LinuxCNC S-Curve Accelerations

In grotius' approach, the jerk is specified as a "dv" value which decides the change in velocity during which the jerk limitation is applied. It does not specify the jerk value directly.

Using "dv" delta velocity = ve-vo.

I had to decide how to specity the curve. In the past i also used a 0-100% curve power value.

Dv represents the curve's velocity steepness,
or determines how much of t2 linear period is
represented in the curve acc period t1,t2,t3. aka dcc period t5,t6,t7
Dv is the delta velocity for a motion from 0 acc to max acc (2*a), back to 0 acc.

The "jm" is then a constant value for all movements of the joint.
This results in a more predictable machine movement.

Where "ct" is curve time.

ct=(ve-vo)/a;
ct=dv/a;

as=2*a;
jm=(2*as)/ct;
I suggest that the topic synchronisation, look ahead, blending belongs to the path planner?

We need a good approach for look ahead and blending etc. And it's almost time to get started
with this.

automata, for me, the whole topic is ill-defined and arbitrarily complex depending on what you include (>3 joints, kinematics, blending, ...).

Maybe limiting jerk in tangential direction is a worthy improvement. I even did some experiments in linuxcnc.

rodw, AFAIK that is exactly what current TP does, it even optimizes blending arcs and feedrate to stay within G64 and machine acceleration limits.

I have no workable ideas regarding path blending nor do I know anything about 5-axis machining.

automata wrote:

For paths that are continuously changing direction like circles or nurbs OR any machine with non-trivial kinematics, the only viable method to limit the jerk on joints would be to sample the path at some small length and check for velocity acceleration and jerk constraints at those points.
-automata

Checking the path is trivial, but that doesn't really help IMO.

Hi Andy,

I have seen this video a few month's ago. It's interesting. Ultimate she want's to design her own spline algo.

Where you able to write some code in c or c++ for the spline example's in the video?
I have a github 3d spline repo containing a cubic spline with vel, acc profile. Maybe this is usable for background info.
This spline works excellent in opencascade, however i have no clue how to use the vel and acc of this spline output?

Today i made a github repository for the current scurve code. github scurve

Currently trying to code a 10 joint planner. 1 joint for interpolating the path, 9 joints for each motor xyzabcuvw.
If this works a bit. I will take a look at mr. Garret's comments for integrating.




 

Hi,

I got the planner running.  It has 2 buttons on the right side of the picture. The left side is unused in this example.
The path has 3 xyz waypoints, using acs (acc at start) ace (dcc at end), vo, ve.



The waypoints are at the blue vertical lines.
Hmm interesting. It's interpolating over xyz now.

Next i will add scurve xyz joints to try to follow the interpolated path. Then we try to trigger a following error by
setting on of the joint's vm a little lower.

Hi,

A little update.

Last post:
Next i will add scurve xyz joints to try to follow the interpolated path.

This is not happened yet. I have been thinking a while about this.
In the past i used a stepgen for this, this worked fine for my steppers.
But the stepgen is based on a linear vel system. I was thinking maybe
make a :

- scurve stepgen
- scurve pid

In the mean time, the planner is able to interpolate 3d lines and 3d arcs.



To hook op lcnc, The connection to gcode input can be made now.
A converter from G2,G3 IJK to arc a waypoint is needed for the scurve planner.
This is needed because the planner uses 3d arcs, and lcnc uses planes.

Next thing on the list is try to code motion reverse, negative and positive adaptive feed.
And mdi input like goto zero etc.

The code is updated, but excedentally i deleted the readme... updated code

Hi,
Any update on the topic? I am eagerly watching this topic and can help with development and testing if needed.
-automata

Hi Automata,

Last day's i reviewed the  engine code. Fixed a few tiny typo's.
An sich i am quite happy with the engine code so far, i have to finish the function : process_curve(..)
This function processes the scurve scenario's.

I made the project structure a little different. It's now divided in dir's with their purpose.

Now working on a planner to look if

• pause
• resume
• vm interupt
• waypoints
• motion reverse

will work. This code is dirty, raw and not finished,
but you can play with it already.

Yesterday i thought how to get rid of the vm sampling, to get a better interupt performance.
I made a line-line intersection of 2 scurves to get a new vm value.
Hmm, this was not going to work. This works only for full curves. Otherwise this would be perfect.

To get rid of the vm sampling, we could try to draw a spline to a few scurve points, to get a new vm value.
But then you have to find the spline y limit. This also is time consuming. If this should take 0.01 ms it's ok.

What you could check is mr. Garrets notes in this thread. Test if his notes setup works.
Check if you can do improvements or anything to the code. Maybe make a test dir in the sc_project.
You could make a hal interface, then scurvelib can be used stand alone. We need a nice readme file for github etc.

You could check this cubic spline lib. It has mutiple files. And is a 3d spline with acc and vel profile.
I don't know how to use the acc and vel results. However, the spline works perfect with opencascade.
Maybe this spline is usable for building scurve's.






 

Hi,

A little update.

Currently finalizing the planner code. A few pause, pause-resume things to solve. Most of it is working now.

I think we get this working soon.