AutoTrackRaymarine_pi: Remote control and route following for Evolution Pilots

[Tehani]

Originally Posted by rgleason

Tehani, thank you for the good link and the downloads. I added the link to our User Manual, under Advanced. I have downloaded and tried the first exe (after restoring from Norton quarantine). When I completed it the screen went blank, so I will have to try it again.

I am using Code :: Blocks as IDE to work with these XTE_test files. Use Gcc as a compiler. It's weird that it doesn't work for you, but in any case, you can compile XTE_test1 directly, because it doesn't need any special includes.

What do you mean by "Sorry, it is not good choice to run too much ..."?

I was very quick to say that Ocpn used the ellipsoid for calculations. As I saw later, the data that completes the RMB sentence is made with 2D geometry.

Also, do I understand correctly, that you think these calculations are better and faster than the ones used in OpenCPN?

I am not saying that calculation with spherical geometry is best in all cases. I just think it's good enough not to create appreciable deviations from the Mercator projection at short distances, and yet I think it's much better for long-distance layouts. These 3D calculations provide an extra advantage which is the XTE sign (- means "To Right" and + "To Left"). Ocpn gets the sign with additional angle calculations.

[rgleason]

Maybe bdbcat will take a look:

...it's good enough not to create appreciable deviations from the Mercator projection at short distances, and yet I think it's much better for long-distance layouts. These 3D calculations provide an extra advantage which is the XTE sign (- means "To Right" and + "To Left"). Ocpn gets the sign with additional angle calculations.

I've posted to the Autopilot testing thread about ShipdriverTracker and AutoTrackRaymarine. Perhaps Douwe will have some thoughts.

[rgleason]

Tehani wrote below:

I am using the procedures of this web:
https://www.movable-type.co.uk/scripts/latlong.html

They are sphere based, not ellipsoid, and are much less heavy than those used in geodesic.cpp and georef.cpp. When used to simultaneously calculate Bearing, Distance, XTE, and VMG, repeated trigonometric functions can be grouped together.
My opinion is that it would be interesting to use spherical geometry for all cases. I have written a single "C" module to compare all three methods, here is the link:
https://drive.google.com/drive/folde...0n?usp=sharing

Note there is another C++ version on github too, with an API

[Tehani]

Douwe, there is another matter that makes me think about the convenience of using many digits for the XTE.

We have seen that moored ships are issuing AIS data, and they often report a SOG of 0.2-0.3 Kn, and sometimes up to 0.5 Kn.
This speed calculation is made by the distance between 2 consecutive coordinates.

If the GNSS sampling time is 1sec, this would mean that the distance, and therefore the position error, would be: 0.3 / 3600 (sec / hour) = 0.00008 Nm. That is, the fifth digit is completely wrong, and it is masked when the ship is sailing.
This position data is the starting point for the XTE calculation.

I think it would be nice to take the HDOP to assess the level of precision of the XTE. Do you see it like that?

[bcn]

Originally Posted by Tehani

I think it would be nice to take the HDOP to assess the level of precision of the XTE. Do you see it like that?

HDOP is an evaluation of the geometry of the GNSS constellation and does not determine the actual view. About expected quality, a statistical parameter - the errors you would expect during an observation period of some hours for example.



Real Time Differential corrections is what is required to nail down precision at a given moment for navigation. Case of few meters precision. Or centimeters if you like so..

[rgleason]

Here is an interesting app for STxMobile for Raymarine ST2000+ Tiller Pilot that works with OpenCPN

https://github.com/xedgecoder/stxmob...k5cZmmzZqia7BI

[akopac]

Just curious how this is coming along[emoji41]

[Tehani]

Originally Posted by akopac

Just curious how this is coming along[emoji41]

Well, this time I have spent designing a low cost remote control for Seatalk, Evolution and Simrad pilots.
It consists of taking the signals from this Chinese 4-channel garage remote control:
https://es.aliexpress.com/item/40011...274263c0Rp4B4Q

and using a microcontroller, translate those signals into Seatalk and NMEA2000 messages for those pilots.
At the moment, I am waiting for the PCB's from China, I have already adapted the software, and I am evaluating the possibility of publishing the basic version (hardware and software), without the gateway section.
The full version also works with a keyboard embedded in a web form over WiFi.

I can't spend the time to compile an OpenCPN branch with spherical geometry computations because I don't have the tools installed to do it, nor am I practiced using Github. If someone wants to collaborate and can compile, I can modify the source files for the RMB, APB and XTE sentences and pass them along.

