Orbital rings and accuracy

[MacaScull]

Hi, I noticed that the demo only seems to create ground tracked orbits for satellites, how would you go about producing a non suborbital/ground tracked orbit, in an instance with a rotating globe and moving satellites? Also how would you bench mark this against accuracy?

I've tried to use some of your code and make orbits that match in start and end positions, that are not ground tracked with the earth rotating also though havent been successful. Each time I try this I seem to end up with accuracy problems as the orbit of the ISS seems to be out a full orbit of the earth.

[dsuarezv]

Hi, by default orbits are displayed in ECEF mode, I guess you want ECI mode. You can display orbits in ECI mode (see below) but in that case you would have to implement the code for earth rotation, it's not in the library. It shouldn't be too difficult, just rotate the earth model the radians for the given time offset that is already used in the orbit prediction.

To display orbits in the ECI frame, set the type parameter of getPositionFromTle to 2 in tle.js. It has a default value of 1, change it to 2 instead, either in the default value or when calling the function.

[MacaScull]

Hi, Ive changed modes into ECI and that obviously now giving me the orbits I want. In regards to performing the earth rotation would that calculation be something along the lines of this:

       deg2Rad((hours * 15) + ((minutes/60) * 15) + (seconds/3600) * 15)

Then with the result of that I can do earth.rotation.y = result.

Would I have to account for any factors before rotating like a prior rotation? Sorry about all the questions about this.

[MacaScull]

Hello again, Ive added some inital rotation code in the _addearth function as: earth.rotation.y = this.deg2Rad(((date.getHours() - 12) * 15) + ((date.getMinutes()/60) * 15) + (date.getSeconds()/3600) * 15)

And some rotation code to match that of the planet itself in the update positions function as: this.earth.rotation.y += this.deg2Rad(15/3600)

Though again ive noticed using this site Tracker to keep track of the ISS and its position, that the path and position between the app and the site become out of sync. This happened when the ISS was coming from around the south up past south america over towards africa, this has got me thinking though, could the prediction code for this be wrong or not account for a different factor such as rotation speed at the equator being different rotation speeds at the poles, or could it potentially by due to the earth map being slightly wrong due to projections.

What are your thoughts on this?

[MacaScull]

I forked the repo and added in my code to do these things which I want, the problems with accuracy are still there though. Any update on how I could fix this?

[dsuarezv]

About these accuracy issues you mention, have you checked the numerical solution? Does it have the error?

El lun., 14 mar. 2022 21:52, MacaScull @.***> escribió:

I forked the repo and added in my code to do these things which I want, the problems with accuracy are still there though. Any update on how I could fix this?

— Reply to this email directly, view it on GitHub https://github.com/dsuarezv/satellite-tracker/issues/16#issuecomment-1067334072, or unsubscribe https://github.com/notifications/unsubscribe-auth/AABV3VZGYTLVCPHW4UENOFDU76YJ7ANCNFSM5P5JWC3Q . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&referrer=utm_campaign%3Dnotification-email%26utm_medium%3Demail%26utm_source%3Dgithub.

You are receiving this because you commented.Message ID: @.***>

[MacaScull]

What do you mean by checking the numerical solution? How would I do this?

I only ever visually see the issue, If you want you can use my forked version and view a ISS tracker on a normal map, you'll be able to see there the difference in positions and inaccuracies.

I apologise if any of this comes across as ignorant, I don't mean to sound like this.

[dsuarezv]

I was referring to the calculated coordinates of the station, the numbers, and check those against what you are comparing with. If they are similar, then the problem might be the texture mapping. If they are different, then the problem might be in how the solution is calculated.

[MacaScull]

Im currently struggling getting the coordinates but due to an async issue, I will work on fixing this. If it were a texture map issue, where would suggest getting a better map? and if it wasnt a map issue but a solution calculation issue, how would you suggest I tackle that?

I have also made sure that the satellites and their orbit are not placed on the earth object and are separate to ensure that the earth and satellites are not directly having any effect on each other.