Prescribed performance fault‐tolerant attitude control for flexible spacecraft under limited communication network

Abstract

Abstract This paper investigates the prescribed performance fault‐tolerant attitude tracking control for flexible spacecraft under limited communication network, where the controller and the actuator are connected via wireless network. The hysteresis quantizer is employed to quantize the control signal, which can reduce the communication burden of the network onboard. First, a novel iterative learning observer is developed by combining with neural‐network approximation method to reconstruct the actuator faults and estimate the unmeasurable nonlinear rigid‐flexible dynamics simultaneously. Then, the signal quantization technique is introduced for control command quantization, and the concerned parametric quantization error is given. Based on the learning observer output, prescribed performance design procedure, and a quantization error compensation method, an active fault‐tolerant control strategy is developed for flexible spacecraft attitude tracking to deal with actuator faults, unmeasurable system nonlinearity, quantization errors, and external disturbances. The stability analysis of the closed‐loop system proves that the quantization errors, modal vibrations, disturbances can be effectively rejected, moreover, the convergence of the transformed performance states and angular velocity states can be guaranteed, which achieves the tracking problem with predetermined transient‐state and steady‐state performance requirements. Finally, a numerical simulation is conducted to illustrate the validness of the proposed control strategy.

Keywords

• attitude control
• fault tolerant control
• iterative learning control
• neural nets