Nonlinear Predictive Control applied to Spacecraft Attitude Control

A class of iterative methods have recently been proposed for the path planning for a variety of fully and under-actuated mechanical systems, including robots and spacecrafts. These methods all involve the basic idea of warping an initial path iteratively to an acceptable final path by using a Newton-type algorithm. Once a path is found off-line, a feedback controller can then be used to follow the path. A modification of the off-line methods transforms them directly into a nonlinear predictive feedback controller, with guaranteed closed-loop asymptotic stability when the system model is known, and certain robustness when the model information is imperfect. This method represents a special class of model predictive control (MPC), since the control action at each time instance is determined based on the future predicted trajectory. Preliminary results are presented illustrating the application of this nonlinear controller to three-axis attitude control of fully-actuated and underactuated spacecraft.