Re: Planar Hexapod Kinematics

[Ray <rehenry-at-up.net>]

Lukas

I can speak only to the home and jog question.

On Mon, 26 Mar 2001, Lukas wrote:
<snip>
> We encountered the following problems:
> Homing:
> 
>      Is homing separated from the kinematics?  When we highlighted
>      the X-axis in tkemc and clicked home, joints[0] of the
>      mechanism moved to the home switch.  I would have thought that
>      homing while on the X-axis display would cause all three axis
>      to move, until an external Home switch is hit.
> 
>      Homing while on the Y-axis display causes joints[1] of the
>      mechanism to run away regardless of the state of its home
>      switch.
> 
> Jogging:
> 
>      Jogging while on the X-axis display, causes only joints[0] to
>      move, but the X- and Y- displays vary.  We were expecting
>      joints[0], joints[1] and joints[2] to move and only the
>      Display to vary.
> 
> MDI:  Since we couldn't home joints[1] and joints[2], an MDI command
> results in following errors.

The correct interface to use with these machines is yemc.  It is the only
one that has implemented the joint and world systems that are required to
start up one of these machines.  

You would need to begin by homing or setting the position of each joint in
joint mode and then switch to world mode for coordinated motions.  I have
only done this on the cable-legged hexapod from NIST but I think that
you would need to jog each axis until you get them very near home and then
use the home routine for that leg. 

For quick testing of coordinated motion, you could figure the
length of each leg wherever it is now and put that value in the ini file as
the home location number for that leg.  Then home all legs and switch to
world mode and try MDI moves.

Hope this helps.

Ray

  

Hope this helps.