Project XR2 Ravenstar - Mk II

[halcyon]

I believe that all atmospheric autopilots today should be able to maintain an assigned altitude, execute FLCH, maintain a specified rate of climb and speed. Many can Autoland. They should also be able to follow waypoints as well. These are basic to any commercial airliner, and are now entering general aviation as well. Any new ship would be remiss without these capabilities.

All these things you mention are available in flight simulators dedicated to atmospheric flying, such as FSX.
Excuse me, but isn't Orbiter a space simulator? You want the XR2 Mk2 to have a working FMC with VNAV, LNAV, level change, V/S hold, and altitude hold modes?
For what? So you can fly over a texture devoid landmass with no weather or other traffic?

The XR2 belongs in space, not crusing in sub sonic flight at 36,000 feet.
Why don't we let the XR2 do what it's designed to do and not ask for features that 1. Aren't necessary in a space simulator where over 80% of the time you're in...space. 2. Adds even more development time on the XR2 which, I'm sorry to say, doesn't need any more time added to it's release.

Sorry but it bugs me when people want features that don't belong in a space simulator. Level change? You really need that to get the XR2 from 3,000 ft to 300,000 ft?

 

[Urwumpe]

Sorry but it bugs me when people want features that don't belong in a space simulator. Level change? You really need that to get the XR2 from 3,000 ft to 300,000 ft?

Also, they forget that a spaceplane is no airliner at all - they are not made for flying straight and level for hours, they prefer other means of cruise.

Autopilots for spaceplanes have little in common with normal airliner autopilots. Stuff like energy management during reentry is unknown to normal autopilots, but crucial for proper spaceplane autopilots.

 

[Keatah]

I'm thinking of those autopilot modes for ferry flights. Surely we're not going to put it on the back of a 747 to go cross country. And we're also not going to expect a pilot to maintain an exact altitude for an hour or two during a transport & ferry flight. Nor will we be doing anything ballistic or high altitude for said flights.

I'm not unaware of even the slightest difference between a modern day airliner and the fictional craft we play with in orbiter. Hardly.

Perhaps some of the finer points of flight level changes and throttle adjustments needn't be coded in. But an altitude hold mode should be.

Either way, coding on the ship has not begun in full. I would expect at least another year or almost 2 before the mk2 is ready. Real aircraft can take years of development work, so the same can apply here too.

 

[ercsim]

I don't think there is any need for any alterations to XR's flight control related systems/autopilots. All it needs are a few more gauges to make it more complicated. Ntg should be added to the AP. Its maneuverability are already unprecedented.

 

[Urwumpe]

I'm thinking of those autopilot modes for ferry flights. Surely we're not going to put it on the back of a 747 to go cross country. And we're also not going to expect a pilot to maintain an exact altitude for an hour or two during a transport & ferry flight. Nor will we be doing anything ballistic or high altitude for said flights.

That is exactly wrong - a ballistic flight would be way more effective for ferry flights for such a vehicle. Airliners would likely do the same, if they just could do that. Which they can't.

Why fight all that benefits the vehicle? For airliners, it makes sense because you usually fly in confined space. There you need accurate altitude keeping to prevent collisions. speed control is needed for getting lined up.

But for a spaceplane, it simply isn't needed to be there. The space above FL690 isn't controlled, the flight time below FL690 very short. Usually you just get below FL690 for the final approach. Your wait pattern is in orbit. if your runway is suddenly not available, you will be redirected during reentry. A commanded to fly a less effective HAC to delay your arrival by a minute or two.

For a XR-2, you could even use your main engines during landing to delay your landing, but then, it is much more punishing as for an airliner. A "go around" wouldn't work well, you consume a lot of your fuel then.

I'm not unaware of even the slightest difference between a modern day airliner and the fictional craft we play with in orbiter. Hardly.

That is just sad. But ignorance is no fact at all.

Perhaps some of the finer points of flight level changes and throttle adjustments needn't be coded in. But an altitude hold mode should be.

