Celestial Navigation with OpenCPN

[Andres_RG]

Originally Posted by otya

... I am leading to a suggestion for a third mode to Andres suggestion. I'd like to be able to enter the azimuth distance pairs (and possible assumed location also)...

Hi otya, in me proposed solutions I do not take into account the traditional Marcq St Hilaire LoP solution because with OpenCPN we can solve the problem without any approximation. http://opencpn.org/node/102
One example of the exact CoP and the tangent LoP using ChartWork and OpenCPN:

1. The circle CoP:
Fecha y hora UT1
01/09/2010
15:52:50
SITUACION ESTIMADA
Latitud: Be = 43.316660 = 43 19.0'
Longitud: Le = -2.000000 = -2 0.0'
ALMANAQUE NAUTICO
Cuerpo celeste: Sol
Polo de iluminacion del astro: Coordenadas Ecuatoriales
GHA = 58.215316 = 58 12.9'
Dec = 8.154656 = 8 9.3'
Correcciones altura:
SD = 15.848866 '
HP = 0.145238 '
CORRECCION DE LA ALTURA
Limbo: Inferior
altura sextante []: Hs = 29.6634 = 29 39.8'
correcion de indice [']: ie = 0.0000
altura del ojo sobre el nivel del mar [m]: heas = 2.5000
temperatura ambiente [C]: T = 10.0000
presion atmosferica [mb]: P = 1010.0000
en []:
altura observada por el sextante: Hos = 29.6634
depresion del horizonte: dip = 0.0464
altura aparente: Ha = 29.6170
correcion por refraccion: R = 0.0285
correcion por achatamiento: OB = 0.0000
Paralaje horizontal: HP = 0.0024
correcion por paralaje: PA = 0.0021
semidiametro: SD = 0.2641
Augmentation of SD = 0.0000
altura observada corregida: Ho = 29.8548 = 29 51.3'





2. The line LoP:
Fecha y hora UT1
01/09/2010
15:52:50
SITUACION ESTIMADA
Latitud: Be = 43.350000 = 43 21.0'
Longitud: Le = -2.150000 = -2 9.0'
ALMANAQUE NAUTICO
Cuerpo celeste: Sol
Polo de iluminacion del astro: Coordenadas Ecuatoriales
GHA = 58.215316 = 58 12.9'
Dec = 8.154656 = 8 9.3'
Correcciones altura:
SD = 15.848866 '
HP = 0.145238 '
CORRECCION DE LA ALTURA
Limbo: Inferior
altura sextante []: Hs = 29.6634 = 29 39.8'
correcion de indice [']: ie = 0.0000
altura del ojo sobre el nivel del mar [m]: heas = 2.5000
temperatura ambiente [C]: T = 10.0000
presion atmosferica [mb]: P = 1010.0000
en []:
altura observada por el sextante: Hos = 29.6634
depresion del horizonte: dip = 0.0464
altura aparente: Ha = 29.6170
correcion por refraccion: R = 0.0285
correcion por achatamiento: OB = 0.0000
Paralaje horizontal: HP = 0.0024
correcion por paralaje: PA = 0.0021
semidiametro: SD = 0.2641
Augmentation of SD = 0.0000
altura observada corregida: Ho = 29.8548 = 29 51.3'
DETERMINANTE DE LA RECTA DE ALTURA
Angulo Horario Local: LHA = 56.065316 = 56 3.9'
Coordenadas Horizontales: Hc, Z
Altura Calculada: Hc = 29.9477 = 29 56.9'
Azimut: Z = 251.4
p = Ho-Hc = -0.092869 = -5.572151 '



The CoP and the Lop:

[Ptizef]

Hi
Since the last GIT version, I can't anymore build the celestial navigation dll
I get the errors message in the file attached
(Windows vista)
thanks
Jean Pierre

[seandepagnier]

Thank you for testing my plugin! I have a patch which maybe works, maybe not, but worth a shot. It's common to get compiler errors on windows if something compiles fine on linux because c++ is stupid


Please test after applying patch

[Ptizef]

Many thanks for the patch
Celestial navigation patched , built and run fine now
Jean Pierre

[Netsurfer]

Ptizef + geckosenator ...

please add such issues (also) to the tracker.
That's (by now) the first place one would look-up for errors, bugs and patches.

