Results 1 to 9 of 9

Thread: XL_Quad

  1. XL_Quad

    I started, to demo the USB2AX Pro and various Dxld Dynamixel devices, a little quad robot umade with XL-320.

    Click image for larger version. 

Name:	IMG_20161102_113858.jpg 
Views:	60 
Size:	108.8 KB 
ID:	6734

    Full Album : https://goo.gl/photos/okZSp7jzWNRkBvD88
    ---
    Personal blog: http://xevel.org
    USB2AX documentation: http://xevelabs.com

  2. #2
    Join Date
    Mar 2015
    Location
    Texas
    Posts
    342
    Images
    15
    Rep Power
    26

    Re: XL_Quad

    Cute! I like it.. lots of parts hey?
    01001001001000000100110001101111011101100110010100 10000001010010011011110110001001101111011101000111 0011

    My Instagram
    My YouTube

  3. #3
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,173
    Images
    27
    Rep Power
    267

    Re: XL_Quad

    Quad with 3-DOF per leg and no servos contributing to the moving mass of the leg and lots of symmetry. Quite nice, but can it turn very well? Forward/backward and strafing should be very easy, but turning seems like it might require lots of bobbing about.
    edit: or not. Now that I think about it, turning should be quite smooth.
    Last edited by tician; 11-02-2016 at 01:38 PM.
    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh

  4. Re: XL_Quad

    You might have a look at Spot Mini for ideas about motion, though it won't be as cool...
    It still misses a few parts, for the head, the lidar and dynamixel devices to read the potentiometers.
    ---
    Personal blog: http://xevel.org
    USB2AX documentation: http://xevelabs.com

  5. #5

    Re: XL_Quad

    That is frickin' beautiful!

    Why did you try potentiometers on the legs, though?

  6. #6
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,173
    Images
    27
    Rep Power
    267

    Re: XL_Quad

    Quote Originally Posted by jwatte View Post
    Why did you try potentiometers on the legs, though?
    Torsional load sensors, maybe? Difference between internal pots and external pots would give rotational displacement of shock material which would give torque if the material properties are well characterized.
    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh

  7. Re: XL_Quad

    Yep, as Tician said the potentiometer ends up serving as a torque sensor among other things.
    The torsional element is a custom piano wire spring, that is described well enough by Hook's law.

    It measures the deflection of the arm from the output of the servo, so I can use it directly to compute torque, and also add its position to the servo position to get the absolute rotation between the servo and the leg. I then plug that into FK equations (simple enough in this case) to get the foot actual position, and mix that with the output of my IMU (in the USB2AX Pro) to estimate body attitude, and ground height at each foot.

    The pots will be read by one of my new Dxld Dynamixel device per leg.

    It should end up at around 480-500g with a 1200mAh 2s1p lipo battery, ssslidar and everything plugged in.
    Last edited by Xevel; 11-03-2016 at 04:16 AM.
    ---
    Personal blog: http://xevel.org
    USB2AX documentation: http://xevelabs.com

  8. #8
    Join Date
    Jan 2008
    Location
    Norway, Stavanger
    Posts
    786
    Images
    276
    Rep Power
    72

    Re: XL_Quad

    Very nice project, creative and looking good!
    I like your setup with the springs and pots, they should give the bot good compliance in the joints. Nice feet, looks like you used a rubber wheel/roller of some sort?

    Looking forward to see it in action.
    Kåre Halvorsen aka Zenta
    ---------------------------------
    Zenta's YouTube channel
    Zenta's Blog
    Zenta's Instagram

  9. Re: XL_Quad

    Thanks all


    The feet are actually one wheel, which is the smallest and softest T40 Banebot wheels.
    All of them are made from one wheel, first cut in two along its natural plane to make two thinner wheels which are then cut again along the prologation of opposite sides the hexagonal hole, to get to that shape.

    Each foot is, at that stage, around 2g (1g is lost in the cutting process done with a cutting wheel and a Dremel).

    Click image for larger version. 

Name:	IMG_20161005_060540.jpg 
Views:	26 
Size:	75.5 KB 
ID:	6735

    The surface of the leg that is in contact with the foot/wheel has the shape of a half-hexagon which mates with the remaining half-hexagon of the foot/wheel (this leg is an old "two-parts + carbon rod reinforcement" design, back when I did not trust the support structures my Up Mini! makes - which is no longer the case).

    Click image for larger version. 

Name:	IMG_20161008_071430.jpg 
Views:	22 
Size:	57.3 KB 
ID:	6737


    In the end I make a small tapped hole in the black part to put a nylon screw in te inside of the leg, just to lock it into place, and which does not bear any of the weight/impact.

    Click image for larger version. 

Name:	IMG_20161031_052125.jpg 
Views:	20 
Size:	115.8 KB 
ID:	6736

    Banebots wheels are the best for good traction, I had already used them for my Xachikoma robot a few years back.
    ---
    Personal blog: http://xevel.org
    USB2AX documentation: http://xevelabs.com

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •