Tractor drives next to the line and does nothing about it

Hello, I am very glad to join the community of AgOpenGPS.

My setup is:

AgOpen v4,
Ardusimple rtk2b,
and I don’t use any IMU.

I have mounted a coulter on the back of my tractor and set the tool’s width at 2m, so after U-Turns the marks after the coulter should be exactly every 2 meters. And the problem is that they ain’t.
On the display i could see that AgOpen knew about that he was driving next to the line but did nothing to fix that.
So, after driving for some time and doing some U-Turns, I measured the distances between the marks and it was regular pattern: 1,8m; 2,2m; 1,8m; 2,2m and so on.
I would be very grateful for every suggestion how can I fix my problem.

Sounds like a calibration issue to me. Could you first clarify if you have observed this behaviour by measuring the coulter marks or can you also see AOG indicate a constant offset from the wayline?

Obvious issues that cause this kind of behaviour are the GNSS receiver position (not centred or incorrect left/right measures stored at AOG settings) or the inclinometer calibration. Obviously also if the implement is not centred or the implement parameters are incorrectly set.

I have not faced the wheel angle sensor calibration issue (have always had it tuned well enough) but I’ve read from the forum that AOG can drive at a constant offset from the wayline if the WAS calibration is off.

Do you have valves or an electric motor?
You need to adjust the PWM and you on the steering settings tab.
Try to increase “minimum PWM”

NorthernFarmer: This behaviour can be also observed in AOG. If the the problem was about inclinometer, would the “sidehill draft gain” be relevant? GNSS receiver is placed in the center of the hood (I tried also placing it on the roof, but it was only worse). WAS is based on a linear potentiometer mounted to the hydraulic cylinder and the calibration has been done.

Baraki: I have got an electric motor and I have tested a lot of Minimum PWM’s settings from around 40 to 90 and it didn’t help.

If placing the antenna higher makes the issue worse, it sounds like incorrect inclinometer calibration. But I don’t think AOG should show the tractor driving 10 cm off the wayline in that case.

Do you have sidehill gain set higher than 0? I have not tried sidehill gain and don’t know if AOG shows the tractor off the wayline or not when compensating for sidehill drift.

Perhaps try to eliminate different error sources. Drive level ground back and forth, inclinometer disabled, sidehill drift disabled etc.

Setting sidehill draft gain for around 15-20 makes the tractor drive very far (50 cm) from the line, so it’s making it even worse, maybe due to incorrect inclinometer calibration.
I will try to eliminate diffrent error sources, as you have written.

While driving try what happen if you change steer point zero. Normally AOG hits then the line. E.G. If you are always left side from the line try to do the steer point zero to the right. And important, the wheel angle sensor construction isn’t allowed to “wobble”.
If AOG then hits the line (driving in both directions) and you have still the offset, your coulter is not centered. This can be fixed in tool settings.
You started without bno and mma on mostly flat land, I hope.

Can you post a picture or two of your AOG screen while driving an AB line? Also maybe a few of your current settings i.e Steer settings and Tractor Measurements?

Yes, I will do it tommorow

Sidehill draft should help, try cranking it up as it really makes a difference.

I thought this case would not be about sidehill draft though. Seems the implement is continuously some 10 cm to one side, same side on the return path.

Sidehill should draft the implement towards alternating sides and where the slope remains constant, the implement would work at a fixed path-to-path distance (2 m in this example).

Thanks you all for suggestions. I’m already in the tractor and I will be trying to make everything work properly. Here are my present settings:
And tractor measruments are: wheelbase 256, antenna from rear axle also 256, and height 175. Turn radius is 550 and in the “Guidance” section I haven’t changed anything.

Could you show us how the guidance tab looks like?

The problem was with WAS. Zero wasn’t well calibrated and i had to do a change in line 527:

From 1533 to 1650 in order to set WAS 0 in a correct place. Now everything works very good, the accuracy is around 5 cm and I’m going to use AgOpenGps to seed dry beans.
Thanks you very much for your help :wink:


Is it a factory mounted WAS or your own?

I have got my own WAS, using a linear potentiometer.


Soooo, can you not just move the position of the WAS so it reads 2.5v when your wheels are centered?

Could an offset happen in V4 that wasn’t there in v3.9? I am using the exact same settings in 4 as 3.9 and I now have an offset to the right of probably 100mm to 150mm. Visible on screen and on the ground.

I would say not since many have been using it successfully. You could try the 4.2.02 that is on github right now. Much better steer control.

Its better because you can crank up the gain, but still have a very control system

1 Like

Very good video. Makes it easy to understand! Not sure why I have the offset but I’ll experiment today. Daren’t try 4.2 today. Yesterday it had communication issues with it. No RTK and other reakky odd stuff going on. My settings obviously but no timey to sort it out till tonight possibly.