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

[smjjames]

Use the last dev release

smjjames can you please stop doing support in the thread ? You re making it worse.

The bug reporting or trying to help do support?

If the trying to help do support, okay I'll ease off on that.

-snip-

In the screenshot I'm at minmus, not the mun, and by worked fine, I mean it did the usual high eccentricity (and high inclination if applicable) return orbit, but from Minmus, it decided to send me into a heliocentric orbit.

It's not a massive deal though. It was probably just the particulars of that specific orbit at that specific time.

Anyway, Sarbian, do you know what's going on with MJ bleeding data into the output log?

Edited May 4, 2015 by smjjames

[KingPhantom]

same rocket. just updated mechjeb2 to dev version. clean orbit and plenty of fuel left. thx!

[diegzumillo]

Hey guys

I'm using KSP 1.02 (so I understand it may not be fully working yet) and mechjeb doesn't seem to be able to calculate delta-V for nuclear engines. Also, there's no vessel info anymore, but I guess this is a different issue, and I can work around it by creating a window myself.

I made a search before doing this post but didn't find anything, so I apologize if it has been answered already.

[moeburn]

Hey guys

I'm using KSP 1.02 (so I understand it may not be fully working yet) and mechjeb doesn't seem to be able to calculate delta-V for nuclear engines. Also, there's no vessel info anymore, but I guess this is a different issue, and I can work around it by creating a window myself.

I made a search before doing this post but didn't find anything, so I apologize if it has been answered already.

You using 2.5 or latest dev build?

http://jenkins.mumech.com/job/MechJeb2/

[porrohman]

If you don't connect the outer lines, the separator holding the rockets in place won't crosfeed fuel, therefore , in order for the fuel to pass from each stage of the asparagus onto the next the fuel lines are needed, and at the final stage, they need to merge onto the main fuel tank.

It's a pretty straightforward asparagus setup and I've been flying it for years without problems (and with the correct dV calculated).

It's not a complicated design at all. I don't know where you get this idea from.

It's a pretty simple asparagus staging. The whole point is to have those fuel lines there.

I've never run into a place where asparagus staging was necessary, but this is not the place for that discussion.

The newest dev build is pretty tight, so upgrade to that.

[jebodiah1976]

@jebodiah1976:could you upload that craft file for someone to have a look at it?

Here is the craft file (dropbox).

I've never run into a place where asparagus staging was necessary, but this is not the place for that discussion.

The newest dev build is pretty tight, so upgrade to that.

I'm sorry but you don't understand what asparagus staging is. Yes, I was aware that I can upgrade to the latest build. Thanks I guess, but basicly what you are doing is "hey my car won't start this morning" and you reply "well ... you are overcomplicating things, I never found a situation where I ever needed a car since I ride a bike".

[diegzumillo]

You using 2.5 or latest dev build?

http://jenkins.mumech.com/job/MechJeb2/

Latest dev build

[Aldmehr]

Two usage questions (dev#442):

1> During atmospheric landings (like Kerbin), is there a way to use the mechjeb landing prediction to calculate a location based on the state of the ship after the next staging?

For example, when the deorbit stage (engine and fuel) is decoupled from the landing stage, this will often dramatically (accurately) alter my landing prediction probably due to the differences in atmo behavior (drag?) of the lesser ship (but now it is too late to correct since the engine is gone).

2> Does the mechjeb maneuver planner adjust the burn start time accounting for the occurrence of an inevitable staging event during the burn?

For example, if a staging event occurs in the middle of a circularization burn, it seems that the resulting orbit often differs from the intended orbit, particularly if the acceleration of the before and after stages is very different.

How do you all handle these scenarios?

I love this mod, thanks for putting in the time to keep it up and running and improving.

Edited May 5, 2015 by Aldmehr

[daver4470]

Hey guys (past and present).... just wanted to do my periodic pop-in to once again thank you all for MechJeb. It's the one thing I couldn't KSP-live without. You're all gods among men (and women). GODS I SAY!!!!

[acerola]

2> Does the mechjeb maneuver planner adjust the burn start time accounting for the occurrence of an inevitable staging event during the burn?

For example, if a staging event occurs in the middle of a circularization burn, it seems that the resulting orbit often differs from the intended orbit, particularly if the acceleration of the before and after stages is very different.

Mechjeb's works fine even when staging occurs while executing the node. But it fails sometimes if your ship is wobbling a lot. Use struts.

- - - Updated - - -

Is there a way to shift the time of the maneuver node so that it lands in the same spot, but in the next loop of the orbit? There is a button in stock KSP that does that. Can it be done with Mechjeb?

[SemMaxim]

How to disable all dependencies from buildings levels and tech tree? I want that all functions be available from beginning.

[sarbian]

Two usage questions (dev#442):

1> During atmospheric landings (like Kerbin), is there a way to use the mechjeb landing prediction to calculate a location based on the state of the ship after the next staging?

