Japan’s Automated Future

Honda’s Asimo humanoid robot can dance, run and even play soccer. Sony’s now discontinued Aibo was a high-profile, poop-free replacement for man’s best friend. Paro the therapeutic seal has been designed as a companion for dementia patients and the elderly. The prototype Robear can lift and carry the infirm. Japan built its reputation as a global leader in robotics on such well-publicized machines.

Robotics, particularly in the field of home care and geriatric nursing, has been touted as a way to deal with Japan’s rapidly aging population. At the other end of the same problem, robots offer a solution to the dilemma of the country’s dwindling workforce.
Robots are also expected to play a part in the hazardous, decades-long cleanup of the crippled Fukushima No. 1 nuclear power plant. To the surprise of many, Japan had no robots ready to deploy immediately after the disaster of March 11, 2011, and the country still doesn’t.

Paulo Debenest (pictured) is one of the co-founders of HiBot, a developer and manufacturer of robots for extreme environments and hazardous operations, such as underwater and land search and rescue and power line inspection. iNTOUCH’s Rob Goss visited the Club Member at HiBot’s head office in Tokyo to glean an insight into the world of robotics in Japan. Excerpts:

iNTOUCH: As someone who creates robots, how would you define them?

Debenest: It’s a very fluid term. The word “robot” was created in 1921 by Karel Čapek, a Czech writer, to denote a fictional automaton in a play, and, in simple terms, it means a machine that does a job instead of a human. If we think about it this way, even a washing machine is a robot, although not a very advanced one. For me, a robot is a machine that does a job with a lot of autonomy. For others, it might mean a machine with some form of artificial intelligence.

iNTOUCH: How accurate is it to say that Japan is a leader in the field of robotics?

Debenest: It’s accurate to an extent. The amount of research done here is incredible, and when it comes to creating new shapes and machines, Japan still leads. That is the research side. When it comes to application, however, Japanese companies tend to be very conservative. They want to see other companies using a machine before they adopt it.

iNTOUCH: So where is research in Japan focused now?

Debenest: There is a very strong emphasis on humanoid robots here. One of HiBot’s co-founders, Naho Kitano, who was doing some very interesting research into robo ethics when we launched the company, explained to me once that this has its roots in Shintoism. People like to put a face on an object to make it feel more human. Japanese tend to want to see a human where there is a robot. The goal of having a humanoid robot doing things and behaving like a human is quite far away and it isn’t something we focus on at HiBot, but we are seeing very useful things like artificial intelligence, touch sensors and visual recognition coming out of humanoid research that can be applied to other machines. There is also a lot of research going into underwater robotics. Deepwater robotics is an area where we will see some very interesting developments over the next few years. There’s a lot going on in healthcare, too, although progress with this is slower because of the high regulatory barriers. In many other fields of robotics, one benefit for developers is a lack of regulations in Japan. Everything is so new that we are helping to create the regulations as we go. We can get a product to market very quickly. On average, we can have a functional prototype ready in a year and a product within two years. It’s a process where we are always upgrading; there’s constant feedback and development, and there’s always something that comes up in the field that we didn’t think about in the lab. But we don’t suffer because of overregulation.

iNTOUCH: Given Japan’s reputation as a leader in robotics, why didn’t the country have any suitable machines to use in the immediate aftermath of the nuclear disaster at Fukushima four years ago?

Debenest: Actually, after the Tokaimura nuclear accident in 1999, the Japanese government decided to allocate more money to emergency robotics, and, in the early 2000s, one of HiBot’s founders, Professor Shigeo Hirose, built, successfully tested and got government approval for several machines for emergency use inside nuclear power plants. The government acknowledged that the machines were functional, but funding was cut because no one expected that a nuclear accident on the scale of Fukushima might ever happen. Instead, after 2004, funding started to get redirected into the development of things such as humanoid robotics. After the accident, Professor Hirose was extremely upset—the machines had been developed, but the backing wasn’t there.

iNTOUCH: What has development been like since 3/11?

Debenest: Since the accident happened, we have seen several machines developed in Japan to operate in the nuclear plant, but still it is very difficult to see anything that has been adopted or recommended by the government. These things have mostly only been developed in the laboratory. At HiBot, we have several ongoing projects in Fukushima in collaboration with other companies, and some of these robots will go into use there this year. One of these robots will be used inside the damaged reactor buildings, where people cannot enter, to get videos and radiation readings. Another is being considered for underwater operation inside the flooded pressure vessels. There is also one machine that was developed at the Chiba Institute of Technology that was actually used in the power plant, but, of course, they found many problems with it as it was developed in the lab. That is part of the cycle of development: you find problems, you make a new version, you use it again and so on, but it takes a lot of funding to do this. The Japanese government is putting some funding into this field, but the tendency is for them to change their focus, and so I think there are very few companies willing to put their effort into the field. Other solutions are coming from overseas. France, for example, has a lot of nuclear plants and they have developed some very interesting machines, and the US has developed machines for military functions that can be adapted.

Photo courtesy of Riken