I have the same setup, but do not think it ran, has gone back to 4.3.10, it looked like a drunk man’s work, could see that for example speed was known with great delay, used 38400 baud for F9P, 115200 for pcb2, Bno085 on pcb2, no relay card. Wondering what I’m doing wrong?
It may only be a matter of adjustment. I was also a drunk at first, I had copy and paste the settings from v4.3.1. Then I started again from zero, and redone everything from the count by degree to the integral . Well it was still not great until I pushed integral to 50.
Definitely don’t use the 4.3 settings for steering. The counts per degree is about 5 times as high and everything is quite different.
Tried testing V5 in the field today, FUZ number was jumping back and forth between 2 values causing the screen to jump around like crazy. GPS and IMU numbers were solid. RTK with Ardusimple F9P connected with USB. works fine using 4.3. Any ideas what would cause this?
are you using the CMPS or bno08x? on board or mounted with extended i2C? Seems really strange that the two numbers that make up FUZ number are steady, but not the result. Config settings all ok?
I have never seen that happening before.
@BrianTee_Admin
Can we still read multiple ADS channels with the v5 ino?
The new code:
adc.setMux(ADS1115_REG_CONFIG_MUX_SINGLE_0);
steeringPosition = adc.getConversion();
adc.triggerConversion();//ADS1115 Single Mode
It seam that we can only read one cannel since we select one, read from the previous loop and trigger for the next?
Or am I missing something?
CMPS, mounted on the board. I will try without CMPS connected tomorrow and see if it makes a difference.
Loaded latest version this morning 4/3/21. Disked 65 acres using JD 4630 2 wheel drive. Running Stanley with Imperial settings. As downloaded this version put the antenna at “3” in front of rear axle. In real life, RTK antenna is 38 inches in front of rear axle so I set it at 38. Tractor would wander left and right overshooting straight A-B line. Changed back to “3” and works much better. Setting says feet/inches. Entering “38” just does bad things. I assume I’m to enter inches because the conversion from metric to imperial seems to convert correctly. Any possibility the Imperial math needs a little work? Also, when testing the U-turn (example is a turn to the right), the tractor would start to swing out to the left correctly but as soon as it got back next to the original A-B line, it seemed to give up on the curve. I need to read the examples of that and test some more.
4/3/21. After 50 acres disking, I started having problems with Autosteer dot dropping green and going to pinky but just when I slowed down at the end of the field. Stopping and restarting AGIO would make it work again. Alternator was failing and must have been low battery affecting steer module. Will retest after alternator repair.
Imperial is always a pain since the whole application is based in meters. i will double check the distance is for sure in inches when in imperial.
Edit: double checked, going from 38 in imperial and back to metric shows 96 cm so all is good. You can also see the cyan colored dot on the tractor where the antenna is. Sounds like your counts per degree and or wheelbase are either too high or too low perhaps, or lookahead too close? Also switch between Stanley and Pure Pursuit to see if any better/worse. Pure is much better on curves.
What kind of a computer are you using?
New version on github where the steer dot shows pink only if steer module gone for at least 3 seconds
I think loopback address is nothing to do with physical network cards, one have localhost (127.0.0.1) even without card.
Atleast “normally”. 
Just to point out, physical network card should have nothing to do with performance with localhost internal connections / speed / reliability.
I’m using a Panasonic FZ-G1. Wheelbase was set at actual 115. Track at 60. For the turn radius, I upped it to 300. I didn’t change the counts per degree. I need to check that. I didn’t try Pure Pursuit much. It wasn’t working well. I’ll try to get that set up better.
My communication problem is probably caused by GPS (simpleRTK2B-F9P) settings.
I did some tests on moving vehicle.
I tried latest version on yet another computer (Surface pro 7) and simpleRTK2B-F9P V3 GPS and it is slow and unstable. On Trimble AgGPS 162 it is smooth and stable.
On my oldest and slowest laptop (just 10 years old) simpleRTK2B-F9P works best and stable, it is slow, but so it is on Trimble GPS. Slower that 4.3.10, witch is kinda understandable on such an old hardware.
Edit: I updated to latest firmware for simpleRTK2B-F9P and used [10Hz simpleRTK2B (Rover) of a simpleRTK2B+heading kit for Ardupilot] -configuration file. No difference, update rate seems like < 1 Hz.
1 Like
think I found the error when I programmed Nano I chose port for F9P by mistake, by this F9P probably got some bad commands, so when I put new config on F9P today just run it. Great program many thanks to Brian and all of you others who work hard for AOG.
1 Like
One fellow on dev team uses the Surface pro, he has been using it a hundred hours, zero problems.
The FZ-G1 seems to be a consistent problem from others as well. If you have any ideas as to why, that would be very helpful.
The other thing you could try, is set to metric. Reset everything and start again. The struggles of a beta and so many different setups.
Oh don’t say that… I just ordered an fz-g1 to have on a second setup🤦♂️ My original is a Dell 7202 and has been going strong for 3 years but I thought I would save a little money on this one.
No one has any idea why at this point.
It “unfortunately” works perfectly on every computer i own.
Hi Brian, First of all thanks for all your efforts, I use Sparkfun bno080, it always shows 0.5 degrees when I first start it up, is this normal? And in simulator mode, there are wavy roads with the bno attached, but when I remove the bno, the lines are straight, is that normal?
Best regards.
