At the Conference and Workshop on Neural Information Processing Systems (NIPS) in Barcelona, Audi demonstrated the latest in artificial intelligence (AI) and autonomous driving with a concept model Audi Q2 with the ability to complete complex parking manoeuvres.
Self-learning systems are a key technology for autonomous cars. The Audi Q2 deep learning concept is a 1:8 scale model car which autonomously searches for and finds suitable parking spaces in the form of a metal frame on a 3x3-metres area and parks itself there.
The Audi Q2 deep learning concept's sensor technology contains two mono cameras, facing forwards and to the rear, along with ten ultrasonic sensors positioned at points all around the model. The data from these is converted in a central on-board computer into control signals for steering and the electric motor.
The car first determines position relative to the parking spaces and , as soon as it perceives the position, it calculates how it can safely drive to its targeted destination. The model car manoeuvres, steers and drives forwards or in reverse, depending on the situation.
The model car's parking ability it made possible through a system that essentially learns through trial and error, or 'deep reinforcement learning'. The car first selects its direction of travel at random. An algorithm autonomously identifies successful actions a continuously refines the parking strategy. A such, the system is able to solve even difficult problems autonomously.
The Audi Q2 deep learning concept is a pre-development project of Audi Electronic Venture (AEV) in Germany. In the next step, the developers are transferring the parking space search process to a real car.
Audi is also working with partners, including the world's leading company in image recognition, Mobileye. The two companies have combined their expertise to develop a deep learning-based software for environment perception systems. Audi will use this software in 2017 in the central driver assistance controller (zFAS) in the new 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. Along with piloted parking, customers will now be able to enjoy piloted driving in traffic jam situations.
Collaborations are intensifying with partners in high-tech industries by integrating more I components into its vehicles. AI is essential for dealing with challenging situations such as urban traffic, enabling piloted driving cars to evaluate their complex surroundings and perform the necessary driving manoeuvres.