Daimler has entered into two independent research partnerships regarding quantum computing, with Google and with IBM. These two companies are among the pioneers in the development of hardware and software for this new type of computer.
"The next big thing?"
This article was originally published in the Daimler blog.
Today the use of the term “quantum computers” is no longer limited to scientific journals and physics conferences. Meanwhile it can also be found in a lot of popular magazines. Daimler is a vehicle manufacturer and a provider of mobility services — so what’s it doing with a quantum computer? And what is a quantum computer, anyhow? That’s what this blog is all about.
7 min reading time
- What can quantum computing do?
- Will our smartphones also work by means of quantum computing in the near future?
- However, there are some limitations due to the quantum computers level of technical development and the method by which it calculates.
- The advantages of quantum computing — One of its weaknesses is also a strength
- Why is Daimler setting its sights on quantum computers?
- Where quantum computing can already help us today
- Other promising long-term topics
- What else is missing for quantum computers at the moment?
What can quantum computing do?
Quantum physics, which principles are the basis of quantum computers’ operation, is known for about a century.
It has taken a very long time for a noteworthy hardware implementation of these quantum effects to become possible for use as a computing machine. Google is now conducting a special experiment (the Quantum Supremacy Experiment) with the goal to make a major breakthrough. The aim is to have Google’s current quantum computer chip solve in fractions of a second a mathematical problem that the world’s highest-performance supercomputer can solve only in hours, months, or years — in other words, not within a practicable period of time. The remarkable aspect of this experiment is that Google is running it with a chip that contains only 72 qubits. A qubit is the smallest unit of quantum information — the quantum equivalent of the bit in a traditional computer. The biggest supercomputers in the world have quintillion of bits.
Will our smartphones also work by means of quantum computing in the near future?
The effects of quantum physics that are being applied here are very sensitive to any kind of disturbance from the environment, such as heat/radiation/magnetic fields. For this reason, the quantum computer chips are protected by several levels of shielding and cooled down almost to absolute zero (-273 °C).
From today’s perspective, a quantum computers structure and mode of operation make it very unlikely that it could ever be built into a mobile device such as a cell phone. However, one or more quantum computers could be set up in todays computer centers and used on site or via the cloud. The energy consumption of a single system is about 20 kW, whether it contains 20 or 2,000 qubits. A traditional supercomputer in a current design requires between 10 and 20 MW — in other words, about 1,000 times more energy.
However, there are some limitations due to the quantum computers level of technical development and the method by which it calculates.
Today’s quantum computer systems are still very imprecise. In spite of the strong shielding and the extremely low operating temperatures, the system can maintain its ability to calculate (i.e. its coherent status) only for microseconds. The individual calculation operations are performed in the nanosecond range. As a result, up to about 1,000 operations can be performed before the system loses its information (decoherence) and can no longer perform meaningful operations. Hardware developers work intense on longer stable systems.
The advantages of quantum computing — One of its weaknesses is also a strength
A quantum computer is optimal for solving optimization problems. It calculates results together with a probability statement. Mathematical precision is the strong point of our standard computers. A standard computer that is given the task 1,000 + 1,000 produces the unambiguous value of 2,000 as the result. The result of the quantum computer might be as follows: 1,000 + 1,000 = 1,997.4 with a probability of 98.3%. This is imprecise, but when it comes to bigger optimization problems, it is supposed to find the best possible result much faster than a traditional computer. In extreme cases, there could be a huge calculation time difference, a few seconds (quantum computer) compared to thousands of years (standard computer).
Because of their respective special characteristics, it is becoming likely that quantum computers will be used in tandem with standard computers so that each system can contribute its own particular strengths. Researchers expect that development times ranging from five to ten years will be needed in order to develop error-free quantum computers with a larger number of qubits and considerably longer computation times. However, with this machines computing operations that could never have been conducted in the past times would then be possible to be done.
Why is Daimler setting its sights on quantum computers?
