Problem LOLA 2010 MFD Abort on Landing

[joertexas]

The LOLA 2010 version autopilot seems to go into an abort mode about 30 seconds before touchdown. I desperately need either a fix, a workaround, or some other MFD to use. I am using a custom designed base near the lunar north pole, with a custom designed lunar lander.

Thanks,

JR

 

[Wishbone]

North Pole? Have you tried it with the same lander on another base (Brighton Beach, for instance)? The state machine in LOLA is quite involved, and may not like the latitude...

 

[joertexas]

North Pole? Have you tried it with the same lander on another base (Brighton Beach, for instance)? The state machine in LOLA is quite involved, and may not like the latitude...

That was a good thought, but the lander did the same thing with Brighton Beach. I did move the base to -55W so I wouldn't have to adjust the lander's orbit. It still gets to 500m, and initiates an abort. After about 30 seconds, Orbiter freezes.

JR

 

[Wishbone]

Thx for the test... During the freeze, does the internal view (glass cockpit) display funny values for altitude and/or speed? If so I suspect division by zero or square root of a negative number teleport you into NaNspace

 

[joertexas]

Thx for the test... During the freeze, does the internal view (glass cockpit) display funny values for altitude and/or speed? If so I suspect division by zero or square root of a negative number teleport you into NaNspace

The screen freezes with heavy lines in several places, and no numbers are readable. The lander always comes to a near hover at about 1.5 to 2km from the pad at near zero velocity and 500m altitude, and then it starts an abort.

I'm wondering if it has to do with my lander. This is the config section of the INI file for my lander:

[CONFIG]
MeshName = "SpaceX\Moon_Shot_Lander_Only"
Size = 5.5
EMPTY_Mass = 713.4
Fuel_MASS = 1086
Main_Thrust = 5338
Retro_Thrust = 40
Attitude_Thrust = 40
Isp = 3300
PMI = (.9,1.14,.93)
CROSS_SECTION = (5.67,7.6,6.86)
COG = 1.5
PITCH_MOMENT_SCALE = 0.00005
BANK_MOMENT_SCALE = 0.00005
LAND_PT1=(1.5,-1.5,-2)
LAND_PT2=(-1.5,0,-2)
LAND_PT3=(1.5,1.5,-2)
FOCUS=1
VISIBLE = 1
CAMERA = (0,1.65,1)
MAIN_PSTREAM1=engine
ATT_PSTREAM1=rcs
MAIN_TEX=MMH_NT_Ex
RETRO_TEX=MMH_NT_Ex
ATT_TEX=MMH_NT_Ex

JR

 

[Wishbone]

Was about to enquire about this very approach (testing a stock DG)... but was distracted by the news about CTD in my MFD...

 

[joertexas]

Was about to enquire about this very approach (testing a stock DG)... but was distracted by the news about CTD in my MFD...

I was thinking of modifying the lander to have hover engines, instead of being a tail sitter. It's distracting to have the view looking straight up on landing, anyway...

JR

 

[C3PO]

Why would looking backwards be better then looking straight up? During the braking phase you will mostly be looking at the instruments anyway. And the final part of landing is is done at a much lower pitch. In a tailsitter you will be looking straight up during the most scenic part of the descent.

 

[joertexas]

Why would looking backwards be better then looking straight up? During the braking phase you will mostly be looking at the instruments anyway. And the final part of landing is is done at a much lower pitch. In a tailsitter you will be looking straight up during the most scenic part of the descent.

You're right that braking is done on instruments. But, as a pilot, I find it much easier to land if I can see the ground. The problem is, while looking straight up, it's nearly impossible for me to judge drift and maintain lateral control.

I think I will try reorienting the main engines to be designated as hover engines. I don't know if it will make a difference to the LOLA program, but it will allow me to use altitude hold, and that may make a difference.

JR

 

[Mandella]

Joertexas, as far as I can tell the current version of LOLA MFD is broken for many of us. The only thing I can get it to do reliably is suborbital transfers, interestingly enough.

What I use is a combination of Basesync and [ame="http://www.orbithangar.com/searchid.php?ID=3057"]AutoHover[/ame] to go from orbit to pad, supplemented with a fair knowledge of orbital mechanics. True, it's not quite a "push a button and land," but it does get you down.

It might also work to use LOLA to get you *almost* there, then shut it off and use AutoHover mixed with seat of the pants flying to get you down, but for me LOLA's tendency to suddenly fire main engines at odd tangents to the correct vector just made me, regretfully, remove it from my flight control toolset altogether.

 

[joertexas]

After some work, and some pointers from the group here, I found the problem:

Put simply, LOLA doesn't like tail sitters. After I changed the lander's orientation to make the descent engines hover engines, it works perfectly. Thanks to everyone for their input


JR

 

[joertexas]

LOLA Bug Report

I have designated a lunar base at 89.55N 30W. When I make my approach from Farside and pass close to the pole, the lander suddenly goes out of control. Even if I switch to manual control, I can't recover the vehicle. If I make the approach over Nearside, it works perfectly.

JR

---------- Post added at 11:42 PM ---------- Previous post was at 11:39 PM ----------

It might also work to use LOLA to get you *almost* there, then shut it off and use AutoHover mixed with seat of the pants flying to get you down, but for me LOLA's tendency to suddenly fire main engines at odd tangents to the correct vector just made me, regretfully, remove it from my flight control toolset altogether.

I've found that it actually works most of the time. The problems I was having, except for the part about it going bats as it passes the pole, turned out to be from other causes.

JR

 

[Mandella]

I've found that it actually works most of the time. The problems I was having, except for the part about it going bats as it passes the pole, turned out to be from other causes.

JR

Interesting. Maybe my problem is that I am using it (or was using it) with a spacecraft3 modeled craft with no main thrusters, only hover thrust available (one of Greg Burch's LSTS's). Those things can be a bit tricky to fly for autopilot or man.

 

[joertexas]

Interesting. Maybe my problem is that I am using it (or was using it) with a spacecraft3 modeled craft with no main thrusters, only hover thrust available (one of Greg Burch's LSTS's). Those things can be a bit tricky to fly for autopilot or man.

For me, LOLA wasn't working with a tail-sitting configuration. When I changed the main thrusters to a hover configuration (thrust along the -Y axis), it worked fine - and still does. I did set up main thrusters anyway, using the same thrust as the RCS units.

JR

 

[Rossco]

The LOLA 2010 version autopilot seems to go into an abort mode about 30 seconds before touchdown. I desperately need either a fix, a workaround, or some other MFD to use. I am using a custom designed base near the lunar north pole, with a custom designed lunar lander.

Thanks,

JR

I think i have a fix for Lola. I have to do a bit more testing and code cleanup and figure out the project dependency dll's but so far I can get my 25 ton tail sitter down on the moon ok, whereas with the current Lola 2010 version it does the "abort" trick straight after entering approach phase. Stay tuned for the posting....

 

[joertexas]

I think i have a fix for Lola. I have to do a bit more testing and code cleanup and figure out the project dependency dll's but so far I can get my 25 ton tail sitter down on the moon ok, whereas with the current Lola 2010 version it does the "abort" trick straight after entering approach phase. Stay tuned for the posting....

The ability to land a tail sitter would solve some other problems for me, too

JR