rules necessitates the acquisition of knowledge and the assembly of a rule-base. A nonparametric technique was designed for the identification of the rules. The identification process was divided into two phases: determination of the scaling parameters, and determination of the stimulation parameters. The scaling parameters, needed for the coordination of movements, were determined in able-bodied subjects. Those depend exclusively on the initial and target positions of the hand. The number of scalings could be reduced by dividing the workspace into 12 zones. The stimulation parameters, needed for the execution of movements, were determined in subjects with paralyzed elbow extensor muscles by identifying triplets: elbow angular velocity, elbow angular acceleration (velocity increments), and the corresponding pulse durations for various classes of movements and loads attached to the hand.
  • Keywords
    "Control systems","Elbow","Automatic control","Muscles","Shoulder","Angular velocity","Feedforward systems","Neuromuscular stimulation","Assembly","Acceleration"
  • Journal_Title
    IEEE Transactions on Biomedical Engineering
  • Publisher
    ieee
  • ISSN
    0018-9294
  • Type

    jour

  • DOI
    10.1109/10.661268
  • Filename
    661268