To be predictable, precise and to align our actions with insights gathered from agricultural data, we have Microsoft Azure FarmBeats.
Microsoft FarmBeats is a data-driven solution for agriculture. It combines the benefit of AI and the Internet of Things (IoT) to provide customized data based on geographic location by leveraging the cloud computing benefit of Microsoft Azure. FarmBeats allows farmers to build digital agricultural solutions. This digital solution needs lots of input data for providing a good agricultural solution. Some of the key data come by analyzing the key driving factors of productivity in farming like soil type, soil moisture, wind speed, wind direction, temperature, rainfall, atmospheric pressure, and sunlight. These data need to be collected and sent to Microsoft FarmBeats. For data collection sensors, drones, satellites, etc. are used. FarmBeats uses AI and IoT to process these data collected and build an agricultural solution for the farmers.
Current agriculture methodology
Due to a lack of technology farmers use the traditional mechanism of farming. They just feel if this ground is dry, I need to water it. So, a lot of farming is based on guesswork. Most of the decisions that a farmer takes about when to water, when to apply pesticides, when to apply nutrients, fertilizer is based on guesswork.
So, instead of using guesswork, if a farmer uses the knowledge coupled with data and data-driven insights, agriculture can be more productive, more cost-efficient, and it’s also better for the environment. That’s what is mean by data-driven agriculture.
Need for technology in agriculture
Agriculture is the backbone of the economy of most countries. There was always a need to leverage innovative technologies to increase productivity. The global population is on the boom and is expected to reach 9.8 billion by 2050 which is predicted by the United Nations. This would need the world’s farmers to grow about 70% more food than is now produced. There is limited land, the population is increasing and due to global warming, unpredictable climate changes are occurring which in turn are decreasing the productivity. There is a major need to leverage technology to increase productivity. For this Microsoft, FarmBeats will serve as a boon. The International Food Policy Research Institute has researched and has predicted that technologies like Microsoft FarmBeats which provides robust data analytics to the agriculture sector can increase the productivity by 67% and also would decrease the resource usage.
Enabling data-driven agriculture with FarmBeats
To get the data about the soil, there are various ways. One of the ways is to get sensors in the farm. Like some of these sensors that Microsoft partners make like Teralytic, Davis Instruments. These sensors can go inside the soil and gather large amounts of data. For example, the sensor has about 26 different types of sensors. It measures nutrients and moisture. The sensor from Davis measures wind and humidity and a lot of other parameters like what’s happening in the farm at that point, at that location. One of the challenges, of course, is to make the sensors communicate to the Azure FarmBeats. To mitigate connectivity problems all the processing of data is done by farmers’ local machine and then uploaded periodically to IoT Base Station. Microsoft FarmBeats leverages the farmers’ internet and TVWS(TV white spaces — the radio frequencies allocated to broadcasting services) to send the data to IoT Base Station which then sends to FarmBeats gateway and then to the Azure FarmBeats Datahub. These data are available on the FarmBeats app and can be accessible to farmers all over the world.
The above sensor solution won’t tell what’s happening 10 meters away. So, if for example, the problem was what is the soil moisture level 6 inches below the soil throughout the farm? You would need a sensor every 10 meters. But putting one of those sensors every 10 meters is expensive to deploy and to manage. So the key technique that has been built as a way to combine ground sensor data with aerial imagery so that with very few sensors and with aerial imagery from drones or robots or satellites maximum necessary data can be gathered. You can then train a machine learning model to then start making predictions in other parts of the farm, where you do not have sensors. The drone collects orthomosaic data and sends it to Azure FarmBeats Datahub. As there are multiple images taken by the drones, these images get stitched and processed by the drone software. Then these images are uploaded by Drones software to IoT Base Station by leveraging TVWS and farmers’ internet connectivity which in turn is uploaded to Azure FarmBeats Datahub. Similarly, satellite images are also collected, stitched and uploaded to Azure FarmBeats. These image data get processed by AI/ML by leveraging Azure FarmBeats Datahub API’s to create a solution for increasing productivity. This solution is available on the Azure FarmBeats app to be used by the farmers.