Completely wrong. Even an altitude hold would look much different - it is simply ineffective to do it like Airliners that are squeezed below 69,000 ft.

Or take Mach/speed hold. The XRs have it, but I consider it a feature that I never needed. For a space plane, it isn't needed. You slow down or you accelerate. or you do nothing at all and fly ballistic. Between that, you have only a desire for acceleration hold. May it be during ascent to throttle engines automatically, or during reentry to automatically follow the envelope.

Autoland might be interesting, but it would still look nowhere like the stuff you know from airliners.

Vertical speed hold? Could be nice during ascent, but then, there are better algorithms around for that dimension, that make more sense. For example PEG.

Either way, coding on the ship has not begun in full. I would expect at least another year or almost 2 before the mk2 is ready. Real aircraft can take years of development work, so the same can apply here too.

yeah, possible. Maybe less.

---------- Post added at 11:37 AM ---------- Previous post was at 11:34 AM ----------

I don't think there is any need for any alterations to XR's flight control related systems/autopilots. All it needs are a few more gauges to make it more complicated. Ntg should be added to the AP. Its maneuverability are already unprecedented.

I think, all that is really needed, is a better clock. Like pressing just one button to start a countdown to the next apsis. Or show the local time at a base. Or indicate "time to go" until you have reached your planned MECO speed.

You can never have good enough clocks in space anyway. Clocks are vital.

 

[ercsim]

That is exactly wrong - a ballistic flight would be way more effective for ferry flights for such a vehicle. Airliners would likely do the same, if they just could do that. Which they can't.

Why fight all that benefits the vehicle? For airliners, it makes sense because you usually fly in confined space. There you need accurate altitude keeping to prevent collisions. speed control is needed for getting lined up.

But for a spaceplane, it simply isn't needed to be there. The space above FL690 isn't controlled, the flight time below FL690 very short. Usually you just get below FL690 for the final approach. Your wait pattern is in orbit. if your runway is suddenly not available, you will be redirected during reentry. A commanded to fly a less effective HAC to delay your arrival by a minute or two.

For a XR-2, you could even use your main engines during landing to delay your landing, but then, it is much more punishing as for an airliner. A "go around" wouldn't work well, you consume a lot of your fuel then.



That is just sad. But ignorance is no fact at all.



Completely wrong. Even an altitude hold would look much different - it is simply ineffective to do it like Airliners that are squeezed below 69,000 ft.

Or take Mach/speed hold. The XRs have it, but I consider it a feature that I never needed. For a space plane, it isn't needed. You slow down or you accelerate. or you do nothing at all and fly ballistic. Between that, you have only a desire for acceleration hold. May it be during ascent to throttle engines automatically, or during reentry to automatically follow the envelope.

Autoland might be interesting, but it would still look nowhere like the stuff you know from airliners.

Vertical speed hold? Could be nice during ascent, but then, there are better algorithms around for that dimension, that make more sense. For example PEG.



yeah, possible. Maybe less.

---------- Post added at 11:37 AM ---------- Previous post was at 11:34 AM ----------



I think, all that is really needed, is a better clock. Like pressing just one button to start a countdown to the next apsis. Or show the local time at a base. Or indicate "time to go" until you have reached your planned MECO speed.

You can never have good enough clocks in space anyway. Clocks are vital.

You said it all! I am with you :thumbup:

 

[flyer]

Just to offer another perspective on this discussion...


I think it would be very useful for any spaceplane to be able to fit in with the existing infrastructure of aviation as it stands today (if you want to pretend you are flying around in the present time). I think it would be difficult (but not impossible) to fit in a "ballistic re-entry approach" with normal aircraft in some cases.


And to say you could just fly a less than optimal HAC might not be feasible in some of the congested airspace around the bigger international airports in the world. Depending on who you had sitting in the back of your spaceplane, you could end up on a scenic tour of the Russian countryside just like all the other "altitude limited" guys trying to land into Moscow on a dark, cloudy night with thunderstorms all around you. For this reason I would say it would be very useful to have altitude/heading hold functions on the autopilot. If you really needed to fit in with the other aircraft on the approach then I would say LNAV would be very useful too. Some of the approach routings can be tricky when flying raw data and some approach routings require some sort of GPS/LNAV system.


