Composite Adaptive Attitude-Tracking Control With Parameter Convergence Under Finite Excitation

Abstract

This brief presents a new class of adaptive controllers for attitude-tracking control problems of rigid bodies. The most important feature of the proposed method is that both instantaneous state data and past measurements (historical data) are introduced into the parameter-adaptation process. Filtered regressor matrices and states are employed in the control formulation, which lay an important foundation for the precision acquirement of historical data and render the resulting parameter-adaptation dynamics to reside within a stable and attracting manifold. A specially designed information matrix is further introduced to encode the composite information into the adaptive law. Under this formulation, state-tracking errors, as well as parameter estimation errors, are guaranteed to converge asymptotically to zero subject to the satisfaction of a finite excitation condition, which is a significant relaxation when compared with the persistent excitation condition that is typically required for these classes of problems. A noncertainty-equivalence term is also used in the adaptation process to ensure the regulation of the tracking error in the absence of finite excitation conditions. Numerical simulations and hardware-in-loop experimental results are illustrated to evaluate the various features of the proposed method.