So you have a set entrance vector to the orbit altitude, which is pretty good besides optimizing efficiency a bit. How do you figure out WHEN for the general case?
@@TheGreatestFez Sorry. With a set ascent function, you can reproduce an ascent path which is pretty good. I was just thinking about how to find out WHEN you'll have completed your ascent+circularization. If you can find that, you can launch straight into docking with anything in orbit.
@@davidelmkies6343 Ahhh yes, so the only real way I know how to do that is to do a test run to a specific orbit. Given a craft, it should take X amount of time for it to reach a specific orbit, so if you don't change any other variables it will always take that amount of time, and you can store that for later use. You can then use the time and also where it finished the circularization to determine what phase angle your target youre trying to rendezvous with needs to be. Unfortunately its the best I have heard of, mostly because its not an easy thing to calculate the time to circularize/where you will finish circularizing without doing a full simulation.
@@TheGreatestFez I have done it in a limited fashion, pretty predictably less than 1 km encounter @ 75k with single stages locked at an acceleration rate. (never did actually get the burn at apo timing just right to fully rendezvous). Air resistance ends up being pretty minor for kinda pointy things and if you need fudge you can always do a little manuvering (which i was not doing) where the air gets thinner. Obviously the trick is to be able to generalize it. The way I did that was to launch at a bunch of twrs, store lots of data, and get acceleration and pitch angle as dependent variables and the time as the independent - and a few other tricks with the rest of the data. Maybe I'll go back and work out the kinks sometime but it seems pretty complex to add staging. But it is possible! :-) In some ways, it really isn't much different than taking one rocket and testing it like you mentioned. But it proves that there is order there that you can find, even if you don't mess with the somewhat complex aero model, which means maybe there's some sensible equation you could find. Mostly I learned a lot about data science and matlab and that the 3d maps I made are essentially intelligence in object form. Completely incomprehensible to me, but my program gained a large amount of understanding from the weird 3d contours I produced from the data in matlab.
Also I was doing gravity turns; with one of the dependent variables being how far I pitched over at like 200ft. So an equation might be more reliable. Crazy to get that kind of accuracy from gravity turns, right?
Your "My Square Root Pitch vs Altitude" plot is incorrectly labeled. The Y axis should be pitch relative to local vertical, not horizon. But otherwise a very solid video.
So you have a set entrance vector to the orbit altitude, which is pretty good besides optimizing efficiency a bit. How do you figure out WHEN for the general case?
Not sure what your asking, what do you mean by 'when'? Like what should the orbital entrance vector be?
@@TheGreatestFez Sorry. With a set ascent function, you can reproduce an ascent path which is pretty good. I was just thinking about how to find out WHEN you'll have completed your ascent+circularization. If you can find that, you can launch straight into docking with anything in orbit.
@@davidelmkies6343 Ahhh yes, so the only real way I know how to do that is to do a test run to a specific orbit. Given a craft, it should take X amount of time for it to reach a specific orbit, so if you don't change any other variables it will always take that amount of time, and you can store that for later use. You can then use the time and also where it finished the circularization to determine what phase angle your target youre trying to rendezvous with needs to be.
Unfortunately its the best I have heard of, mostly because its not an easy thing to calculate the time to circularize/where you will finish circularizing without doing a full simulation.
@@TheGreatestFez I have done it in a limited fashion, pretty predictably less than 1 km encounter @ 75k with single stages locked at an acceleration rate. (never did actually get the burn at apo timing just right to fully rendezvous). Air resistance ends up being pretty minor for kinda pointy things and if you need fudge you can always do a little manuvering (which i was not doing) where the air gets thinner. Obviously the trick is to be able to generalize it. The way I did that was to launch at a bunch of twrs, store lots of data, and get acceleration and pitch angle as dependent variables and the time as the independent - and a few other tricks with the rest of the data. Maybe I'll go back and work out the kinks sometime but it seems pretty complex to add staging. But it is possible! :-)
In some ways, it really isn't much different than taking one rocket and testing it like you mentioned. But it proves that there is order there that you can find, even if you don't mess with the somewhat complex aero model, which means maybe there's some sensible equation you could find. Mostly I learned a lot about data science and matlab and that the 3d maps I made are essentially intelligence in object form. Completely incomprehensible to me, but my program gained a large amount of understanding from the weird 3d contours I produced from the data in matlab.
Also I was doing gravity turns; with one of the dependent variables being how far I pitched over at like 200ft. So an equation might be more reliable. Crazy to get that kind of accuracy from gravity turns, right?
Your "My Square Root Pitch vs Altitude" plot is incorrectly labeled. The Y axis should be pitch relative to local vertical, not horizon. But otherwise a very solid video.