Listen to this article

Not only does autumn in the Northern Hemisphere mean the return of pumpkin-flavored food and drink; it also provides an opportunity for agricultural robotics companies to demonstrate technical progress. Cognitive Pilot said that its Cognitive Agro Pilot system has harvested more than 720,000 tons of crops across 35 regions of Russia.

Moscow-based Cognitive Pilot is a technology joint venture of Sberbank and Cognitive Technologies Group. The company recently announced a three-year agreement with EkoNiva Holding Co. to add Cognitive Pilot’s autonomous driving system to farm equipment and create a service network across Russia.

“Controlled by the Cognitive Agro Pilot system, machines have worked over 230,000 hours, passing over 950,000 km [590,303 mi.] this year,” said Olga Uskova, CEO of Cognitive Pilot. “These high figures prove the setup is reliable and called-for among farmers. [Our] partnership with Sberbank and leading agricultural holdings allows us not only to scale the use of technology efficiently, but also to guarantee that our clients enjoy high-quality local service.”

Cognitive Agro Pilot designed to add autonomy to agriculture

Cognitive Agro Pilot is an artificial intelligence-based driving system for farming equipment such as grain combine harvesters, tractors, and sprayers. The system analyzes images from a single video camera. It uses a deep learning convolutional neural network tuned for agronomic purposes to understand the types and positions of objects facing the machinery, said the company. Cognitive Agro Pilot then builds movement trajectories and sends commands to perform maneuvers.

Cognitive Agro Pilot includes an autonomous unit to manage farm equipment, a video camera, a display, a set of connecting cables, and other elements of the control system. The user interface is available via an Android mobile app for smartphones or ruggedized tablets, said Cognitive Pilot.

Cognitive Pilot claimed that its system’s comprehension of its surroundings offers an advantage over other approaches that use multiple tools, such as GPS navigation with real-time kinematic (RTK) positioning for their control models.

The edge capture when controlling a harvester by the Cognitive Agro Pilot system is stable at no more than 10 cm (3.9 in.), which helps farmers avoid unnecessary passes and fuel losses, said the company. With the autopilot mode on, the setup avoids obstacles and stops when detecting a threat of collision with other equipment or people, it said.

“Equipping harvesters with autonomous control systems has increased the efficiency of harvesting significantly,” stated Bjorne Drexler, first deputy CEO at EkoNivaTekhnika-Holding. “Considering the wide dealer network of EkoNivaTekhnika-Holding and the fact that the system can be installed on any combine, regardless of the manufacturer’s brand and model, we can confidently say we’ve taken part in what is currently the largest industrial project aimed at agtech robotization.”

“Combine operators were among the first to appreciate the capabilities of Cognitive Agro Pilot,” he said. “The system really helps in their work, reduces the load, and allows you to focus on other harvesting parameters. All this enhances the efficiency of business processes and saves resources.”

Cognitive Pilot scales to industrial use

From June to October 2020, more than 350 New Holland, John Deere, and Claas combines equipped with the autonomy system farmed over 160,000 hectares (395,000 acres) of fields.

With the help of Cognitive Agro Pilot, as many as 590,000 metric tons of grain crops — such as wheat, soybeans, barley, oats, sorghum, and buckwheat — were harvested over 130,000 hectares (321,000 acres). About 130,000 metric tons of row crops and roll crops such as corn and sunflower were harvested over 30,000 hectares (74,000 acres).

Equipment using Cognitive Agro Pilot operated in the Kaliningrad Region, Kaluga Region, Kursk Region, Belgorod Region, Tambov Region, Penza Region, Rostov Region, Tomsk Region, Kurgan Region, Krasnodar Territory, Krasnoyarsk Territory, and Stavropol Territory.

Cognitive Agro Pilot enabled stakeholders to save money this harvesting season because of less consumption of fuel and other related materials, shorter harvesting time or machine hours, and reduced equipment depreciation, said Cognitive Pilot. It also extended the active use of equipment before capital expenditures, cut the number of human errors, and optimized business processes for total savings of over $6.5 million, the company said.

Cognitive Pilot said its clients include Russia’s largest agribusinesses, such as EkoNiva, the Resource Group of Companies, Steppe Agroholding, Yug Rusi GC, Tkachev Agrocomplex, Peschanokopskaya Agrarnaya Gruppa, and Yugo-Vostochnaya Agrogruppa. Other customers include Hyundai Mobis, Russian Railways, Transport Systems PC, and major international vehicle makers. It also recently partnered with Chinese smart city technology firm Fitsco.

“A few years back, artificial intelligence in agriculture was a technology much talked about, but there were no practical examples of its widespread adoption in Russia,” said Anatoly Popov, deputy chairman of the executive board at Sberbank. “Now, once the first harvesting campaign with smart combines is over, we can say with confidence that the future is becoming a reality before our eyes, and what’s especially pleasing is that this is happening with the direct participation of Sber.”

“Russia is one of the leading countries in terms of agricultural production and is a key exporter of wheat,” he added. “Therefore, given the scale of production, the Cognitive Agro Pilot system has the potential to save the domestic agro-industrial complex millions of dollars.”

According to the estimates of project members, in the next three years, every tenth harvester in Russia may become autonomous.