[1.12.x] Anatid Robotics / MuMech - MechJeb - Autopilot - [2.14.3] [4th March 2023]

[sarbian]

7 hours ago, maculator said:

The Link on the CurseForge page points to nowhere.

Fixed but it will take a while to actually show up

[maculator]

Since I commented here I found it anyways

I however got a question:

To me it seems that MJ takes the window possition of the first flight with it installed on the vessel as the default windows setup.

Last time, months ago, I always had my three "HUDs" active and a window I didn't want whenever I launche a vessel. This time I got 0 windows as "default" and have to open my three HUDs manually each time I launch.

Is there something I'm missing?

It would be pretty usefull to be able to flag windows as "default" so theyre always open when you launch.

Thanks.

[Wingnutt]

Has anybody had any luck with the rover autopilot?

So far mine seems to ignore way points and will just go in the direction of whatever the heading is set to.

[aluc24]

@sarbian , after the last big update (2.7.0), I can't seem to get the Maneuver Planner - Advanced Transfer to Another Planet to work with targeted asteroids. The porkchop plots, the limited time option also seems to be working, but when I press "create node", it says "Ejection Optimization failed; try manual selection". Not sure what "manual selection" means here. Never had this problem with the previous Mechjeb version.

Not including log, because I don't think it would be relevant in this case... Savegame is pretty heavily modded. Somehow I suspect you would be able to re-create it in vanilla too. But if you need anything from me, let me know.

[sarbian]

8 hours ago, aluc24 said:

@sarbian , after the last big update (2.7.0), I can't seem to get the Maneuver Planner - Advanced Transfer to Another Planet to work with targeted asteroids. The porkchop plots, the limited time option also seems to be working, but when I press "create node", it says "Ejection Optimization failed; try manual selection". Not sure what "manual selection" means here. Never had this problem with the previous Mechjeb version.

Not including log, because I don't think it would be relevant in this case... Savegame is pretty heavily modded. Somehow I suspect you would be able to re-create it in vanilla too. But if you need anything from me, let me know.

Manual selection means you have to click on the porkchop to select a solution because MJ failed to find a good one by itself. I can't promise I can do much here because that part of the code is not my forte.

16 hours ago, Wingnutt said:

Has anybody had any luck with the rover autopilot?

So far mine seems to ignore way points and will just go in the direction of whatever the heading is set to.

Select a waypoint while holding Alt to properly initialize it. I need to fix that...

[aluc24]

3 hours ago, sarbian said:

Manual selection means you have to click on the porkchop to select a solution because MJ failed to find a good one by itself. I can't promise I can do much here because that part of the code is not my forte.

Well, if I click on the porkchop, and then press "create node", I get the same error... It appears that MJ is having trouble in the final stages of producing a node.

[AlonzoTG]

finally remembered there was ADVANCED transfer to another planet, finally got my interplanetary missions on their way... Anyway, I'm trying to build a probe cluster to spam a bunch of moons with scanner probes, the problem is that they're carrying the stock blanket solar arrays. I'm like OK, I'll just be sure to not select "auto deploy solars" on the ascent control so they don't cause problems while they're packed. So I send the rocket up, the fairing deploys, the circularization happens, then all the solars deploy... I go in and manually retract the ones that are clipping through the other probes in the cluster... I then test deploy one of the probes (this was a test flight), and immediately it's solars deploy... NOOOOO!!! and much destruction ensues... They have a backup RTG so they have power, the solar is just to help out but it is absolutely critical that they remain stowed until after the probe has cleared the transfer vehicle... Mechjeb does not understand how the transfer vehicle works and only computes it's DV assuming the probes are deployed before the transfer vehicle's engines fire... I think mechjeb is thinking "Ok, this is dead weight I can ditch before firing the engine.." =\ The only other mod that I know affects these solars is warpTech+, but I can't nail down the blame for this. =\

[Witwenmacher]

Hey there, as most of the mods I usually utilize are now up-to-date I want to re-live the KSP moments with all it's failures and start a new career. However, I read that the combination of remote-tech and mech-jeb has some problems. Does anyone has some information or experience whether that limitation is still true or if it has been resolved? 

