This forum uses cookies
This forum makes use of cookies to store your login information if you are registered, and your last visit if you are not. Cookies are small text documents stored on your computer; the cookies set by this forum can only be used on this website and pose no security risk. Cookies on this forum also track the specific topics you have read and when you last read them. Please confirm whether you accept or reject these cookies being set.

A cookie will be stored in your browser regardless of choice to prevent you being asked this question again. You will be able to change your cookie settings at any time using the link in the footer.

Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
New tinypilot image
#1
Hello everyone newby here although I've been using pypilot for a few years now. First on the main steering wheel with a 3d printed sprocket and a window wiper motor which was nice in calm waters, but recently added a tiller pilot to the hydrovane controlled by the pypilot which works really well!
 I've recently added a wind sensor to my system and now the pypilot crashes every once in a while (tinypilot bought from Sean in 2021) I've tried all different kind of things but it's definitely when it gets the wind information through opencpn. Doesn't matter if it's connected to the raspberry pi or open cpn on a windows machine it gives the same results. So I've decided to give the Tiny pilot a reflash with the latest image. Now it says in the instructions to save your current pypilot.conf file and after a bit of searching I've found it. But the manual doesn't say how to change it or at least get the permissions to do so. I've tried without changing the config file, but then I need to do all the calibrations again including the accelerometer calibration. And another thing the remote also wasn't recognized and I'm guessing these settings are in the hat.conf file? So probably need to change that too I guess?
So how do I change this file? Probably in the terminal, but I'm a bit stuck....
Reply
#2
(2024-02-23, 09:03 PM)viggy Wrote:  but then I need to do all the calibrations again including the accelerometer calibration. 

I`ve seen the calibration fie in there but can't remember where & don't have tinypilot up at the moment - anyone know where it is?
Reply
#3
(2024-02-24, 09:13 PM)PaddyB Wrote:
(2024-02-23, 09:03 PM)viggy Wrote:  but then I need to do all the calibrations again including the accelerometer calibration. 

I`ve seen the calibration fie in there but can't remember where & don't have tinypilot up at the moment - anyone know where it is?

Hi Paddy. I know where the file is and I can copy it to another place. But then on the new image I can't replace it, delete it, or modify it as I don't have the permissions to do so, but I can't find how to change the folder or file permissions. Well I've found some instructions how to change permissions through the terminal, but then in the terminal I don't know how to find the files on a seperate SD card.
Reply
#4
Hi,

With a Windows PC, to be able to access and modify the pypilot.conf file, I use the Linux Files System ror Windows application by Parangon Software which allows you to mount the Linux disk in WIndows and modify the pypilot.conf file

When you go from version 0.2 to 0.4, there are new parameters in the file. It's complicated to retrieve old important data to update the new pypilot.conf file.
I prefer to remove the computer case to install the last image on the SD card then recalibrate the accelerometers. To do this, I supply the calculator with 5V and I connect it via Wifi to a tablet. After unlocking the accelerometer calibration, I tilt the computer case on all sides on a table. All that remains is to lock the accelerometers before continuing with the configuration.
Reply
#5
(2024-02-23, 09:03 PM)viggy Wrote: Hello everyone newby here although I've been using pypilot for a few years now. First on the main steering wheel with a 3d printed sprocket and a window wiper motor which was nice in calm waters, but recently added a tiller pilot to the hydrovane controlled by the pypilot which works really well!
 I've recently added a wind sensor to my system and now the pypilot crashes every once in a while (tinypilot bought from Sean in 2021) I've tried all different kind of things but it's definitely when it gets the wind information through opencpn. Doesn't matter if it's connected to the raspberry pi or open cpn on a windows machine it gives the same results. So I've decided to give the Tiny pilot a reflash with the latest image. Now it says in the instructions to save your current pypilot.conf file and after a bit of searching I've found it. But the manual doesn't say how to change it or at least get the permissions to do so. I've tried without changing the config file, but then I need to do all the calibrations again including the accelerometer calibration. And another thing the remote also wasn't recognized and I'm guessing these settings are in the hat.conf file? So probably need to change that too I guess?
So how do I change this file? Probably in the terminal, but I'm a bit stuck....

Thanks for the feedback.    I was unaware this older image was somehow unstable with wind input.    I wonder if it is somehow the particular wind sensor?   I have not heard about this issue, so I am wondering if the one you have somehow generates wind data this is different which somehow causes a crash.  When it does crash, does it simply reset (go back to standby) or?   In any case, the newer image has better debugging information via logs in the web interface, so if you did come across a crash you should report it with the logs.    There are a lot of changes since 2021, and it is likely this problem does not exist anymore, but if it does it is a high priority to fix.    The wind modes have the least amount of testing.

As for restoring the settings,  honestly it is probably just as well to simply recalibrate the sensors.    Some of the older sensors (mpu9255) did require accelerometer calibration, but the icm90248 do not.   They are precalibrated, so calibrating them has a very slight effect.    You may have the older sensors, in which case, it should be straightforward to recalibrate by holding the sensors in each orientation.    I really need to make a video about this process and I will try to do it in the next week or two.