For example, when the deorbit stage (engine and fuel) is decoupled from the landing stage, this will often dramatically (accurately) alter my landing prediction probably due to the differences in atmo behavior (drag?) of the lesser ship (but now it is too late to correct since the engine is gone).

I already thought a bit about that and it may be doable with stock with the ideas I had. But first I need to fix the current landing bugs.

2> Does the mechjeb maneuver planner adjust the burn start time accounting for the occurrence of an inevitable staging event during the burn?

For example, if a staging event occurs in the middle of a circularization burn, it seems that the resulting orbit often differs from the intended orbit, particularly if the acceleration of the before and after stages is very different.

It does if auto-staging is on. MJ uses the deltaV stats to know how much thrust and how long each stages last. If it need to stage it adds the staging time set in the auto-staging config. It times the burn so half the dv is spent before reaching the node, so you if you have staging with different engines you can have the burn start earlier or later than half the burn time.

The math itself could use some love (Someone provided a better solution) but it is good enough for now.

Anyway, Sarbian, do you know what's going on with MJ bleeding data into the output log?

Yeah, I m trying to fix bugs and I ask for logs. So I print usefull stuff in the logs...

diegzumillo : strange, it should work fine with the last dev...

And thanks to those who sent craft. It did not help me to figure the bug yet but it seems my test craft was somehow magic...

What I found out is that with some of those craft the simulation gave widly different result in manual mode and in automated landing. Which makes no sense since it is called the same way. It was getting too late for me to check more but I guess the initial parameters are somehow different.

What I will most likely do is release a new official build with a warning about the landing AP. That way users who don't follow the thread will get a version without the big bugs I already fixed.

Edited May 5, 2015 by sarbian

[pargentum]

In latest dev builds "Limit to terminal velocity" is mostly fixed. At least, it does not try to limit to some funny speeds like 95m/s. But one annoying behavior remains.

When I start the gravity turn, drag of the vehicle increases, so MJ turns the engine off. This is understandable, but on vehicle with small lift it means that I will turn much slower than possible, and the speed drops considerably.

I understand that it is hard to decide what to count as a "terminal velocity" at high angles of attack. May be you should implement a feature in Accent Guidance to temporary disable "Limit to TV" on early stages of gravity turn?

[78stonewobble]

Well, I don't know how applicable this is, in regards to tweaking the landings and with the new atmosphere model.

Currently trying to get mechjeb to do a pinpoint canyons landing on the mun and having problems with it either overshooting, undershooting or just not allowing enough time to brake (smashing into the ground at between 50-110 m/s). It will land fine sometimes, just not necessarily in the right position. From different altitudes. 100 km, 25 km and 15 km.

Tried 2.5 and the dev build 442.

But it's been quite a while since I played ksp and I might have designed a crappy ship. It has a rather crappy thrust to weight ratio (1.11) and im relearning both kerbal and mechjeb again so...

Anyways, I just meant this as possibly helpfull feedback and not a complaint... I love mechjeb

[pargentum]

I mean, a feature that calculates optimal trajectory for a rocket standing on the ground to reach a desired point on the surface of the same body. Would be very cool for:

1. Missions to "measure a temperature flying above 18000m above ..."

2. Make a lander to jump to other points of the moon instead of using rovers.

Ideally it could look like landing gudance UI: you just select a desired point on a map or enter coordinates, and whoosh! May be there can be options to minimize dV or to have apoapsis no lower than X km.

Effectively it is a combination of accent guidance and landing guidance. At least, as I understand, it could massively reuse algorithms from both. For example, landing guidance already can calculate a landing point with consideration for planet rotation, and accent in a right direction would be the same gravity turn, you just have to end it much earlier than during orbital accent.

[bdragon9222]

I mean, a feature that calculates optimal trajectory for a rocket standing on the ground to reach a desired point on the surface of the same body. Would be very cool for:

1. Missions to "measure a temperature flying above 18000m above ..."

2. Make a lander to jump to other points of the moon instead of using rovers.

Ideally it could look like landing gudance UI: you just select a desired point on a map or enter coordinates, and whoosh! May be there can be options to minimize dV or to have apoapsis no lower than X km.

Effectively it is a combination of accent guidance and landing guidance. At least, as I understand, it could massively reuse algorithms from both. For example, landing guidance already can calculate a landing point with consideration for planet rotation, and accent in a right direction would be the same gravity turn, you just have to end it much earlier than during orbital accent.

I like that idea.

[ss8913]

so it looks like the bad-framerate-using-landing-predictions thing I'm running into is an incompatibility between the dev build of nuFAR I'm using, and MJ2. One of the folks on the FAR thread said, "It NaN's closest approach distance (which tanks FPS as a result)" - makes sense, and it also kind of makes sense that MJ devs wouldn't chase this until nuFAR is released, but just in case it was *completely* off the current radar/priorities list, well, now you know. This seems to be the only feature-killing incompatibility between the 2 mods that I've found so far, though - everything else still works pretty much like it did in 0.90.

[smjjames]

Well, I don't know how applicable this is, in regards to tweaking the landings and with the new atmosphere model.

