A novel system architecture for automated field-based tent systems for controlled-environment agriculture and experimentation

Abstract

Experimentation within the field of agronomy relies upon maintaining a controlled operating environment to determine various environmental factors’ effects upon a crop. These experiments are carried out in small growth chambers and can control limited variables such as light, temperature, and humidity.Space is a premium inside the chambers which limits the capacity for additional sensors and other equipment. The chambers are set to specific environmental parameters and maintained throughout the experimental cycle. Field conditions are more complex than a growth chamber, which makes it difficult to analyze the effect of factors in a more realistic scenario. A system architecture for a field-based controlled environment for agriculture and experimentation is proposed. First, the overall architecture is proposed for integrating a multitude of wired and wireless sensors, different controllers, small unmanned aerial vehicles (UAVs) and unmanned ground vehicles (UGVs), and actuators to assess and maintain environmental variables. Second, each component is detailed for its role and responsibilities within the system. Next, a set of commercially available components are examined and compared for their strengths and weaknesses in the proposed system. Then, scientific applications of the system are proposed and explored.

We prototype and analyze a simplified implementation of the architecture in a wheat heat stress experiment, and demonstrate the capability of our proposed architecture in studying the effect of different environmental conditions on crops in CEA settings. The architecture is simplified to a set of two controllers connected over wireless local area network (LAN): one for heated/experimental tents, and one for the outdoor/control tents; three pairs of such controllers are incorporated into the experiment. A singular Raspberry Pi within each tent implemented the functionality of most of the components in the architecture via software modules. In the case study setup, each pair of controllers communicate with six temperature sensors and one carbon dioxide sensor and continuously maintain the environment by managing relays to actuate a heater element by comparing the environment temperature with a pre-set temperature threshold.