Abstract：In the design of unmanned aerial vehicles, the control surfaces and servos are connected via a flexible linkage device. When traditional proportional integral differential (PID) methods are used for control, mismatches between the parameters and the system model can cause control surface flutter, and parameter tuning is challenging, resulting in poor adaptability to different scenarios. To address this issue, this paper establishes a dual-inertia system mathematical model for the electric servo system of unmanned aerial vehicles and analyzes the causes of control surface flutter. An extended state observer is employed to estimate and compensate for the total disturbance of the servo system in real time. A nonlinear state observer for state error is designed, and an engineering-practical active disturbance rejection controller is developed. By upgrading the control algorithm of the servo system product and validating it through bench tests, the response results of the improved servo system are obtained. The results show that the designed controller not only rapidly tracks position commands but also avoids resonance, thereby eliminating control surface flutter.
2026, Vol. 27 
