Energy-Optimal Attitude Control Strategies with Control Moment Gyroscopes

In this work, an optimal spacecraft maneuver planner is developed for rest-to-rest attitude transfers using single gimbal control moment gyroscopes (CMGs). In contrast to conventional optimization approaches developed using simplified dynamical models, this work examines the optimal performance and unique control strategies available to a variable speed CMG array under comprehensive physical models for its dynamics and power consumption. This formulation employs a dynamical model which preserves the array's (conservative) momentum exchange dynamics, a power model directly tracking the usage of the individual CMG motors, and typical operational safety constraints on input saturation, angular velocity, and camera exclusion cones. On average, the optimal control strategies produced under this comprehensive formulation present a 35% reduction in mean required electrical energy and a 44% reduction in maneuver time over the classic singularity robust (SR) control law. These improvements are observed to correlate with several specific control behaviors. To extend these improvements to practical spacecraft design restrictions, suggestions on how to reproduce these behaviors using existing feedback control methods are provided.