Brain-Actuated Speed Modulation for Position Control of an Artificial Robotic Limb for Rehabilitative Application

Abstract

The paper provides a novel approach to design and develop a brain-actuated 2-loop position control with an inner loop to control speed and an outer loop to control position of an artificial robotic limb for neuro-motor rehabilitation. The subject utilizes motor imagery brain signal to drive the robotic limb-motor at a fixed speed until he finds the limb crossing the desired target position. The subject releases an ErrP brain signal, when the robotic limb crosses the target position first time. The brain-computer interface system recognizes the target position by identifying the location on the trajectory of the robot arm, at the instance when ErrP is released by the subject. The position of the robot arm at the said instance corresponds to zero-crossing in positional error. Depending on reaction-time of the subject in releasing ErrP, the positional offset-error usually has a non-zero (variable) magnitude, which needs corrections by the subsequent mode of automatic control. The brain-actuated speed-modulator sets the speed of the limb to half of its value after each zero-crossing, however, with an opposite sign. Novelty of the work lies in brain-actuated controller design and stability analysis of the system using Root Locus theory. The merit of the proposed scheme lies in reduced cognitive load of the subject as he is set free after releasing the ErrP at the first zero-crossing of positional error.