I gathered the same errors as Jean Pierre at first when trying to compile from latest GIT source and finding related posts here in the forum is very much like winning in the lottery ... .

Thanks,

Gunther

PS: It also might be a good idea to test compiling the different versions after pushing new/ changed code to GIT ...

[Andres_RG]

The celestial navigation plugin in "OpenCPN Version 2.2 Beta Build 1007" under WXP has a bug.
I'm trying to compile the whole project but as I have not spent much time I have not yet achieved.

Looking at the source code, even though I think I know something about celestial navigation , I do not understand the meaning of the azimuth sight, geckosenator could you explain it in detail?
The lunar sight do not have the correct focus, information is available at google books, see Mendoza y Rios and Chauvenet.
And I believe that the treatment of uncertainties is not correct, is not known a priori, is the result of statistical treatment of observations.

To achieve something, I propose the following:
First focus on the sextant sights and validate it:

• The code provided by Don for altitude correction is OK.
• We must need to check the built-in almanac: GHA, DEC, HP, SD
• With this the circle of equal altitude can be plotted.

This is the most simple and easy way to perform celestial navigation using the power of OpenCPN.

After this work, you can go about doing extensions.

[seandepagnier]

Originally Posted by Andres_RG

The celestial navigation plugin in "OpenCPN Version 2.2 Beta Build 1007" under WXP has a bug.
I'm trying to compile the whole project but as I have not spent much time I have not yet achieved.

If you could post more info on the build error in opencpn.org/flyspray

I maybe can fix the problem, unfortunately with c++ and cross platform I am always running into things that build on one operating system but not the next. It would be a great help if you could test the current plugin for accuracy and report on the specific errors and tendencies, so we know where to change the code.

Looking at the source code, even though I think I know something about celestial navigation , I do not understand the meaning of the azimuth sight, geckosenator could you explain it in detail?

We went over what it does before, if you don't understand it , don't worry about it. For now we need to focus on the accuracy of altitude shots.

The lunar sight do not have the correct focus, information is available at google books, see Mendoza y Rios and Chauvenet.

What do you mean correct focus? It is probably possible to improve my moon handling in other ways though

And I believe that the treatment of uncertainties is not correct, is not known a priori, is the result of statistical treatment of observations.

All I am doing is solving for the line of position on the earth when you add the error to the measurement, and when you subtract it. Then the area between is filled. I dont really know how this could be wrong, but if it is, please explain and how I could make it right, I don't think it would be hard.

To achieve something, I propose the following:
First focus on the sextant sights and validate it:

• The code provided by Don for altitude correction is OK.
• We must need to check the built-in almanac: GHA, DEC, HP, SD
• With this the circle of equal altitude can be plotted.

This is the most simple and easy way to perform celestial navigation using the power of OpenCPN.

After this work, you can go about doing extensions.

I already have many more extensions in mind. In any case, the boring code is in place. The user interface, the ability to plot the lop, almanacs etc.. I can extend this as needed. I think we can achieve great accuracy just by tweaking a few lines of code for altitude corrections. I would be glad to accept patches :-P Otherwise, maybe you can point me to don's code again?

[LeaseOnLife]

I played with the celestial navigation plugin in opencpn 2.2v918 and can't get good results. Before posting this to bugspary, I thought I discuss this here, maybe I do something wrong.

Using other tool, specifically "Teacup Navigation v1.8" and "Navigational Algorythms" and I get down to a few miles error with these tools.

Debugging, I think I found one flaw: Using the time entered, the plug in seems to calculate the geographical position of the sun by many (80-90) miles wrong.

Here is a way to see the error, repeatably.

