This paper studies sensor fault detection and isolation in continuous-time switched systems subject to unknown but bounded output or measurement disturbances. The proposed approach utilizes the asynchronous switching phenomenon, in which there is a lag between subsystem and observer switching. The aim is to design a bank of sliding mode observers for fault isolation, with robustness to residual provided by H∞ performance. For this purpose, we first transform the sensor faults into pseudo-actuator faults using a coordinate transformation with a low-pass filter, making the design of the observer framework more convenient. Then, a series of sliding mode observers is designed for effective fault isolation such that each residual signal is affected by its corresponding sensor fault. Next, to provide an optimal solution in the sense that the effect of measurement disturbances on the residual signal is minimized, the H∞ optimization index is utilized for the problem of asynchronous switched systems. To ensure stability, a piecewise Lyapunov function along with an average dwell time switching constraint is utilized for both matched and mismatched periods of asynchronous switching. The conditions for a feasible solution lead to sufficient conditions that are formulated in terms of linear matrix inequalities. Moreover, the evaluation of the residual signal and threshold is computed to enhance fault detection and isolation capability. Simulations of a three-tank system for multiple sensor faults are utilized to demonstrate the effectiveness of the proposed approach.
This study addresses the trajectory tracking control problem for an underactuated hovercraft subject to additive bias and multiplicative loss-of-effectiveness thruster faults under environmental disturbances. In these systems, actuator degradation structurally breaks the differential flatness mapping, driving nominal controllers to generate control actions that induce severe actuator saturation and cause instability. To resolve this challenge, a hierarchical physics-informed neural adaptive control (PINAC) framework is proposed. First, a gated-recurrent-unit physics-informed neural observer (PINO) is designed to isolate thruster faults from exogenous hydrodynamic disturbances. Second, a constrained Safe-TD3 reinforcement learning agent functions as a supervisor, computing an online dilation factor to slow down the mission timeline, thereby reconfiguring the reference trajectory to accommodate degraded actuator boundaries. Third, a low-level non-singular terminal sliding mode (NTSM) controller is implemented as a tracking-guarantee layer. Unlike classical asymptotic schemes where convergence is only achieved as time approaches infinity, or finite-time controllers where the settling time depends on the initial state, the proposed PINAC framework guarantees practical fixed-time stability, ensuring that the settling-time bound is independent of initial conditions. Simulation results demonstrate that the designed controller prevents actuator saturation, provides smooth trajectory adjustment, and reduces tracking errors under severe composite faults.