Arquitectura IoT para la automatización agrícola: captura del conocimiento ancestral y gestión en entornos de conectividad limitada

Abstract

AAgriculture is a fundamental sector in developing countries such as Ecuador, where it is still practiced in a traditional way with limited access to technologies that optimize production. IoT-based solutions have shown potential to automate tasks such as irrigation, fertilization, and environmental monitoring; however, they often rely on constant connectivity and energy supply, which limits their effectiveness in rural areas with deficient infrastructure. Moreover, these solutions usually focus solely on automation, without considering mechanisms that preserve the ancestral knowledge of farmers. This work aimed to design a distributed software architecture for IoT applications in agricultural contexts with limited connectivity, oriented toward task monitoring and the preservation of ancestral knowledge. The proposal includes: (1) detecting agricultural practices through environmental data captured by sensors, (2) recording farmers’ decisions and techniques, and (3) correlating both sources to digitalize ancestral knowledge, validate patterns through machine learning, and generate adaptive recommendations that integrate traditional wisdom with precision agriculture. The methodology applied was RUP, and the architecture was documented using Kruchten’s 4+1 views model, also applying SOLID principles. A prototype was implemented with Flutter and Isar for the mobile application, Angular for the web environment, RabbitMQ for asynchronous messaging, Spring Boot with PostgreSQL for the central server, and a local node developed in Python on Raspberry Pi. The results demonstrate autonomy under intermittent connectivity, fault tolerance, scalability, energy efficiency, and a modular, resilient, and replicable solution that promotes technological equity and sustainability in rural contexts.