Right: Universal Robots
has been a pioneer in
the development and
adoption of cobots
COVER STORY | COBOTS
More than that, he continues, once
a designer commits to a speci c gear
drive they are tied into suppliers of
other components like motors, brakes
and electronics.
“The Automata Drive is central to
our robot.” Says ElSayed. “Its patent
protected technology is very similar
to a harmonic drive and gives about
80% of the performance for around
20% of the cost. But, it’s purpose
built, it doesn’t need to perform all
kinds of use cases. It’s all about
manufacturing tolerances, that’s what
helps bring the price down, not the
price of the components.”
Programming Eva’s movements
is simply done by pressing one of
the two buttons which releases the
brakes and allows operators to lead it
to a ‘waypoint’ where a command is
to be carried out. Clicking the second
button sets the waypoint, a process
called ‘back driving’. This movement
is transmitted to the operator’s
device, running Choreograph
on a browser, and saves those
points.
“If I want to locate an
object on this table I can
simply drag the robot there,”
Chandra explains. “This is
something that would have
taken much longer, much
more measurement and much
more CAD skills previously.”
Editing the movements on
Choreograph is also simple.
Waypoints can be edited
graphically on the screen
of the device and those
waypoints can be dragged
and dropped into a timeline,
similar to video editing
software, in the order the
robot needs to move.
Given that they are
likely to become ever more
common, what is the next
stage of development
for cobots? Almost
certainly greater
levels of sophistication
will emerge. One
suggested possibility is
reinforcement learning,
a form of machine
learning where a
computer learns to
complete a task by
having repeated
interaction with
a dynamic environment. Through an
iterative trial-and-error approach, the
machine explores the environment.
This exploration generates data,
which is used by the machine to
determine the best course of action
to complete its job. This happens
without human intervention and
without having to programme
the machine to perform a
speci c task.
Reinforcement
learning differs
from supervised
machine learning
in that in the latter,
algorithms are built using
data sets that contain
the correct answer to a given
problem. In reinforcement
learning there is no answer
– the machine has to nd
one by trying different
courses of action and
eventually selecting
the one that gives
the most
reward with
the least
effort.
Neil
Ballinger,
head of
EMEA
sales
at EU Automation explains:
“Reinforcement learning algorithms
encourage a machine to act in
a similar way, interacting with a
dynamic environment – for example
a factory oor with several production
lines – until it nds the most
convenient way of proceeding.”
In industrial manufacturing,
reinforcement learning is used in
processes where complex decisionmaking
skills are required, especially
where machines need to cope with
changes in dynamic environments.
Says Ballinger: “For example, a
cobot can be trained to nd the best
path to avoid interferences, such
as objects or the limbs of human
workers, while continuing to
perform its task. This would be
simple for a human, but
for machines it is an
incredibly complex
process that requires
a careful analysis
of an unpredictable
environment. If successful,
the cobot will be more
productive, because it won’t need
to stop to avoid impact.”
Reinforcement learning can also
be used to streamline production, an
approach used by researchers at the
Industrial AI Lab at Hitachi America.
The researchers designed a virtual
shop oor as a bidimensional matrix
and used reinforcement learning
algorithms to repeatedly interact with
this virtual environment. By doing
this, they were able to determine the
best set up to increase productivity
and reduce delays in servicing their
customers. !
14 WWW.EUREKAMAGAZINE.CO.UK | JULY 2020
/WWW.EUREKAMAGAZINE.CO.UK