[1.9-1.10] Throttle Controlled Avionics

[allista]

Bug with the set point text window in v18, if you delete all of the text, it goes and puts the current set point back in there. It might not be a bug, but it is somewhat annoying.

Ouch! I was actually trying to fix another bug this way

[allista]

Here's the

next BETA version.

• Fixed radar issues caused by invisible colliders (like triggers on docking ports, etc.)
  
• Fixed altitude editing problem introduced in the previous version.
  
• Added height threshold for auto-retraction of landing legs as well.
  
• Added engine/group profile name editing.
  
• Greatly improved HSC control over manual engines for translation.
  
• Improved near-target navigation: crafts should now slow down more accurately as to not overshoot, especially in low gravity.
  
• Improved altitude control slightly to prevent vertical overshooting due to obstacle avoidance.
  
• Reworked debug lines to be GL.Lines, not triangles.
  
• Edit: also fixed the bug that caused followers to not gather around an idly staying leader.
  

Note: this is a debug version which will show CPS and AutoLander state vectors.

Edited October 3, 2015 by allista

[allista]

Here's the next

BETA version

and BETA debug version

• Fixed horizontal overshooting of followers in low gravity.
  
• Improved altitude control in low gravity; still overshooting, but that may be fixed only by reimplementing the radar.
  
• Fixed several issues with squadron commands.
  
• Optimized CruiseControl for heavy ships.
  

Also, I'm leaving for a week

[qm3ster]

@allista what would you say about Davon Throttle Control systems mod?

And what would you say about MechJeb differential thrust utility?

[allista]

@allista what would you say about Davon Throttle Control systems mod?

And what would you say about MechJeb differential thrust utility?

I haven't used DTC myself, but users of TCA reported that they complement each other well. At least until I've also introduced engines profiles; but that's still in BETA.

I'm also not aware of the differential thrust utility, but I seriously doubt that it is as sophisticated as TCA has become; the whole MechJeb, however, is beyond any competition

But what is it that you've really wanted to know?

[NeoMorph]

It would be useful to have a part that controlled TCA. For example I use smart parts to activate stuff at specific altitudes and times. You could set a lander crane to hover and decend slowly at 1000m and then stop, extend the cable and drop off the rover, then tell TCA to zoom off and crash far from the rover. What is needed is a part like the AR202 that lets you access the controls of TCA to set the parameters via Smart Parts.

Currently I am using the Translatron to hover but it's not as good as TCA. TCA is far more stable. What I want to do eventually is do a Seven Minutes Of Hell landing without touching any controls.

[allista]

It would be useful to have a part that controlled TCA. For example I use smart parts to activate stuff at specific altitudes and times. You could set a lander crane to hover and decend slowly at 1000m and then stop, extend the cable and drop off the rover, then tell TCA to zoom off and crash far from the rover. What is needed is a part like the AR202 that lets you access the controls of TCA to set the parameters via Smart Parts.

Currently I am using the Translatron to hover but it's not as good as TCA. TCA is far more stable. What I want to do eventually is do a Seven Minutes Of Hell landing without touching any controls.

So basically you need a TCA Controller part that exposes TCA controls through actions and part menue to control TCA via smart parts? If you're talking about all the controls (including autopilot, navpoint navigation, etc.), that's a lot of work. BTW, are you playing with stable version or with beta?

[qm3ster]

But what is it that you've really wanted to know?

DTC most recently introduced "throttle steering functionality". Haven't tried that mod at all, but sounds like it's the same idea the third time over.

Mechjeb introduced differential thrust centuries ago, in .90, but it seems to be quite limited in effect. Setting MechJeb desired attitude to 175 degrees off the currently pointed direction, limiting max thrust to like 1%, and burning still burns radial engines on the wrong side of the CoM with non-zero thrust.

As far as I know, all 3 are currently in active development, so I was wondering what amount of collaboration takes place. Perhaps there are some shared backend modules to be taken out of these mods, to prevent the self-excitation of a "MechJeb &/| DTC &/| TCA" system, to improve the transition between hovering, MechJeb autopilot, and manual control with throttle steering augmentation in flight, to improve performance by having only one module run derivatives of flight/engine data, and to use the most precise and stable (in real ksp practice) algorithm for all 3 cases.

[Diazo]

I've been talking with qm3ster in the DTC thread about this and with the recent updates to the mods there is now significant overlap between the different systems.

First, I have to warn that I have not played with the most recent versions so a lot of this post will be based off what I see in the change logs. (My second baby arrived recently so I don't have playtime, to post on the forums I cheat and post on my mobile while out of the house for work.) Second, I'm ignoring MechJeb, never used it.

For a one-line comparison, TCA is plug-and-play, DTC is complex but comes with all the bells and whistles. Or maybe a better way to put it is TCA is more general purpose while DTC is more advanced, but more complex to use?

