In keeping up with technology in precision agriculture, I often read articles mentioning both drones and IoT as distinct areas of study. While they are certainly two topics that deserve their own focus, this all too often comes at the expense of understanding how drones fit into the IoT space now and in the future.
The primary application for drones in the precision agriculture space to date has been the mapping of fields. This mapping allows a grower to determine what areas of the field may need fertilizer, herbicide, pesticide, or other forms of attention. Currently these images are looked at by an agronomist, a field visit may be scheduled for inspection, and a course of action such as applying fertilizer is done in the traditional manner. In short, the process is still largely manual and not all that different from a generation ago, but with more accurate imaging information. How this scenario plays out in the future is likely to be significantly different using hyperspectral imaging, drones, and IoT.
Hyperspectral imaging, as summarized by reading through the Wikipedia definition, is the process by which images are taken and a numerical value assigned for every pixel. This imaging considers a variety of light wavelengths from visible to infrared. Analytics can then be performed on the images to help determine plant health. Using drones to take hyperspectral images of fields is a growing approach in agriculture. As this method has been researched and improved, people are now able to determine various plant stressors such as drought, pathogen infection, and fungal infection in crops. In fact, in at least one study it was found that hyperspectral image analysis showed it was possible to detect plant pathogens in wheat before it was visible on the plant to anyone visually inspecting infected plants.
While drones are currently used to take hyperspectral images by attaching the appropriate sensor, it’s the future use of drones and IoT connected devices that is particularly exciting. As with any process, hyperspectral image analysis will become more routine, computing power will increase, and the location of computing can go from the desktop or cloud to being done on device. That is to say, the drone itself will have the computational horsepower to analyze the data obtained from the onboard sensor taking the image and determine the problem. At that point, a drone that is network connected can transmit this information to an agronomist for further analysis and consideration, or in the future it may relay that information to another drone. That drone can be tasked with collecting a sample plant and delivering it to an agronomist for analysis. In this scenario, the time it takes to determine the existence of plant pathogens is decreased and the treatment time is sped up. This reduces the cost to the grower in terms of chemicals used for treatment and lost yield due to crop damage. Additionally, with this type of technology an agronomist can support a larger number of acres and save both time and money.
To get a second perspective, consider the scenario of drought. In the event a field has a dry spot, a drone flies over and captures a hyperspectral image. The hyperspectral image analysis can be done on the drone (it could still be uploaded to the cloud for historical purposes). The drone can then communicate over the network to an irrigator. The irrigator, with its remote computing power from onboard electronics, uses the mapping information and connected GPS sensor data on device to automatically plot a path and plan for irrigating the dry land. All done without the need for interaction from an individual except for the occasional software update. While not the reality right now, this level of sophistication is where the future of precision agriculture is headed as it leverages drones and IoT.
Do you have a precision agriculture application that will help growers of the future? If so, contact us and we’ll help make your project a success.