In any case, I have already decided to incorporate those calculations into the AP that I am also developing. That way it can be autonomous if the navigator fails or if the user wants great circle navigation. (It will only be necessary to enter destination waypoint coordinates, or receive them from RMB in NMEA0183, or from PGN 129284 in NMEA2000 following a route).

Rick, in the link you have put, I can't find the app to install on the mobile, but looking at the source files, it would be easy to receive the messages from that app and translate them to any Seatalk or N2k pilot.

For my part, here's the thing...

[rgleason]

José Luis
I have a clone of OpenCPN here which has been fetched and is current with Master.
https://github.com/rgleason/OpenCPN/tree/master

How would you like to work on this?

Tehani wrote:

I can't spend the time to compile an OpenCPN branch with spherical geometry computations because I don't have the tools installed to do it, nor am I practiced using Github. If someone wants to collaborate and can compile, I can modify the source files for the RMB, APB and XTE sentences and pass them along.

-------------

I don't quite see how to get the app either, perhaps Dave would have a suggestion, but first I can ask the author about it. What would be the best way to translate them to Seatalk or N2k pilot?

Tehani wrote:

Rick, in the link you have put, I can't find the app to install on the mobile, but looking at the source files, it would be easy to receive the messages from that app and translate them to any Seatalk or N2k pilot.

[rgleason]

Tehani, I wrote Marco re stxmobile.

[rgleason]

Marco,
Very nice! I like the interface configuration and the confirmatory press required for changing modes. I trust that you don't mind if I post this?

Thank you for your excellent diagrams, instructions and youtube's and information about compiling.
The link to autopilot_rm is from a github repository that I found, and added to the Beta Plugins List and compiled for Windows. I have not heard from BenrdC for sometime and I have not updated the compilation for some time.

Sincerely, R

--------

Hi Rick, thank you for contacting me.

The application is an Android application written in Java and it has been developed using the NMEA 0183 Wi-Fi Router YDWR-02 for routing NMEA data from openCPN to both the Raymarine ST2000+ and to the application, as well as to route SeaTalk data between the application and the Raymarine ST2000+.

I didn't publish on Google Play. One can just clone the repository and use Android Studio to build the app and deploy it on an Android phone.

Note that the interface is very much done for my Samsung Galaxy S21 plus.
You might need to modify the app to fit your mobile screen.
Here are few videos I made:

https://youtu.be/fPmrYc350-0
https://youtu.be/HTNjK66iiMQ
https://youtu.be/nx-qNSi9m54

I found that somebody else has done a similar thing as OpenCPN plugin.
Take a look at the following link and let me know what you think
https://opencpn.org/wiki/dokuwiki/do...utopilot_rm_pi

/Marco

------

Dear Marco,
I help with OpenCPN Plugins, Tehani thinks your Android app stxmobile on github can be augmented for AP's using N2K, and would be an excellent tool for Opencpn users. https://opencpn.org/ Tehani is an experienced programmer/hardware engineer who has great skills.

Please See https://www.cruisersforum.com/forums...ml#post3585704

Also Is there a way to get the app? Do we need to compile it under android? for build for android?
Best, R

[rgleason]

Marco wrote:

Feel free to post what you need.
My app/code is a PoC. Just a draft and improvements are more than welcome.
It is just something done in my spare time for fun.

[rgleason]

Marco suggested that he had found these links as well, to Madman Marine App

I also found this: "AP-WRC3iF Autopilot Remote Control" by Madman marine
https://youtu.be/YUADyh8bHdY (min 1:36)

https://www.madmanmarine.com/product...remote-control

iOS e Android:https://play.google.com/store/apps/d...ne.neil.isteer


Also on that website is Calypso Instruments

https://calypsoinstruments.com/shop

[rgleason]

Marco has sent another development he is working on.
stxMobile wrist watch. I think this has some merit.
Don't what watch or know how it is controlled. It isn't a touch screen watch is it?
It would also be useful to have a configurable screen for instrument data that would work with Opencpn somehow.
He sent this image.

[Tehani]

Originally Posted by rgleason

Marco has sent another development he is working on.
stxMobile wrist watch. I think this has some merit.
Don't what watch or know how it is controlled. It isn't a touch screen watch is it?
It would also be useful to have a configurable screen for instrument data that would work with Opencpn somehow.
He sent this image.

Rick,
Unfortunately, a gateway between OpenCPN and anything else (WiFi, NMEA2000, Seatalk, ...) will always be necessary.
That is why I have made a universal and economical one.
I have to find out how OpenSource works...

J.L.