Page 4 of 6 FirstFirst 123456 LastLast
Results 31 to 40 of 60

Thread: MX-Phoenix hexapod and some more

  1. #31

    Re: MX-Phoenix hexapod and some more

    Hi Kåre,

    Looks like you are making some great progress. Like your new YouTube video!

    Kurt

  2. #32
    Join Date
    Jan 2008
    Location
    Norway, Stavanger
    Posts
    784
    Images
    276
    Rep Power
    71

    New video and some outdoor pics

    Quote Originally Posted by KurtEck View Post
    Hi Kåre,

    Looks like you are making some great progress. Like your new YouTube video!

    Kurt
    Thanks Kurt!
    I'm glad you liked it! I've not had time to update this post until now.
    Yesterday the weather was good enough to do some real terrain walking. At this time of the year we doesn't have much sun either..

    Actually my first attempt yesterday didn't go so very well. For some reason the cold outdoor temperature affected 3 of the foot sensors. It was a quick fix though. I enlarged the hole through the tibia part with 0.1mm for the shaft holding the footpad. I also added an extra spring since little spring on the switches was a little to weak to give a reliable function. So far I've not had one faulty foot sensor. A close-up picture of the foot sensor:
    Click image for larger version. 

Name:	MX_Phoenix_Foot_01.jpg 
Views:	23 
Size:	90.0 KB 
ID:	6815

    I took my six years old son outdoor again for more testing and some photos and a video!

    Click image for larger version. 

Name:	MX_Phoenix_Outdoor_02.jpg 
Views:	26 
Size:	140.6 KB 
ID:	6816

    There are still a lot of work left to make the terrain adaptation and the gait algorithm to work perfect. Some photos of MX-Phoenix posing outdoor.

    Click image for larger version. 

Name:	MX_Phoenix_Outdoor_04.jpg 
Views:	15 
Size:	212.4 KB 
ID:	6817

    Click image for larger version. 

Name:	MX_Phoenix_Outdoor_01.jpg 
Views:	15 
Size:	245.8 KB 
ID:	6818

    Click image for larger version. 

Name:	MX_Phoenix_Outdoor_03.jpg 
Views:	16 
Size:	259.6 KB 
ID:	6819

    I also made a little video of it in action. Will hopefully post a couple more for demonstrating all the new features. I'm using the Phoenix code as a base but I've written a new gait algorithm and done several changes all over.. One day if I find the code reliable and good enough I'll share it (if anyone find it interesting).



    Enjoy!
    Kåre Halvorsen aka Zenta
    ---------------------------------
    Zenta's YouTube channel
    Zenta's Blog
    Zenta's Instagram

  3. #33
    Join Date
    Mar 2015
    Location
    Texas
    Posts
    338
    Images
    15
    Rep Power
    25

    Re: MX-Phoenix hexapod and some more

    That terrain video

    Such nice work Kåre! It moves so smooth and really keeps a level body. Are you using any type of inertial sensing or does the body remain level with the foot sensors and new gait engine?

    The tension springs seemed to have helped with your backlash. One thing I've been somewhat curious about is if you could get away from mechanical switches and use something like a short range IR sensor inside the tip of each foot. Regardless it is certainly cool to see your solution in action. Looking forward to seeing more.
    01001001001000000100110001101111011101100110010100 10000001010010011011110110001001101111011101000111 0011

    My Instagram
    My YouTube

  4. #34
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,159
    Images
    27
    Rep Power
    265

    Re: MX-Phoenix hexapod and some more

    I would expect cheap hall effect sensors to be a better choice than optical sensors. Had been planning last year to make some trigger-type buttons using some small rare-earth magnets and Allegro A1304 sensors. Never got around to ordering any parts since I've been low on cash, but was hoping to make fully-sealed dual-trigger boards for controlling Scooty-Puff (one board for each hand: one throttle and dead-man switch; one brake and dead-man switch).

    Surprisingly, I did keep some records (notes and octave code, but not McMaster part numbers for the springs in the triggers) of that project in the eagle board file:
    B_r = 14800;

    L = ((1/8)*25.4);

    R = (((1/8)/2)*25.4);

    X=[3.75:.125:16];

    Fs = (B_r/2).*( ( (L+X) ./ sqrt(R^2+(L+X).^2)) - (X ./ sqrt(R^2+X.^2)) );

    plot(X,1.65-(Fs.*0.004))



    http://www.kjmagnetics.com/proddetail.asp?prod=D22-N52

    Allegro A1304

    3.75~16mm from sensor

    6~20mm from board?
    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh

  5. #35
    Join Date
    Jan 2008
    Location
    Norway, Stavanger
    Posts
    784
    Images
    276
    Rep Power
    71

    Re: MX-Phoenix hexapod and some more

    I'm sorry for my very late reply.
    Quote Originally Posted by r3n33 View Post
    Such nice work Kåre! It moves so smooth and really keeps a level body. Are you using any type of inertial sensing or does the body remain level with the foot sensors and new gait engine?
    Thanks for your nice comment r3n33! Currently I'm not using any IMU sensors yet. What you saw on the video was a very crude way to solve TA with several bugs. I'm currently struggling inside a rabbit hole.. What I'm currently working on is to make the hexapod orient the body based on every legs position and then equalize the leg values. This is done by calculating how much the body is rotated based on the current leg positions (I'm currently using the balance mode for that), then use these values as an input for a rotation matrix. Then the gait positions gets adjusted by the result of the rotation matrix. Now the body is "level" to the terrain. I do this process every time all six legs are on the ground at the same time. My next step is to add an IMU so that the hexapod can determine how it is oriented. Should be useful to shift the body when walking uphill/downhill.

