Abstract: The aperiodic event-triggered prescribed performance control problem based on reduced-order observer is studied for a class of switched nonlinear systems with unknown control directions and arbitrary switching signal. First, based on sampling quantization output, a new reduced-order neural network quantized observer is designed to estimate unmeasurable states and reduce system observation costs. Next, a novel aperiodic event-triggered control strategy dependent on switching signals is proposed to overcome the assumption of synchronous switching between controllers and corresponding subsystems in existing literature. And it successfully avoided the adverse effects of asynchronous operation of factor systems and matching controllers on system stability in conventional triggering mechanisms. The output feedback controller obtained through backstepping design ensures the boundedness of all variables in the closed-loop switched systems under arbitrary switching, and the system tracking error trajectory can converge to any small neighborhood near the origin, achieving arbitrary switching aperiodic event-triggered control for switched nonlinear systems with unknown control directions. Finally, simulation analysis verified the correctness and effectiveness of the proposed method.