Shop OBEX P1 Docs P2 Docs Learn Events
Lynxmotion's new PROPELLER robot control board — Parallax Forums

Lynxmotion's new PROPELLER robot control board

vanmunchvanmunch Posts: 568
edited 2011-10-06 19:28 in General Discussion
Just thought some people would like to know that Lynxmotion is finishing up their new PROPELLER board that they'll be selling along with their robot kits (like the one in my avatar) :)

http://www.lynxmotion.net/viewtopic.php?f=49&t=7531&p=75777#p75777

Comments

  • GordonMcCombGordonMcComb Posts: 3,366
    edited 2011-10-03 15:18
    Looks good. Did Jim design this from scratch, or is it based on your stuff?

    For those reading this who don't know Lynxmotion's products, they're top-notch. Their servo erector set parts are simply mahvelous.

    -- Gordon
  • vanmunchvanmunch Posts: 568
    edited 2011-10-03 18:38
    Hey Gordon,

    The Bot Propeller, as they're calling it, is all Jim's work. There's a thread where they're talking about porting Xan's IK engine from the Atom to the propeller. I don't know how far they are with it, but that's the same/ foundation that they use for all their hexapods including the ones Zenta is famious for (Phenanx, A-Pod, ect.)

    Phoenix code running on Parallax Propeller
    http://www.lynxmotion.net/viewtopic.php?f=8&t=6872
    http://www.lynxmotion.net/viewtopic.php?f=3&t=7346

    I originally wrote my hexapod walking program when I was still using Basic Stamps so I didn't have enough memory or variables space to calculate everything out the way you do with IK so I created a program that uses the relationships between the leg movements. I really don't know what you would call it, but it also has a bit of what I would call physical programing. I've posted the code here so that people can use it in their own hexapods.

    Code:
    http://forums.parallax.com/showthread.php?129632-Hexapods-and-Code-for-BS2-BS2sx-and-Propeller-post-your-own&highlight=Prospero

    But back to Lynxmotion. I agree, their parts are top notch and can get you up and running in a couple of hours. They're also great for trying out ideas and everyone is very helpful. I think that it's a huge win for everyone that they're building this board. They've even created a forum especially for the Bot propeller:
    http://www.lynxmotion.net/viewforum.php?f=49&sid=0e7c6ff7643e2ccb70cf80d5db5a291e

    I can't wait to see what those guys are going to create with it! :):):)
  • sylvie369sylvie369 Posts: 1,622
    edited 2011-10-03 20:11
    vanmunch wrote: »
    I originally wrote my hexapod walking program when I was still using Basic Stamps so I didn't have enough memory or variables space to calculate everything out the way you do with IK so I created a program that uses the relationships between the leg movements. I really don't know what you would call it, but it also has a bit of what I would call physical programing. I've posted the code here so that people can use it in their own hexapods.

    "Embodied Mind", perhaps?

    This book talks quite a bit about control without a central processor representing the state of the world:
    http://www.amazon.com/Being-There-Putting-Brain-Together/dp/0262531569/ref=sr_1_9?ie=UTF8&qid=1317697765&sr=8-9
  • vanmunchvanmunch Posts: 568
    edited 2011-10-04 17:34
    Hey sylvie369,
    I hadn't heard of that before, but it sounds like the same idea. My formal background is in biology and insect behavior and I think most people would be very surprised to know how many complex, natural processes function without any kind of central controller. At the subcellular level you have self arranging protein complexes, syntheses pathways, etc. At an insect level you have complex swarm behavior that emanates from just a couple of simple rules. I think that this has really influenced how I approach robotics and programming.
    Dave
  • Bill HenningBill Henning Posts: 6,445
    edited 2011-10-06 18:34
    I picked up an 8DOF quadbot to play with; I am hoping I'll have time to assemble it and try to make it walk with RoboProp this long weekend.
  • vanmunchvanmunch Posts: 568
    edited 2011-10-06 18:58
    Cool, is it one of the kits or your own design?
  • sylvie369sylvie369 Posts: 1,622
    edited 2011-10-06 19:28
    vanmunch wrote: »
    Hey sylvie369,
    I hadn't heard of that before, but it sounds like the same idea. My formal background is in biology and insect behavior and I think most people would be very surprised to know how many complex, natural processes function without any kind of central controller. At the subcellular level you have self arranging protein complexes, syntheses pathways, etc. At an insect level you have complex swarm behavior that emanates from just a couple of simple rules. I think that this has really influenced how I approach robotics and programming.
    Dave

    I think then that you would very much enjoy Clark's book (the one I linked to above). I certainly for most of my life assumed that robot controls would have to involve representations of the world and a central controller. This despite having been convinced by the old "Game of Life" programming exercise that complex behaviors can emerge from simple, mindless, non-representational rules. Clark's book (which I'm only about halfway through) really shatters that assumption. I can't really wrap my head around the ideas needed to make practical applications, but I accept now that truly complex, flexible robots will probably use nonrepresentational methods.
Sign In or Register to comment.