Desenvolvimento de um controlador de voo para aeromodelo com aquisição de dados e análise dinâmica

Abstract

This work presents the development of a flight controller for a model aircraft based on the STM32F103CBT6 microcontroller, chosen for its high processing capability and low cost, integrating the MPU-6050 inertial sensor for attitude angle measurement and an SD card logging module for in-flight data acquisition. The proposed embedded architecture is capable of reading PWM signals from the radio control receiver, generating PWM outputs for actuators (roll and pitch), and performing sensor fusion through a nonlinear complementary filter, providing realtime estimates of pitch and roll angles with low latency. The STM32 implementation followed best practices for configuring peripherals, timers, I2C and SPI buses, as well as FATFS integration, enabling a structured and efficient approach to microcontroller programming. The system records the aircraft’s inputs and outputs in CSV format for later application of system identification techniques, aiming to obtain the transfer function representing the aircraft dynamics. The proposed method includes bench testing for electronic and functional validation, controlled excitation tests, and flight evaluations, following established procedures for dynamic system analysis. Preliminary results indicate that the prototype shows stable reading of multiple PWM channels, consistent attitude estimation, and reliable data logging, providing a promising basis for controller tuning and dynamic analysis of the model aircraft. Additional flight tests will be carried out to confirm the technical feasibility, low cost, and reproducibility of the acquisition–processing–analysis workflow, aiming at its application to other unmanned aerial platforms.