DENSO Game - Changing Technologies
Optimize the Moment.
DENSO Game - Changing Technologies
Optimize the Moment.
In a future where everything is connected, there will be an expanding range of possibilities.
The solution that appears to be the best will be derived instantly from this vast range of possibilities.
We are imagining the future utilization of big data.
Taking up challenges is part of DENSO’s culture. We have continually created new value by accepting difficult challenges.
This culture will not change. We will continue to give form to new value through co-creation with our excellent partners.
Dr. Kadowaki was one of the first scientists in the world to propose quantum annealing. With the automobile industry entering a once-in-a-century phase of change of innovation, conditions are right to take up the new challenge of finding applications for quantum computing. DENSO is ready to support this challenge, and Dr. Kadowaki has already taken the first steps in Japan and globally.
Dr. Tadashi Kadowaki
Quantum computing will change our world with its astonishing computation speeds.
Quantum computers are expected to provide solutions in seconds to problems that would require massive amounts of time on conventional computers. Richard Feynman proposed the concept in the 1980s, but he was told that it would take decades to develop practical quantum computers. However, the Canadian company D-Wave Systems took a totally new approach, and in 2011 it launched the world’s first commercial quantum computer. Major semiconductor manufacturers, IT companies, and numerous venture companies have stepped up the pace of development, and there is fierce competition for leadership in the development of core technologies for next-generation computers.
Overlapping of “0” and “1” in the Quantum World
A major difference between quantum computers and conventional computers is the application of quantum theory to the smallest unit of information. In a conventional computer, the state of information is expressed as either “1” or “0”. Known as a “bit”, this “1” or “0” represents the smallest unit of information. In a quantum computer, however, information is processed using quantum bits, in which both “1” and “0” overlap. While a conventional computer processes information based on whether a bit is a “0” or a “1”, in a quantum computer it is possible for the bit to be both “0” and “1”.
This quantum bit is the key to the amazing computational speeds of quantum computers. For example, with a single bit, the computer only needs to process a “0” or a “1”, but with two bits it needs to calculate among four possibilities: “00”, “01”, “10”, and “11”. For this reason, the number of calculations required expands explosively as the number of bits increases. In contrast, a quantum bit can be both “0” and “1”, allowing it to express all combinations at once so that they can be processed in parallel.
Two Types of Quantum Computers
Quantum computers can be broadly divided into two types: gate-type computers, and annealing computers. Gate-type quantum computers were first proposed in 1985 by the physicist David Deutsch. This system expands the logic gates that comprise conventional computers into quantum gates. Quantum annealing was proposed as a metaheuristic method for solving optimization problems through the use of general-purpose algorithms that can be applied to a variety of problems. It has been implemented in quantum computers manufactured by the Canadian company D-Wave systems.
Changing the World with Technology that Originated in Japan
At present, the only quantum computer available commercially is a quantum annealing computer developed by D-Wave Systems. Quantum annealing made its world debut in 1998, when it was jointly proposed by Professor Hidetoshi Nishimori of the Tokyo Institute of Technology, and Tadashi Kadowaki, then a postgraduate student and now a DENSO employee, as a quantum algorithm for solving the combinatorial optimization problem*.
“Simulated annealing” is an algorithm inspired by a natural phenomenon used in metallurgy, whereby a heat treatment based on the gradual cooling of heated metal over a long period of time results in a more uniform and stable structure. Its inventors used the similarity between thermal fluctuation, an optimization problem discovered though basic research into advanced statistical mechanics, and quantum fluctuation to create quantum annealing.
Quantum computers will become essential tools for coping with increasingly complex problems through the use of big data. A proof-of-concept trial is already under way with the aim of using quantum computers to solve the problem of traffic congestion. We will continue to work with world-changing technology to create a new future.
Examples of utilization in the next-generation mobility society
Our challenge is to enhance the quality of people’s lives and create a safer and more secure social environment through the real-time optimization of many transport-related services. Proof-of-concept trials and other activities are currently in progress in Bangkok, one of the world’s most congested cities. We are working to develop practical uses for quantum computers from such perspectives as transport, logistics, and emergency services.
By responding flexibly to traffic conditions and customer needs, we aim to reduce repeat deliveries or delayed deliveries and improve loading ratios.
Our goal is to create transport systems that closely match people’s lifestyles by seamlessly integrating shared mobility with core transportation systems.
