Design and Implementation of Fuzzy-Mode-Based Fault Isolation and Fault-Tolerant Control for Aircraft Electric Braking Systems

This paper addresses the fault isolation, estimation, and fault-tolerant control scheme for the aircraft electric anti-skid braking system (EABS) in the presence of actuator and sensor faults. First, the inherently nonlinear dynamics of EABSs are represented by a Takagi-Sugeno (T-S) fuzzy model, incorporating immeasurable antecedent variables to capture the time-varying characteristics. Second, based on the output equivalence principle, a fuzzy observer with unmatched antecedent variables is proposed to achieve isolation and estimation of actuator and sensor faults. The designed observer can guarantee the sensitivity to specific faults while enhancing the robustness to disturbances. The estimated fault information is then utilized to develop a fault-tolerant control strategy, ensuring effective fault compensation and tracking performance. Subsequently, the design of separate and integrated frameworks for the estimation and control units is considered, taking their interaction into account to achieve state and fault isolation, estimation, fault compensation, and tracking control. Finally, hardware-in-the-loop experimental results verify the effectiveness and real-time performance of the proposed fault isolation and fault-tolerant control method, demonstrating the practical applicability of the proposed framework. Note to Practitioners—The aircraft anti-skid braking system (ABS) is crucial for ensuring the safety during landing, taxiing, and other ground movements. This paper focuses on developing reliable fault isolation and fault-tolerant control strategies to maintain ABS performance and efficiency in the presence of faults. The proposed approach employs a fuzzy model to analyze the effects of various faults on system outputs, enabling precise fault isolation and estimation for simultaneous multiple faults. The reconstructed fault information is then integrated to enhance the fault-tolerant control mechanism. This ensures that braking performance can be maintained, even in the presence of multiple simultaneous faults, thereby enhancing system robustness and safety. Moreover, the proposed strategy holds potential applications in other safety-critical domains, such as rail transportation and aerospace vehicles. Future research will explore the integration of historical data to further enhance the accuracy of the fault diagnostic and accommodation units.