Professional looking dog robot was actually [Martin's] Master’s Thesis

If you think this thing looks good you should see it move. [Martin Smith] hit a home run on the project, which was his Master’s Thesis. Fifteen servo motors provide a way for the bot to move around. Having been modeled after a small canine the gait is very realistic. The tail is even functional, acting as a counterweight when moving the legs.

The project was meticulously built in a 3D environment before undertaking any physical assembly. The mechanical parts are all either milled from aluminum or 3D printed. Two mBed boards mounted on its back allow it to interact with its environment. One of them handles image processing, the other drives the array of motors. And of course it doesn’t hurt that he built some Larson Scanners in as eyes.

Don’t miss the video after the break which shows off the entire project from planning to demonstration. We can’t help but be reminded of the rat-thing from Snow Crash.

27 thoughts on “Professional looking dog robot was actually [Martin's] Master’s Thesis

      1. Any day now of course.

        Seriously, this thing is too overdone, even if someone cured cancer this video would be annoying, this isn’t some revolutionay thing that will change mankind.

      1. Yeah, I don’t get why everyone is hating the video. I think it was the choice of music. Does that song conjure up thoughts of world changing grandiose things? I wonder where else it was used to make people think that way. I liked the video. It was clear and to the point. Not just some dude shakely holding up a camera ( I can invent words if I want to ) for 14 minutes barely talking in an inaudible voice explaining next to nothing. Good job on the robot and the video.

  1. “Having been modeled after a small canine the gait is very realistic”. Not really. Otherwise cool project, visual and speech recognition were probably very hard, but now please learn it to walk better :)

    1. learn it to walk better? Oh god.

      First of all it would be teach it to walk better, but this is a robot. He needs to program it to walk better.

    1. I did too until I saw it walk. Looks like it’s nearly falling over, and the tail may be the only thing keeping this from happening. As such, I see it more as a workaround for poorly implemented kinematics, rather than a nice touch.

      Hate to be negative, as there’s obviously a lot of skill and care that went into other aspects of the design. But we are living in the age of Boston Dynamic’s Big Dog. Or in the hobby domain, that ant robot (with the creepily-realistic inverse kinematics and Coke can, if you’ve seen it you know the one). Maybe those are impossibly high standards, but to focus so much on form while neglecting kinematic function still seems wrong. And the HAD editorial did prep us to expect much better, with the very first sentence – “If you think this thing looks good you should see it move.” Or was that meant sarcastically?

      1. I agree with everything you said. It’s an amazing dog, lots of skill and love went into it. I was just expecting a Gallop or Trot. If it were me Id just copy the Big Dog in every way possible. Why re-invent the wheel right? Then I’d sell it to the highest bidder and rule the World. _joke_

  2. That walking speed, is it that low because of hardware or software? Just wondering if the servo’s aren’t quick enough or the microcontroller just isn’t quick enough to calculate the movement on the fly.

  3. @Squirrel You owe me a new keyboard :-)

    A robot cat would be useful, use some videos of a few cats playing as training material for the neural net and 3D model to get lifelike motions.
    Ideal for people who live on their own who can’t have a pet for hygiene and/or mobility reasons.

    With modern CPUs such as the RPi and Exynos 4 you could build all sorts of extra features such as the ability to monitor biosigns and alert the authorities in the event of a fall etc.
    Much less invasive than one of those alert bracelets methinks.

  4. Anyways, that’s quite a nice build,
    liked the music, loved the counter-balancing tail, wondering in what specific subject he did his thesis.

    Plain old ordinary computer science, system integration or even industry design ?

  5. It is a very impressive build and is very easy on the eyes. Have you considered using Bayesian, GA or neural nets to evolve a better walk pattern? Seems there should be a way to move more than one servo at a time to speed up the walk without it falling over, and neural nets can do amazing things in this application.

  6. I dont understand some negative comments.. Do you really think you could do it better? i doubt it.
    I know the guy , and he is an excellent electronics engineer, making this from scratch proves that he is awesome engineer.
    Dont bother about bad comments Mart, they are probably jelous. Congratulations!!!!!!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s