As for restoring the file,  it is possible mostly via ssh, or by inserting the sdcard into another computer.    It might be easier to simply take note of the gain settings, and just manually set them in the new image.   If you do connect via ssh, the username is "tc" password "pypilot"   the files are in /home/tc/.pypilot    You can overwrite them with sudo if the permissions did not allow, but the permissions should not be set to prevent this.     The pypilot.conf file contains calibration and gains settings while hat.conf contains the remote control codes.    I think it is probably easier to open the web interface and under "configuration" look for reprogramming remote controls,  or connect to http://192.168.14.1:33333    in a browser.    I will make a video detailing this process as well.
Reply
#6
(2024-02-24, 11:48 PM)Jean-Marc Douroux Wrote: Hi,

With a Windows PC, to be able to access and modify the pypilot.conf file, I use the Linux Files System ror Windows application by Parangon Software which allows you to mount the Linux disk in WIndows and modify the pypilot.conf file

When you go from version 0.2 to 0.4, there are new parameters in the file. It's complicated to retrieve old important data to update the new pypilot.conf file.
I prefer to remove the computer case to install the last image on the SD card then recalibrate the accelerometers. To do this, I supply the calculator with 5V and I connect it via Wifi to a tablet. After unlocking the accelerometer calibration, I tilt the computer case on all sides on a table. All that remains is to lock the accelerometers before continuing with the configuration.

Thanks for your reply. I did notice the different parameters in the config file and already wondered how this would work with changing the new file to the old file. I guess I'll have to do the recalibration then. Not the easiest task for me as it's built into the boat and have to take the ceiling apart to disconnect it and take it off. Then take it to shore as I'm in a rocking anchorage. But If it's the better option I'd rather go for that and do it properly.
Another question. I've downloaded two images. The latest stable one, and the beta. I haven't looked at the beta one yet, but what are the differences between the two? Would it be worth to try the beta one? I could also prepare two SD cards and have them both calibrated and ready to go so if I want I could swap between the two.

(2024-02-25, 01:11 AM)seandepagnier Wrote:
(2024-02-23, 09:03 PM)viggy Wrote: Hello everyone newby here although I've been using pypilot for a few years now. First on the main steering wheel with a 3d printed sprocket and a window wiper motor which was nice in calm waters, but recently added a tiller pilot to the hydrovane controlled by the pypilot which works really well!
 I've recently added a wind sensor to my system and now the pypilot crashes every once in a while (tinypilot bought from Sean in 2021) I've tried all different kind of things but it's definitely when it gets the wind information through opencpn. Doesn't matter if it's connected to the raspberry pi or open cpn on a windows machine it gives the same results. So I've decided to give the Tiny pilot a reflash with the latest image. Now it says in the instructions to save your current pypilot.conf file and after a bit of searching I've found it. But the manual doesn't say how to change it or at least get the permissions to do so. I've tried without changing the config file, but then I need to do all the calibrations again including the accelerometer calibration. And another thing the remote also wasn't recognized and I'm guessing these settings are in the hat.conf file? So probably need to change that too I guess?
So how do I change this file? Probably in the terminal, but I'm a bit stuck....

Thanks for the feedback.    I was unaware this older image was somehow unstable with wind input.    I wonder if it is somehow the particular wind sensor?   I have not heard about this issue, so I am wondering if the one you have somehow generates wind data this is different which somehow causes a crash.  When it does crash, does it simply reset (go back to standby) or?   In any case, the newer image has better debugging information via logs in the web interface, so if you did come across a crash you should report it with the logs.    There are a lot of changes since 2021, and it is likely this problem does not exist anymore, but if it does it is a high priority to fix.    The wind modes have the least amount of testing.

As for restoring the settings,  honestly it is probably just as well to simply recalibrate the sensors.    Some of the older sensors (mpu9255) did require accelerometer calibration, but the icm90248 do not.   They are precalibrated, so calibrating them has a very slight effect.    You may have the older sensors, in which case, it should be straightforward to recalibrate by holding the sensors in each orientation.    I really need to make a video about this process and I will try to do it in the next week or two.

As for restoring the file,  it is possible mostly via ssh, or by inserting the sdcard into another computer.    It might be easier to simply take note of the gain settings, and just manually set them in the new image.   If you do connect via ssh, the username is "tc" password "pypilot"   the files are in /home/tc/.pypilot    You can overwrite them with sudo if the permissions did not allow, but the permissions should not be set to prevent this.     The pypilot.conf file contains calibration and gains settings while hat.conf contains the remote control codes.    I think it is probably easier to open the web interface and under "configuration" look for reprogramming remote controls,  or connect to http://192.168.14.1:33333    in a browser.    I will make a video detailing this process as well.
Hi Sean thanks for your reply.

I'll see if I can get the nmea stream from the wind meter for you. It's the Calypso ulp pro from Calypsoinstruments.com. I had it connected to both the rpb and the windows laptop with a nmea to USB converter. It should be possible to change the nmea settings of the windmeter, but I can't get the provided program to talk to the windmeter. 

