Implementación de un payload basado en microcontrolador para cálculo a bordo de CubeSats.
No Thumbnail Available
Date
2026
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
El diseño y fabricación de placas electrónicas para la implementación de algoritmos de prueba con el propósito de realizar payloads para sistemas espaciales como CubeSats es de suma importancia para el desarrollo satelital de una organización estatal, privada o académico. Sin embargo, este tipo de avances no se ha presentado en el ámbito universitario. Por esto, en la presente memoria se propone el diseño e implementación de un PCB para el desarrollo de algoritmos de prueba y de identificación de imágenes. Para esto se utilizarán softwares de desarrollo de componentes electrónicos, desarrollo de código y herramientas para la caracterización de parámetros eléctricos como potencia, corriente y voltaje, en conjunto con el desarrollo de un código para la realización de on-board processing.
El trabajo realizado permitió el diseño, fabricación e implementación junto a una selección de sensores para el desarrollo de algoritmos de prueba para la medición de parámetros relevantes en el área espacial, tales como, temperatura, corriente, voltaje, aceleración y giroscopio. Además, se realizó un algoritmo de análisis de imágenes, donde mediante la detección de los colores de cada pixel que componen una imagen, se logró determinar la cantidad de pixeles blancos presentes. Esto como preámbulo para la implementación de sistemas de aprendizaje como redes neuronales, para la detección de nubes, con el propósito de solucionar el cuello de botella producido entre los segmentos terrestres y espacial producto de la limitación en el download bandwith.
Con los resultados obtenidos se tienen las siguientes conclusiones relevantes.
•La iteración y trabajo en conjunto con los fabricantes de placas electrónicas es relevante al momento de realizar diseños y fabricar PCBs, debido a la poca experiencia en el desarrollo de estos sistemas. Por otro lado, la retroalimentación producto del trabajo entre áreas distintas de estudio permite la optimización de diseños al momento de fabricar.
•La selección de sensores mediante la utilización de sus parámetros relevantes (sensibilidad, resolución, rango y consumo) es necesaria al trabajar con sistemas complejos como satélites, donde la potencia total es limitada y los rangos de operación son bajos.
•La utilización de microcontroladores de bajo costo, tales como la Raspberry Pi Pico, son útiles en el desarrollo de prototipos y para la realización de algoritmos de prueba. Sin embargo, esto conlleva a un hardware de menor capacidad computacional, lo que a su vez no permite el desarrollo de algoritmos complejos o la utilización de grandes archivos de datos.
The design and fabrication of PCBs for the implementation of test algorithms with the purpose of creating payloads for space systems like CubeSats is relevant for the satellite development of a state, private, or academic organization. However, this kind of advancement has been stuck in a university setting. Therefore, this thesis proposes the design and implementation of a PCB for the development of test and image identification algorithms. For this, electronic component development software, code development and tools for the characterization of electrical parameters such as power, current and voltage will be used, altogether with the development of code for on-board processing. The work carried out involved design, fabrication and implementation, along with a selection of sensors, for the development of test algorithms to measure relevant aerospace parameters, such as temperature, current, voltage, acceleration and gyroscope readings. Furthermore, an image analysis algorithm was developed where; by detecting the colors of each pixel composing an image, the number of white pixels present was determined. This serves as a preamble for the implementation of machine learning systems, such as neural networks, for cloud detection, with the purpose of addressing the bottleneck between ground and space segments caused by the limitations in download bandwith. The results obtained lead to the following relevant conclusions: •Iteration and collaborative work with electronic board manufacturers is relevant when designing and fabricating PCBs, due to the limited experience in developing these systems. Furthermore, the feedback from working across different areas of study allows for the optimization of designs during fabrication. •The selection of sensors using their relevant parameters (sensitivity, resolution, range and consumption) is necessary when working with complex systems like satellites, where total power is limited and operating ranges are low. •The use of low-cost microcontrollers such as the Raspberry Pi Pico, is useful in prototyping development and for running test algorithms. However, this leads to hardware with lower computational capacity, which in turn does not allow for the development of complex algorithms or the use of large data files.
The design and fabrication of PCBs for the implementation of test algorithms with the purpose of creating payloads for space systems like CubeSats is relevant for the satellite development of a state, private, or academic organization. However, this kind of advancement has been stuck in a university setting. Therefore, this thesis proposes the design and implementation of a PCB for the development of test and image identification algorithms. For this, electronic component development software, code development and tools for the characterization of electrical parameters such as power, current and voltage will be used, altogether with the development of code for on-board processing. The work carried out involved design, fabrication and implementation, along with a selection of sensors, for the development of test algorithms to measure relevant aerospace parameters, such as temperature, current, voltage, acceleration and gyroscope readings. Furthermore, an image analysis algorithm was developed where; by detecting the colors of each pixel composing an image, the number of white pixels present was determined. This serves as a preamble for the implementation of machine learning systems, such as neural networks, for cloud detection, with the purpose of addressing the bottleneck between ground and space segments caused by the limitations in download bandwith. The results obtained lead to the following relevant conclusions: •Iteration and collaborative work with electronic board manufacturers is relevant when designing and fabricating PCBs, due to the limited experience in developing these systems. Furthermore, the feedback from working across different areas of study allows for the optimization of designs during fabrication. •The selection of sensors using their relevant parameters (sensitivity, resolution, range and consumption) is necessary when working with complex systems like satellites, where total power is limited and operating ranges are low. •The use of low-cost microcontrollers such as the Raspberry Pi Pico, is useful in prototyping development and for running test algorithms. However, this leads to hardware with lower computational capacity, which in turn does not allow for the development of complex algorithms or the use of large data files.
Description
Tesis presentada para optar al título de Ingeniero/a Civil Aeroespacial.
Keywords
Láminas (Ingeniería), Satélites artificiales, Microcontroladores, Sensores