All of these things are aids that can reduce workload and increase safety which I'm sure everybody agrees is a good thing. And I wouldn't be too happy flying into Chambery with weather on minimums from a ballistic approach (performance issues aside, some rich spaceplane owner would definately want to do this at some point and so I would say a nice controlled approach based on established procedures would be the best way to go, until the next generation of navigation gets certified etc etc).


So leaving current aviation aside, it really depends on the way you run your simulations in your own mind. If you're happy making a ballistic flight from Paris to London and you don't mind saying "ATC can clear me for the ballistic approach" then that's fine, you don't need your "airliner autopilot". But for those that want to run their simulations taking this stuff into consideration then they should not be limited just because someone else says "it wouldn't happen that way so you will not be allowed to have that opportunity".


You could also envision your simulations being run in a future where only spaceplanes are flying around in which case I'm sure a different system of air traffic control would be in place which would allow for the more optimal ballistic trajectories.


At the end of the day it is up to the devs if they add these functions, all we can do is debate and offer opinions for them to read.

It's also worth saying that with LUA (the sdk has the examples) it is possible to create very quickly an autopilot tuned to the XR2's characterstics (or whichever spaceplane takes your fancy). It is also possible to create something like this in c++, or look on orbithangar etc.


Anyway, this is a very interesting discussion so I just thought I would offer another view on it.


Cheers,



Flyer

 

[Urwumpe]

And to say you could just fly a less than optimal HAC might not be feasible in some of the congested airspace around the bigger international airports in the world. Depending on who you had sitting in the back of your spaceplane, you could end up on a scenic tour of the Russian countryside just like all the other "altitude limited" guys trying to land into Moscow on a dark, cloudy night with thunderstorms all around you. For this reason I would say it would be very useful to have altitude/heading hold functions on the autopilot.

Short counter-argument: For flying level, you need a higher speed than all the other planes. You can't just enter a standard wait-pattern anyway, because of it. Its a spacecraft. You can't be lined up in a nice chain with all the other planes, you would BADLY disturb the other planes regardless how much you put effort into it.

You are right, being able to fit into the existing infrastructure would be great. Even greater would be a space-plane that is safe and robust in space. You can't get it all, it is all a number of trades. you want to be more mobile in low atmosphere and at slow speeds, you need either wings, that are terrible during reentry, or you need variable geometry, which adds complexity and can cause serious trouble - just watch how many variable geometry planes fail in 10 years because of variable geometry faults (not every year, but often enough).

 

[Keatah]

While the current airframe layout may not be optimal at slower landing speeds, it is certainly capable. And as far as autopilots go, in real life this would be an electronics package with sensors and software. Much of which is already in the existing xr2 anyways. So mostly it would be software.

As far as variable geometry goes, the only thing that should be thought about (for spaceplanes) is the 'waveriding' wingtips that move up and down. Similar to the xb-70.

 

[flyer]

You are right, being able to fit into the existing infrastructure would be great. Even greater would be a space-plane that is safe and robust in space. You can't get it all, it is all a number of trades. you want to be more mobile in low atmosphere and at slow speeds, you need either wings, that are terrible during reentry, or you need variable geometry, which adds complexity and can cause serious trouble - just watch how many variable geometry planes fail in 10 years because of variable geometry faults (not every year, but often enough).

That's true and I agree.

Short counter-argument: For flying level, you need a higher speed than all the other planes. You can't just enter a standard wait-pattern anyway, because of it. Its a spacecraft. You can't be lined up in a nice chain with all the other planes, you would BADLY disturb the other planes regardless how much you put effort into it.

I hadn't thought about the speed issue and I agree that it could be difficult to utilise the current routings/procedures used in aviation. Perhaps special arrival routes could be developed for these craft which provide the required protected areas on any turns that need to be made whilst restricting the tight turns made in the more congested airspace closer to the field.