*I tried the search but nothing meaningfull came up - but I'm glad for any help

[alexilahio]

hello there ! 53 pages so I admit I havent been reading all of them !! Sorry : I was watching the FH liftoff ^^

 

Anyways, I have been using mechjeb with 1.3.1. Quite efficient with basic ships. Quite not with big ships but useful nonetheless

Yesterday I have addes SSRSS to enjoy a better game exp and better views. Now, I tried a simple orbit around earth via mechjeb ascent guidance.... it crashed my big ship and failed to circularize properly a basic vessel (KerbalX). Worse, it sent me on an interstellar course instead of the supposed moon encounter....

 

What happened ? Did SSRSS completely changed the gameplay so that mechjeb can't deal with it ? Cuz I've seen friends playing KSP with that mod and mechjeb and doing fine.

 

Any idea ?

[Cdodders]

2 questions about MechJeb (be quiet hardcore manual flyboys). Can it be set up so that Ascent Guidance remains on when switched to another vessel, in the same way Landing Guidance does? Did a BFR test launch using Kerbalised SpaceX and while judicious useage of BDAC's vessel switcher and physics range extender, was able to set the spent(ish) first stage to land at the pad, noting that when switching between vessels the LG stayed engaged and the AG didn't. Luckily Time To Land was shorter than Time to Circ. Burn.

2nd, is there a way to make MJ Landing Guidance more accurate? Said First Stage clipped the VAB itself (set the pad for landing), resulting in both RUDing

[sarbian]

1. The AG should not disengage when switching vessel if you are still in range. I have seen videos of players doing just that. BUT MJ can only stage the active vessel (KSP code limitation  ).

2. Not really currently. The current sim got better after I found a code error but aerodynamic simulation has some precision limits. A solution I am thinking about is allowing to use steeper descent profile (like a Falcon booster) since a more vertical entry is more precise.

[Cdodders]

47 minutes ago, sarbian said:

1. The AG should not disengage when switching vessel if you are still in range. I have seen videos of players doing just that. BUT MJ can only stage the active vessel (KSP code limitation  ).

2. Not really currently. The current sim got better after I found a code error but aerodynamic simulation has some precision limits. A solution I am thinking about is allowing to use steeper descent profile (like a Falcon booster) since a more vertical entry is more precise.

The AG only seems to disengage once initial burn is complete, ie before circ burn. Maybe its to do with range. Two stage rocket so staging isnt an issue

[eberkain]

At the end of the first burn on Asecent Guidance using the classic profile MJ seems to make one hard pull up putting the rocket into a flip.  I'm thinking this is just a side effect of having the Mandatory RCS mod.  Any way I can tweak that?

[The Aziz]

Technical issues here, there... I tried to search but found nothing.

I have a problem with the AR202 part itself. So far happened two times, in 1.2.2 and 1.3.1, on (as far as I can tell) stock aero.

So I had the module attached to my vessel. Returning back to Kerbin. Everything burned it the atmo, only AR202 survived, and started falling. But instead of reaching terminal velocity and crashing at decent speed, it was slowing down? At let's say 1km above the surface the speed was about 3m/s.

Is this part somehow extremely resistant for everything including air or I messed something up?

Not to mention that regular, dunno, plane crashes can't destroy it either, even if everything else blows up.

If more info is needed I'll try to provide it.

[vardicd]

2 hours ago, The Aziz said:

Technical issues here, there... I tried to search but found nothing.

I have a problem with the AR202 part itself. So far happened two times, in 1.2.2 and 1.3.1, on (as far as I can tell) stock aero.

