Continental, a German
multinational automotive parts manufacturing company, has invested in setting
up its own supercomputer for artificial intelligence (AI), powered by NVIDIA
InfiniBand-connected DGX systems to develop innovative technologies around autonomous
driving more efficiently.
has been operating from a data centre in Frankfurt, Germany, since the
beginning of 2020. It is offering computing power as well as storage to
developers in locations worldwide. AI enhances advanced driver assistance
systems, makes mobility smarter and safer, and accelerates the development of
systems for autonomous driving, the company noted.
supercomputer is built with over 50 NVIDIA DGX systems, connected with the
NVIDIA Mellanox InfiniBand network. It is ranked as the top system in the automotive
industry, according to the publicly available list of TOP500 supercomputers. A
hybrid approach has been chosen to be able to extend capacity and storage
through cloud solutions if needed, the company said. The supercomputer has
every detail planned precisely in order to ensure the full performance and
functionality today, with scalability for future extensions, it added.
assistance systems (ADAS) use AI to make decisions, assist the driver and
ultimately operate autonomously. Environmental sensors like radar and cameras
deliver raw data. This raw data is processed in real-time by intelligent
systems to create a comprehensive model of the vehicle’s surroundings and
devise a strategy on how to interact with the environment. Finally, the vehicle
needs to be controlled to behave like planned. But with systems becoming more
and more complex, traditional software development methods and machine learning
methods have reached their limit. Continental noted that Deep Learning and
simulations have become fundamental methods in the development of AI-based
In the case of Deep Learning, an
artificial neural network enables the machine to learn by experience and
connect new information with existing knowledge, essentially imitating the
learning process within the human brain. Continental said that a child is
capable of recognizing a car after being shown a few dozen pictures of
different car types. However, several thousand hours of training with millions
of images and enormous amounts of data are necessary to train a neural network
that will later on assist a driver or even operate a vehicle autonomously. The
NVIDIA DGX POD not only reduces the time needed for this complex process, it
also reduces the time to market for new technologies, it explained.
To date, the data
used for training those neural networks comes mainly from the Continental test
vehicle fleet, the company said. It added that currently, they drive about
15,000 test kilometres each day, collecting around 100 terabytes of data.
Already, the recorded data can be used to train new systems by being replayed
and thus simulating physical test drives. With the supercomputer, data can now
be generated synthetically, a highly computing power consuming use case that
allows systems to learn from travelling virtually through a simulated
environment, it explained.
The advantages of this process
come in the form of firstly making recording, storing and mining the data
generated by the physical fleet unnecessary over the long run, as necessary
training scenarios can be created instantly on the system itself. Secondly, it
increases speed, as virtual vehicles can travel the same number of test
kilometers in a few hours that would take a real car several weeks. Thirdly,
the synthetic generation of data makes it possible for systems to process and
react to changing and unpredictable situations. Ultimately, this will allow
vehicles to navigate safely through changing and extreme weather conditions or
make reliable forecasts of pedestrian movements – thus paving the way to higher
levels of automation.
Christian Schumacher, Head,
Programme Management Systems, Advanced Driver Assistance Systems business unit,
Continental, “The supercomputer is an investment in our future.” The
state-of-the-art system reduces the time to train neural networks, as it allows
for at least 14 times more experiments to be run at the same time, he added.
Schumacher noted that Continental selected NVIDIA after intensive testing and
scouting, and the project was implemented in less than a year.
NVIDIA DGX systems give
innovators like Continental AI supercomputing in a cost-effective,
enterprise-ready solution that’s easy to deploy, observed Manuvir Das, Head,
Enterprise Computing, NVIDIA. He said Continental is engineering tomorrow’s
most intelligent vehicles, as well as the IT infrastructure used to design them
by utilising the InfiniBand-connected NVIDIA DGX POD for autonomous vehicle
“Overall, we are estimating the
time needed to fully train a neural network to be reduced from weeks to hours,”
said Balázs Lóránd, Head, AI Competence Centre, Continental, who also works on
the development of infrastructure for AI-based innovations together with his
groups in Continental. Lóránd added that the development team has been growing
in numbers and experience over the past years. With the supercomputer, the team
is now able to scale computing power even better according to its needs and
leverage the full potential of the developers, he noted.