Why is Daimler already getting involved with quantum computing in the form of research projects at this early stage? The simple answer to this question is that we have very many computation tasks within the company that ultimately are optimization problems. That applies, for example, to production issues, worldwide component and data logistics, and the optimization of vehicle production planning in Daimler’s plants all over the world. Today we still don’t know with any certainty whether quantum computers, alone or in combination with a standard computer system, can profitably use their special attributes to address these issues. Quantum computing technology is still in its infancy. But we want to conduct initial tests based on simple problems of this kind to find out.
Another very important topic that is just developing in major cities and overpopulated areas is a new variant of transportation (compared to conventionally driven cars): “Mobility as a Service” (MaaS). Thanks to the rising of autonomous driving technologies, the Internet-based software solutions for routing and the linkage of different transportation systems, MaaS will become very attractive in the future. However, this will result in new challenges for mobility suppliers. For example, how can we optimally distribute transportation capacities in a very dynamic urban system with millions of transportation requests for people and goods at short time intervals? How can we make the transportation system highly reactive to disruptions, for example if traffic accidents cause congestion, a commuter-train line breaks down completely, or an extreme weather event is taking place? Such occurrences would have a strong impact on current transportation processes and lead very dynamically to new transportation requests.
These are questions that we must answer soon if we want to stay on top in our competition with other suppliers. Quantum computers may one day play a key role for us in this area.
Where quantum computing can already help us today
In any case, in view of the state of development of quantum computers, such problems can be addressed in the medium term at the earliest. The nearer tasks for a quantum computer are problems related to materials — for example, calculating the formation of molecules made up of certain chemical elements. That’s because nature also functions according to the principles of quantum physics. As a result, the best way to mimic or simulate nature is a system that itself operates according to the principles of quantum physics.
Electric mobility is mainly based on a well-functioning cell chemistry of the batteries. We are very interested in predictions about future materials, precise ways of simulating battery-cell chemistry, and the aging processes and performance limits of battery cells. In these areas, there is justified hope that quantum computers will yield initial results in the years ahead.
Other promising long-term topics
In the long term, artificial intelligence will also be a topic relevant to a quantum computer. For example, researchers are discussing whether machine learning can be conducted at high speeds on a quantum computer.
If “data is the new oil” — in other words, the key “raw material” for a well-functioning economy — and if the optimal processing of this data, e.g. for “Mobility as a Service,” will play a key role in the future, then all companies have to adapt to this new reality. In any case, Daimler will also move along this path in order to develop from one of the most renowned vehicle manufacturers into one of the best mobility providers. Strategic partnerships can play an important role in this process and accelerate it.
It will be important to find out how large volumes of data can be stored and processed quickly and reliably. In other words, the refineries of the “new oil” will no longer be chemical plants but highly efficient computer centers that may one day combine conventional supercomputers with high-performance quantum computers. These new computer centers will need suitable software and algorithms in order to make the capacities of a quantum computer usable. And of course we would like to have quantum memory as soon as possible.
What else is missing for quantum computers at the moment?
There are several promising possibilities for building qubits. A number of specialized companies are using various approaches to develop the necessary hardware. In addition to the big players Google, IBM, Intel, and Microsoft, there is also a number of startups such as D-Wave and Rigetti active in the field.
At Daimler, we are focusing on the applicability of this research to our calculation tasks. We also believe it’s important to make sure that the QC hardware manufacturers understand our needs and support us in translating our specific questions into the quantum computer language. We’re also interested in making sure they work on developing the high-level languages that will make quantum computers usable in the future by the programmers who are being trained today. That’s because only that will make the industry-wide introduction of quantum computers successful in short time.
The way a quantum computer functions, the underlying quantum-physical effects, and the approaches to solving our calculation tasks are huge challenges for all the members of this team. Those who have not studied theoretical physics have a hard time to integrate the deep impact of quantum physics with their previous understanding of physics. For them, it’s at least slightly comforting to know that even Einstein had problems with its implications (as testified by his remarks “God does not throw dice” and “spooky action at a distance”). There are many indications that we are here opening up a new chapter in the history of technology. This is a really new field that is on the threshold of implementation. To the question of what the “next big thing” could be in about five to ten years, the quantum computer team already has a hunch…