EkoNiva is an agricultural holding, and Cognitive Pilot is a software corporation. These two Russian companies are embarking on the world’s largest autonomous driving project for agricultural machinery. Together they aim to equip up to 10% of the country’s grain harvester fleet of 100,000 machines with autonomous driving systems during the coming three years.
Cognitive Pilot autonomous steering systems
The companies have recently inked an agreement, under which EkoNiva takes over promotion and sales of Cognitive Pilot’s Cognitive Agro Pilot autonomous driving systems. This system features an AI-based autonomous driving system known as C-Pilot. The agreement includes installation, adjustment, servicing, and engineering maintenance of the self-driving grain harvesters. The project is planned for 35 Russian regions in 10 climate zones.
“Such large-scale maintenance work on servicing combines spread over a large territory will allow EkoNiva and Cognitive Pilot to establish the world’s most comprehensive video database for the further training of neural systems. These systems are already being deployed on combines not only in Russia but also in the USA and Latin America, China and other countries,” said Olga Uskova, CEO of Cognitive Pilot.
In the coming months, all technical specialists of EkoNiva’s service department in Russia will undergo specialised training to work with the autonomous driving system.
Autonomous driving to make harvesting more efficient
“Equipping our customers’ machines with Cognitive Agro Pilot systems will make harvesting more efficient and reduce the production costs of grain for them by 3 to 5%. Because we are represented in large parts of the country and run an extensive network of modern service centers we will be able to quickly scale up the use of artificial intelligence technologies in Russia”, said Bjorne Drexler, CEO of EkoNivaTehnika-Holding.
Autonomous driving system to work with all brands
The system is designed to work with all brands of grain harvesters. It could be installed both on brand new machines and the existing fleet of used combines. It is still mandatory for an operator to stay in the cabin during harvesting. However, the system allows the operator to focus more on the quality of the harvesting process and better control the angle of the header and other settings.
Cognitive Agro Pilot is an AI-based autonomous driving system for agricultural machinery, which includes grain harvesters, tractors, and sprayers. It’s designed to let machinery operators focus on the quality of harvesting while leaving the robot assistant to run the machinery itself.
Just one video camera
The system analyses images from just one video camera and by using a convolutional neural network designed for agronomic purposes understands the types and positions of objects facing the machinery, builds the correct trajectory of the combine, and sends commands to perform maneuvers. This sets Cognitive Agro Pilot apart from foreign solutions, which generally use a whole set of sensors in their models, like laser scanners for moving along the field’s edge, stereo cameras for windrows, etc.
Cost savings
“The use of that system cust harvest losses in half. With our system of computer vision, only one camera is needed to achieve the same results other leading suppliers achieve with three or four sensors, including lidar,” said Olga Uskova.
The system is securing safe autonomous movement in both plowed or unplowed ground and mown or unmown fields. According to the company, it is able to detect obstacles, such as other machinery, trees, roads, animals, and people.
“The autonomous driving system is always running with the same accuracy. A human is working with one level of accuracy at the beginning of his shift and with a different one when he gets tired. Plus, we all know that drivers have different skills, so some work better than others, while the system is working perfectly all the time,” commented Yuri Minkin, director of the autonomous driving machinery department of Cognitive Pilot.
Fuel savings
“The increased accuracy also leads to fuel savings,” he added. “The camera sees in the same way as a human eye does. It passes information on to the convolutional neural network designed for agricultural tasks. This system determines the types and positions of objects, builds a movement trajectory, and adjusts operation if needed.”
Conversion cost
The conversion cost reportedly is estimated at $ 9,000 per combine, although the final price depends on numerous different factors, according to the company. So far, the system is designed only to navigate combines across the field. Still, in the future Cognitive Pilot aims to develop a comprehensive solution that would make combines driverless altogether.
Rusagro Group is ahead of the curve
Russian agricultural giant RusAgro Group is planning to install Cognitive Pilot’s autonomous driving system on 242 combines for the 2020/21 harvesting campaign. The company is one of Russia’s biggest pork and sugar producers, operating a land bank of 665,000 hectares.
The equipment to be installed on Rusagro’s harvesters includes an automatic control unit called Agrodroid, a video camera, a control display screen, and a set of connecting cables. Rusagro plans to equip itsentire fleet of 800 combines with autonomous driving systems during the next few years.
“The use of an autonomous control system during harvesting will minimise the risks of negative human factors and will optimise the use of combines,” said Roman Shkoller, top manager of Rusagro.
The Cognitive Agro Pilot system has already been successfully installed and tested in the United States, Brazil, and China, as well as in several Russian regions, including the Tomsk Region, Kurgan Region, Belgorod Region and the Republic of Tatarstan.
