Tools & data

News

Actionable insights with FluroSat analytics engine

Australian FluroSat sets out to develop an interoperable agronomic analytics engine that can combine locally- and remotely-sourced agricultural data.

The Australian agricultural data startup FluroSat and industry partners Agworld, PCT Agcloud, CSIRO, and AgLink Australia, have been awarded a CRC-P project grant to support and build an interoperable agronomic analytics engine, reports AgFunderNews.

Extension to FluroSat’s FluroSense platform

The engine will be built as an extension to FluroSat’s FluroSense platform (see video below this article) and should be available worldwide in 2021. It is to combine locally- and remotely-sourced agricultural data. The goal is then to detect crop stress and generate nutrient and chemical application recommendations tailored to each farm’s unique cropping history, operations, soil and climatic conditions.

Text continues underneath tweet

Cross-collaboration

According to Anastasia Volkova, CEO of FluroSat, the FluroSense engine will accommodate the various datasets that exist now but currently, reside in “silo’s.” "Observations and farm records in farm management software like Agworld enable cross-collaboration by both agronomist and farmer," she told AgFunderNews.

"PCT Agcloud is well experienced in transforming machine and field data sets into robust data layers that are currently viewable in farm management software such as Agworld. Modeling information by CSIRO is also a well-utilized tool by agronomists to help understand the likely impact weather, moisture, cropping history and nutritional inputs have on projected yield."

At the moment, if agronomists and farmers want to create actions based on available data sets, this creates the need for multiple browser tabs, says Volkova.

Text continues underneath image

Right now, it is impossible for an agronomist to systematically analyze all data sets and generate a recommendation in context with all this data, says FluroSat. Irt's new engine aims to change that. - Photos Jan Willem Stad
Right now, it is impossible for an agronomist to systematically analyze all data sets and generate a recommendation in context with all this data, says FluroSat. Irt's new engine aims to change that. - Photos Jan Willem Stad

Ad hoc analytical approach

"When a remotely captured image suggests that one area of a field may be low in Nitrogen, the agronomist has to open multiple software platforms, and use an ad hoc analytical approach to decide what the best recommendation is to make," Volkova says.

According to her, that makes is impossible for an agronomist to systematically analyze all data sets and generate a recommendation in context with all this data.

Text continues underneath image

The FluroSense engine coupled with CSIRO knowledge (through existing decision tools or models) is to contextualize data layers and carry out advanced analytics and modeling to create actionable insights.
The FluroSense engine coupled with CSIRO knowledge (through existing decision tools or models) is to contextualize data layers and carry out advanced analytics and modeling to create actionable insights.

Well informed recommendation

FluroSat claims that the project will bring data layers that currently reside with the project partners - AgLink, Agworld, and PCT Agcloud. "By doing so, FluroSense can engineer a well informed (by data) and modeled (insights from CSIRO) recommendation based on weather forecast, historical information and real-time observations. A truly contextualized actionable insight."

FluroSense engine

The FluroSense engine coupled with CSIRO knowledge (through existing decision tools or models) is to contextualize data layers and carry out advanced analytics and modeling to create actionable insights.
Observations made by agronomists (in Agworld) and additional processed data (from PCT Agcloud) will be introduced into the analytics engine to continue building “intelligence” creating scalable, automated, accurate and contextualized insights for farmers and their agronomist.
According to FluroSat, the use of machine learning techniques and data sets such as soil characteristics, identification of crop stress will enable FluroSense to automatically adapt to every farm’s unique management history and climatic conditions. "This can make the system truly scalable across the crop types, geographies, and crop management scenarios," writes FluroSat CEO Anastasia Volkova on AgFunderNews.

The video underneath shows a short impression of the FluroSat platform

Or register to be able to comment.