Multi-microcontroller sensor data acquisition to enhance UAV attitude control, predictive maintenance, and healthcare solutions

Abstract

This paper studies multi-microcontroller data acquisition systems in various industries, such as UAV attitude control, predictive maintenance, and healthcare solutions, specifically human-fall detection. We use state-of-the-art microcontrollers, L475 and L485I, from the STM32 family to capture sensor data. We propose a five-layered data acquisition architecture, which includes the development of a graphical user interface to make data collection more efficient. We then pre-process the collected data before feeding it to machine learning models and deep neural networks. We use a decision tree classifier for UAV stability control, a gradient boosting classifier for human fall detection, and a multi-layer perceptron classifier for predictive maintenance. The results show the robustness and reliability of the proposed architecture, and it offers promising implications for optimizing maintenance practices, providing safer options, and enhancing patient care.