[PART, 1.0.2] Anatid Robotics / MuMech - MechJeb - Autopilot - Historical thread

[Hedd]

So I take it MJ finally needs to tackle the issues with the variable thrust now that even stock engines are variable at sea level?

[jztemple]

Sorry, I did do a search but couldn't find the answer, what is the meaning of "Suicide Burn" in the Warp Helper?

[Sarxis]

That's what the TV limit does, though it's rather harsh and overshooting up and down on the throttle when hitting TV speed. Sarbian has been getting that sorted, which is a very good thing. I'm sure I'll get to like the new aero, especially after I CFG up some better performing clones of the ultra-nerfed KR-2L, Poodle and LV-909.

It may have been some kind of glitch honestly. As the culprit rocket was accelerating through the atmo, the TV started getting really low, almost down to 0m/s. I re-flew the rocket today and it was fine (TV rose steadily higher as I gained altitude).

[fairytalefox]

Sorry, I did do a search but couldn't find the answer, what is the meaning of "Suicide Burn" in the Warp Helper?

I guess it's when you have to start a suicide burn in order to not crash. It makes sense in case you are on a collision course with a celestial body, otherwise it should do nothing (or do dumb things if coded poorly).

I may be wrong, I'm not an expert.

[TinfoilChef]

I guess it's when you have to start a suicide burn in order to not crash. It makes sense in case you are on a collision course with a celestial body, otherwise it should do nothing (or do dumb things if coded poorly).

I may be wrong, I'm not an expert.

A "suicide burn" is the practice of delaying your braking burn for a landing until the last possible second and then firing at full thrust. If it's timed right you'll come to a full stop *just* as you touch down. If it's timed wrong there will be explosions and dead pilots.

[Fyrem]

it is also worth noting that a "suicide burn" is the most fuel efficient method.

[Gruhl]

Sorry, I did do a search but couldn't find the answer, what is the meaning of "Suicide Burn" in the Warp Helper?

A suicide burn is a full power burn that brings your velocity down to 0 exactly when you hit the ground. I guess it's called suicide burn because any mistake in the timing of it is likely to lead to the usual consequences for the poor kerbals involved. I don't think this is the right thread to ask about common terms in ksp, but as the answer was short, and slightly concerning MJ?, I gave it anyway.

[VaporTrail]

If it's timed wrong there will be explosions and dead pilots.

Only if you start late. Starting a suicide burn early doesn't have too much in the way of negative consequences... just being less fuel efficient.

Unless of course, you only *had* enough fuel for the suicide burn. Then, yes, explosions, screaming, black and white footage of the event being played for more than seventy years...

Reminds me, the current version of Mechjeb isn't fully compatible with 1.0.2 yet, right? It works... but the control isn't all it could/should be.

Edited May 6, 2015 by VaporTrail

[futrtrubl]

Only if you start late. Starting a suicide burn early doesn't have too much in the way of negative consequences... just being less fuel efficient.

Unless of course, you only *had* enough fuel for the suicide burn. Then, yes, explosions, screaming, black and white footage of the event being played for more than seventy years...

Or if your engine only has so many ignitions, or if your target is a small barge in the ocean and your engines can't throttle down...

[pargentum]

How high an angle of attack are you doing at the start of your gravity turn?

It can happen even with very small angles of attack, if the ship already was at or near the terminal velocity.

I had this on autopilot ascend with AoA limit 5, and also on manual ascends.

[Honeybadga]

How do I delete a target from the landing guidance module?

Edit: Never mind, I simply targeted something else.

[Coneshot]

Just want to say a big THANK YOU! to Sarbian for all the hard work and all the hours you have put into getting this thing running right!

Coneshot

[sarbian]

Sarbian, if I'm understanding correctly the landing AP should now be working for airless bodies but still has issues with ones with an atmosphere. Is that correct?

Yes, that should be the case.

Hm, any idea why the trajectory correctly started at the atmosphere for Duna and not Kerbin as I've observed before? I haven't tested out Kerbin, Eve, or Laythe yet with the lander. Going to actually launch it to the destinations though instead of hyperedit.

