Does KDL deal with redundant manipulators?

Submitted by snrkiwi on Tue, 2008-05-06 16:28.

We're getting a seg fault from within KDL::svd_HH.cpp when doing positional inverse kinematics of a 7DOF manipulator. Before I dive down the rabbit hole, I figured that I would ask the basic question about redundant manipulators (and hence non-square jacobians, etc).

Thanks
S

‹ Need help with DH parameters setting up KDL chain non periodic and blocking ›
» login or register to post comments | printer friendly version

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
Ruben Smits's picture

Does KDL deal with redundant manipulators?

Submitted by Ruben Smits on Wed, 2008-05-07 08:15.

On Tuesday May 6 2008 18:28:12 snrkiwi wrote:
> We're getting a seg fault from within KDL::svd_HH.cpp when doing positional
> inverse kinematics of a 7DOF manipulator. Before I dive down the rabbit
> hole, I figured that I would ask the basic question about redundant
> manipulators (and hence non-square jacobians, etc).

Yes,

but for now we only use a non-weighted pseudo inverse to solve the problem at
velocity level, this means that the velocity is distributed equally over all
axis.

I could reproduce the bug, it is not there using the other svd implementation.
I'll fix it.

Ruben

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm


» login or register to post comments

Fixed for me

Submitted by snrkiwi on Wed, 2008-05-07 17:19.

Confirming that r29235 and r29236 solve this problem for me. Seg fault disappeared - thanks!
S

» login or register to post comments