Where Have All the Robots Gone?

Right after the Japanese earthquake, tsunami, and nuclear disaster, I began jokingly mentioning to friends: “At least Japan has lots of robots that can help.” Then, as the days have worn on, and the tragedies mount, I started seeing comments on Facebook and blogs, asking: “Where are the robots!?”

Where are the robots? Japan is known for its cutting-edge robotics development and deployment — it’s even known for its development of emergency robots. In all of the coverage, I haven’t seen any discussion whatsoever about bringing in bots. After the Three Mile Island and Chernobyl accidents, Red Whittaker, of Carnegie Mellon, and later RedZone Robotics, developed robots specifically for nuclear emergency applications. If such robots exist (and can handle some of the tasks being currently undertaken by humans), then why are they needlessly cooking plant workers and JSDF (self-defense forces) if they don’t need to be?


62 thoughts on “Where Have All the Robots Gone?

    1. List of power systems that failed:
      This reactor, all four other reactors, several separate connections to ‘the grid’, the grid itself, several -sets- of diesel generators and then the eight hours of batteries were exhausted. The earthquake killed some, but the tsunami wiped out the diesel generators and the fuel storage for the generators.

      So I’d amend “Why no backup generators?” into “Why weren’t the backup generators a couple stories off the ground?” Assuming sturdy concrete buildings to support them… hindsight: 20/20. Sigh.

      1. Or fuel cells. For the past few years, Japan has been rolling out some serious fuel cell advancements on a regular basis, to the point where I was surprised there were no fuel cell backups at Fukushima.

      2. From what I have read the power cables they’re laying to power the pumps are around half a mile long–unreasonably long for a tethered bot (Though there are fire fighting bots that might be able to pull a cable that long) and if the bot was battery powered it would spend a huge amount of its time and energy commuting to and from the worksite.

        Frankly virtual presence bots are still incredibly slow and unwieldy and time is of the essence at the reactors. An actual human has vastly better situational awareness and ability to respond and manipulate equipment.

    1. I meant specifically at the power plant. But then, I’ve only seen one image of a rescue robot, altho I assumed more were being used.

  1. The had backup pumps and generators the ocean took care of them. I wish you wouldn’t use the term cook. The radiation levels are not high enough to cause radiation sickness at this time. No need to make it worse that it is.
    As to where are the robots? Well maybe none of them are good enough to do what is needed.
    Kaman is talking about sending this to japan. They demonstrated a UAV version of it in 2008 and it can carry 6000 lbs. It could help with the water drops but it can not carry as big of a load as the Chinooks they are using can.

  2. Well, I’m sure there are robots performing all kinds of tasks, we’re just not seeing them. Western coverage of the quake+tsunami+aftermath has been plainly abysmal. It has been largely overshadowed by nuclear doom+gloom FUD reporting, of which 5% is reliable information and 95% is speculation. And I think I’m being generous with my percentages.

    I know from friends in Japan that there are all sorts of rebuilding/relief efforts being made on the ground, all over the country, but I haven’t seen any of that in the Western media. Watching the NGK coverage on UStream is the only placed I’ve seen it discussed at all. And just now I read that CNN is sending 8 times as many “journalists” to cover Prince William’s wedding (400) as there are on the ground in Japan (50).

    Oh, and we’re about to go to war with Libya.

    It seems it’s a good time to be an ostrich.

    1. I bailed on US coverage pretty quickly. I’ve been watching Japanese and European coverage and have still not heard about anything related to robots being used or planned. But maybe I’m just missing it.

  3. Since the reactors weren’t designed for robots and two are now twisted steaming masses of concrete and metal, robots would have to be custom designed for this. While we don’t yet have a general purpose replacement for a human, some specialized robots sould be useful here:

    1) Recon bots: – the robot in your photo, some of the French Intra robots and bomb disposal robots might be able to climb near #3s cooling pool. The U.S. drones also help with this. Would balloon or kite deployed cameras lowered into the reactor also help? How about the rovers being designed for mars or the moon, are there any prototypes which would work here?

    2) Cooling robots: Google earth shows that the reactors are no more than 200 meters from where the harbor was before the tsunami. Wouldn’t it be possible to configure a remotely controlled barge or boat with a powerful (solar/wind powered?) pump and steer it into the harbor where it could fire sea water continuously at reactors #3 and #4?

    3) Hose/tube carrying robot: Do any robots exist which can carry firehoses into a building? A firehose or series of small hoses might be able to supply a slow steady water/boric acid supply to reactor #3.

    4) Fuel removal robots: Once things have cooled down enough, we may have time to design a specialized robot which can take individual fuel rods out of the reactor, suck them into a holding device and bring them into another intact storage pool.

    1. On #4, Good God! Once you’ve got the fuel rod out, dispersing them would be a better idea (drive them into the harbor maybe?). I would think the last thing you would want in this situation would be to pile more nuclear material into a storage pool (esp. one that was just hit by a quake/tsunami) next to one that might be in the process of catastrophic failure.

  4. Robots are still ancillary and frills tools. Not to bash robots, but the most effective drones and best bomb defusing robots are used in static or overwhelmingly “asymmetric” situations. Look at it from the other end, how many robots do you think are planned to be REbuilt as the result of this tragedy? Knowing what we know now (20/20 hindsight) would you beef up your robot rescue reserves, human rescue reserves, consumable goods reserves, quake/tsunami (infra)structural reserves? Going forward (foresight), what order would you rank them? I know what I’d put at the bottom of both lists.

