<p>This paper addresses the control problem of a nonlinear singularly perturbed switched system (SPSS) by designing a filtered backstepping controller. The proposed approach integrates a high-gain filter to reduce system complexity while maintaining stability and performance. To further enhance the robustness of the system against external disturbances, a neural network disturbance observer (NNDO) is employed. The NNDO leverages the approximation capability of neural networks to estimate and compensate for unknown disturbances in real time, ensuring improved disturbance rejection. The combination of the high-gain filter and NNDO significantly reduces the computational burden associated with traditional backstepping methods, making the control strategy more practical for real-world applications. The Lyapunov method, incorporating average dwell time, has been applied to establish the conditions for ensuring closed-loop stability during mode transitions in the system. The effectiveness of the proposed control methodology is validated through its application to a single-link manipulator.</p>