Skip to content

eGreenhouse

Kenneth Kang edited this page Apr 28, 2020 · 8 revisions

eGreenhouse | Updates | Github

Project Leads: Ariel Stroh, Kenneth Kang, Isaac Salchenberg, and Liam Duncan

Overview

eGreenhouse is a lightweight and low-cost greenhouse sensor package integrated with the OPEnS HyperRail. The sensor package travels through the greenhouse and transmits a live stream of spatial environmental data to the internet.

Description

Advances in gas sensors and open-source hardware (e.g., Arduino) are enabling new options for low-cost and light-weight gas sampling devices that are also robust and easy to use and duplicate. Oregon State University’s Nursery Research and Extension and the Openly Published Environmental Sensing (OPEnS) Lab, have developed the ‘HyperRail’ a combined gantry and sensor, imaging, phenotyping package. A modular smart greenhouse in-a-box system that can potentially be configured for a wide variety of sensing, phenotyping, and automation purposes. The 3D HyperRail moves in three directions moving over a one-meter rail on the x-axis continuously, two-meter on the y-axis, and one-meter on the z-axis, with the sensor package attached on the z-axis in a water-resistant container. This device can be programmed to visit a suite of locations to gather information from the on-board sensor package such as high-accuracy measurements of temperature, relative humidity, luminosity, CO2, and dust/pollen concentrations. The system is assembled from off-the-shelf products, with a total cost of less than $300. The system is deployed on the OPEnS Lab HyperRail, a modular rail conveyance system. Validation of the sensor package is conducted in a greenhouse.

Objectives

  • Collect and analyze spatial data in greenhouses to provide useful metrics to growers.

  • Lower the cost of expensive sensors by condensing multiple sensor locations into one moving one.

  • Adjustable framing capability.

  • Adjustable length in all axis.

  • Carry a sensor package continuously throughout the frame.

  • Carry a sensor package to any desired coordinate location in all three axes.

Outcomes

eGreenhouse Installation The figure above shows time series results from a five-day deployment in an OSU greenhouse. Temperature, RH, and Luminosity trends match expectations following the diurnal cycle. This graph highlights the temperature dependence of the O2 sensor, a problem which requires further investigation. Diurnal oscillations in CO2 due to photosynthesis are below the tolerance of this sensor (± 30 ppm ± 3 % of measured values) and therefore cannot be detected in this model. Deployment in a greenhouse with higher plant density, such as commercial vegetable production, will likely render more relevant CO2 results.

Future

The eGreenhouse is being deployed with the new 2 and 3-dimensional HyperRail. This will allow the collection of multi dimensional spatial data and coverage of a larger greenhouse area with one sensor package. The addition of new sensors to measure ndvi and soil moisture will allow for fine-grained health metrics for individual plants. Finally, integration with Spool and Loom will simplify the code base and allow online data visualization and remote device logging and device management for greater ease of use.

Resource List

See the eGreenhouse GitHub for technical information.

See the eGreenhouse GitHub Wiki for information.

Tutorials

See the eGreenhouse GitHub Wiki for information.

Keywords

CO2 sensor, O2 sensor, greenhouse monitoring

References

Clone this wiki locally