It could be possible to create holing patterns for these craft that are away from the existing traffic patterns and then allow the space planes to hold there whilst the traffic sequencing was worked out, then once cleared for the approach, a long straight-in could be given. I think this holding area would be necessary as things can quite easily get messy very quickly for a variety of reasons around busy airports. It's maybe not quite a straightforward as I originally thought though...

But regardless of that, I still think to have the ability to maintain an altitude/heading etc would be very very helpful if you want to consider yourself coming into a present-day busy international/interplanetary airport. Again this goes back to where/when you consider yourself landing your space plane and how you visualise your virtual world operating...

And to extend what Keatah has just said, the systems/electromagic needed to get it all working would be straightforward. I guess there are tweaks that could be used to make the XR2 better lower down and it would again be the trade-off scenario. But as it stands right now, the current XR2 would seem to be a bit too tricky to just "slot in" to a standard arrival/approach procedure philosophy without some clever systems/procedure modifications.

 

[Urwumpe]

While the current airframe layout may not be optimal at slower landing speeds, it is certainly capable.

No you can't - a shape that is good for Mach 3 or higher is pretty bad for slow speeds. A fighter that should get to Mach 2 can still do some compromises there to fly good at subsonic speeds and not that bad at Mach 0.85 and higher, but for spaceplanes, we are talking about up to Mach 28, and most of the time, Mach 10-15.

A shape, that has pretty good Lift/Drag ratio at high speeds, and thus is better performing, has very low L/D at subsonic speeds. The Shuttle for example is bad everywhere, because of the compromises needed. L/D of 1.6 at supersonic speeds isn't that bad, but L/D of 4.5 at subsonic flight is literally a flying brick.

And the Deltaglider (XR or not) have all already pretty unbelievable good aerodynamics. Such a stable performance would be impossible with current aerodynamic knowledge, not even active aerodynamics can provide such good values at all speeds.

If I would go really sci-fi, there could be magnetic augmentation during high speed flight to provide better lift to drag. But... there we are talking about stuff, that is so extremely esoteric that I can't even find scientific papers about it.

As far as variable geometry goes, the only thing that should be thought about (for spaceplanes) is the 'waveriding' wingtips that move up and down. Similar to the xb-70.

The XB-70 has not been that effective there, as far as I remember, its landing behavior was even with the variable geometry pretty annoying. Especially the wing flutter was named to be a problem.

 

[Cras]

The only autopilot that I could possibly suggest on a space craft is LVLH. But then again, that is no use unless you can stick the object you are tying to dock with into LVLH as well.

An atmospheric autopilot would be a utter waste of time to develop, as about 1% of the Orbinauts would use them.

A proper DAP would be excellent, in any Orbiter add-on. And that would be many many times more useful than any altitude hold or fletch.

 

[Coolhand]

I think Dougs provided a great set of automatic controls, no it's not the pre-programmed effortless, sit back and relax etc. experience but its not meant to be and if you think it should be like that then you're missing the point. XR's put the human back into human spaceflight, if that means we end up in a million peices somewhere over the med or upside down in a crater, so be it...:lol: Learn to fly, learn to use the instruments and the MFD's, full computer control is great and safer and so many other things but its not engaging in the same way, its more like watching a replay or a video.

While its impossible to design something thats perfectly optimised for all these different flight regimes - even with VG, unless you have something like Trimaxian technology. I think it's fair to say the MK2 as the slickest shape of this family of designs should be a little trickier to fly at low speeds but slightly faster or more efficient at getting to orbit. Also it perhaps might not heat up as much on ascent and re-entries might be handled slightly differently.

There may be complaints that it doesn't have enough wing or wont be as good at low level but I think overall the MK2 is a better shape at speed than the MK1, which in itself had a better shape than the other XR's and DG's which are more optimised for the return/gliding stage of flight rather than actually getting up there.

