Design and Experimental Evaluation of a Modular and Extensible Flight Control Board for Unmanned Aerial Vehicles

Abstract

Commercial flight control units used in unmanned aerial vehicles are typically delivered as closed, black-box platforms, which restricts their utility for academic research and rapid prototyping. This study presents the design, fabrication, and multi-stage evaluation of a modular and extensible flight control board for a four-rotor unmanned aerial vehicle. The board is built around a high-performance microcontroller with a hardware floating-point unit and a real-time operating system that schedules sensor acquisition, signal processing, and data transmission as concurrent tasks. A six-degree-of-freedom dynamic model of the vehicle is derived and used to design a cascade proportional-integral-derivative controller, in which an outer loop regulates attitude and a faster inner loop regulates angular rate. Raw inertial measurements are conditioned through a hybrid signal-processing chain combining an 81 Hz infinite impulse response (IIR) Biquad low-pass filter with a recursive state estimator, suppressing vibration-induced noise while limiting phase lag. On the fabricated hardware, the implemented IIR Biquad filter reduced the root-mean-square noise of the gyroscope signal by 60.2\% (from 1.002$^\circ$/s to 0.399$^\circ$/s) while adding a 2.78 ms phase delay. In 6-DoF numerical simulations, the cascade controller achieved rise times below 0.77 s for the roll and pitch axes and eliminated steady-state error. Comprehensive disturbance analyses demonstrated that the controller maintains stable closed-loop behavior, restricting peak attitude deviation to under 1$^\circ$ during a 3.0 N$\cdot$m external wind disturbance torque. The telemetry subsystem sustained a stable 100 Hz communication link with a 0.18\% packet loss rate and an 18 ms latency, confirming its suitability as a transparent, modular, and reproducible platform for advanced avionics research.