Human-actuator collaborative control by a novel frequency-division technique for linear maneuverability of control moment gyroscopic actuators

Abstract

Control of control moment gyroscopic actuators (CMGs) has been used for regulating the attitude of satellites, ships, vehicles, robots, and even humans for a long time. Generation of directional torques often causes the singularity problem that is strongly dependent on the direction of the momentum vector. Cooperative performances of controlling the magnitude and direction of CMGs can be achieved by both a human operator and a programmed system controller. A human-actuator collaborative control framework for CMGs is proposed by introducing a novel frequency-division control technique, which controls the magnitude and the direction of the torque separately in the frequency domain. The magnitude is controlled by the programmed system controller in the high frequency and the direction is controlled by a human operator in the low frequency. To realize such a frequency-division collaboration method between an actuator and a human operator, two CMGs are developed, and one is collaboratively controlled by another CMG manipulated by a human operator like a haptic device. The goal is to manipulate the direction of the momentum by transferring human inputs to the CMG. Defining impedance models of both actuators forms the bilateral transformation matrix that provides haptic and manipulation capabilities. The proposed method is verified through the experimental studies of two CMGs.