My previous replies was too simplistic. The trajectory starts at the point where the sim actualy starts. The start point change with 4 different conditions. One of those and the most common is hitting the atmo, the other common one is the simulaiton thinking that the ship should start to use the engine to slow before hitting the ground. To know that we used a method that gave us an aproximation of velocity lost to drag while travelling a specific distance in the atmo. That function was "easy" in .90 with the mass related drag and simple alt/atmo relation.

Duna having a different atmo and gravity the sim does not start at the same point.

I think you left in the first 'if' block section that should also have been removed. (in dev #444)

Can you PM me those whole fixed method. I don't want to spent time on that.

So I take it MJ finally needs to tackle the issues with the variable thrust now that even stock engines are variable at sea level?

In what context ? all that was in the new MJ 1H avefter 1.O got out.

[smjjames]

Okay sarbian.

What about the deltaV discrepancy that I reported? You did ask to report anything with the deltaV window and they were the same numbers under 442. Plus I did ask whether the discrepancy was okay or not.

Edited May 6, 2015 by smjjames

[flyboy67109]

Hey Sarbian,

Thanks for working so hard on this mod. You're doing a fantastic job and really appreciate everything you're doing. I downloaded the latest dev build until you get the 1.0.2 compatible version available and it's working fine so far. What issues should I be watching out for until you get the official version out? I've created a backup save file in case things go south.

[daawgees]

Hey Sarbian,

Thanks for working so hard on this mod. You're doing a fantastic job and really appreciate everything you're doing. I downloaded the latest dev build until you get the 1.0.2 compatible version available and it's working fine so far. What issues should I be watching out for until you get the official version out? I've created a backup save file in case things go south.

Might be easier and faster if you just read a page or two back. I can promise you that in 5 minutes you'll find mention of some of the current issues/bugs.

Edited May 6, 2015 by daawgees
Wording

[mcdjfp]

I have been using MechJeb ever sense I started with KSP. I like both the numbers and the ability to automate things that while fun the first ten times, get tedious after a point. I especially liked the advanced transfer tab of the maneuver planner.

I do have one thing that I would like to report that I didn't see in a thread search. It is not really a bug, and it might be desired behavior. Unless the MechJeb case is mounted very near the centerline its drag will cause the plane (I have not checked rockets yet) to eventually spin out of control. Even placing the case within a cargo bay did not help, it needed to be very near the centerline or the plane would eventually lose control. I normally mount it near the centerline near the cockpit, but I wonder if it is contributing to rockets tumbling after the gravity turn.

[daawgees]

I have been using MechJeb ever sense I started with KSP. I like both the numbers and the ability to automate things that while fun the first ten times, get tedious after a point. I especially liked the advanced transfer tab of the maneuver planner.

I do have one thing that I would like to report that I didn't see in a thread search. It is not really a bug, and it might be desired behavior. Unless the MechJeb case is mounted very near the centerline its drag will cause the plane (I have not checked rockets yet) to eventually spin out of control. Even placing the case within a cargo bay did not help, it needed to be very near the centerline or the plane would eventually lose control. I normally mount it near the centerline near the cockpit, but I wonder if it is contributing to rockets tumbling after the gravity turn.

Right-click the command pod of your craft and select, "Control from here."

[smjjames]

The MechJeb case is supposed to be weightless and dragless, but the new aero treats weightless parts differently, however, that's really all I know. No idea how the new aero would treat dragless parts.

I'm using a config that applies MJ to all command parts, so I can't give any observation on the drag since I don't use the case.

Try what daawgees suggested though.

[mcdjfp]

I have watched MechJeb fly craft nose first into the ground enough times that the first thing I check is the control from here. The plane flies fine at first, then as the speed increases the craft begins trying to rotate toward where the MechJeb unit is placed. Eventually the drag force (at least that is what it looks like to me) becomes so great that it overcomes all available control input (anywhere from Mach 1 to when the flames begin to appear depending on placement and craft design. In one of my tests the unit was facing backwards, and another time it was upside down, so I really doubt it would have flown at all if the MechJeb case was the control point.

