Cockroach Pimps A Sweet Ride

This giant Madagascar hissing cockroach rides proudly atop his three-wheeled robotic platform. This project from several years ago is new to us and our reaction to the video after the break is mixed. We find ourselves creeped out, delighted, amazed, and saddened.

The cockroach controlled robot uses a trackball type input. A ping-pong ball is spun by a cockroach perched on top. The lucky or tortured (depending on how you look at it) little bug has an array of lights in front of it that illuminate when obstacles are in front of the robot. The roach’s natural aversion to light should make it move its legs away from that part of the display, thereby moving the robot away from the obstruction.

We’ve seen some bio-hacking in the past. There were robots that run off of rat brain cells and remote controlled beetles. But none of these projects make us want to get into this type of experimentation. How about you?

[vimeo=http://vimeo.com/2398096]

[Thanks Scottman]

36 thoughts on “Cockroach Pimps A Sweet Ride

  1. Does it count as torture if the bug doesn’t have self-awareness? and either way, its a bug. are there any rights for bugs? no. if it were a human controller being forced into it, then that would be bad, but bugs being forced to do our work, why not?

  2. What cockroach wouldn’t want to control a giant (to them) robot? Add a hiss triggered, auto-tracking gun or laser turret to the robot, and send it on it’s way to world domination.

    Honestly, you could scale this up to human size, and geeks would be lining up around the block to be velcroed into it. So wouldn’t call this any more torturous than whatever other form of captivity you would be keeping the roach in.

  3. Totally what I’m saying NatureTM! I hope we never give the cockroach robots power to self replicate other cockroach robots…but then again I really would like to see it. Unstopable.

  4. to djrussell:
    good point, this seems to be more of an exercise in cockroach research then anything else, i mean, anytime we use a remote controlled vehicle (or drive a car for that matter), it’s our brains that control it, they just used a smaller packaged biological to be the controller. they definitely put in the work on this, good job on that.

  5. I think more time was put into the BS “reasoning” than the actual implementation. The idea is certainly interesting, but:

    A.) That distance sensor light-array “immersion” setup does not seem to work very well.

    B.) We learn/are told nothing about the cockroach’s response to the stimuli independent of the robotic platform. A more useful starting point would be a virtual environment feeding the array, with some mechanism of determining whether the cockroach was reacting to the array in the way that the researcher hypothesized. If I remember correctly from biology cockroaches have far more senses than vision, and the huge assumption in this project is that the cockroach will react primarily to the display and not other environmental factors.

    I like the concept, but wish that the project was executed/explained in a stronger manner. It’s like Skinner without training, trying to piggyback on natural response.

    http://en.wikipedia.org/wiki/Project_Pigeon

  6. Don’t get me wrong, this is a neat project but I don’t see current application.

    “The robot and insect display attributes like: unpredictability, laziness, irrationality, and emotional response.” When designing a robot usually those “attributes” are not on my list of intended features(or are very low on my list). Now if the robot had a uC deciding when to run the lazy() routine, that would be more interesting to me.

    We know that living things can control robots; we’ve built ROVs that dive to the depths of our oceans, diffuse explosives, and go to mars. Hell we’ve even trained monkeys and other animals to control robotic arms and the like.

    I do like how the guy essentially augmented the roach’s sense of sight, and it would be more interesting if that wasn’t the roach’s only sense.

    Sure it’s neat, but this doesn’t seem that significantly different than me controlling my 2 ton vehicle on a daily basis. Sure I may have higher cognitive abilities, and a different way of sensing than the roach, but it’s a very similar concept.

  7. Is there an entomologist on the team? Couldn’t you somehow incorporate the antennae as a sensory input instead of only visual cues? Look at its antennae “flail” about when it crawls. It has to be a confusing experience for the bug. Maybe if some objects could be placed in and out of reach of the antennae as it moves, it might get more info about obstacles vs. small objects it can walk over, etc.

  8. “Couldn’t you somehow incorporate the antennae as a sensory input instead of only visual cues? Look at its antennae “flail” about when it crawls. It has to be a confusing experience for the bug.”

    Speaking from experience as an amateur entomologist(admittedly, Im more arachnologist, but whatever) no. Antennae are *very* complex sensory devices relying not only on touch but also chemical sensing, changes in air currents, and vibrations to do their jobs. I highly doubt the roach is capable of true “disorientation” but perhaps it is missing a facet of its normal locomotion…its worth noting that they often lose their antennae anyway, by the end of their lives, so they are more accessory sensory organs than necessary ones.Chapter 12 in Entomology by Gillott might be of interest to you, though, Joe.

  9. Never in my life have I had saddened feelings for such a vile creature. But after watching this it just makes me think that this is more torture than it’s worth. I mean really? Does he have to be strapped in like that?

  10. Heratiki, I emailed the guy shortly after he originally put this project up on the internet(06 according to my email archives) and he only keeps the roaches on the machine for short periods, and the velcro really shouldnt effect them at all, especially because he is very careful to keep it on just a single segment of the body. I dont see any reason why any part of this project would cause them problems…

  11. Is any one else reminded of the Daleks from Doctor Who? If that thing starts saying “Ex-termin-ate! Ex-termin-ate!” I’m the hell out of here. Anyone up for building a Tardis?

  12. I think the distance sensors and 8×8 matricies should be replaced with some lcds and a couple webcams near the bottom of the robot. Then he would have a better feeling of a real environment than just light intensity (but still lacking input for his other senses). It doesn’t seem like they thought this through very well.

  13. I don’t think I’ve ever said this, and I’m probably never going to say it again, but “That poor cockroach!”

    Also, the ‘ROACH’ on the schematic drawn like a chip might be one of the funniest parts of this whole thing.

  14. This would be fun just to see how it turns out. I can’t see any other real use for this. The light sensor isn’t enough stimuli for the roach imo.

    I bet it was sure fun building a simple trackball rover and letting the roach control it.

  15. Seems a bit cruel, and surely they could remove the cockroach, and with minor modifications it would work without it – just have the light input control the direction directly.

  16. Is it just me or does his voice sound like it should be in a consperacy therory vidio.
    I really don’t get the desire to put live animals/parts there off into a computer system as the controller.
    it just adds another level of difficulty in the build.
    feeding the thing would be a pain. how would you program it in the field if you can indeed program it at all.
    on the other hand i would love to see a project using live muscle tissue. doesn’t have the argument about sentience. also gets around battery life since you can pack a lot of energy into sugar solutions/pppt.
    and a hell of a lot of strength.
    and suddenly I’m a mad scientist.

  17. When closing in on the wall the sensors react to late and instead of only lighting the middle light it puts on all the lights. The machine goes too fast, I think. And it should be able to climb walls, because it’s possible according to the roach.

  18. yeah, they should supersize and weaponize, but leave the madagascar cockroach at the controls. that would make for some awesome news coverage. deploy the cockroachdroids and evacuate the troops. let the bugs eradicate the humans for once. just don’t let it loose on our turf.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

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