scurve trajectory planner

Hi,

i think you need to update local git clone first. It needs latest updates.

Then build_cmake.

S curves. I neeeds it.

So how stable is this at this point for "real work"? I have not been paying attention the last 2-3 months much.

The best way to know is testing for yourself, it all depends on what you mean by "real work", if you find issues for your use case politely post them here.
I did my own testing, and it is getting closer to my "real work", it is very exciting, but there is still a lot of work, and I'm extremely impressed and grateful to Grotius and co to be tackling this, much appreciated...

 Hallo,
I tested the new version today and noticed that when I round the corners P64 P10.0 the speed of the axis (joint) does not behave as expected (yellow circle).
The acceleration (green circles) is interrupted by (blue circles) whereby the change in acceleration (jerk) appears to be too large.
At P64 P0.0 the axis (joint) behaves as expected.
Can this behavior be set via parameters?

 

Gruß
Rüdiger

Hi, is there any more testing and progress on the scurve planner to report?
-automata

@Rüdiger
May this be a result of the missing (?, the screenshot is hard to read) Q parameter? From the   readme :

Important Runtime NotesWhen using the S-curve trajectory planner, always use the 

Code:

G64 P[x] Q[x]

 command combination. If 

Code:

Q[x]

 is not specified, the interpreter may:

1. Remove or filter out small segments.
2. Convert arc segments into line segments.

A standard input should look like this:

Code:

G64 P0.01 Q0.0

As far as I know, the Scurve Trajectory Planner plans the tool's trajectory using jerk, acceleration, and speed, which works very well, but doesn't take the constraints of the individual axes into account.
For example, the trajectory speed is constant in a 90° corner, while the axis speeds change without considering constraints.
This can be seen in my previous post.
I hope I explained the connections correctly.

gruss
Rüdiger

but doesn't take the constraints of the individual axes into account.
This is correct. In the source code the trajectory path is used to calculate xyz coordinates.
There it doesn't apply any calculation to reduce any speed for x or y or z. So if Z maxvel is very low, it does not reduce
speed for z.

Hi,
Any update on the project?

No, not any news.
Maybe coming winter time.