Continuous Position Control of 1 DOF Manipulator Using EMG Signals

One of the main problems of robot manipulations by using bio-signal interface is discontinuous position control. This discontinuity causes undesirable effects on users such as lack of accuracy and an overall degraded performance. This paper presents the control strategy for the continuous position control of 1 DOF manipulator using electromyography (EMG) signals. We designed the control system which makes possible that 1 DOF manipulator replicate the wrist movements by estimating the wrist joint angle continuously from EMG signals measured from the four muscles. In order to estimate the joint angle, we transformed raw EMG signals to quasi-tensions, which represent the actual muscle tension, using low-pass filtering. A supervised multi-layer neural network trained by a back-propagation algorithm which estimates a joint angle from quasi-tensions, so that the proposed system controls the manipulator continuously with the obtained time-varying angle. The proposed methodology can be applied to improve the control strategy for tele-operated robotic manipulator.