To clarify by after gravity turn I mean at the point that MechJeb reports between Mach 0.95 and 1.05. Any rocket that survives that might fight, but it stays in control.

I wish to add that many of my rockets have been pulling slightly towards the south, which is where the MechJeb case tends to live on my rockets. (Use to deorbit top stage along with SAS of some type).

[BARCLONE]

Sarbian,

LGAP: I'm seeing a problem in build 444, where MJ is spamming the logs. The effect of spamming on my machine is to drag the whole machine down to a point where it cannot respond to other requests for CPU time. There is a long period of HDD access, sometimes lasting for many minutes. The point in the game where this occurs is when some ship/probe/lander touches down on the surface of any other planet or moon.

I've seen similar behavior throughout the 0.90 lifetime as well, but could not immediately connect it to MJ.

Here's a link to my most-recent log files. The "NoMJ" files are with the exact same mod list, but without MJ: FOLDER LINKY

Linux Mint 17.1, 64-bit, using the 64-bit executable of KSP.

As for the landing approach itself, the landing sequence appears nominal on non-atmosphere bodies like Mun and Minmus. Landings on Kerbin are routinely non-nominal...

[Inside Outside]

Quick question: how can I use Mechjeb to get a satellite into a particular orbit for a contract? I've been using the AN/DN to line up my inclination during ascent, but I don't know how to line them up exactly beyond eyeballing when to launch and hoping I get lucky. You can't seem to set the designated orbits as a target and inclination changes in orbit use up far too much dV, so any suggestions?

[smjjames]

Heres my testing for Laythe, which was fail all over the place.

Craft file: http://sta.sh/0cwk610imhq

Save+output log (which got really big): http://sta.sh/0o53f8tpbi2

Let me know if you guys want the screenshots in spoilers.

Attempt1:

It decided that the landing point would be a lot closer than it actually is. If it was going to make a low orbit type burn, this would actually make sense.

Uh, MechJeb, shouldn't you be BRAKING right now and not speeding up??

...

*aborts landing AP*

Attempt2:

Near exact replay of attempt 1

Attempt 3:

I decided to put a landing spot exactly on the equator to see what it would do.

It initially looked like it was going to do the exact same thing as before, but then I decided to let it keep going.

It somehow got stuck in this orbit, wanting to do a course correction but never actually doing it.

Reverted and tried again, it did differently.

But ended up 270km+ offtarget still :/

Attempt 4:

Tried to do a polar landing site, forgetting that I needed a whole lot more deltaV, nothing to see here.

Attempt 5:

Tried to set one almost right under the ships location at that moment to see what it would do.

Eh? Why are you thinking I'm exiting the atmosphere when I'm already outside of the atmosphere?

It got stuck in this orbit again, I'm manually warping to see if it would do anything, it didn't, so I reverted.

Attempt 6:

Same starting conditions as 5. It started doing it's thing and then started being confused about atmosphere.

Uh, MJ, are you mixing the Apoapsis with atmospheric exit? Certainly looks that way.

Aerobrake node seems to be working, except it's on the other side of Laythe....

Um, why are you doing an inclination change?

MJ?.... (At this point, I decided to hyperedit the fuel tank back to full to see what it was trying to do, it just continued on northwards with the orbit leaving laythe.)

Attempt 7:

I had it just land somewhere, which went fine, although it expended way more fuel than neccesary, IMO.....

Edited May 6, 2015 by smjjames

[billkerbinsky]

So just wanted to say that the MJ dev build I installed on KSP 1.0.x seems to be doing a much better job in the docking autopilot than the older version of MJ on 0.90; docking seems to go smoothly with much less back-and-forth oscillation, less RCS fuel consumption, and generally less fussiness. Thanks!

[sarbian]

New dev build. I think I fixed the atmo landing this time. It landed me 6km of my target which I consider not too bad for now.

- correct version of Aldmehr change

- fixed landing for Atmo

- warp for land anywhere

The remaining quirks are :

- the landing AP not using the low atmo values for thurst for some estimation

- some part of the landing AP may think we are landing at sea level (need to check that one)

- no parachute sim yet

Since the parachute sim should not be too long I think I ll finish that before releasing a new official.