(I'm sure you'll still be able have lots of fun with it at low level -with those powerful rocket engines, how could you not)

It's clear that a genuine SSTO as we would build it today has very little wing and they probably do not make for good low speed gliders as they have to be more optimised for the hypersonic stages of flight - due to the more limited energy available to them. But even they are compromises and not perfect for every stage of high speed flight. and of course are not the 'Jeeps' of the orbi-verse; the go anywhere do anything vessels that the DG's/XR's are.

For making something which operates well in very different regimes, Variable geometry, swing wings, etc are very useful, they add complexity and weight but really all the concepts - valkyrie, tomcat. etc. - which you might think of of are 40+ year old... In fact in a sense, we already have a spaceplane that changes shape. Spaceship one, not only 'extends its wing for greater climb performance' but also has that high drag configuration for use on 'orbit'

Besides, if we get EVERYTHING crammed into the Mk2, what will be left for the Mk3 ??

Make the most of the MK2, because I'm not coming back to this one again. I've attempted to foresee some coming developments and made some provision for those so it's possible you may see future improvements but I probably wont have anything to do with them.

Whether I make any other models for Orbiter or not I don't know, but what would rather have, an incremental update for the XR2, or something completely new. I think like many people here if i realised how much effort I would put into it, I wouldn't have started it in the first place. I suppose once a certain amount of time is invested, its harder not to see it through than to put the actual work in.

Having said that, we're still not done with the model here, perhaps I'll have some updates ready later on.

 

[Rtyh-12]

I think Dougs provided a great set of automatic controls, no it's not the pre-programmed effortless, sit back and relax etc. experience but its not meant to be and if you think it should be like that then you're missing the point. XR's put the human back into human spaceflight, if that means we end up in a million peices somewhere over the med or upside down in a crater, so be it...:lol: Learn to fly, learn to use the instruments and the MFD's, full computer control is great and safer and so many other things but its not engaging in the same way, its more like watching a replay or a video.

I totally agree with you Coolhand. The DGIV is a great ship with all its autopilots for everything, that's how I learnt to fly. Then I switched to the XR-2. Now, when I can easily fly an XR-2 into orbit, flying a DGIV is weird. I want to fly it into orbit myself (and I can), but even though it has autopilots for a lot of things, it lacks the very autopilot it should have: attitude hold. Which is present in the XR-2.

I think that the current autopilots that the XR-2 has are just perfect. Some people may prefer it to have an ascent or reentry autopilot, but then why don't you just watch a video? I can record one if you want... KSC to Brighton Beach, is that fine? :lol:

 

[Eli13]

Most of the time, I just don't have the time to do certain things. Like ascent, I launch and go do my HW or something more important. I watch it and as soon as its done, just quicksave it and/or pause it and figure out what to do with it later.

 

[markl316]

Quick question to Coolhand:

I'm an engineering student, so I work with CAD software, so I'm used to that interface.

Out of curiosity, about how many features (revolve, extrude, etc.) will there be in the final model?

Thanks!

 

[Coolhand]

I'm really not sure what you mean... How are modelling tools like those considered features? Are you curious about the processes involved in building it?

 

[Screamer]

re-entry autopilot.

Correct me please if I miss the point here.
In my opinion the only reliable re-entry autopilot is AutoFCS.
In fact, I tried it with the current XR2, and it work.
LOLA mfd is also accurate, but it is meant for body's without an atmosphere.

 

[Izack]

Correct me please if I miss the point here.
In my opinion the only reliable re-entry autopilot is AutoFCS.
In fact, I tried it with the current XR2, and it work.
LOLA mfd is also accurate, but it is meant for body's without an atmosphere.

A practised brain can still do better with simpler tools, but that's another matter entirely.

I'm satisfied with the current XR-2 autopilots. IMO maximum automation would be nice if I made a living flying one of these, but would make for a very boring hobby otherwise!

 

[Eli13]

My only request is auto-docking. I'm a rather errr slow manual docker. In other words, I take a good, non-realistic amount of time to get everything just right. Oh well. It'd probably be better if someone compiled the DG auto-docker to a MFD.