Both parties of the agreement expect production of fully self-driving tractors to start by 2024. The companies note that the main advantage of the joint solution is that when AI controls the key processes, it allows farmers to achieve a fundamentally different level of quality and efficiency.

Besides autonomous driving, in its current version, the Cognitive Agro Pilot system makes it possible to automatically transmit progress reports to the messenger apps of tractor owners, as well as to transmit telematic data about the state of the machine and progress of work with photographic and video data. Based on the information from the smart tractor sensors, the system makes a comprehensive analysis of soil and crop conditions, such as assessment of vegetation, disease damage and more.

"With a huge arsenal of breakthrough solutions based on artificial intelligence, Russia has all the chances to become a trendsetter in the global agritech market," says Sergey Serebryakov, CEO of Peterburgsky Traktorny Zavod. "Our joint solution with Cognitive Pilot will allow local and foreign farms to significantly increase their efficiency, save resources and move to a fundamentally new level of farm management.

“Together with the Ministry of Agriculture of the Russian Federation, we expect to promptly develop and adopt a set of needed measures to support farmers who purchase such innovative solutions that will ensure the proper pace of our project development and will also have a significant impact on the dynamics of the entire domestic agricultural industry."

Olga Uskova, CEO of Cognitive Pilot, says, "By combining the efforts of two market leaders in agricultural machinery and artificial intelligence, we will be able to create a truly unique and breakthrough product that is widely in demand, not only in Russia, but in foreign markets as well. It’s important that the schedule for our project calls for fully self-driving tractors to appear on the market by early 2024, which will obviously precede the appearance of industrial self-driving cars."

The signed agreement between Peterburgsky Traktorny Zavod and Cognitive Pilot sets out three main phases of the project and the production of three versions of Cognitive Agro Piot agricultural machinery control system for tractors.

The robot-tractor with the first version of Cognitive Agro Pilot, currently installed on PTZ machinery, is capable of following a trajectory in autonomous mode, automatically controlling the optimum speed, gear settings, slippage and engine load, stopping before obstacles and operating in low visibility conditions.

An important feature of the Cognitive Agro Pilot solution for tractors will be the presence of an additional high-precision positioning system (GNSS) in the steering system, in addition to the computer vision technologies. This is needed for operations in areas where no visual guidance points are available and farm fields are demarcated. It also enables automatic interpretation of satellite images, plotting of harvesting patterns, automatic machine turns and driving directions, all of which can be corrected by the computer vision system.

Testing of autonomous tractors with the first version of Cognitive Agro Pilot is expected to be completed by the end of October 2021. Serial equipment of Kirovets K-7M tractors with the autopilot system is expected to start in February 2022.

In the second phase of the project the functionality of the system will be expanded with automatic turns, automatic calculation of the best route for work, refining of the course by visual landmarks (edges of cultivated soil, technological tracks, etc.), automatic control of a given set of trailed implements (ploughs, mowers, harrows, sprayers, etc.), as well as automated video analysis and control functions (rock mapping, assessment of vegetation, damage to plants from diseases or pests). Finally, the tractor will be equipped with side-view sensors to monitor lateral areas on the move.

The third phase of the project envisages a fully self-driving tractor control mode without the need for human involvement. It includes group autonomous operation with a leader, the movement of a group of self-driving tractors on non-public roads, the autonomous movement of a robo-tractor paired with a combine harvester not equipped with an autonomous control system, and the use of an expanded set of farm implements with automated control and functionality to monitor their condition during operation.

Cognitive Pilot is an autonomous driving technology joint venture of Sber and Cognitive Technologies group. The company specializes in the development of AI-based solutions for ground transport automation and machine vision. The Cognitive Pilot /C-Pilot/ multi-platform autonomous driving system is used in agriculture (for combine harvesters, tractors and sprayers), rail transport (for road switchers, locomotives and trams) and city transport (LRVs, cars, and buses). Besides software the company also develops its own NextGen radar systems and sensors.

Cognitive Agro Pilot is an autonomous AI-based driving system for farming equipment such as grain combine harvesters, tractors, and sprayers. It includes an autonomous unit to manage farm equipment, a video camera, a display, a set of connecting cables, and other control system elements. The user interface is available via an Android mobile app for smartphones or rugged tables. From June to October 2020, over 350 New Holland, John Deere and Claas autonomous combines equipped with the Cognitive Agro Pilot system farmed over 160,000 hectares of field and harvested more than 720,000 tons of crops.

With the help of Cognitive Agro Pilot, as many as 590,000 metric tons of grain crops (wheat, soybeans, barley, oats, sorghum, buckwheat, etc.) were harvested from over 130,000 hectares; and some 130,000 metric tons of row crops and roll crops (corn, sunflower, etc.) were harvested from over 30,000 hectares in various regions across Russia. The Cognitive Agro Pilot system allows to reduce the time of harvesting by an average of 25% and reduce direct crop losses from 8 to 13%. The daily output of the machine operator increases by up to 25%.

In 2021, the number of system installations has already exceeded 700 units.