Automating The Disinfection Of Large Spaces With Robots

What do you do when you have to disinfect an entire warehouse? You could send a group of people through the place with UV-C lamps, but that would take a long time as said humans cannot be in the same area as the UV-C radiation, as much as they may like the smell of BBQ chicken. Constantly repositioning the lamps or installing countless lamps would get in the way during normal operation. The answer is to strap UV-C lights to a robot according to MIT’s CSAIL, and have it ride around the space.

As can be seen in the video (also embedded after the break), a CSAIL group has been working with telepresence robotics company Ava Robotics and the Greater Boston Food Bank (GBFB). Their goal was to create a robotic system that could autonomously disinfect a GBFB warehouse using UV-C without exposing any humans to the harmful radiation. While the robotics can be controlled remotely, they can also map the space and navigate between waypoints.

While testing the system, the team used a UV-C dosimeter to confirm the effectiveness of this setup. With the robot driving along at a leisurely 0.22 miles per hour (~0.35 kilometer per hour), it was able to cover approximately 4,000 square feet (~372 square meter) in about half an hour. They estimated that about 90% of viruses like SARS-CoV-2 could be neutralized this way.

During trial runs, they discovered the need to have the robot adapt to the constantly changing layout of the warehouse, including which aisles require which UV-C depending on how full they are. Having multiple of these robots in the same space coordinate with each other would also be a useful feature addition.

29 thoughts on “Automating The Disinfection Of Large Spaces With Robots

    1. At room temperature, it appears that after a reasonably short period, there is insignificant viable virus. The question is still “how long”. The early numbers weer guesses. More current numbers are still mostly extrapolations from similar viruses, but there is getting to be sufficient data to give a value for the present virus of note.

      See: for one, and for another

      The National Institute of Allergy and Infectious Diseases found viable virus in air for three hours (NEJM, 17mar2020)

      Surface viability numbers are a bit less specific, but it appears up to three day on smooth surfaces that are non hostile (copper is maybe four hours, by several studies. Copper is hostile to most life, including us) ad a day or so on cardboard and similar that have not been specially treated.

      Shut the door for the weekend is fine, on the weekend. But as of 8AM monday (figuring typical US 7AM start), the facility is probably contaminated again. Basic good hygiene procedures are the primary protection, but ongoing disinfection is good practice. Whether UV-C in an open environment is worthwhile is a different question, but a 90% reduction in viability is a believable number. Unfortunately, I would suspect the surfaces untouched are also lifely to be the biggest concern for surface transmission, and, as the UV-C source can’t be where the people are, I doubt it will do much for what appears to be the primary transmission route- aerosols between those occupying the same space

    2. It depends on who you ask, a few months back it got up to 72 hours, and even briefly peaked at 96 plus in some circles, but yeah, seems to be back down to 24 or so now. It’s still 4 to 6 weeks for packages from China though, or so I am told.

      1. Don’t worry about Chinese packages, even if they can deliver the virus overnight. The Chinese spread their pandemics using infected people flying around in aluminum cans full of soon to be infected people – much more efficient.

    3. odd. My reply with references seems to be hung up.

      Short summary: it isn’t simple, and the difference between detectable virus and viable virus is kinda important. Overall, it depends, and 24 hours is as good a number as any for many things. Of course, locking the doors for the weekend only works until the door is opened. Once the monday first shift shows up at 7AM, the clock starts, and before 8AM everything must be presumed contaminated again.

      1. @cliff claven said: “odd. My reply with references seems to be hung up.”

        It seems to me that if you post more than one link, the comment goes to “moderation” for a human (or robot) to check the links for SPAM, and there’s no message warning you about this. If you’re lucky your comment will eventually reappear. But then again, I’ve had completely legitimate comments with a couple of on-topic legit links just go POOF – and vanish into the big bit-bucket in the sky. To say the least, this comment system on HaD is a crap-shoot. I long for the days of all lower-case.

        1. This, kinda. The robots bring the “questionable, comments to us, we have a look, and push them out to the world. Of course, what constitutes “questionable” is an open question.

          Sometime about a month ago, the robots went crazy. We’ve raised tickets, but the response so far is “it’s an AI algorithm, so just keep moderating and it will work itself out.”

          Anywhoo, the human moderation can take some time sometimes. Apologies. But if a comment doesn’t make it through after a few hours, please let us know:

          1. Thanks for the detail. I figured it would work itself out of not, and either way, so be it. Lets be honest: NOTHING here is likely to be that critical.

            I will say that “just keep moderating and it will work itself out.” sound more like advise for constipation that administrating a tech system, but given what pops up in the comments at times, maybe that is not purely coincidental.

          2. Thanks for the insight. It seems a bit tweaky about some brand names, I guess because of spam bots posting things similar to “get free eye-pads” and other scams.

          3. I’ve been happy with the moderating and metamoderaring system on Slashdot. I think SlashCode is free and open too.

            I also note that many of the good stories, and good commenters, seem to be on HaD now :-)

            So I guess my dream is to see Slashdot style moderation, with the Hackaday community.

            Just don’t copy their mobile site – it sucks.

    4. Depends on a lot of factors. Porosity of the surface, temperature, and amount of UV-C exposure to name just a few.

      Smooth things in outdoor environments with direct sun exposure probably nuke the virus in less than 24 hours. Something with a lot of nooks and crannies (e.g. used face mask or tissue) in a dark, cold location might have the virus last a very long time.

    1. Yes, that’s pretty unavoidable. But if it’s significant, depends on how much exposure there will be over a longer time period.
      Luckily, in warehouses almost everything is metal or cardboard, neither of which is much harmed by the uvc. Paint may discolor but probably won’t disintegrate.
      Polystyrene cups will rapidly succumb, though! Gotta make sure the coffee machine’s cup stock is neatly stowed away before sterilizing.

    2. There’s also the opposite problem. The reason they don’t use UV light in hospitals is because it doesn’t reach nooks and crannies and so does a terrible job of disinfecting surfaces.

    1. Like typical warehousing tasks I assume they have more than one and they pull shifts while one charges others are about doing the business. These industrial robots can go anywhere between 2 and 5 hours on a charge (not counting the lamps which would be a considerable load). Exposure times at around 2-3 meters would be around 10 minutes to get 99.9% disinfection rates (obviously depends on the lamps, just eyeballed those). Would definitely take more than 1 robot to cover such a huge space in 8 hours of downtime.

    2. Do you think that the people who designed built and tested the machine might have thought about how much power would be needed to do the job at any point during the building development and testing process?

    3. Forget batteries.

      Build this thing with a small propane engine and a generator. Might as well turn the wheels directly from the engine, skipping the electricity while you are at it.

    4. If you run an electric motor slowly it does not use much power. The video clearly shows that this robot has a charging point which it can dock to. So how about that they have already worked out how long it can run between charges and allowed for that? Maybe you should, watch the video and read their write-up before commenting at random?

  1. I can’t look at this and not see an easy, lower level video game villain. Or maybe a new kind of Dalek.

    Well, anyway. What about all the surfaces that are not exposed? Each box has 6 sides and they are stacked together.

  2. What about putting the UV-C tubes on the bottoms of forklifts that are already in use in the warehouse? Surround them with a brush-like gasket to contain the UV rays. Then as your forklifts move around the warehouse they’re also spraying UV-C.

    Probably not a huge benefit, but every little bit helps, right?

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.