So I had the module attached to my vessel. Returning back to Kerbin. Everything burned it the atmo, only AR202 survived, and started falling. But instead of reaching terminal velocity and crashing at decent speed, it was slowing down? At let's say 1km above the surface the speed was about 3m/s.

Is this part somehow extremely resistant for everything including air or I messed something up?

Not to mention that regular, dunno, plane crashes can't destroy it either, even if everything else blows up.

If more info is needed I'll try to provide it.

That has been a thing with that part for as long as I can remember. the MJ casing is pretty much indestructible, and is made out of an anti-physics, anti-gravity metal. Lol.  

[canisin]

38 minutes ago, vardicd said:

That has been a thing with that part for as long as I can remember. the MJ casing is pretty much indestructible, and is made out of an anti-physics, anti-gravity metal. Lol.  

Does it cause drag though?

[Loren Pechtel]

3 hours ago, The Aziz said:

So I had the module attached to my vessel. Returning back to Kerbin. Everything burned it the atmo, only AR202 survived, and started falling. But instead of reaching terminal velocity and crashing at decent speed, it was slowing down? At let's say 1km above the surface the speed was about 3m/s.
Is this part somehow extremely resistant for everything including air or I messed something up?

Simply being very light for it's size would cause the slow fall you're seeing.  A high temperature tolerance would cause the former, I've never looked at what it's temperature tolerance is.

[The Aziz]

2 hours ago, vardicd said:

That has been a thing with that part for as long as I can remember. the MJ casing is pretty much indestructible, and is made out of an anti-physics, anti-gravity metal. Lol.  

So it is. But watching it for few minutes straight at full warp until it gently drops on the ground isn't particularly interesting..

Is there a way to maybe edit the config so it acts as regular part? I mean, it's not really useful when 99% of the craft is gone

Edited February 9, 2018 by The Aziz

[Kuszotke]

I have a problem. When using Landing Guidance, MechJeb overshoots the KSC Pad by about 20km. Everything else goes well, it just lands in the sea instead of landing on the KSC Pad.

There's a deceleration burn, then it stops, i fly over the KSC and decelerate again well past the KSC.

I open landing guidance, click on KSC Pad, click SET and then Land At Target. 

Am i doing something wrong? 

LOGS

 

 

[vardicd]

11 hours ago, Kuszotke said:

I have a problem. When using Landing Guidance, MechJeb overshoots the KSC Pad by about 20km. Everything else goes well, it just lands in the sea instead of landing on the KSC Pad.

There's a deceleration burn, then it stops, i fly over the KSC and decelerate again well past the KSC.

I open landing guidance, click on KSC Pad, click SET and then Land At Target. 

Am i doing something wrong? 

LOGS

 

 

As far as I can remember, the landing guidance has never been 100% accurate on planets/moons with atmosphere, though I'm not the mod creator/maintainer so maybe it was working at some point I'm unaware of, but I know in all the years I've used MJ I have never gotten pinpoint landing with it, in an atmosphere.

17 hours ago, canisin said:

Does it cause drag though?

It did at one point. I remember I edited the config files manually, because it would generate a stupid amount of drag, and I would get some bizzare rocket/plane flip-outs
because of it.

16 hours ago, The Aziz said:

So it is. But watching it for few minutes straight at full warp until it gently drops on the ground isn't particularly interesting..

Is there a way to maybe edit the config so it acts as regular part? I mean, it's not really useful when 99% of the craft is gone

I'm sure it could be, if you knew what you were doing, or if you wanted to copy the temp/impact tolerance from some other part. I'd imagine though that with all the other work the mod author/maintainer is putting in, changing the config file to fix a relatively minor issue like this is not a high priority on his list.

