Many different types of robots are being developed in companies, universities, polytechs, high schools and elsewhere in Japan. Exhibitions, fairs, and competitions bring together those in various categories to determine which has the best performance. Among the participants are an increasing number of robot models designed to assist disabled people or to perform searches or carry out similar tasks during emergencies. These designs are already producing substantial benefits as they emerge from laboratories and begin handling real-world applications.

Development in robotics in recent years has made extensive use of embedded systems know-how, and the hardware has come to rely on a wide range of different microcomputers. Many different Renesas microcomputers, including models from the SuperH and H8 product lines, are being used in a diversity of designs.

To get some expert perspective on the current state of robotics research and the role played by microcomputers, we talked to Shin'ichi Yuta of the University of Tsukuba, who is preeminent in the field of robotics research, and Hideo Inayoshi, Executive Director at Renesas Technology. The discussion was moderated by Motoko Takenishi, Editor of "ROBOCON Magazine," a specialist magazine covering robots.

		Shinich Yuta Dr.Eng.,IEEE fellow Professor,Intelligent Robot Laboratory University of Tsukuba School of Systems and information Engineering Profile: Graduated from the Ph.D program in Gradate School for Engineering (Electrical Engineering) at Keio University in 1975. Research associate at Faculty of Engineering, Tokyo University of Agriculture and Technology in the same year. Appointed lecturer in Electronics and Information Science at the University of Tsukuba in 1978. Made Associate Professor in 1983 and Professor in 1992. Appointed Professor at the Institute of Engineering Mechanics and Systems at the University of Tsukuba in 1999. Was a Director and Vice-president of the University from April 2004 to March 2006, responsible for research, for liaison between industry academia and government, and for social service. His specialty is robotics engineering. Over 25 years, he has been working in the field of mobile and has developed the "Yamabico" self-contained mobile robot platform for research. Fellow of the IEEE and the Robotics Society of Japan. Member of the Japan Society of Mechanical Engineers, the Society of Instrument and Control Engineers, and other professional bodies.
		Hideo Inayoshi Senior Vice President Profile: Joined Hitachi Ltd. after graduating with a Master of Engineering degree in applied physics from the University of Tokyo in 1971. Held a succession of important roles in development of microcomputer chips at Hitachi. Became Senior Vice President, Board Director of Renesas Technology Corp. in April 2003. Subsequently served for SoC development and business expansion. Responsible for Corporate Planning now.
		Motoko Takenishi Editor in chief, ROBOCON Magazine Publishing Department Profile: Editor of ROBOCON Magazine since July 2003. Also works as a journalist on stories such as reports on robot contests. Has served as a judge at many contests, including ROBO-ONE and the Kawasaki Robot Competition.

Takenishi (moderator): "Astro Boy" in the popular Japanese science-fiction comic launched decades ago was depicted as having been born on "April 7, 2003," a date that was then far into the future (see Note 1). I mention this fact as a starting point because I think that his great popularity in Japan is partly responsible for the recent unprecedented surge in discussions here about robots. However, despite the heightened awareness among the general public, it must be said that robots are still a long way from creating a market for themselves like the thriving markets that exist for other consumer appliances. In fact, it seems that we have still to decide what a robot is and what we want robots to do for us.

Yuta: Our idea of what a robot is still rather vague. A robot does not necessarily have to walk bipedally and it does not have to look like a human the way that Astro Boy does.

The robot that is the most realistic and familiar to me is the fully automatic washing machine. Looked at through the eyes of someone who lived back in a time people did their washing in the local stream, the washing machine is a true robot. You put your clothes into it and press a switch, then 30 minutes later your clothes are clean and it has even wrung out the excess water for you.

Takenishi: That's right. The level of sophistication of home appliances has come a long way. We are surrounded by home appliances that it would not be inappropriate to call robots. However, we do not normally think of home appliances as being robots.

Yuta: Another example is an excavator. If someone from the past saw one of these construction machines, they would see it as a robot being controlled by a person. Cars, too, are much more automated than they were in the past. Cruise control is a function that would have been thought of as being like a robot in the past.

Looked at in this light, the idea of a robot seems to be a "dream," something that does not yet exist. Once it does exist, it is no longer called a robot. Instead, it is given a different name. As a result, the word robot has been redefined to convey an image of something that does not yet exist.

Inayoshi: That is a very interesting idea. Industrial robots are already in wide use in many applications. They are now practical solutions for tasks such as welding and painting, and are mainstays in the production of automotive, mechanical, electrical, and other parts (see Note 2). Regardless of this, what we think of as a robot is different, depending on our point of view.

Takenishi: I agree. I have often noticed recently that the image conveyed by the word "robot" is different for different people.

Yuta: Yes. But in the other side,the readers of ROBOCON Magazine (See Note 3)have a common understanding of what a robot is. Certainly, nobody is going to say "that's not a robot" about the machines that appear in robot contests (see Note 4). Yet, if the rules gave contestants complete freedom to design a "robot," that would actually make building robots harder. Instead, robot contests need to pose specific challenges. By progressively changing the nature of these challenges, you can drive progress in robotics. Indeed, having an objective for the future is very important. For example, the RoboCup (Robot Soccer World Cup), in which robots play soccer against each other, has the objective of "developing a RoboCup team that can beat the winning team from the human World Cup in 2050." Certainly that's an ambitious goal!

