LEGO robots in the laboratory

We've talked here before about the extremely important (and often-overlooked) DIY aspect of science. Scientists are makers. They have to be. The tools they need often aren't available any other way. Other times, the tools are available, but they're far more expensive than what you could construct out of your own ingenuity.

In this video, researchers at Cambridge build LEGO robots that automate time-consuming laboratory processes at a fraction of the cost of a "real" robot.

Video Link

Via Karyn Traphagen


  1. Several years ago i actually had the utter gall to suggest a lego robot to an HTS ‘expert’.  i was very impressed with the color change in his face before he managed a “you’re trying to be funny, are you?”  “no, i wasn’t”, said i; and then the conversation was politically and quickly changed …others thought i was being funny too.  -sigh-  if it doesn’t have a logo: TurnkeyTron Inc (and come with a microsoft windows box) and isn’t clad in shiny then the suits won’t even consider it.

  2. When I was in graduate school, we built or improvised or built nearly all of our own equipment, even fairly sophisticated lasers, vacuums, and control systems. This really is how science works! Of course we also very excited when we had the funding to buy a commercial laser, for example, because they would be much cooler and less finicky than the things we built ourselves. 

    At least in our case, I’d  say 80% of the ‘making’ was because we were doing something novel or so specific the products didn’t exist, and maybe 20% of the making was because we were in an environment where the labor (grad students and undergrads) was cheaper than the hardware. Also,  the educational benefit of doing things yourself was probably part of the value equation; sometimes people forget (professors included) that if you are working at a university, training grad students, even if time consuming and inefficient is part of the job.I work for a fairly well funded engineering company now. Even with all our equipment and rules, a lot of test situations still involve coming up with something quickly from what you already have lying around.

    The biggest surprise in all of this to me: As someone who has not played with lego(s) in nearly 20 years… they’ve come a long way since then. I didn’t even realize they were computer controlled now. I wish those existed when I was younger.

  3. “meatspace macros” – the physical equivalent of scripts and macros in software.

    The Lego NXT system is more powerful than some of the early PLCs I used to program. Not surprising when it was developed in partnership with NI, a PLC manufacturer.

  4. I’ve seen Lego mindstorms used in the university nanotech lab I used to work in. Very cheap and versatile for making something bespoke exactly as in this video. Was used similarly where I was, a laborious repeatable process.

  5. It’s a neat setup, and I totally  with the ideas of making do with what you have, and how amazing the capabilities of some things most people would think of as toys can be.

    BUT… from what they show, this looks like a simple every-day job for an autostainer, nothing custom. A top of the line new one might cost $10K, but at a glance it looks like functioning ones on eBay are going for less than the cost of a new set of Mindstorms.

  6. I had a friend that studied Robotics at a university in Melbourne (not sure which sorry). He tells me that for virtually all of the classes where actual assembling/designing/testing of robots was involved, used lego for the first 3 years of the 4 year course.

    In the fourth year they got to use some more “high-tech” robots. His conclusion was that the lego robots were often just as good for their tasks and that the computer systems and programming they used were pretty rad.

    Sadly, he now designs conveyor belts for factories in a job he describes as mind numbing.

Comments are closed.