Currently trying to get mechjeb to do a pinpoint canyons landing on the mun and having problems with it either overshooting, undershooting or just not allowing enough time to brake (smashing into the ground at between 50-110 m/s). It will land fine sometimes, just not necessarily in the right position. From different altitudes. 100 km, 25 km and 15 km.

Tried 2.5 and the dev build 442.

But it's been quite a while since I played ksp and I might have designed a crappy ship. It has a rather crappy thrust to weight ratio (1.11) and im relearning both kerbal and mechjeb again so...

Anyways, I just meant this as possibly helpfull feedback and not a complaint... I love mechjeb

Yea, sounds like you're using a lower thrust engine (that or you have a really heavy lander) and don't have enough thrust for MJ to do what it needs to do. I don't know whether you're playing stock or with part mods, but I would suggest something with a max thrust of 120 or more, or just stick on more engines. Just trying to help a bit.

Yeah, I m trying to fix bugs and I ask for logs. So I print usefull stuff in the logs...

Oh, I didn't know whether it was intentional or not and ok.

And thanks to those who sent craft. It did not help me to figure the bug yet but it seems my test craft was somehow magic...

What I found out is that with some of those craft the simulation gave widly different result in manual mode and in automated landing. Which makes no sense since it is called the same way. It was getting too late for me to check more but I guess the initial parameters are somehow different.

Welcome. As for the wildly different results, most of the ships are modded to an extent, however, that shouldn't affect anything. I think I'll try with a pure stock ship (no tweakableeverything) to remove all possible variables coming from part or part tweaking mods (still going to use stockbugfixes though) and use hyperedit to place the ship in a precise orbit.

[terahurts]

I think I'll try with a pure stock ship (no tweakableeverything) to remove all possible variables coming from part or part tweaking mods (still going to use stockbugfixes though) and use hyperedit to place the ship in a precise orbit.

That's pretty much what I did. Stock ship, and HE'd to a 75km orbit (My previous post & save). It's almost as if MJ is getting wildly different results each time it runs the landing calculations, especially when using 'Land at Target' and also that it's trying to perform a non-aerobraked landing as even a 'Land Somewhere' landing will cause a throttle up.

[Ethanadams]

Mechjeb is incompatible with current version of ksp?

Why?

[daawgees]

Mechjeb is incompatible with current version of ksp?

Why?

Read the OP; in addition to the release version of MJ there are also dev builds that are released. If the main version is not working for you then try one of the dev builds.

[smjjames]

That's pretty much what I did. Stock ship, and HE'd to a 75km orbit (My previous post & save). It's almost as if MJ is getting wildly different results each time it runs the landing calculations, especially when using 'Land at Target' and also that it's trying to perform a non-aerobraked landing as even a 'Land Somewhere' landing will cause a throttle up.

I agree with the observation that it seems to be trying to do a non-aerobraked landing sometimes and it seems to be confused about it's orbital trajectory and the trajectory it shows. Would be cool if the trajectories mod worked for 1.0.2 so that I can get a second reference point for trajectory so that I can get insight into what MJ is doing with it's trajectory.

Edit: Actually, sounds like they're having a hard time getting it working.

I'll do saves around a bunch of different bodies, give sarbian some data points from places other than Kerbin, although the pre-set KSP pad coordinates does allow for 100% repeatable coordinates without having to write down the coordinates.

Sarbian, small idea here, could you put an indicator onto the trajectory at the point where you enter/exit atmosphere? It seems like the trajectory line starts at the atmosphere where it initially makes the orbital trajectory greyish (or purpleish I think) until it diverges from it, but not sure if that's actually where it starts.

Edited May 5, 2015 by smjjames

[sarbian]

The trajectory is drawn from the point of atmo entry. The "World trajectory" option draw the trajectory in a fixed frame if enabled and rotating frame if disabled (I have to rename that option).

[smjjames]

The trajectory is drawn from the point of atmo entry. The "World trajectory" option draw the trajectory in a fixed frame if enabled and rotating frame if disabled (I have to rename that option).

Oh I see, thanks for confirming that.

hm *wonders how the trajectory would look with a more spaceplane type craft*

Edit: Actually, those that make spaceplanes, did the landing AP work for craft that are meant to reenter like the shuttle does back in 0.90? I know it's a different aero, but just wondering if it works for spaceplanes as I haven't made one before.

Edited May 5, 2015 by smjjames

[78stonewobble]

Yea, sounds like you're using a lower thrust engine (that or you have a really heavy lander) and don't have enough thrust for MJ to do what it needs to do. I don't know whether you're playing stock or with part mods, but I would suggest something with a max thrust of 120 or more, or just stick on more engines. Just trying to help a bit.

Heavy and 1 small engine... but mechjeb did manage to land it quite good on the last attempt and I even managed to pick a spot that was flat enough, but I'll keep a better TWR in mind for my next generation of lander ...

In any case it doesn't get any more kerbal than adding moar/more engines.