Update on [James'] bipedal robot

From the looks of the latest update [James] has made quite a bit of progress on his bipedal robot. He added to the top of the post just a few days ago, but didn’t include the video link which you’ll find embedded after the break. There’s about ten minutes of explanation before he gets down to demonstrating the static and dynamic balance which can be chosen using the buttons on a TV remote.

We looked in on the project about one year ago. The most notable change is the control electronics anchored in the torso of the robot. At first it makes us a bit nervous that he hasn’t built a protective cage around the components. But after seeing the latest stability demonstration we guess it’s because this thing is fantastic at staying upright. The torso is connected at the hips in such a way that no matter where each leg is it will always remain upright. All together the thing stands twenty-six inches tall, but that will grow when he gets around to building a head for it.

Comments

  1. Kaaaaaang says:

    I like that rowbot, because that rowbot is the coolest rowbot I’ve ever seen.
    P.S.: Rowbot! Rowbot!

  2. garym53 says:

    These days just about any animated object is being called a robot.
    In my opinion a robot is, effectively, a mechanical human being, I may allow mechanical animals or insects in some circumstances – but for sure and certain a robot does not move on wheels! (unless it can sprout legs or something to get it up a flight of stairs) This being a biped is getting there, but the use of R/C servos drags it back into the mechanical toy realm unfortunately.

    I note he says something like the electronics for the wiper motors were horrendous and that the R/C servos are much easier to use – I suppose that is true in the sense there are off the shelf solutions – but obviously, with a little effort, a wiper motor, or any other motor, can be made to work as a servo, so I don’t really see the point – it may be that using a “model” is a good way of testing software and designs etc.

    Having said that we don’t see many hacks turning cheap powerful motors into servos for some reason – I wonder why?

    • octel says:

      “In my opinion a robot is, effectively, a mechanical human being, ”

      You’re thinking of androids.

      • andar_b says:

        Yeah, a robot is usually thought of as an automaton of some sort which can carry out actions without human intervention. Android does indeed seem to be the more appropriate term for humaniform robots.

        • James Bruton says:

          I guess that’s why I called the project ‘Android 11′. The aim was to implement bipedal locomotion as cheaply and easily as possible, and build the android as a development platform – so it’s going to get further development.

      • garym53 says:

        Nope, the term “robot” comes from the play R.U.R by by Karel Čapek.

        R.U.R. stands for Rossum’s Universal Robots

        the word robot comes from the czech word robota, meaning literally serf labour…

        • xef6 says:

          Etymological origin and modern operational meaning are two different things. If you look up the modern meaning of robot, you would know that robot does not necessarily imply humanoid form. It’s fun to nerd out about the origin of words, but modern lexicon is also a “real thing”.

          • garym53 says:

            Well you can change the meaning of any word you like but that doesn’t make it correct, just popular – the two are NOT synonymous. Hitler and ABBA proved that.

            Of course I know the modern dictionary meaning of the term.

            I have given you one reference, you have given me another that I just happen to disagree with.

          • garym53 says:

            Well, in a way, that is my point. Dictionaries don’t just define something off their own bat – they follow common usage – if the common usage is wrong – as in this case – the dictionary is wrong also. I don’t need a dictionary to tell me what a robot is – I have lived and breathed them for over 50 years

    • Hitek146 says:

      He actually does have a write-up on how he made his own linear actuators out of threaded rod and dollar-store cordless screwdrivers, with feedback and servo control through the circuits he built. He also mentions adding feedback to the DC motor gearboxes he uses for some of the joints that require more torque than a standard servo, and less torque than the linear actuators and wind-shield wiper motors he used in the past…

      • garym53 says:

        Indeed, I have been following his work for years and it looked promising – that is why I am disappointed at this effort – OTOH if it is just a “model” to develop the necessary software then that is fine.

  3. chrisc says:

    I understand linking the servos directly to the moving joints (I reluctantly had to do so recently on a project of mine), but I’d like to think it not beyond the hacking community to do some passive dynamic walking, and servos with elastic elements. It’s harder but can be much more efficient, and once the control mechanism is cracked could be more acrobatic as well.
    Maybe it would require some sort of evolutionary or neural network approach.
    Anyway good work James. Always look forward to your vids.

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

Follow

Get every new post delivered to your Inbox.

Join 91,306 other followers