Audi, Autonomous Cars, high tech, Motoring News, road safety, vehicle technology

Artificial intelligence: Audi taking lead on highways

  • California hosting talks on artificial intelligence
  • ‘Neural network’ generates 3D environments
  • Driving further down the road to autonomy
AUDI A8: Focus on big sedan at California conference on artificial intelligence. Image: Audi / Newspress
AUDI A8: Focus on big sedan at California conference on artificial intelligence. Image: Audi / Newspress

LONG BEACH, California – Automaker Audi is exhibiting a pre-development project at what is described as “the world’s most important symposium for artificial intelligence” here.

The conference has been co-sponsored by Audi and will run from Dec 4-9.

The project uses a mono camera that uses artificial intelligence (AI) to generate precise 3D models of a vehicle’s current environment.

IMAGES ‘PRECISE AS POSSIBLE

Audi says its A8 is the first car to be developed for conditional Level 3 automated driving with a “traffic-jam pilot” controlling progress in traffic moving no faster than 60km/h if so requested by its human.

A requirement for such driving is a “precise as possible” mapped image of the approaching environment and AI is the key technology.

A project team from Audi Electronics Venture presented a monochrome camera at the Conference and Workshop on Neural Information Processing Systems (NIPS) that uses artificial intelligence to generate an extremely precise such image which makes it possible to capture the car’s surroundings.

READ MORE Audi features on Carman’s Corner

Audi explained: “A conventional front camera acts as the sensor. It captures the area in front of the car within an angle of about 120 degrees and delivers 15 images/second at a resolution of 1.3 megapixels which are then processed in a neural network.

“This is where “semantic segmenting” occurs: each pixel is classified into one of 13 object classes to identify and differentiate nearby cars, trucks, houses, road markings, people and traffic signs.

“The system also uses neural networks for distance information visualised by virtual boundaries to define a constant distance. This combination of semantic segmenting and estimates of depth produces a precise 3D model of the current environment.”

‘GREAT POTENTIAL FOR TRAFFIC’

Audi engineers had previously trained the neural network to “learn” through numerous views of road situations recorded with a stereo camera. “As a result, the network learned to independently understand rules to use to produce 3D information.

“The project holds great potential for the interpretation of traffic situations.”

Two Volkswagen Group partners at the conference were scheduled to present their AI topics: the Fundamental AI Research department was to focus on unsupervised learning and “control through variational inference” – described as “an efficient way to represent probability”.

Then the Audi team from the Electronics Research Laboratory of Belmont, California, was to demonstrate a solution for purely AI-based parking and highway driving with “lateral guidance” (read: steering) controlled by neural networks.

HEAD-HUNTING, TOO

Autonomous cars, Audi believes, benefit from a large network in AI technology which includes companies in Silicon Valley, Europe and Israel.

It was in 2016 that Audi became the first automakers to participate at NIPS with its own exhibition booth. This year, as a sponsor of NIPS, Audi is seeking to further develop its network in California – and perhaps find more staff.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s