Improving yield for farmers around the world.
For one of the world’s leading agriculture technology companies, helping farmers become as efficient and successful as possible is what drives them. And, for assistance reaching that goal, they turned to RIIS. Together, we developed a custom interface that utilizes satellite and drone imagery to analyze crops all the way down to individual fields and plants. The system makes soil and compound recommendations based on multiple factors, including crop type, yield history, crop growth stage and overall plant population.
The client is a true global leader in the creation, support, and advancement of agricultural hardware, including just-in-time crop health management, crop sensing, nutrition measurement, connectivity and data management. In addition to precision agriculture, their products advance industries such as construction, geo-positioning, mapping, surveying, forestry, mining, utilities, forensics and education.
Their hardware meets RIIS custom software.
Farmers utilize our client’s agriculture hardware and enabled machinery to conduct soil tests, upload crop health status and data, record crop yields and much more to help minimize the use of expensive fertilizers, nutrients, and other compounds. With constant access to their own custom dashboard, farmers can monitor specific fields, crops and even individual plants for moisture and nutrient levels, as well as harvest readiness. They can even automate vehicle steering and location functionality. Next, we help take all this functionality to the next level.
Custom solutions for a one-of-a-kind partner.
The opportunity to assist with this project was truly extraordinary. We immediately got started honing the analysis of nutrient recommendations by painstakingly refining equations based on real world testing and farmer input. The result was a customized program that produces unique recommendations for the specific areas in question on individual properties. In addition to specifying which crop is in the field, farmers can even input many other factors, such as growth stage and the exact location where the field is located anywhere worldwide.
Better information leads to the best results.
All of the variable parameters farmers can now input have a clear impact on final planting, maintenance and harvest recommendations. Additionally, soil test results can be uploaded and analyzed almost instantly. They can test for specific compounds like sulphur, potassium, nitrogen and other nutrients that can significantly change the environment in which the crops grow. With nutrient management being the key to maintaining healthy crops until harvest, as well as producing bountiful yields, this information is invaluable.
Images from the clouds. Servers in the cloud.
By utilizing drone and satellite imagery based on field boundary coordinates sent from Topcon, we were able to map specific fields with stunning accuracy. Then, testing and processing is all handled in the cloud by our servers and reported directly back for data aggregation, reporting, and year-over-year analysis. Farmers can continue to improve their yield from season to season. And all of us at RIIS can continue to solve the next great challenge.
Key technologies used.
From a technology perspective we used a Python-based website using Flask with OAuth for authentication. For database management we went with PostgreSQL. Virtual Machines from Azure were utilized in the cloud. For queuing up async jobs we trusted RQ Worker. We also worked with the client to install a back-end API for external calls. For the user interface, layouts, buttons and grids the choice was Semantic. And lastly, analytics, automatic code reviews and security are so important that we trusted SonarQube.
How did we do it?
We work in one way, as efficiently as humanly possible. To meet client needs and expectations, Scrum and Agile methodologies were used extensively with daily scrums and two-week sprints. We also instituted reviews and retrospectives at the end of every sprint, planning session and backlog grooming. There were 12 total sprints for this project, and they were executed by two different team configurations over a period of seven months.
It takes the best data.
To map fields as accurately as possible, we experimented with multiple sampling algorithms, from distance-based options centered around input points, to averaging around different areas of the field until we perfected the system. The engine to process all the jobs was fully dependent on the custom equation editor we created in the user interface. It allows for user-defined constants that will be used across potentially many equations, as well as the ability to use the output of an equation within another equation. A given equation includes data such as type of test being conducted, specific crops being grown in a given field boundary, crop growth stage, exact location worldwide and much more.
Key Services Provided:
- Agile Methodology
- Web Development