Examples of utilization in the manufacturing revolution
We will take up the challenge of creating “living factories” that can supply products reliably to meet diverse customer needs, by optimizing the continual changes that occur in factories in real time. We are working to create practical systems from various perspectives, such as the planning of factory operations and equipment, and production planning, using actual data from our 130 factories around the world.
We aim to improve transportation efficiency by flexibly responding to orders for automated guided vehicles (AGVs), and to conditions on transportation routes.
We aim to provide our customers with shorter delivery times and support for large product ranges by analyzing facility operating conditions and operational plans for manufacturing staff in real time, in order to maximize the use of management resources.
Examples of utilization in the smart society
In addition to problems relating to transportation and factories, we will also take up the challenge of social issues in such areas as the environment, energy, business management, health and education. We will explore the potential of quantum computers with our many partners who share our belief in the amazing possibilities of this technology.
We will accumulate experience and knowledge of the emerging technology of quantum annealing through applications, theory, implementation, and other activities with various partners. Our challenge is develop quantum annealing into a technology that can be used in the real world. We will share information about possibilities and issues through forums for discussion about quantum computers, and we will take up new challenges with a wide range of people.
We have assessed the performance of the D-Wave machine and existing algorithms through proof-of-concept trials targeting taxi dispatch problems in Bangkok.
We have worked to improve control algorithms using quantum mechanics in order to improve the speed and accuracy of solutions produced by the D-Wave machine.
A key step toward practical applications will be the development of an efficient way to load the problems that we want to solve into quantum bits on the D-Wave machine. We have developed technology to allow large-scale problems to be mapped rapidly into quantum bits.
Can quantum power be used by conventional computers? We have developed an algorithm to simulate quantum behavior on an conventional computer and applied to training for machine learning.
Assistant Project Manager
Electronics Research and Innovation Division,
Collaboration with Colleagues within and beyond DENSO the Key to the Evolution of Business and Technology Building the Next Stage in the Future of Quantum Computing
In 2007, I completed a postgraduate degree in quantum mechanics at Nagoya University’s School of Engineering. During my postgraduate studies, I was involved in research into the use of superconductivity to create a new type of integrated circuit that would be superior to semiconductors. My passion for motorcycles and my experiences of travelling the world inspired me with a determination to bring new technologies into the world and provide society with new forms of mobility. These ideas led me to join DENSO, which supplies a wide variety of products to automobile manufacturers around the world. Initially I worked on the development of in-vehicle microprocessors and communication technology. In 2011 I was transferred to Germany, where I was involved in the co-creation of industry standards with overseas manufacturers and worked on the Single-Edge Nibble Transmission (SENT) protocol for automotive communications. I also helped to open up a path for standardization activities as the first Japanese to participate in the PSI5 Consortium.
Basically, I aim to apply quantum computing to new business creation. Using quantum computers developed by D-Wave Systems, I am engaged in proof-of-concept experiments focusing on problems relating to mobility IoT and factory IoT, and in the co-creation of applications. There are two main obstacles to the development of practical quantum computers. First, there is the market-technology gap. The problems that we solve with applications cannot be solved directly by quantum computers. A special type of modeling is required, but there are few people with knowledge of both the markets and the technology. For that reason, I have been engaged in quantum computer research for several years with people in various organizations, including Tohoku University and Waseda University. I have already accumulated knowledge about obstacles to the efficient implementation of applications, and approaches to the elimination of those obstacles. Second, it will be necessary to create worthwhile markets. Because of the lack of applied research relating to quantum computing, we are still not sure how quantum computers can be used. For that reason, I am carrying out proof-of-concept trials, while also actively sharing information and working to spread knowledge about quantum computing in ways that are easy to understand. More people than ever are becoming interested in quantum computing, and the field is already being debated from various perspectives. I believe that by expanding this spiral, we will be able to create markets to exploit the unique characteristics of quantum computing.
I am convinced that we should continue to take up new challenges without fear of failure. To open up new markets for the first time in the world, we need to create not only the how (technology), but also the what (value). When we had only mobile telephones, nobody imagined a smartphone, and few people would have responded that they wanted such a thing. As this suggests, the creation of the “what” is never easy. My approach is to start small with a lean start-up, and to develop applications technology with my colleagues while monitoring the mood of society. This method sometimes leads to failure. However, my attitude is to learn from failures, to value the opinions of those around me, and to continue to take up the challenge of creating truly valuable markets.
