Chattering-free adaptive iterative learning for attitude tracking control of uncertain spacecraft

This paper aims at developing chattering-free adaptive iterative learning for repetitive attitude tracking tasks of spacecraft subject to initial state errors. Thanks to the proposed reduction mechanism in the parametric learning law, iteration-varying initial state errors, as well as the unknown inertia matrix and external disturbances, can be addressed effectively. Moreover, to avoid the chattering phenomenon of the control signals, we introduce an approximation of the sign function. Of note is that this approximation leads to a class of non-negative definite problems. A new analytical method is consequently exploited with a Lyapunov-like theory based on contraction-mapping and composite-energy-function, which rigorously shows the boundedness and convergence of the iterative learning process in the presence of initial state errors and non-negative definite problems. These observations are validated by implementing the proposed chattering-free adaptive iterative learning method to achieve the specific attitude tracking task of an uncertain spacecraft.