    If I had the choice of engineering a nuclear rescue robot or engineering and installing an ‘invincible’ cooling pipe and pump…

      1. Silicone on Sapphire and other SOI technology preventing stray currents caused by radiation from spreading to nearby circuit elements. Affected transistors still can be damaged or go transient loosing bits. The first industrial implementation of SOI was announced by IBM in August 1998. July 2006 TSMC claimed no customer wanted SOI.

        When satellite technology is damaged by radiation, they send human (into radiation) to repair.

  5. This is a good question, I was wondering the same thing. Perhaps the radiation levels are too high for the robots to work correctly?

  6. Portable power generators can be driven to the site by robots, which can also plug themselves in for a battery charge. The latter is a capability that even vacuum cleaner robots have.

    From the wikipedia article on radiation hardening, I gather than even CMOS circuits can last a while even under radiation levels measured in krads. That would be thousands of hours at the maximum levels that humans could tolerate. And of course they could be shielded.

    If telepresence systems are too unwieldly, then simple systems which show ordinary video and operate simple claw arms should suffice. Communications can be over fiber optics cables, so again radiation shouldn’t interfere.

    Little that humans do in a reactor plant is beyond the capability of a teleoperated robot. You watch gages, you push buttons, you twist valves. Right now the pumps are dead, so I’m not even sure what the humans are doing in that reactor control room. The real action is in getting water into the reactor vessels and cooling pools and then circulating it. And humans can’t go into the places where that needs to be done. Robots could go, but I see nothing in the news about that. If there were robots there, I would think the news media would have pictures plastered all over the internet.

    So, I’m as puzzled as everyone else as to why robots aren’t being used.

    1. IMO, the whole question seems grossly out of place and backwards (even the title makes it seem as if robots should ride in on a white horses and save them). If we/they lived in a world like iRobot where the consumer market for humanoid robots was considerable and robots could be stamped out by the thousands daily, I think the question might be valid. As it stands, it takes a team of humans to build and operate a single robot with a fraction of the operational capabilities of even one of the humans that built it and relatively zero intellectual capacity.

      Given the sparsity of ubiquitous and unspecialized robot labor, you seem to overlook the obvious, the humans don’t read the gauges or twist the valves, automated systems already do that. I’m not intimately familiar with nuclear regulations, but I assume multiply redundant systems do that. Why would you have multiply redundant automated systems AND a idle (or even active) army of emergency bots on top of that?

  7. There certainly are automated systems to monitor the plant. The question then is if monitoring the plant is all they can do, why are humans there? Are they there to take readings? Then why not set up a panning camera in the control room and send everyone outside the danger perimeter until they are actually needed, so that they don’t burn through their radiation allowances unnecessarily?

    “As it stands, it takes a team of humans to build and operate a single robot with a fraction of the operational capabilities of even one of the humans that built it and relatively zero intellectual capacity.”

    Okay, so it takes a team of humans to build a robot. Let’s assume that no robot has ever been built by a lone human being and always requires a team. So? There are a hundred million people in Japan, three hundred million in the US. Virtually every university has a robotics team of experienced robot builders. Teamwork is not a problem, it’s a solution.

    ” . . . with a fraction of the operational capabilities of even one of the humans that built it and relatively zero intellectual capacity.”

    I think you don’t understand that we are not talking about artificially intelligent autonomous robots. We are talking about robots that would be operated by humans via remote control. In that sense, the robot would have the full intellectual capacity of the human operator and the advisory staff.

    Many years ago, as a civilian employee of the Department of Defense, I worked around naval nuclear reactors built around the same time as these civilian reactors on much the same principles. To operate a nuclear power plant requires the ability to push buttons, flip switches, and twist valves. These are capabilities well within those of teleoperated robots built by college and even many high school teams.

    Repair of reactor plant mechanical and electrical systems would be more complicated, as would laying down planks and rugs to enable wheeled or treaded robots to navigate over debris-strewn floors. But there are no laws of physics being violated here. It’s mathematically possible. I admit that I’ve never built a robot in my life, but I’ve seen enough videos of college and high school students putting together incredible machines to know amazing things can be done. I’m not one of those people who say, “Well, I couldn’t do it, so it can’t be done.”

    I hope you’ll reconsider using phrases like ‘robots riding in like knights on white horses.’ That almost sounds mocking. There are extremely bright but shy kids who could put together astonishing machines in a matter of hours, but if we accuse them of trying to be pretentious or indulging in foolish fantasy when they’re only trying to help, you’ll intimidate them into passivity. And what we need is for them to be inspired and motivated. If there are knights, they’re not the robots, they’re the robot builders.

  8. Us students living here in Japan hasn’t seen any use of robots in the nuclear problem at Fukushima being shown in TVs here. Personally, methinks it would be a real help in inspecting the situation inside the reactors and a guide in the cooling efforts.

  9. Reality check:
    First of all most robots in Japan are in a toy stage, I’m sorry but they, they cannot be easily deployed in numbers or even one by one for complex operations like nuclear emergency. Second of all, robotics cannot operate in high radiation. Chips and mother boards last about 15 min in the chernobyl sarcophagus. The radiation levels are not as high at the jap plant but never the less if you have a prototype robot you will be sending it for a one time use like a paper plate and get minimal use out of it losing time and effort you could of applied else were in the this type of emergency.

Comments are closed.


Gareth Branwyn is a freelance writer and the former Editorial Director of Maker Media. He is the author or editor of over a dozen books on technology, DIY, and geek culture. He is currently a contributor to Boing Boing, Wink Books, and Wink Fun. And he has a new best-of writing collection and “lazy man’s memoir,” called Borg Like Me.

View more articles by Gareth Branwyn