c2 smooth velocity profile

I'm not sure how I can help. If you are comfortable running self-built master you could just checkout my branch from github and play around with it. Please keep in mind that this branch does NOT change the trajectory, it is only manipulating accelerations.

Trying different trajectory models is a whole new can of worms and IMHO best POCed and developed outside of linuxcnc. I personally like scipy for this kind of stuff.

I can't really help with git or running development branches, but these things are covered elsewhere.

Hi rmu,

I appreciate your work. I think linuxcnc will benefit greatly with this new feature.

Do you have any news?

Greetings.

hello any news of this job?

github.com/synthetos/TinyG/wiki/Jerk-Con...led-Motion-Explained

I'm curios about this project they have jerk controlled motion.
can we learn about it and implement in Linuxcnc?

The tinyG 6th order bezier curves inspired me to the code in the branch github.com/rmu75/linuxcnc/tree/rs_6otp .

I cloned your repo but I fear it will be while before I can try it..

Just a warning, it does exceed acceleration limits while it should not, and the linuxcnc version it is based on is outdated. Do not run on something that moves.

I didn't pay attention and upgraded all my machines to ubuntu 20.04, now I'm a bit in a pickle regarding linuxcnc development... probably will need to setup a VM with debian buster.

If somebody wants to look at it, I think the first thing to do would be to rebase it onto recent master and look for acceleration excursions in chips.ngc. On the "theoretical" side, jogging (should be easy) and feed override (probably some tedious math involved) need to be figured out.

rmu wrote: I didn't pay attention and upgraded all my machines to ubuntu 20.04, now I'm a bit in a pickle regarding linuxcnc development... probably will need to setup a VM with debian buster.

forum.linuxcnc.org/9-installing-linuxcnc...-nearly-easy-install
This should work for Ubuntu 20.04 also, might want to give it a try.

Need Sam on this - he loves testing this stuff!

rmu wrote: Just a warning, it does exceed acceleration limits while it should not, and the linuxcnc version it is based on is outdated. Do not run on something that moves..

I had thought it might be possible to do something similar to the way external offsets is managed. That is allocate a portion of max acceleration (say 20%) to the s curve which reduces the machine's acceleration to a lower limit (80%). Maybe that override can be calculated?