HUMAN INTERFACE USING SURFACE ELECTROMYOGRAPHY SIGNALS

We have been working toward the construction of a forward model arm em ploying a neural network model with electromyographic signals as the c ontrol input. We have succeeded in estimating 1) the joint torque duri ng isometric contraction in a plane from the electromyographic signals , 2) the path of motion from the degree of joint acceleration as well as 3) the path of motion from the joint torque and 4) the position in three-dimensional space. In this paper, we present a new human interfa ce employing a model of an arm, robot control of an artificial hand, a nd the learning of motion capability. In addition, based on the fact t hat we can estimate the position of the arm in three-dimensional space , we discuss as an example of the human interface the results of gener ating motion from electromyographic signals in a hypothetical arm usin g a neural network model.