04The Advantages Offered by DENSO
A key advantage for me is the fact that I have colleagues from various backgrounds and experience in wide range of fields. When I am talking to one of my colleagues, other people join in and the scope of the conservation expands. DENSO also offers fertile ground for a one new initiative after another, and an environment in which innovation happens within the company as well as outside. I was also attracted by the opportunity to carry out proof-of-concept trials of large-scale IoT systems using DENSO’s
Project General Manager in Charge of
the Information Electronics Laboratory
Electronics Research Division
We are ready to seize the opportunities offered by this once-in-a-century phase of change and innovation by taking up this new challenge in the global arena.
In 1998, Tadashi Kadowaki, then a postgraduate student at the Tokyo Institute of Technology, and Professor Hidetoshi Nishimori published quantum annealing in the transverse Ising model. They were the first in the world to propose quantum annealing as a method for solving combinatorial optimization problems. After graduation, Kadowaki worked on the development of field programmable gate arrays (FPGAs) at ROHM Semiconductor. He later switched to bioinformatics, and after working for a venture company and carrying out post-doctoral work, he was employed by Eisai, where he was involved in genome analysis and the analysis of cancer bio-markers. Most recently he undertook research in the Data Science section with the aim of using artificial intelligence in drug discovery. In May2018 he joined DENSO in search of an environment in which he can take up new challenges in the field of quantum computer applications, at a time when the automobile industry in Japan and globally is on the threshold of a once-in-a-century phase of change of innovation.
Kadowaki first touched a computer in 1983 and began programming with the aim of creating his own games. His interest in the enhancement of his personal computer sparked a passion for electronics. He also became interested in the hardware programming capabilities of FPGAs. At university he studied physics, which is the operating principles of computers. He also became interested in bioinformatics and neuroscience, which are totally different from semiconductors, and has experience of working in those fields. Kadowaki continues to work on computer-related problems on various levels, both professionally and as a hobby. For example, he still writes programs whenever he has time and maintains his home server. He also designs circuits and circuit boards.
Computer development has hitherto been supported by Moore’s Law (*), which is now reaching its limit. Kadowaki aims to create a quantum computer that will go beyond the limits of conventional computers through wide-ranging efforts at all levels from the basics through to applications. Currently he is working on a methodology for understanding the characteristics of quantum bits, which are vulnerable to noise. One quantum computer that can already be used is the quantum annealing computer developed by the Canadian company D-Wave Systems. While experimenting with this computer, Kadowaki is also engaged on research using both theoretical analyses and computer simulations, while debating with his co-researchers at Tohoku University and Waseda University. At the applied level, he is not only working on the combinatorial optimization problem, but also carrying out surveys on trends in cutting-edge research that could be used within DENSO, such as quantum simulations and quantum machine learning.
Kadowaki thinks that today’s quantum computers are being used primarily as research tools that will help us to understand the practical quantum computers of the future. The strategic directions emerging from research around the world are likely to determine the shape of practical quantum computers and the tasks for which they will be used. His goal is to develop core technologies that will give DENSO products and services a competitive advantage, while also verifying the usefulness of those technologies in meeting real-life social needs within and beyond DENSO. The development of practical quantum computers will involve many challenges at both the basic and applied levels. We will need to select targets and determine the level of effort to be devoted to each under a strategy that also encompasses collaboration with outside partners.
For around 25 years, Kadowaki has engaged in research and development in a wide range of fields in both universities and industries. By looking back on major and minor failures and successes, he has learned that it is never possible to predict in advance whether a research project will succeed. Perhaps he has simply become aware of the obvious fact that we carry out research because there are things that we cannot predict. Kadowaki believes that the potential for serendipity can be enhanced by discovering the fun of research within oneself, and by giving priority to producing diversity. Depending on the situation, some research fields are likely to be given priority. However, we all need to deepen our knowledge by widely cultivating unexplored lands on the frontiers of science, in the hope of discovering knowledge that will be valuable for industry.
DENSO Develops New Electric Power Steering Motor Control Unit for Improved Vehicle Handling, Safety
Sep. 3, 2020
DENSO Establishes Pittsburgh Innovation Lab to Accelerate Automated Driving R&D
Aug. 5, 2020
To Achieve AI-based Fully Automated Driving—R&D Project on Elemental Technologies at DENSO
Jul. 21, 2020