Question The XR2 normal/anti-normal autopilot

[JWise]

Just wondering if others experience this problem and what's the best solution.

In the XR2 I noticed when I turn on anti-normal or normal autopilot and use a full main engine burn the autopilot is unable to hold the craft steady and drifts to the side. This ruins my Apoapsis and Periapsis, especially on a long burn.

I noticed it does not drift so much when only applying a little thrust but it seems inefficient to do so and I thought it might have something to do with trimming the engine?

Does anyone have a solution for this or is low-thrust and translation thrusters the best way to align orbit planes?

 

[Nemoricus]

That's not the autopilot drifting, but the plane of your orbit changing. At high thrust, the autopilot can't keep up with the change.

If you want to use the orbit normal autopilots to change plane, you either need to burn slow or at high power for short periods of time.

However, it's more efficient to make a single, high power burn, yes. This requires that you orient yourself in a direction other than orbit normal, I'm afraid, but you can use TransX and IMFD to plan such a burn.

 

[JWise]

Thanks for the quick reply. I forgot to say I noticed this problem does not occur on the DG4, does the DG4 have a more precise autopilot that avoids you messing up you apa and pea on a burn?

 

[Cras]

try using Attitude MFD or Pursuit MFD to hold attitude. That might help.

 

[dbeachy1]

XR vessels use the default Orbit Normal, Orbit Antinormal, Prograde, Retrograde, and Kill Rotation autopilots built into the Orbiter core.

Attitude Hold, Descent Hold, and Airspeed Hold, on the other hand, are custom XR autopilots.

 

[JWise]

Thanks for replies and dbeachy it is a fantastic craft and given me many hours of enjoyment (and enjoyable stress).

I'll just have to learn to fly it better!

 

[johan]

Thanks for the quick reply. I forgot to say I noticed this problem does not occur on the DG4, does the DG4 have a more precise autopilot that avoids you messing up you apa and pea on a burn?

Could it simply be that the DG-IV has more powerful attitude thrusters in relation to its mass, than do the XR vessels?

Or perhaps it has a non-standard orbit-normal autopilot program.

 

[JWise]

Actually I've now noticed the dg4 orbit-normal autopilot also can't hold it steady during a full burn, but only drifts very slightly and doesn't hurt the ApA/PeA that much. The dg4 seems to drift pitch upward (from a orbit-normal autopilot perspective) while the XR seemed to go to the side.

Perhaps it's mathematically impossible to do a orbit-normal burn without affecting your ApA/PeA and I'm unaware of this?

 

[johan]

Perhaps it's mathematically impossible to do a orbit-normal burn without affecting your ApA/PeA and I'm unaware of this?

While I'm nowhere near being a mathematician, I doubt that... it's what someone else said earlier... as you apply thrust anti-normal, you're changing the plane of your orbit, which of course is the point of the manoeuver. As the plane changes, the direction called "anti-normal" changes. The faster your orbital plane changes, the faster this direction drifts. Applying more power, causes faster change.

Of course the autopilot compensates by rotating the spacecraft, but it inevitably has a small lag. You'd need a slightly more clever autopilot to get it perfect. It needs to be able to notice the rate of plane change, use that to predict the rate of drift, and match that perfectly, then increase the rate of rotation slightly until your craft is perfectly lined up on anti-normal, then match the rate of drift again perfectly.

From describing that process and having some programming background, I can tell you there's a reason the autopilot doesn't do that - it's not exactly going to be simple to code! Especially since it also has to account for the rate of change, changing - as your vessel's mass changes, or as you fiddle with the throttle setting.

 

[csanders]

It's funny. One thing I don't like about some things in orbiter is all the precision. The sensors can know your location, speed and attitude far more accurately than any real sensor. Real sensors have a margin of error. Most of Orbiter's margin of error boils down to floating point precision.

I like that it's not "perfect."