Inayoshi: Renesas has helped with the development of these robots by supplying microcomputers from our extensive product range. I am happy to say we enjoy a significant market share in electronics for robotics applications.

Yuta: In my laboratory, we have changed away from Motorola and INMOS to use mainly Renesas microcomputers recently.

Takenishi: Among hobbyists, devices in the SuperH family and H8 family are widely used. SuperH microcomputers predominate in robots that use large numbers of motors and H8 microcomputers are very popular in the educational field.

Takenishi: When you build a robot, you learn a wide range of skills, including mechanical design, control system design, and electrical design.

Inayoshi: This is because robot development draws on many different aspects of engineering, including advanced control and computational processing.

Yuta: This is what interests the people who aspire to robotics, including the readers of ROBOCON Magazine. In this respect, they can be expected to be particularly active, even by the standards of the scientific community. Generally, when the systems grow in scale, the work of designing is shared, instead of being performed by a single individual. However, robots require a balance of functions, so it is very important that should understand every function of the robot. So, we are trying to develop complete systems in our laboratory.

In university laboratories, it is an advantage to have the person next to you working on the design of a different part of the robot. Being involved with a multi-faceted development project provides students with a simulated private experience, which is very valuable. Instead of hearing that a task work was done by so and so from such and such laboratory, students can see first hand that "if that person can do it, then there's no reason why I can't, too."

Further, learning how something can be designed by combining available components is also very important for robot technology. If a better electronic or electromechanical component appears, people will naturally tend to use it and we have to be able to adapt to this.

Takenishi: Microcomputers have now become essential components in robotics. Are there any particular things you would like to see in future models?

Yuta: It may be a rather detailed point, but the number of I/O ports that are available is quite important. For example, if a device has a multi-channel up/down counter that we can make extensive use of, we are very happy. On the other hand, even a slow sampling speed of 1ms for subsystems such as analog-digital converters (A/D converters) is fast enough to satisfy the requirements of real mechanical control systems. Consequently, the amount of CPU power that microcontrollers can deliver is currently enough for us than it is for many embedded control systems in other fields. Even though, it is not yet clear how much more processing power we will require when we try to realize advanced robotics functions such as real-time smart-sensors and imaging.

Inayoshi: Although it seems that you are largely satisfied with our products, the role of Renesas as a volume producer of microcomputers is to design chips with a well-rounded set of specifications, accommodating requests from many different customers in many different markets. However, this can sometimes result in a chip that is "neither one thing nor the other." While we manage our business carefully to avoid this problem, we also deliberately aim to include some "pointy bits" that are distinctly Renesas within our well-rounded designs.

Yuta: The fact is, unlike in the past, modern microcomputers are packed with functions. Accordingly, in the lab we are satisfied so long as we can make use of these functions. However, that is becoming increasingly difficult. I feel that the problem is not in the microcomputer functions themselves, but rather in the software that must be developed to use the functions.

In the past we created our own developing enviroment. Thus, we could create an environment tailored for our own purposes and could change anything that irritated us. Now that developing systems have become much larger and complex, it is extremely difficult for us to build them on our own. Instead, we have to use systems produced by others. There are many functions for which we would be very pleased if someone would produce software that we could use in our development work.

Inayoshi: We get exactly the same requests from customers in other fields, such as digital consumer electronics. The relative importance of the software used for development is steadily increasing. Renesas is currently working on various measures to provide optimum software development solutions for microcomputers.

Takenishi: How does the level of robot technology in Japan compare to the rest of the world?

Yuta: Although the answer to that question depends on how you chose to measure the level, I think that Japan is the overall leader in the field of humanoid robots and bipedal robots. On the other hand, the USA is particularly strong in software.

Takenishi: In what areas do you think Japan needs to become stronger?

Yuta: Our country is considered to be strong in the embedded systems technology. Elements of control system design and mechanical design that are important in embedded systems are also essential elements in robotics applications. Therefore, many beneficial crossovers are possible between areas in which the Japanese information technology industries and Japanese robots are strong.

In the future, the growing strength of Japanese embedded system technology will be of considerable significance to robot development. Conversely, I also believe that research in robotics can contribute at least indirectly to progress in the field of embedded systems.

Inayoshi: Considering factors such as the extensive technical resources available from the field of embedded systems and the large amount of research going into humanoid robots, there are many people who think that robots will, in the future, become a huge market for Japan. I have high hopes that a whole new industry—and a market with a significant affect on our daily lives—will emerge from the robotics research being carried out in our country's universities, such as the work being done in Professor Yuta's laboratory.

Of course, a key objective of Renesas is to further strengthen the technology used in embedded systems and robotics. We intend to take responsibility for providing the required microcomputers and development systems.

Yuta: This is how robotics technology will come to fruition and make its contribution to society. This is what we are really aiming to achieve.

Takenishi: Thank you for your time today.