of freight. A single image can capture multiple barcodes and the
mission can be paused or aborted at any time. As soon as a barcode
is recognised, the data, including the location and the number itself,
is automatically updated to the dashboard. If no barcode is found,
the system will alert the user. A live video showing the drone’s field
of view is also visible on the dashboard, which is accessible via a
handheld tablet. The drone’s immediate location is also identifiable.
The machines operate for a duration of 20-25 minutes, meaning
they can complete the inventory of a full aisle on a single charge
before automatically returning the charging pad when power falls
below 20%. It then takes 45 minutes to an hour for the drone to
recharge fully and it will then resume any unfinished missions.
“I’m pleased to say, so far testing has been successful. The drone
is flying autonomously; it’s conducting missions; it’s taking off and
landing from its charging pad very effectively; it’s detecting
obstacles reliably and from different angles. And we’re achieving
an accuracy in air waybill reading of 82%, which is a fantastic start.
The next few months are going to be spent refining and improving
on that accuracy figure and also helping the drone to understand
anomalies,” says Morris. “We’re not expecting 100% straight away,
but we’ll just continue improving. If we can get it up beyond 90%,
I think that would be an even better achievement.”
The challenges
The testing period focussed on four main challenges, explains
Founder and CEO of FlytBase Nitin Gupta: accuracy of AWB
reading, recognition of empty positions, obstacle detection and
autonomous navigation. The drone had to be able to fly between
and over the racking, as well as to land and take off from the
charging pad.
Gupta details some of the specific challenges relating to the
indoor application. “Because there is weak or no GPS signal
indoors, position becomes one of the key challenges. When you’re
operating outdoors, because of the GPS, it is a lot easier to navigate
from point A to point B.” Other challenges include the abundance
of static and dynamic obstacles. “Outdoors, once you’re beyond a
certain height, the number of obstacles will be much less or none,
but in the case of an indoor environment like a warehouse, you
have a lot of racking, forklifts, people going around,” he highlights.
As well as the aisles being closed to personnel during a drone
mission, for safety reasons the machines have been programmed to
detect obstacles four to five metres away and to fly upwards at
speed to avoid them.
Stability also becomes a
challenge when operating in a
confined space, continues
Gupta. “The system should be
very stable. You cannot afford
to make errors larger than 10
centimetres or so. More than
that and you are likely to hit
the racking or go off track. And
weight and size also become
very important: you need to
have something which is very
lightweight, very small in size,
so that you can effectively
navigate.”
Ensuring full automation
was also important. “Flights
need to be fully autonomous,
but also the battery charging
needs to be automated because
if you have to send somebody
to change the battery or charge
the battery every time, it
defeats the purpose,” he adds.
As well as collecting data,
the solution must be able to
automate the entire workflow,
he stresses, from recognising
where the barcode is and
reading that barcode to
automatically feeding that data
into a warehouse management
system. Customisation of the
user interface to reflect the
shape and size of the warehouse
must also be possible, and
cloud connectivity is a must, he
FlytBase’s technology works in combination
with off-the-shelf drone hardware from DJI
relates, to enable complete
visibility of multiple
warehouses across different
locations.
Old problems, new solutions
If any other carriers are doing
the same thing, ALI is not aware
of them. Morris believes that the
success of this trial is proof of
the huge potential drones have
in the aviation and logistics
sectors and is confident that a
Plan B won’t be necessary.
“We have evaluated other
technologies and we are
continuously looking at the
market to understand what
other technologies are being
created. That’s the best part of
innovation – there’s always
something new. But at the
moment, we’re very focused
on drones and we’re
optimistic that this can be a
great solution.”
This year, testing of the
drone will continue, with a
focus on four main areas of
operation: accuracy, speed,
scalability and reliability. This
trial is just one aspect of IAG
Cargo’s continuous programme
of innovation, which will see
further new technology trialled
in 2020, aiming to tackle
age-old industry problems.
www.airlogisticsinternational.com February 2020 13
/www.airlogisticsinternational.com