Cognata, the world’s leading provider of ADAS and autopilot simulation
software and training data, recently released a new generation of an
agricultural and off-road simulation platform to help agricultural autopilot
vehicles operate more accurately and overcome more challenges, blind spots,
and limitations in field operations.
Autonomous driving has been used in commercial vehicles and public
transportation for only a few years, but its application in agriculture can be
traced back to 20 years ago. Agricultural Vehicles usually work on private land,
or in wilderness or unmanned areas, therefore, the application scenario of
automatic driving is simpler and easier to land. Now, those automatic operation
vehicles loaded with GPS and sensors, or self-driving vehicles that are currently
active in the field of agriculture are various, from planters, transplanters,
harvesters, tractors, and so on.
An unmanned automatic harvester was harvesting in a field in Jining
City, Shandong Province in 2019
XAG released an R80 agricultural unmanned vehicle, which can
automatically avoid obstacles and spray in 2019
Compared with autopilots, agricultural vehicles encounter more uneven roads
and more dead angles. However, since they do not need to drive at high speed,
and the environment is relatively closed, the technical requirements for
automatic driving are not as high, and therefore, easier to be commercialized.
In order to refine the operation and make these advanced operation vehicles
adapt to more and more complex work, the giants of agricultural automatic
driving vehicles have aimed at AI and robot fields, hoping to play a role in more
operation fields, such as transportation docking in operation areas, field
exploration, bricklaying, etc.
Cognata’s AI based agricultural and off-road simulation platform provides a
wider range of application training scenarios for these agricultural automatic
driving vehicles, overcomes various perception problems such as field terrain,
space and path judgment, and visual blind area, and helps agricultural automatic
operation machinery speed up the training and testing of its automatic driving
algorithm, and improves its operation ability.
For agricultural off-road simulation, perception is the key to planning. What are
the challenges of off-road AV stack awareness? To measure the size and angle of
the path and passage to evaluate whether it is passable or not; to detect the
unexpected end of the driveable path; to compensate the displacement of the
sensor relative to the vehicle caused by road turbulence; Limited visibility caused
by other objects, such as tail dust, vegetation and ground blind area, etc; and to
overcome the blind spot caused by the sensor being placed in its shelter. The
excellent performance of Cognata’s agricultural off-road simulation platform in
perception can perfectly cope with various situations of field operation, and
effectively help agricultural automatic driving vehicles to complete the new
generation of the business mission.
Cognata is a leading global supplier of large-scale automotive simulation for the
Advanced Driver Assistance System (ADAS) and autonomous vehicle markets.
Working with leading automotive technology companies around the world,
Cognata’s end-to-end platform accelerates time to market by delivering simulation
solutions for the entire automated driving product lifecycle, from training to
testing to deployment.