    My main problem now is that the balance mode doesn't work properly (it never had actually). I did compensate the problem by simply lowering the center of rotation. (You can see the "PosY +150" in the single leg balance part in the Phoenix code) I know what the cause is, but I'm not sure how to solve it. It drives me crazy.. Today it somehow work but not as good as I want it to. My math skills are limited. Maybe I need to learn how inverse rotation matrix really works, or if that is the solution.


    Quote Originally Posted by r3n33 View Post
    The tension springs seemed to have helped with your backlash. One thing I've been somewhat curious about is if you could get away from mechanical switches and use something like a short range IR sensor inside the tip of each foot. Regardless it is certainly cool to see your solution in action. Looking forward to seeing more.
    Good idea. I find a simple mechanical switch to be pretty reliable. But using it in combination with a short-range sensor to determine the actual distance to the ground or to detect obstacles would be useful. Then it also start to be really complex. I believe in real world when walking outdoor on different surfaces like tall grass or moss a simple foot switch should be reliable enough (KISS).

    Quote Originally Posted by tician
    I would expect cheap hall effect sensors to be a better choice than optical sensors.
    I'm not sure how a hall effect sensor can detect ground, don't you need a magnetic field to activate the sensor? Or did you think instead of a physical switch?
    Kåre Halvorsen aka Zenta
    ---------------------------------
    Zenta's YouTube channel
    Zenta's Blog
    Zenta's Instagram

  6. #36
    Join Date
    Sep 2010
    Location
    ಠ_ಠ
    Posts
    2,159
    Images
    27
    Rep Power
    265

    Re: MX-Phoenix hexapod and some more

    I was expecting the optical sensor to be used with the existing foot and sliding rod, and not used to directly measure the distance to the ground.

    Basically, just place a small magnet on the rod currently actuating the switch and replace the switch with a hall effect sensor. It would be better protected against dust and would also give you an output proportional to the distance of the magnet from the sensor. If you know the spring constant of the spring resetting the foot tip, then you could get a fairly accurate measure of the loading on each foot. A non-contact alternative to using small FSRs in the feet.
    Please pardon the pedantry... and the profanity... and the convoluted speech pattern...
    "You have failed me, Brain!"
    bleh

  7. #37
    Join Date
    Jan 2008
    Location
    Norway, Stavanger
    Posts
    784
    Images
    276
    Rep Power
    71

    Re: MX-Phoenix hexapod and some more

    Thanks, that's a really good idea Tician!
    I might try that one day. I think I have some in my workshop, somewhere..
    Kåre Halvorsen aka Zenta
    ---------------------------------
    Zenta's YouTube channel
    Zenta's Blog
    Zenta's Instagram

  8. #38
    Join Date
    Jan 2008
    Location
    Norway, Stavanger
    Posts
    784
    Images
    276
    Rep Power
    71

    Re: MX-Phoenix hexapod and some more

    Hi,

    I've done lots of work on the code lately. Did post some short vids on my Instagram.

    This video demonstrate smooth gait transitions while walking. It start with wave gait, then ripple, mechatronic and tripod gait.

    The Terrain Adaptation is also improved and faster. I've also added a feature that makes the leg restart the swing cycle if it doesn't hit ground (walking into a hole or something to far away).

    Until recently the hex walked at constant speed. As an option I've added the accelerated walking that was default on the original phoenix code.

    Running a gait cycle with lots more steps than used in the Phoenix code, its much easier to do really slow walk and still perform fast body rotation.

    There are still some issues I need to solve.
    Having fun though.
    Kåre Halvorsen aka Zenta
    ---------------------------------
    Zenta's YouTube channel
    Zenta's Blog
    Zenta's Instagram

  9. #39

    Re: MX-Phoenix hexapod and some more

    Great stuff! I really appreciate the smooth transitions and adaptation code!

    Gets me inspired to get my PhantomX back in one piece and start playing again!

    Glad you are having fun!
    Kurt

  10. #40
    Join Date
    Jan 2008
    Location
    Norway, Stavanger
    Posts
    784
    Images
    276
    Rep Power
    71

    Re: MX-Phoenix hexapod and some more

    Thanks Kurt!
    Glad you liked it. If others haven't already made it, I'm thinking of making a 3D printed foot with sensor for the PhantomX. Would be fun to try the code on it too.
    Kåre Halvorsen aka Zenta
    ---------------------------------
    Zenta's YouTube channel
    Zenta's Blog
    Zenta's Instagram

Thread Information

Users Browsing this Thread

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

Similar Threads

  1. Question(s) Phoenix Hexapod
    By nimda in forum Humanoids, Walkers & Crawlers
    Replies: 24
    Last Post: 04-06-2016, 11:40 AM
  2. KURT< HELP!!!!!!!!! Phantom X Phoenix code to run on my hexapod.
    By gregorio_626 in forum Projects For Request and/or Sale
    Replies: 8
    Last Post: 01-26-2016, 05:51 PM
  3. Sale: Black Phoenix Hexapod (complete)
    By BlackDogCpt in forum Garage Sale
    Replies: 4
    Last Post: 02-11-2015, 10:10 AM
  4. For Sale: Complete Phoenix Hexapod
    By Tyberius in forum Garage Sale
    Replies: 4
    Last Post: 01-06-2010, 10:28 PM

Posting Permissions

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