scurve trajectory planner

Hello,
I tested further and apparently the acceleration is not limited. In the ini it says 'MAX_ACCELERATION = 5000' in the halscope at approx. -10000.

Gruß Rüdiger

 

@ Rüdiger,

Max acc scurve = 2 x acc.
at scurve inflection point the
Scurve acc = 2 x acc.
see  scurve lib abstract.
 

Hi all,

Today is activated the abc, uvw axis for the scurve planner.
It should work with look ahead.

git updated

Hi all,

Today was a lucky day. We managed to fix acceleration spikes at the 4-clothoid-spline junctions.
The problem was solved by using a new gauss legendre and minor function updates to the clothoid3d lib.
The iteration for gaussian legendre is now set for a 16 for loop.
We can expand to 32 or 64 loops to even increase accuracy.

Now i changed the back color off halscope to plot a white background as this is nicer for websites and documents
to show pictures.

Below are some halscope results, where motion is passing trough 3d-clothoids. The acceleration spikes on
joint 0 & 1 are gone. This tells us the interpolation off the clothoid is now absolute. And without any numerical errors.
So now the clothoid interpolates just as a line or arc.

This was really a bad situation. Wich is now solved.

 
 
 

Git is updated.

/me feeds grotius and co more cookies and pizza

nice work

@IHaveNoFish,

Thanks!

Today i started investegating something new.  experimental code

The question is, can we use a clothoid curve for motion, just like the scurve is a curve used for motion?
The idea came up a while ago. Scurve motion uses derivates, the clothoid also uses derivates.

Somehow i used the clothoid curve for the velocity profile.
The theta start of the clothoid curve is a vector, this can be transformed to acceleration start and visa versa.

1. The x axis of the clothoid is then time.
2.  The y axis of the clothoid is then velocity.

During the curve's velocity profile, we printed the acceleration values along the curve. What i didn't know is,
the acceleration profile is not a straight line as like the trapezium or scurve motion profile.

Here is a plot of acceleration stage, mention the acceleration line. It's curved. This involves less Jerk?