All of this technology is aimed to help
improve the process. “The opportunities
for ef ciency are substantial,” claims
Rob Zemenchik, global product marketing
manager of the company’s Advanced
Farming System (AFS), which markets a
range of sensing technologies.
One of the two user trials showing
the bene ts was at the Gallo winery
in California, where the autonomous
guiding system technology NHDrive was
installed in Case New Holland T4.110F
vineyard tractors (with cab) performing
maintenance and crop production tasks.
FIVE LEVELS
At the 2018 Louisville, Kentucky farm
show, the company broke down agricultural
automation into ve levels. The rst is
automatic guidance, in which a navigation
system (its own is branded Accuguide)
steers the tractor to help reduce the
likelihood of missing areas or processing
them twice. The system, which has been
available for 20 years, relies on a guidance
correction signal broadcast from one or
several radio base stations to a receiver
mounted on the roof of the tractor, or
possibly satnav signals. That information
guides steering actuators through what it
calls ‘deferential corrections’. It can be
retro tted to manual tractors. A recent
(2017) extension to the system was
AccuTurn, which calculates turning radii for
tractors and attachments.
An example of category two is vehicleto
vehicle communications, such as the
company’s sign-up to ISOBUS Class
3, in which the tractor can control the
implement, which can also feed back to
the tractor (limiting speed, for example).
“This is where industry is coming out
with solutions for customers today,”
comments Lukac.
Category three is operator-assisted
autonomy, in which vehicles operate
themselves, but underneath a human
driver who remains in the vehicle as a
supervisor. Adds Lukac: “The overall theme
here, why there is more autonomy, is really
because the problem in the marketplace
is the lack of skilled labour, not the lack of
labour overall. If you don’t have someone
skilled in the community to drive the
combine or a tractor with a planter, you can
use lower-skill labour” in combination with
the additional technology.
Category four is supervised autonomy,
in which a single person manages
multiple (two or three) vehicles working
independently; perhaps from the cab of
one of the tractors. “This can be done
using a leader-follower concept,” points
out Lukac; most likely the two are working
together. “This is the focus of where
we think we are going to see driverless
tractors; it’s about how to take operators
out of the vehicle,” he concludes. Case
IH is trialling this concept at carrot grower
Bolthouse Farms, using Steiger Quadrac
tractors pulling harrows. Finally, category
ve is completely independent work.
Although the development of roadelement,”
roadgoing
autonomous vehicles may
have captured public attention, in
fact the challenges that agricultural
VEHICLE AUTOMATION
work pose to autonomous vehicles are
sometimes greater than road-going
vehicles, according to Lukac. He states
that the paths that road-going cars will
take have already been developed by
the route planning industry. They are
xed-path plans with a few choices. But
in agriculture, path plans are variable
and based on the implement that the
tractor is pulling, by how many vehicles
there are, and by the shape of the eld.
In obstacle detection, for highway-based
vehicles, everything that they detect is
an obstacle. But in a cultivated eld, that
is not necessarily the case; the tractor
could be facing weeds, a standing crop
or an obstacle, and need to discriminate
between them to take the right action.
This is why the current focus of the
Small Robot Company, a start-up of Ben
Scott-Robinson and Sam Watson-Jones,
is to generate data for farmers with its
scouting robot ‘Tom’. It was due to begin
trials in wheat and barley elds toward the
end of last year, with plans to work toward
serial production this year. The idea is
that this data could then be acted upon
by other robots: the weeder ‘Dick’ and
planting robot ‘Harry’ (pictured, below).
Both are still under development.
“Once you can gather, and understand,
data, acting on it is the most simple
says lead mechanical
element,”www.ied.org.uk 21