Audi is a sponsor this year of the annual Conference and Workshop on Neural Information Processing Systems (NIPS) and is showcasing its expertise at the conference on artificial intelligence for the first time. Throughout this week, the automaker is showing, with the aid of a 1:8 scale model—the “Audi Q2 deep learning concept”—how a car develops intelligent parking strategies. On an area measuring 3 x 3 meters, the Q2 autonomously searches for and finds a suitable parking space in the form of a metal frame, and then parks itself there.
Self-learning systems are a key technology for piloted-driving cars, and Audi has already built up a body of know-how in machine learning. The company is the only automaker represented at NIPS 2016 with its own stand and a showcase. (Although engineers from Daimler are presenting a demonstration at NIPS 2016 on the detection of obstacles by an autonomous car.)
The Audi Q2 deep learning concept’s sensor technology consists of two mono cameras, facing forward and toward the rear, along with ten ultrasonic sensors positioned at points all around the model. A central on-board computer converts their data into control signals for steering and the electric motor.
On the driving surface, the model car first determines its position relative to the parking space. As soon as it perceives the position, it calculates how it can safely drive to its targeted destination. The model car maneuvers, steers and drives forward or in reverse, depending on the situation.
The model car’s parking ability is made possible by deep reinforcement learning. In other words, the system essentially learns through trial and error. To begin, the car selects its direction of travel at random. An algorithm autonomously identifies the successful actions, thus continually refining the parking strategy. So in the end the system is able to solve even difficult problems autonomously.
The Audi Q2 deep learning concept is a pre-development project of Audi Electronics Venture (AEV), an AUDI AG subsidiary in Gaimersheim, Germany. In the next step, the developers are transferring the parking-space search process to a real car.
The Audi global network encompasses not only research institutes, but also companies from hotspots in California’s Silicon Valley, Europe and Israel. The premium manufacturer is working with partners including Mobileye, the world’s leading company in the field of image recognition. In this partnership, the two companies combined their expertise to develop a deep learning-based software for environment perception systems.
Audi will use the software for the first time in 2017, in the central driver assistance controller (zFAS) in the new generation of the Audi A8. NVIDIA, a leader in the field of hardware systems with an associated development environment, was an important partner in the development of the zFAS. (NVIDIA is also a NIPS 2016 sponsor.) These technical solutions will enable the customer to enjoy piloted driving in traffic jam situations as well as piloted parking.
Audi is further intensifying its collaborations with partners from high-tech industries through an increasing degree of integration of components with artificial intelligence (AI). These forms of artificial intelligence are important for dealing with challenging situations such as urban traffic. It enables piloted driving cars to evaluate their complex surroundings and perform necessary driving maneuvers accordingly.
Also at NIPS to gain insights into these and other exciting developments will be AI specialists interested in working on innovations at Audi. Specialists and HR experts from the company will be at the event to provide them with information on a range of career opportunities. At Audi the specialists will have opportunities to help shape the role of AI in the automotive industry by applying their knowledge in the areas of machine learning, cloud computing, data analytics and vehicle architecture.NIPS. The Neural Information Processing Systems (NIPS) Foundation is a non-profit corporation the purpose of which is to foster the exchange of research on neural information processing systems in their biological, technological, mathematical, and theoretical aspects. Neural information processing is a field which benefits from a combined view of biological, physical, mathematical, and computational sciences.
The primary focus of the NIPS Foundation is the presentation of the continuing series of professional meetings known as the Neural Information Processing Systems Conference, held over the years at various locations in the United States, Canada and Spain.
The Thirtieth Annual Conference on Neural Information Processing Systems (NIPS 2016) is a multi-track machine-learning and computational neuroscience conference that includes invited talks, demonstrations, symposia and oral and poster presentations of refereed papers. Following the conference, there are workshops which provide a less formal setting.