Finite-Horizon Active Fault Isolation and Identification for Spacecraft Attitude Control Systems With Multiple Constraints

Abstract

This article proposes a finite-horizon active fault isolation and identification scheme for spacecraft attitude control systems. Considering initial state uncertainties, noises, and diverse faults, the linearized discrete-time model of spacecraft attitude kinematics and dynamics is built. The attitude mandatory and forbidden constraints for respectively achieving observation missions and ensuring operational safety are modeled as linear constraints. Then, we formulate the finite-horizon active fault isolation problem for the spacecraft attitude control systems under different faults with the consideration of attitude constraints and angular velocity constraints. Applying the robust optimization method and the Karush–Kuhn–Tucker (KKT) conditions, the overall finite-horizon active fault isolation problem is converted into a tractable mixed-integer quadratic programming (MIQP), which is solved to obtain an optimal input sequence for active fault isolation. Furthermore, to identify the actual fault of the spacecraft and fulfill the systematic active fault diagnosis (AFD) frame, we also propose a fault identification approach based on the acquired input sequence and corresponding output data. Finally, the effectiveness and superiority of the proposed finite-horizon active fault isolation and identification mechanism are illustrated through numerical simulations of spacecraft attitude control systems under various faults.