Use the following date and time: 10/10/2010 05:00 UTC

• using a paper almanac or Teacup navigation, I come up with GHA = 258 deg 13.4' and DEC S 6 deg 35.9'.
• GP should then be at 006 35.9 S and 102 13.4 E, drop a mark in opencpn on this position
• Zoom with opencpn to this position, you should see Sunda Strait, Indonesia
• In opencpn / celestial-plugin enter the time from above and a sextant reading of 90 degrees, change the accuracy from 1 to 0.1
• the celestial plugin should draw a circle 70 miles out of Sunda Strait, diameter about 10 miles (0.1 degree)
• the center of this circle is about 006 35.836 S / 103 41.025 E, put a opencpn mark there
• create a route between the two marks
• check the route properties
• Distance (Error) between the two points is 87.6 miles!

Geckosenator: It would be great if one could see the results of various calculations. I would like to use your plug in as a learning and training tool. Having a tab which lists all the calculations performed will help debugging and learning!

[LeaseOnLife]

No reply ==> Entered the above text as a bug in flyspray

[ChuckSK]

Originally Posted by LeaseOnLife

I played with the celestial navigation plugin in opencpn 2.2v918 and can't get good results. Before posting this to bugspary, I thought I discuss this here, maybe I do something wrong.

Using other tool, specifically "Teacup Navigation v1.8" and "Navigational Algorythms" and I get down to a few miles error with these tools.

Debugging, I think I found one flaw: Using the time entered, the plug in seems to calculate the geographical position of the sun by many (80-90) miles wrong.

Here is a way to see the error, repeatably.

Use the following date and time: 10/10/2010 05:00 UTC

• using a paper almanac or Teacup navigation, I come up with GHA = 258 deg 13.4' and DEC S 6 deg 35.9'.
• GP should then be at 006 35.9 S and 102 13.4 E, drop a mark in opencpn on this position
• Zoom with opencpn to this position, you should see Sunda Strait, Indonesia
• In opencpn / celestial-plugin enter the time from above and a sextant reading of 90 degrees, change the accuracy from 1 to 0.1
• the celestial plugin should draw a circle 70 miles out of Sunda Strait, diameter about 10 miles (0.1 degree)
• the center of this circle is about 006 35.836 S / 103 41.025 E, put a opencpn mark there
• create a route between the two marks
• check the route properties
• Distance (Error) between the two points is 87.6 miles!

Geckosenator: It would be great if one could see the results of various calculations. I would like to use your plug in as a learning and training tool. Having a tab which lists all the calculations performed will help debugging and learning!

Hi Dirk

I just posted a fix on flyspry #186.

Chuck

[LeaseOnLife]

Chuck, Thanks! And I though this might be a dead thread.....

I am not set up for compiling, hopefully this goes in the next Beta for XP.

Dirk

[LeaseOnLife]

That's a shame. I was just getting started with testing this and now it's gone.

Is the geckosenator anywhere to be seen?

[MvN]

I can't find the plugin any more and nor a link to GPX Chart work. I would really like to try the Celestial plugin. Any news???

[Andres_RG]

The plugin is no longer available for the stable release OpenCPN 2.3.0.
If you read it history here, it is no well focused for practical navigation, and has a lot of problems.
Geckosenator, its creator give no signs of life. He would not heed the advices and the offer of collaboration for this to have ended well.

[Andres_RG]

Now, the only way to use celestial navigation with OpenCPN is using my software, available for free at my Web page, (running only under Windows).

For Chart work, see: Supplementary Software | Official OpenCPN Homepage

If you want a complete celestial navigation solution, you can try also my program “Celestial Fix”. Or wait till January 2011 when the new fully integrated release will be launched: Astronavigation https://sites.google.com/site/naviga...attredirects=0 This generates a file "cop.gpx" with all the observations, the fix, and the confidence ellipse. This GPX file it can be open in OpenCPN, Google Earth or other compatible applications.

Regards
---
If someone want to use this programs in his own language, I can send some files text for translation, and I shall adapt the GUI.