When it crashes it just looks like the tinypilot is turned off no screen anymore. Once or twice it would come back into standby mode. It didn't matter in what mode it was steering or if it was steering at all. It was during my Atlantic crossing and I also added a rudder input just before this trip so also tried with and without the rudder input. But as soon as I took the wind sensor off it didn't crash anymore and had it steering for a few days straight again and without any flaws. I had it steering in windmode for a while too beforehand and when it worked it worked really well and in the plots plugin I could see it was doing a better job than the hydrovane itself . I'm really exited about it! Really exited about the pypilot in general as with this tillerpilot to hydrovane setup the power usage is so little it's really neglectable. 

Your last bit of explanation is a bit over my head as I'm not much of a computer wizzard. No idea how to connect through ssh. I'll see if I can manage to program the remote again in the new image. Regarding accelerometer calibration I guess I have one of the older sensors as when I looked at the calibration in the new image the dots where all in the same spot on the sphere.
Reply
#7
I've made a screenshot of the nmea debug window in open cpn with only the wind input to show the nmea sentence. Hope that is of any help.
Edit: don't seem to be able to upload a picture. Well it uploads, but then disappears somehow. I'll try something else
Edit: picture uploaded here: https://ibb.co/Xp3ghHn
Reply
#8
So you had the wind sensors going into a windows laptop that then relayed the wind data via wifi to the tinypilot computer? Or?

I am trying to understand the exact setup. If you had the sensors plugged in by usb to the tinypilot it is possible to cause a hardware reset.

You said the screen went blank on the tinypilot computer. Did it stay blank forever? It is very peculiar. Even if the autopilot itself crashed, it should respawn, and even if it could not, the process that writes to the screen should say "connecting to server..." or something rather than completely blank. A blank screen might indicate rebooting or a power cycle, but then it should boot up again. Could you confirm the exact behavior?

As for 2 sd cards, yes it is possible to have one with each image. The beta has all of the changes and improvements since the stable image but it is still in testing.
Reply
#9
(2024-02-26, 12:13 AM)seandepagnier Wrote: So you had the wind sensors going into a windows laptop that then relayed the wind data via wifi to the tinypilot computer?  Or?

I am trying to understand the exact setup.  If you had the sensors plugged in by usb to the tinypilot it is possible to cause a hardware reset.

You said the screen went blank on the tinypilot computer.  Did it stay blank forever?  It is very peculiar.  Even if the autopilot itself crashed, it should respawn, and even if it could not,  the process that writes to the screen should say "connecting to server..." or something rather than completely blank.    A blank screen might indicate rebooting or a power cycle, but then it should boot up again.  Could you confirm the exact behavior?

As for 2 sd cards, yes it is possible to have one with each image.  The beta has all of the changes and improvements since the stable image but it is still in testing.

I have open plotter running on a raspberry pi 400 and open cpn on a windows laptop. I have a vesper Marine ais transponder that provides them with gps and ais signals either through an nmea to USB converter or wifi. I added the windmeter to the system that also feeds the nmea signal into open cpn through an nmea to USB Converter. The Tiny pilot connects to open cpn through wifi. I had the exact same results connecting to the laptop or the raspberry pi. Can't recall if the tinypilot ever crashed before I added the windmeter. In the end I had both the raspberry pi and the laptop running at the same time, but the windmeter only connected to the raspberry pi and the laptop just to provide the gps signal to the tinypilot. I have a steel boat and it's difficult to get a proper compass calibration so I prefer to have it run in gps mode. I had no crashes in this setup anymore in almost a week continuously sailing . I used the autopilot for a few days straight, but because of wind conditions about half the time with the normal windvane.
When the tiny pilot crashed it was as if it was turned off. Blank screen and no response. Just once or twice it would reboot by itself and come back in standby mode. Otherwise I had to do a power cycle to get it running again. At first I suspected the power feed to have a problem so did some changes to that as my electrical switchboard could do with some upgrading and gives some problems every once in a while. Since that didn't make any difference I made a temporary power feed straight from the battery to eliminate this all together. But only after the tiny pilot didn't receive the wind information anymore through open cpn it ran stable again.

I have now completed all the calibration sequences on the latest stable release except compass calibration and will do a test sail soon. I will report back if I run into any problems.
Reply
#10
Thanks for the lengthy clarification!

So I understand that you finally wired the tinypilot directly to the battery. You said it still crashed in this configuration with a blank screen? It is only stable regardless of power without wind input correct? This is very unusual, since tinypilot runs linux, a crash rebooting the system would indicate more than just the software crashing. Perhaps the log output became so long it ran out of memory, I am just trying to imagine what could cause this. How long until it occured?

Please try to get the new image to crash! Feed it whatever data you can! If you can repeat it, check the logs by going to the web interface http://192.168.14.1/logs and check the pypilot output log before it actually crashes. I hope the new image helps.
Reply


Forum Jump:


Users browsing this thread: 2 Guest(s)