Anyone who drives a car through busy streets to work in the morning would love to know how heavy the traffic is going to be ahead of time. Experience might help, but it can't predict a traffic jam. Volkswagen wants to change this: The carmaker's IT specialists are working on a solution that informs drivers 45 minutes in advance where gridlock looms.
Scientists and companies have been working on a solution to prevent traffic jams for decades. But predicting the behavior of a system as complex and dynamic as road traffic has proven too complicated time and again.
Now Volkswagen wants to make the breakthrough using quantum computers. While not yet fully developed, these computers promise performance that surpasses today's systems by several orders of magnitude.
"The quantum computer opens up a whole new horizon," Martin Hofmann, Volkswagen's Chief Information Officer (CIO) told Handelsblatt, adding that their immense computing power is ideal for the calculation of extremely complex processes.
Indeed, the device Volkswagen is using in its battle against traffic jams has little in common with a conventional PC. The D-Wave 2000q is a black box about three meters high, which contains a computer chip that fits on a fingernail. In order to fully unfold its enormous computing capacity, it needs to be situated in a vacuum with a temperature near absolute zero. This is a prerequisite to exploit the laws of quantum physics.
The device is not located at VW headquarters in Wolfsburg, but 8,000 kilometers away in Vancouver, where the Canadian manufacturer D-Wave operates. Volkswagen accesses it via the internet, so the carmaker does not have to buy the machine, which costs around €15 million ($16 million).
For testing, the VW Group uses freely available GPS data of thousands of taxis in the Chinese capital of Beijing. 2000q processes the taxi data in seconds and calculates nearly traffic-free routes. "The traffic jams were gone," said Florian Neukart, who manages the test series for Volkswagen from San Francisco. VW emphasizes that this is the first time a calculation for traffic jam avoidance with real life data was done beyond research purposes.
Volkswagen's tests also offer a glimpse of what might one day be possible. In theory, quantum computers have immense computing capacity. What today's supercomputers need half an hour for, they can manage within a few seconds. Which opens the door for myriad real-world applications, such as advancing artificial intelligence, simulating chemical molecules, optimizing trading on the exchanges, or making traffic forecasts in real-time.
In reality, however, such scenarios are still a long way off, as quantum computers need to produce more reliable results. Robert Ewald, President of D-Wave, compares the situation to that of the computer pioneers in the 50s when researchers created the foundations for the digital era. "We are at a point where quantum computers are soon going to be superior for selected academic problems."
For Volkswagen CIO Martin Hofmann, this is all part of the bigger picture. The carmaker is looking beyond simply selling cars toward large scale mobility services. One day in the not too distant future, customers could be paying for the company to get them to work while avoiding traffic jams.
This year, Volkswagen subsidiary Seat is to carry out tests with the quantum computer. "Now use is starting on a larger scale," said Mr. Hofmann. If the experiments succeed, it would be a quantum leap for Volkswagen.