Digitization and new business models
Digitization is a decisive driver in the transformation of mobility. Ultraprecise environmental data and artificial intelligence take networking to a whole new level. Audi has set itself the aim of setting up a fully integrated platform for digital services and networked premium mobility on which new business segments can be developed.
Digitalization is a key driver behind the transformation of mobility. More and more customers are also looking to be ‘always on’ in their vehicle so they can be part of the community. Audi sees digitalization as a huge opportunity for new technologies and business models, as two examples illustrate.
Roaming without limits – possible thanks to the Audi connect SIM, which was introduced in 2016. This is a SIM card which is permanently installed in the vehicle. The card supports numerous Audi connect services on-board, including navigation using Google Earth and Google Street View, travel, traffic and parking information, as well as access to Twitter and an e-mail inbox. The Audi connect SIM includes unlimited data volume to use the services and supports convenient roaming in most European countries.
Independently of the integrated Audi connect services, data packages for individual internet use for the Wi-Fi hotspot can also be booked as an option. This is a component of the optional navigation systems. As such, users surf the internet with their mobile devices with attractive rate plans and can use other Audi connect services. If they selected a European data package, the data transfer also automatically works at the same fixed rate after crossing an international border.
On the A9 between Nuremberg and Munich, Audi is testing solutions aimed at delivering safety and convenience in piloted driving in real traffic conditions. The “Digital Highway Test Bed” is a joint initiative between the German Federal Ministry of Transport, the Free State of Bavaria, the automotive and supply industry, and the IT sector. On several sections of the A9 between Nuremberg East and Munich North, transmitters and sensors connect cars with their surroundings as well as with other vehicles.
This communication will play a decisive role in the piloted driving of the future. Audi is also involved in the “Car2Infrastructure” project on the A9. This project connects the car with variable message road signs that indicate speed limits, traffic jams or lane closures. The information gathered is uploaded to the Audi cloud via the mobile network, then transferred from there back to the test cars – keeping them constantly up to date on new traffic situations.
Thanks to the future mobile communication technology LTE-V, the data transfer modules in the cars will also be directly connected to each other. This ad hoc communication enables cars to communicate with one another even in regions without mobile coverage. Furthermore, LTE-V allows new safety functions such as black ice warnings, or even “platooning,” where piloted driving cars form an energy-saving convoy.
The piloted cars that Audi is developing for the near future must be capable of navigating down to the centimeter. The company HERE, in which Audi holds a stake, generates the digital map required to do this: the HERE HD Live Map.
AUDI AG joined a consortium at the end of 2015 with the BMW Group and Daimler AG to purchase the HERE maps database from the Nokia Corporation. The central project of the Berlin-based company is the HERE HD Live Map, which forms the digital basis for the piloted driving of the future. The new data platform describes the traffic space as a three-dimensional model with unprecedented precision – it is accurate to the centimeter rather than the meter, and dynamic rather than static. Worldwide it uses around 80,000 different sources to keep its data permanently up to date.
The HERE HD Live Map has three layers: The first of them, the HD Map, incorporates a digital image of the surroundings. Guardrails, street signs, curb edges and similar fixed objects form the reference points on which the self-driving cars of the future will be able to orient themselves precisely. There is also a database with information on hotels, businesses and restaurants.
The “live roads” constitute the second layer of the HERE HD Live Map – a dynamic layer that provides up-to-date information virtually in real time, regarding things like roadworks, accidents, emergencies or black ice. The live roads content comes from various sources, but primarily from the sensors of participating cars. The intelligent swarm generates constantly updated information about the traffic, such as green waves in city traffic, changing speed limits and available parking spaces.
The third layer is the “humanized driving” section. The piloted cars of the future learn from the HERE data pool how the owner responded in a certain situation similar to the current one. As such, these cars can tailor their behavior to the owner’s preferences.
The HERE HD Live Map is divided up into square tiles, with each edge measuring two kilometers; this enables the immense data volume to be broken down into manageable chunks. The data is hosted on a HERE backend; transfers to and from the cars are largely handled via the cellular phone network. Today’s LTE standard already supports countless possibilities, while future solutions promise higher data transfer rates and faster connection setups.
The HERE HD Live Map is still under construction; many cars in North America and Western Europe are already using partial functions. For Audi customers, the map makes complex driver assistance functions like the traffic jam assist and predictive efficiency assist even more exact and powerful.
Self-learning systems are a key technology for piloted cars. Consequently, Audi has built up extensive knowledge on machine learning. This is brought to life with a 1:8 scale model car – the Audi Q2 deep learning concept demonstrates an intelligent parking process.
Audi took part for the first time in the NIPS (Conference and Workshop on Neural Information Processing Systems) in Barcelona at the end of 2016. This industry conference focuses on machine learning and computational neuroscience. It was here that the company used a scale model to demonstrate how a car develops intelligent parking strategies: In an area measuring 3 x 3 meters, the “Audi Q2 deep learning concept” autonomously searches for and finds a suitable parking space in the form of a metal frame, and then parks itself there.
The sensor technology on the 1:8 car consists of two mono cameras, facing forward and toward the rear, along with ten ultrasonic sensors. A central on-board computer converts their data into control signals for steering and the electric motor. On the driving surface, the model car determines its position relative to the parking space. Once it detects a parking space, it calculates how to get to its destination; it maneuvers, steers and moves forward or backward depending on the situation. It acquires this ability using deep reinforcement learning – that is, it learns based on trial and error. To begin, it selects its direction of travel at random. An algorithm autonomously identifies the successful actions, thus continually refining the parking strategy. Ultimately, the system is able to perform even difficult tasks on its own.
The Audi Q2 deep learning concept is a pre-development project of Audi Electronics Venture (AEV), an AUDI AG subsidiary. In the next step, the developers aim to transfer the parking-space search process to a real car. Using artificial intelligence (AI), piloted cars will correctly assess their complex environment in urban traffic and carry out suitable driving maneuvers.
The greater the degree of integration of AI components, the more Audi will intensify its cooperation with partners from the high-tech industry. The company’s global network encompasses research institutes as well as companies from hotspots in California’s Silicon Valley, Europe and Israel. These include Mobileye, a leader in the field of image recognition, as well as long-standing Audi partner Nvidia. In 2017, Audi will premiere environmental-perception software based on deep learning in the new generation of the Audi A8.
Audi is also increasingly embracing high-tech simulation tools as part of sales and marketing. Customers can experience their chosen Audi virtually in the Audi City cyberstores and with the Audi VR experience.
State-of-the-art virtual technology enables visitors to Audi City stores to experience the entire range of Audi models and equipment options first-hand and configure their own dream vehicle. In the Audi City London, opened five years ago near Piccadilly Circus as the company’s first cyberstore, customers can themselves put together and view their Audi on powerwalls in a showroom.
The Audi VR experience provides new technology for conventional dealerships and Audi City stores. This technology will be gradually rolled out from 2017. With virtual reality glasses, customers can now virtually explore every last detail of the car of their choice – in three dimensions and from a 360-degree perspective, with all sound effects and all available equipment and trim.
Two different versions of the Audi VR experience are available to cover various applications. In the compact version, customers sit in an armchair – while in the virtual world they are sitting behind the steering wheel. The consultant can also place the customer at various positions outside the vehicle. The large version takes up around 5 x 5 meters, allowing the customer to walk freely around the virtual car. Audi has already designed a number of unusual VR environments. One setup takes the customer to Paris, others to Iceland or even to the moon.