Delayed feedback attitude control using neural networks and Lyapunov-Krasovskii functionals

This paper addresses the regulation control and stabilization problem of spacecraft attitude dynamics when there exists an unknown constant discrete delay in the measurements. Radial basis function neural networks are used to approximate the kinematics and inertial nonlinearities while a back propagation algorithm is employed to update neural network weights. By employing a Lyapunov- Krasovskii functional, a delay independent stability condition is obtained in terms of a linear matrix inequality, the solution of which gives the suitable controller gains. Finally, to show the effectiveness of the proposed controller, a set of simulations are performed and the results of the proposed control strategy are compared with results obtained using the method for delayed attitude control suggested by Ailon et al.¹.