Kalibrieren

[SefanH]

Im Hauptmenü kann man doch unten die Console aufrufen. Da steht drinnen was der Mäher gerade macht.

 

[Richtel]

OK, now I've gone down to the garden 20 times

So first screen video with 2 coils and the second video with a coil, but I stopped because it didn't stop over the wire.

 

[Richtel]

Same driving behavior as seen in the videos above. I also think that if you make an arc to find the wire again, this circle or radius is too small. It no longer reaches the wire, but is already at 360 degrees in front of it. The arc would have to be larger. but doesn't explain why he always wants to get away from the wire instead of following it....

 

[Bernard]

Can you deactivate the sonar in the setting.
message LOOP NOT OK is not good , certainly issue in Sonar or I2C line (Deactivate screen ??)

 

[Richtel]

Yes, screen is deactivate.

 

[SefanH]

Ich stimme Bernard zu, könnte ein Problem mit der ich Verbindung sein. Daher auch die overload Meldung. Bei overload ist kein vernünftiges Tracking möglich.
Bei mir war es damals der imu. Ich habe die Kabel dann direkt verlötet, nicht gesteckt, dann funktionierte es.

 

[Bernard]

Here a version of robot.cpp including other management for detecting wire on 2 coils setting
Simply keep a backup of your robot.cpp and test this one

 

[Richtel]

Oh I would like to test that, my mower had a collision with a stone today that it shouldn't have been over
What shit.
.....so printed a new disc

 

[Richtel]

Hello,
Seems to work quite well but it only reads the left perimeter receiver!?

 

[Bernard]

Hello,
Seems to work quite well but it only reads the left perimeter receiver!?

It's not OK something is wrong.
Try to set again YES and NO in the Read the 2 coil setting.
You need to see the 2 perimeter coil value.
I make a video.

 

[Richtel]

OK, I turned it back on after a few hours and now both work.
Great, now it’s time to continue testing. but great job so far!!!

 

[Bernard]

OK, I turned it back on after a few hours and now both work.
Great, now it’s time to continue testing. but great job so far!!!

Here the video.
First of all the motor setting need correct value according RPM vs PWM.
It's mean that PWM and RPM real time value need to be near the MAX one when mower run on flat ground
View attachment setting motor.mp4
In the video it's near RPM 30 and PWM 80 real time value for RPM 34 and PWM 97 max value

 

[Bernard]

Tracking Video IN THE 2 COIL Version:
View attachment tacking1.mp4
View attachment tracking2.mp4
Mower roll CW (clockwise) when coil left is outside .
If mower reach perimeter with the right coil first and if left coil is inside at the end of the braking phase mower roll CCW (counter clock wise)

Here the console data:
In straight line MAG left coil is near -14000 and right one oscillate between 2000 and -2000 (IN/OUT transition)
View attachment tracking console.mp4

 

[Bernard]

Start from station using button.
When button is pressed:
Mower send wifi data to sender to start (if no wifi ,sender start only when mower leave the station)
Wait 6 secondes (Optional door garage opening duration : need dev on the sender side code)
Reverse
Roll
TEST if the sender signal is OK
Start to mow or error if no signal

 

[Bernard]

Go to Dock.
One RFID tag is locate near station to reduce docking speed,You can see that at second nr 27
View attachment dock1.mp4
in the video

 

[Bernard]

Hello,
Seems to work quite well but it only reads the left perimeter receiver!?

In your video i can see that you have not a straight line before the docking station, so PID tune have trouble to have a perfect straight docking .