Тёмный

XR2 RAVENSTAR VC (WIP) - Singapore to London - Testing SCRAM ascent and runway landing autopilots. 

dgatsoulis
Подписаться 301
Просмотров 285
50% 1

More testing of the scram ascent autopilot and introducing the runway landing autopilot. There are some bugs to iron out, but for the most part it's going ok. Also animated the furry dice :)
I am getting closer to the end of the list of features that I want to add.
Mainly an LPAD landing AP and a hover take-off AP.

Опубликовано:

 

21 окт 2024

Поделиться:

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист
Посмотреть позже
Комментарии : 14   
@dgatsoulis
@dgatsoulis 4 месяца назад
Here is how the flight goes from the user's perspective: You set up the destination and hit execute in the screen next to the HUD. The script runs and carries out the entire ascent at the correct heading, accounting for the Coriolis effect and using all the scram fuel. The ascent script stops the engines when the groundtrack is ~95% of the distance to the target. It keeps running a bit longer (altitude 75 km), at which point it opens the radiator and then gives the ship control back to the user. The user only has to use aerobrakeMFD and make sure that the aerobrake will take them to ~ 50 km away from the target. The user runs the landing script when they close the radiator just before reentry. It doesn't do anything until the ship is 100 km away from the end of the runway, so up to that point it's the user's job to adjust the AoA and bank in order to arrive at ~ 50 km away from the runway coordinates (given on screen by the script for now). At ~ 150 km way from the target, the user disengages the AerobrakeMFD HoldAoA and HoldBank and lets the ship glide. The landing script is already running, but will not assume control before reaching 100 km from the target. Once that distance is reached, it assumes control and manages the altitude, heading and speed, until it gently touches down on the runway, at which point it applies the brakes. Control is given back to the user at "wheel stop".
@spitzerysbertjose4431
@spitzerysbertjose4431 4 месяца назад
Thank you so much for what you are doing. This certainly looks next level.
@WingC3
@WingC3 Месяц назад
Just thought I’d pop by and see if we were any closer to getting our little hands on this?
@NorbertWoeller
@NorbertWoeller 4 месяца назад
Hello Dimitris that's great again! I admire your work and your motivation and hope that it continues for a long time...
@kerravon4159
@kerravon4159 3 месяца назад
Very impressive, great work!
@maievshadowsong5436
@maievshadowsong5436 4 месяца назад
This is shaping up to be an absolute beast of an addon! I like the Helmet Of / Off function. Not sure if it's new i noticed it only today heheh.
@predattak
@predattak 4 месяца назад
Aircraft carrier style landing :) It could be useful if you had a manual "insert target base" for the scram ascent autopilot just so you don't have to scroll through the whole list of destinations.
@dgatsoulis
@dgatsoulis 4 месяца назад
Agreed, I will definitely add the ability to directly type the base's name or even the coordinates.
@72TheScarface
@72TheScarface 4 месяца назад
amazing!
@DavidCourtney
@DavidCourtney 4 месяца назад
Looks like you set up the flight on the ground, and a script carried out the entire ascent? Then you set up the reentry, and another script carried out reentry and landing? Or did I miss something?
@dgatsoulis
@dgatsoulis 4 месяца назад
Hi David, I pinned a comment describing how the flight goes from the user's perspective. Is short, the scram ascent AP takes you towards the target, you take over after MECO at 75 km altitude to make sure that AerobrakeMFD will get you to ~50 km away from the target and slightly earlier than that (100 km from the target) the landing script takes over and lands you on the runway.
@glados1073
@glados1073 4 месяца назад
Hello Dimitris, you know this sim inside and out, i have a quick question for you. Iv'e been browsing on youtube looking for people that play orbiter to see what others do with the sim etc and i found a guy that talked about how orbiter works. He said this about how orbiter calculates trajectories and orbits : ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-q9lXGg3kc88.htmlsi=hKw7wCp0E-eqSxFu&t=363 (the link is with timestamp but if it fails the thing starts at 6:04 ) don't worry it's under a minute. I am curious about this because if it is true, how is orbiter accounting for gravity influence of other bodies on trajectory and orbits because from my point of view it clearly does, if a go to jupiter (or saturn for that matter) my orbit is "pulled" by all the moons and jupiter itself so my trajectory is all over the place because of the insane moon tugging. Also if you are in orbit around one of them, the orbit will get pulled and "Decay" in time Is something else at work here and i'm just imagining things? Also if the sim itself doesn't care about multiple bodies for trajectory calculation what is this in IMFD map settings when it tells you how many bodies it takes into account ex : 4+7 of 62 planets used. I would have asked him but it's been a long time and also another user raised the same question and his answer was in essence " leave me alone i know better than you"
@dgatsoulis
@dgatsoulis 4 месяца назад
Hi @glados1073, thanks for the link. I agree with you that it would be pointless to raise a question on a 9 year old video. As your intuition and experience in Orbiter have already shown you, he is not right about Orbiter using strictly 2 body Keplerian calculation to propagate the state vectors of a vessel in the simulation. The 2-body method is only used as a baseline for further calculations. My source is the dynamics.pdf that is located in the Doc\Technotes folder of your Orbiter installation. If you're so inclined, it is an interesting read. Orbiter takes into account the gravitational influences of multiple celestial bodies when propagating the state vectors of a vessel. While the pdf describes the use of a "2-body orbit" as a baseline for certain calculations, it explicitly mentions the inclusion of gravitational potentials from relevant celestial bodies. Here's a more detailed explanation according to the content of the dynamics.pdf written by Orbiter's creator Dr. Martin Schweiger: Gravitational Forces: The gravitational acceleration 𝑔(𝑟,𝑡) at any position 𝑟 and time 𝑡 between 𝑡0 and 𝑡1 is computed, which involves the summation of numerically calculated gravitational potentials from relevant celestial bodies. This indicates that the gravitational influence of multiple bodies is considered. Perturbations: In cases where time intervals become too large for accurate numerical integration of state vectors, Orbiter switches to "orbit stabilisation," a variant of Encke’s method. In this method, only the perturbations on top of a baseline 2-body orbit (defined by its osculating elements) are integrated numerically. This further suggests that while a 2-body model may serve as a baseline, additional gravitational perturbations from other bodies are incorporated. Numerical Methods: Different numerical integration methods, such as Runge-Kutta and symplectic integrators, are used to solve the equations of motion, which include gravitational forces from multiple sources. For example, in the symplectic integrators section, the methods are noted for their stability in long-term integration, which is crucial for accurately modeling the effects of multiple gravitational influences over time. Simulations: The numerical simulations section mentions that the gravitational point sources (Sun, Earth, Moon) were considered in the computational cost analysis of time propagation methods. This clearly indicates that multiple gravitational sources are factored into the simulations. Therefore, Orbiter does not rely solely on 2-body calculations; it accounts for the gravitational influences of other celestial bodies in its state vector propagation. Hope this answers your question. Best regards, Dimitris
@glados1073
@glados1073 4 месяца назад
@@dgatsoulis By the Gods this was a lightning fast response! Thank you, it indeed answered my question, now i have something interesting to go read (even though i'm not sure how much i will understand from it) AND most importantly i know i'm not crazy heheh.
Далее
XR2 Ravenstar Scram ascent autopilot VC WIP
26:22
DIY Pump Solutions
00:18
Просмотров 1,2 млн
Working Multiple positions in a small Airport
1:06:38
A35K arrival OPLA
26:03
Просмотров 8
MSFS 2024 - GAMEPLAY FOOTAGE - Virtual Reality!
27:00
Orbiter 2016 - Direct Lunar Landing
8:05
Просмотров 353