Advanced Search

Just installed the latest RTAI deb set (5.4.290-rtai-amd64) (packaged with linuxcnc 2.9.4) from the usual place: onelaboratories.us/abc/ and i must say the latency is about half what 5.4.279 was on my pc. Also the GUI doesn't lag on boot like it did with 5.4.279.  Running a couple tests:

1: histogram with 4 glxgears, stress-ng & 2 yt the servo thread was 16.7us and 12.9us base thread.

2: histogram & unigine heaven the highest 23.8us base and 23.4us servo thread as seen in the histogram here.

Win! Win!

Die 5.0.1 ist weiter als die Develop, da ich an der V5 nicht weiter entwickel. Die V6 habe ich noch nicht auf Stand um daraus die neue Develop zu machen.

20 iterations vs 1000
Nice.

It would be better to have some protection of course, the same as with a Raspberry. The pin header can be used via the driver made by Guglielmi:
forum.linuxcnc.org/18-computer/48079-can...cnc?start=200#273554

For Rockpro64 (rk3399?), I built an image before for a Orange Pi 4, but that didn't seem to work well, spikes to 130us, could work with Mesa though. You can try to build one via Armbian, from 6.12 and up you don't really need to patch it, just the right configuration. Then experiment to isolate cpu1 and move the rt tasks to that core as explained in the following post:
forum.linuxcnc.org/18-computer/48079-can...cnc?start=280#286432

 Hi Arciera,

Here is a basic  powel example. It uses a equatation with 2 unknown's.
This example run's ok.
powel basic example

The residual  = distance ( guess position result from first calculation up to final end position. )

I just found out why to use the gaus method....
gaus calculates clothoid length

I found it by accident, when computing the endpoint of the clothoid compound in a optimized function.
it turn's out that the Gauss method is used to get very high accuracy in a few iterations.
My current approach has the property to drift off. As values are incremented including nummerical errors.

So now finally i know why gauss is used.

Gaus vs Drift:
 

I guess we could simply calculate the length of our 3D residual vector and pass that back as the error value to be minimized.
Yes. Excact. That's going to be tested.

Some pictures of your machine would help. tommylight and I have both built hot wire foam machines.
Personally I would use ladder logic for the pause control. I am confused however as
to why your wire breaks often enough that you need a system to pause the program when it happens.
Some questions.-
What type of wire are you using Titanium? NiChrome?
How much tension is the wire kept under?
What gauge is the wire?
How hot are you running the wire?
What voltage is the wire powered with?
What frequency is your PWM signal?

 And in other dumification news we have this . I am sure a Nigerian prince is ready to export all over the world
for a small transaction fee that will be reimbursed once you receive the vehicle.
enjoy
 

Should I just download the bim from pyodc and put it in stlink v2 and after that I will be able to record remora on nven? Could you detail me step by step?

I guess we could simply calculate the length of our 3D residual vector and pass that back as the error value to be minimized.

Very nice this will be perfect for base thread.
The pin header that has the board can drive 4 close-loop steppers ?


(i have a rockpro64 any hope for this for an image with preempt-rt kernel?)

yes, eq 14 :
 

because we set s1=s2=s3=s4 we need to integrate from 0 to 4*s1.

It's still unclear to me how dogleg optimizes a system of 3 equations as it would not get back a single residual value but rather a vector with 3 residual elements.
 

@Arciera,

Thanks for your responce.

Indeed, the error is likely the endpos. I was thinking the same yesterday night.

error = distance "requested endpos" - "guess endpos"

When playing with different Y11, 21 gues values using the same length s1:
Changing the guess values Y11, Y21 will influence the rigid of the clothoid compound.

Picture values from left to right :      
Y11=1 Y21=1
Y11=0.5 Y21=0.5
Y11=0.1 Y21=0.1
 

Now i will try to setup the dogleg method. Using our new criteria.

The Jacobian usage seems to be more effiecient in finding the solution. But without is also no problem and
will result in more iterations to find the fit solution.

Your point at eq44 if solved here as a example : eq14
This eq14 is not a optimized function yet.  And for calculating the clothoid endpoint, i want to make a
few one liners today.

Plan :
1. Create a function to calculate the spline endpoint efficient, using y11, y21, s1.  Eq14 current code is not effiecient for this.
2. Create a Dogleg fit function to optimize y11, y21, s1 values that the spline compound fit's the given endpoint. (Magic function).

For info, to be aware off:
In linuxcnc we use G2,G3 as helix. This abstact isn't taking account of helixes. Only flat arc's for input,
We will create a solution that uses a end vector of any type of curve.

For info:
Deepseek is better in reading pdf files than Chatgpt.
Also the code example from Deepseek are better.
So i often just upload the clothoid_3d.pdf to deekseek and start asking questions.
This really helps the process. Deepseek just throw's out a full dogleg c implementation. It's amazing.

Let's hope the Magic Function will find the solution for endpos.