Introduction

Overview

This project aims to develop a decentralized automation system for Controlled Environment Agriculture (CEA).

The system follows a modular structure, where each module has its own subnetwork of accessory devices, such as sensors and actuators, that autonomously monitor and control environmental conditions without relying on a central hub.

The system prioritizes interoperability, enabling the integration of devices from different manufacturers and the replacement of existing ones without major updates. It also implements security measures to ensure safe data transmission and access control, maintaining system functionality and scalability.

Motivation

Advancements in IoT-based monitoring and control systems have improved automation in agriculture, increasing efficiency and productivity.

Precision Agriculture enhances field farming by using sensor data, GPS, and automated equipment to precisely manage inputs like water, fertilizers, and pesticides. Controlled-Environment Agriculture extends this approach by eliminating limitations such as external environmental factors, soil variability, and large-scale deployment challenges through enclosed systems where temperature, humidity, light, and CO₂ are actively regulated. This allows for year-round production, reduces reliance on weather conditions, and ensures consistent crop yields, making it more reliable than traditional open-field farming.

However, many farming automation startups struggle to succeed, primarily due to the high cost of research and development. Building advanced automation systems requires substantial investment, which raises food production costs and makes it difficult for startups to compete with traditional farms that operate at lower expenses.

A major factor driving these costs is the reliance on centralized servers or cloud resources for data processing and control. This setup requires complex communication infrastructure, increases operational expenses, and complicates scalability, particularly for larger farms. Additionally, centralized systems create a single point of failure - if the central server or cloud service fails or is compromised, the entire operation can be disrupted. For example, an AWS IoT outage in 2020 caused significant downtime for applications relying on AWS IoT Core, demonstrating the risks of centralized dependency.

Another issue is the difficulty of integrating components from different manufacturers. Automated farms use a variety of sensors, actuators, and equipment, but differences in APIs and communication protocols can make it challenging to connect these systems smoothly.

Discussion

The system leverages the Matter protocol, which ensures compatibility between devices from different manufacturers. This standardisation simplifies integration, replacement, and future expansion of devices. However, the high costs associated with Matter certification – including a $7,000 annual membership fee for the Connectivity Standards Alliance (CSA) and $3,000 per product certification – pose significant challenges for small manufacturers and farms operating on tight budgets.

Future advancements in this system could include the adoption of microfluidics within hydroponic systems. Such developments would enhance resource efficiency and allow for precise control over environmental conditions tailored to specific crops. These innovations align with the system’s modular design philosophy, which prioritises adaptability and scalability in agricultural automation.