Optimizing a Reusable Rocket Trajectory!
HTML-код
- Опубликовано: 7 июн 2024
- Today we're perfecting a gravity turn for a reusable rocket! I'm once again improving my code to make it better and including the ability to track and land the first stage on a barge. Then I'm using the simulation to find the BEST* gravity turn possible.
*Results may not actually be perfectly realistic or even fully optimized, sorry I'm trying my best.
Music from Bensound.com
(0:00) - Intro
(2:09) - The Program
(4:24) - The Rocket
(7:01) - The Results
(9:44) - Flight Profile
(12:08) - Animation of the Flight
(12:57) - The Data
(17:39) - Outro 
Наука
I'm gonna do it: there are ways you could approximate stagnation point heating which require very little effort. Look at some of the stuff by Sutton and Graves
I ended up cutting some of my discussion about heating but yes stagnation temperature is pretty easy to get, the hard part is 1. determining the heat transfer coefficients (how much of that temperature is transferred to the rocket as thermal energy through convection and radiation) and 2. How much energy can the rocket absorb over what period of time. It’s not as simple as a max temperature because you can sustain a high temperature over a short time but may not be able to handle a lower temperature over a longer time if it ends up transferring more energy. The numbers for that aren’t readily available as far as I know
Looking at your velocity graphs, I think you forgot the centripetal acceleration effect of your x velocity on your y acceleration. I highly recommend doing some of your modeling in KSP and your programming with KRPC.
You don’t add centripetal acceleration back in, it emerges from modeling a point source of gravity. I tested the simulation against the 2-body problem by starting the rocket in orbit and removing all the thrust. It remained stable over multiple orbits. Just remember I’m using Cartesian coordinates not polar (i.e. y is not the same as altitude)
@@ConHathy Ah OK. A few years ago, I modeled a constant-altitude landing on the Mun in Excel, and it was something i had to account for to get the solution to work. I ran about 40 tests in game at various TWRs and I was getting to within 1% of the predicted dV expenditure.
Are you using simulink to do these simulations?
No, I’ve been writing them directly in Matlab (actually the last two have been in GNU Octave now that I’ve lost access to my school’s Matlab license but they’re mostly interchangeable)
3:50 Wouldn‘t it be possible to just say rocket = cylinder (probably a hollow one) and get first degree approximations that way? I would assume you can either find or calculate the lift behavior of a cylinder rather easily. I mean it doesn‘t have to be super crazy amounts of data anyway since you can make some reasonable assumptions about your maximum AoA and limit the complexity through that (and the simplified cylinder model).
I’m sure there’s an analytical approximation for the lift slope but I’m not familiar with it. It’s probably less effective than a flat plate but by how much? Also I would have to see how it changes at supersonic and transonic speeds