The Institute - September 2021 - 67

AUGMENTUS
spray-coating or polishing parts, or
loading and unloading inventory.
By making industrial robots easier
to program, Lim says, the software can
help businesses increase efficiency and
reduce costs-which would in turn
help retain local manufacturing jobs.
" We want to lower the time, skill,
and cost barriers for companies to
adopt robotic automation, " Lim says.
Augmentus cofounders [from left] Chong Voon Foo (CTO), Yong Shin Leong (CEO), and
Daryl Lim (COO), programming an industrial robot using the software they developed
STARTUP
Program Industrial
Robots Sans Coding
Augmentus uses AI and computer-vision algorithms
to make the process more intuitive
BY PRACHI PATEL
ROBOTS ARE EXCELLENT at repetitive,
tedious, and time-consuming
jobs-which makes them beneficial in
manufacturing. But training industrial
robots requires substantial coding
skills and knowledge.
Singapore-based startup Augmentus,
founded by IEEE Member Daryl
Lim, Yong Shin Leong, and Chong
Voon Foo, is trying to make automation
more accessible with its intuitive
robot-programming platform.
The platform's software has
a graphical interface that allows
nontechnical users to program
industrial robots in minutes, says Lim,
the startup's chief operating officer.
It also has an integrated artificial
intelligence model that lets clients
train the software to identify objects
such as car parts. The model uses
computer-vision algorithms, like the
ones used for object and facial recognition
in digital images.
Augmentus in December was
named one of four IEEE Entrepreneurship
Stars at Slingshot 2020, one of
the biggest startup competition events
in the Asia Pacific region. The award
recognizes budding ventures driven
by engineering innovations that align
with IEEE's core mission. Awardees
become honorary IEEE members for a
year, and they receive mentorship and
support from the IEEE network.
Most Augmentus clients are
advanced industrial manufacturing
companies that produce automotive
or machinery parts. The companies
use robots for quality inspections,
Increasing efficiency
Industrial robots can be costly beasts
to tame. Teaching a robotic arm to do
a seemingly simple task, like sorting
objects or moving them from a bin
to a conveyor belt, typically requires
thousands of lines of code, Lim says.
The arduous coding process has to be
repeated every time the arm must be
reprogrammed for a different task.
To add to the problem, robots made
by different manufacturers often use
different programming languages. And
programmers with the requisite coding
skills are in short supply.
It all translates to higher expenses.
" Close to 70 percent of the cost of
an industrial robot is software- and
programming-related, " Lim says.
Augmentus software does not
require the user to create any code.
Instead, factory technicians can
program robots or robotic parts with
an iPad and an Apple Pencil stylus.
The technician selects a robot and
equipment from the software's menu,
uses the iPad's camera to scan the area
in which the robot works, and then-
with the stylus-plots points on the
screen to map out the path the robotic
arm will take for its task. The software,
which runs in the cloud, then automatically
generates code to create the
optimal path for the bot. Users can test
and verify the code via virtual simulations
before deploying it on a factory
floor. They can edit it if need be.
Compared with the traditional
coding route, the startup's technology
allows companies to develop and
deploy robots 10 times faster and for a
10th of the cost, Lim says.
The software is mostly being used
now by manufacturing companies for
spraying and inspecting parts, but the
team is updating it for additional applications
such as welding and sanding.
SEPTEMBER 2021
THE INSTITUTE
67

https://entrepreneurship.ieee.org/2020_12_14_stars-slingshot/ https://www.augmentus.tech/ https://entrepreneurship.ieee.org/2020_12_14_stars-slingshot/ https://entrepreneurship.ieee.org/session/2020-slingshot-switch/ https://www.augmentus.tech/ https://www.linkedin.com/in/daryllimyq/ https://www.linkedin.com/in/daryllimyq/ https://www.linkedin.com/in/voonfoo/ https://www.linkedin.com/in/voonfoo/

The Institute - September 2021

Table of Contents for the Digital Edition of The Institute - September 2021