EDIT: @The Aziz Unless I am sorely mistaken, the MJ part is essentially coded to be its own heat shield. I've found a line in the code "thermalMassModifier = 100" which, if I understand how it works means it can absorb and hold a lot of heat before it starts to overheat and explode, easily tanking the heat of atmospheric entry. The Squad heat shields only have a thermal mass modifier of 1. Removing this line from the config file, might, and I mean might, remove it's heat tolerance and allow it to blow up if the rest of the craft does.

EDIT2: probably will need to also delete line: "skinThermalMassModifier = 100"

EDIT3: as for making it so it doesn't slow down in the air, and will explode on ground contact.... I am clueless.

EDIT4: Nope, removing those lines from the config doesn't make much of an effect at all. the MJ unit does get hot enough you can see a thermal bar when entering from the Mun, but it cools down fast enough that it doesn't explode. I'm not sure how to do this, actually.

Edited February 10, 2018 by vardicd

[Kuszotke]

4 hours ago, vardicd said:

As far as I can remember, the landing guidance has never been 100% accurate on planets/moons with atmosphere, though I'm not the mod creator/maintainer so maybe it was working at some point I'm unaware of, but I know in all the years I've used MJ I have never gotten pinpoint landing with it, in an atmosphere.

It's not even about the atmosphere or being accurate with reentries. 

The landing could be easily corrected if only MechJeb decel burned for a bit longer. All it has to do is correct the course, rather than fly over the KSC Pad and decel burn past it.

[Norcalplanner]

On 2/9/2018 at 7:56 AM, The Aziz said:

So it is. But watching it for few minutes straight at full warp until it gently drops on the ground isn't particularly interesting..

Is there a way to maybe edit the config so it acts as regular part? I mean, it's not really useful when 99% of the craft is gone

I used to be annoyed by this as well. One solution is to add the MJ and KER for all mod, which adds partless MJ functionality to every probe core and capsule.

[The Aziz]

13 hours ago, vardicd said:

EDIT: @The Aziz Unless I am sorely mistaken, the MJ part is essentially coded to be its own heat shield. I've found a line in the code "thermalMassModifier = 100" which, if I understand how it works means it can absorb and hold a lot of heat before it starts to overheat and explode, easily tanking the heat of atmospheric entry. The Squad heat shields only have a thermal mass modifier of 1. Removing this line from the config file, might, and I mean might, remove it's heat tolerance and allow it to blow up if the rest of the craft does.

EDIT2: probably will need to also delete line: "skinThermalMassModifier = 100"

EDIT3: as for making it so it doesn't slow down in the air, and will explode on ground contact.... I am clueless.

EDIT4: Nope, removing those lines from the config doesn't make much of an effect at all. the MJ unit does get hot enough you can see a thermal bar when entering from the Mun, but it cools down fast enough that it doesn't explode. I'm not sure how to do this, actually.

I tried and I guess I succeeded, at least partially. I did not delete the lines but changed some of them to more or less same values as the smallest batteries. And it did blow up on my uncontrolled reentry from the Mun, so I guess i'm fine now.

[vardicd]

4 hours ago, The Aziz said:

I tried and I guess I succeeded, at least partially. I did not delete the lines but changed some of them to more or less same values as the smallest batteries. And it did blow up on my uncontrolled reentry from the Mun, so I guess i'm fine now.

Can you please share the specific changes you made? my own efforts have not yielded results. would appreciate it.

14 hours ago, Kuszotke said:

It's not even about the atmosphere or being accurate with reentries. 

The landing could be easily corrected if only MechJeb decel burned for a bit longer. All it has to do is correct the course, rather than fly over the KSC Pad and decel burn past it.

true, but while a change like that would be simple for you or me, programming that into the MJ module is probably a fairly complex operation, and not a 5 minute fix.

[drhay53]

On 2/4/2018 at 11:58 PM, sarbian said:

Select a waypoint while holding Alt to properly initialize it. I need to fix that...

I confirm that holding alt while clicking the 'drive' button fixes the rover autopilot issue of not setting and following headings. This has been bugging me for like 6 months, and I even made a github issue about it