Interest in artificial intelligence is growing every year: in 2011, 67 deals were made with companies that develop AI technologies (Artificial Intelligence), and already in 2015 – about 400. In the first quarter of fiscal year 2016, more than 140 deals were made – a record indicator for the history of the sector.
In 2012, the Autonomous Tractor Cooperation (ATC) showed a prototype of the unmanned Spirit tractor, claiming it was “the first fully unmanned agricultural tractor.” ATC is now continuing to work on improving the prototype. Spirit is equipped with the company’s developed AutoDrive technology, which combines radio navigation and laser gyroscope together with artificial intelligence in agriculture.
Main features
In May 2016, the Russian company Cognitive Technologies tested an unmanned tractor with a computer vision system. The tractor was equipped with video cameras, navigation and inertial GLONASS and GPS sensors and a computing unit.
“The computer vision system allows detecting dangerous objects with high accuracy, determining their size and coordinates for drawing high-precision maps,” says Olga Uskova, President of the company. “Thanks to the accurate knowledge of the position of objects on the field, many of them become possible to remove even before the harvesting stage, when they can pose a real threat to the mechanical elements of agricultural machinery.”
The tractor bypasses objects that cannot be removed during harvesting, and thanks to video cameras and sensors, it receives information about the appearance of new objects in real time. According to Olga Uskova, the technology should be ready for sale to consumers in a year and a half.
In August 2016, CNH Industrial unveiled a self-driving tractor concept based on the existing Case IH Magnum. The tractor can work without the participation of the driver, using lidar (a technology for receiving and processing information about distant objects using systems that use the phenomena of light reflection and scattering – ed.) And a camera in order to detect and avoid obstacles. Self-driving vehicles will help improve precision and productivity in the agricultural sector, the company said.
Blue River Technology launches LettuceBot devices that use machine learning technology to recognize weeds among healthy plants and spray them with herbicides. According to the DataScience UA developers, LettuceBot photographs more than 5,000 plants per minute.
The PlantVillage website has collected a database of more than 50 thousand photos of healthy and diseased plants:
- The goal of the developers is to launch an application where farmers will upload images of diseased plants, and algorithms will determine their diagnosis.
- The publication Wired writes that farmers spend a lot of money and time due to incorrect diagnostics, and artificial intelligence will be able to quickly and accurately determine the problem.
What’s next?
Artificial intelligence technologies will be applied in vertical farms, according to Neil Jacobstein, head of the artificial intelligence and robotics course at Singularity University in Silicon Valley. Vertical farming is the practice of growing crops in greenhouses, where all the necessary conditions are artificially created: light, temperature, humidity
“AI will help optimize crop growing in vertical farms. It will control moisture, heating and atmospheric pressure in greenhouses, ”says Yakobstein.