Page 24 from: Previous editions >>

TECHNOLOGY EXPERTS

Artificial intelligence and smarter sorting deliver
greater reliability

Together with better

performance, a crucial element

for the economic viability of a

sorting facility is simplified

maintenance. Through new

software and the fifth

generation of its tried-and-

tested UniSort PR, Steinert

offers improvements in both

respects and is paving the way

to the next era of waste sorting

technology.

Some sorting applications are far more com-
plex than others and cannot be handled using
conventional methods – for example process-
ing silicone cartridges. Since they comprise a
polyethylene (PE) outer wall, they are valuable
materials for recycling. But silicone residues
inside the cartridge can contaminate pure PE
products, even rendering them unusable and
requiring removal from the material flow.

GOING BEYOND WHAT IS POSSIBLE
The Intelligent Object Identifier has been
developed for precisely this application: an
Artificial Intelligence (AI)-supported system
that detects and isolates these cartridges dur-
ing sorting using optically detectable charac-
teristics. These additional distinguishing fea-
tures ensure a more stable sorting process
while also improving sorting results.
In the future, this technology will also support
other sorting tasks or indeed make others pos-
sible for the first time. For example, Steinert is
working on an addition to the sorting program
to separate polyethylene terephthalate (PET)
bottles and trays, improving sorting reliability
thanks to a new object detection feature.
What makes the Steinert solution so special
is that it can be integrated without any addi-

tional sensors and is compatible with UniSort
machines dating back to 2018 with a combi-
nation of near-infrared (NIR) and colour camer-
as.

LEARNING FROM MACHINES
This improvement has come about through
enhanced software and the latest develop-
ments in the field of machine learning, espe-
cially artificial neural networks.
Machines learn through the use of algorithms
processing information without being explicitly
programmed to do so. In its most basic form,
data can therefore be analysed by a machine
to learn recognition and distinguishing fea-
tures and then – in the case in question – come
to a conclusion about any silicone cartridges
potentially present.
The optimum precondition for this training is
comprehensive and detailed data. Since hyper
spectral imaging (HSI) technology was intro-
duced in 2012, Steinert has been generating
this data and uses it today to create the train-
ing conditions for algorithms that set the bar
in this industry. The data offers better sorting
and, in the long term, allows users to develop
digital strategies, for example in collaboration
with customers and suppliers.

Users can then be provided with the best pos-
sible sorting result without having to configure
machinery themselves or master programming.

TAKING THE ALL-ROUNDER TO ITS NEXT
LEVEL
This is exactly how the latest version of the
UniSort PR, the UniSort PR EVO 5.0, came
about.
The model has been put through its paces in
practical trials since the start of 2019 and is
the next logical step in the evolution of sorting
machines. It is also a showpiece of modern
technology and robust engineering work.

Alongside a host of improvements, the latest itera-
tion features a design that is much easier to main-
tain and provides the basis for AI-supported object
detection.

The first application of the AI-supported approach,
Intelligent Object Identifier, is the detection of
silicone cartridges to achieve PE products free of
silicone residue.

The UniSort PR EVO 5.0 reflects Steinert’s years of
experience gained across several generations of
the brand and through mass data processing.

24-25_steinert.indd 24 05-03-20 12:26