What I mean by that is that TCA tries to do everything needed for flight control in one mod, but that means that each individual function of the mod is less complex because the mod is doing so much. Note that this isn't a bad thing, if you are just looking for a mod to help you fly unbalanced designs and stop crashing so much, TCA will do all that and more for you.

However, DTC only does engines, throttle, unbalanced thrust compensation, etc. and that specilization allows it to offer more options in that area then TCA does. The two biggest being that DTC does not required SAS activated to work and it offers multiple throttles so when building a VTOL you can control your lift and horizontal thrust engines separately. Note that each throttle does its own unbalanced thrust compensation, throttling up your horizontal thrust engines will not flip your VTOL over. And all this is toggleable so you can get very fine control over your engines.

In exchange however, DTC does nothing with altitude control or horizontal speed control. You need other mods for that. I'm going to mention my Vertical Speed Control and Horizontal Speed control mods, I've worked with the DTC dev to make sure they integrate smoothly with DTC. Note that they will work just fine with TCA as well, although there will be feature overlap.

So it boils down to what you are looking for, I would say TCA is easier to use, but DTC offers fine-tune control and more options to tweak.

D.

Edited October 8, 2015 by Diazo

[allista]

DTC most recently introduced "throttle steering functionality". Haven't tried that mod at all, but sounds like it's the same idea the third time over.

Mechjeb introduced differential thrust centuries ago, in .90, but it seems to be quite limited in effect. Setting MechJeb desired attitude to 175 degrees off the currently pointed direction, limiting max thrust to like 1%, and burning still burns radial engines on the wrong side of the CoM with non-zero thrust.

As far as I know, all 3 are currently in active development, so I was wondering what amount of collaboration takes place. Perhaps there are some shared backend modules to be taken out of these mods, to prevent the self-excitation of a "MechJeb &/| DTC &/| TCA" system, to improve the transition between hovering, MechJeb autopilot, and manual control with throttle steering augmentation in flight, to improve performance by having only one module run derivatives of flight/engine data, and to use the most precise and stable (in real ksp practice) algorithm for all 3 cases.

I don't know about DTC-MJ collaboration, but I personally have not contacted either devs and haven't been planning to for two reasons: first, collaboration is time-expensive; I barely have time to develop TCA as I see fit (and even that is at expense of the development of the Hangar), but to coordinate three projects, to convert them to a common codebase... not untill Squad hires us, I'm afraid Second, I don't even see a technical way for such collaborations other than actually building a team to work on the same plugin (or a set of plugins with the common codebase). Otherwise it just won't work.

I've been talking with qm3ster in the DTC thread about this and with the recent updates to the mods there is now significant overlap between the different systems.

First, I have to warn that I have not played with the most recent versions so a lot of this post will be based off what I see in the change logs. (My second baby arrived recently so I don't have playtime, to post on the forums I cheat and post on my mobile while out of the house for work.) Second, I'm ignoring MechJeb, never used it.

For a one-line comparison, TCA is plug-and-play, DTC is complex but comes with all the bells and whistles. Or maybe a better way to put it is TCA is more general purpose while DTC is more advanced, but more complex to use?

What I mean by that is that TCA tries to do everything needed for flight control in one mod, but that means that each individual function of the mod is less complex because the mod is doing so much. Note that this isn't a bad thing, if you are just looking for a mod to help you fly unbalanced designs and stop crashing so much, TCA will do all that and more for you.

However, DTC only does engines, throttle, unbalanced thrust compensation, etc. and that specilization allows it to offer more options in that area then TCA does. The two biggest being that DTC does not required SAS activated to work and it offers multiple throttles so when building a VTOL you can control your lift and horizontal thrust engines separately. Note that each throttle does its own unbalanced thrust compensation, throttling up your horizontal thrust engines will not flip your VTOL over. And all this is toggleable so you can get very fine control over your engines.

In exchange however, DTC does nothing with altitude control or horizontal speed control. You need other mods for that. I'm going to mention my Vertical Speed Control and Horizontal Speed control mods, I've worked with the DTC dev to make sure they integrate smoothly with DTC. Note that they will work just fine with TCA as well, although there will be feature overlap.

So it boils down to what you are looking for, I would say TCA is easier to use, but DTC offers fine-tune control and more options to tweak.

D.

When we, with smjjames, are done with the testing and bugfixing and TCA-3 is published these differences will be almost completely erased. But the overall divergence will only increase as TCA becomes more and more about automation and AI than manual control and finetuning.

- - - Updated - - -

It would be useful to have a part that controlled TCA. For example I use smart parts to activate stuff at specific altitudes and times. You could set a lander crane to hover and decend slowly at 1000m and then stop, extend the cable and drop off the rover, then tell TCA to zoom off and crash far from the rover. What is needed is a part like the AR202 that lets you access the controls of TCA to set the parameters via Smart Parts.

Currently I am using the Translatron to hover but it's not as good as TCA. TCA is far more stable. What I want to do eventually is do a Seven Minutes Of Hell landing without touching any controls.

I thought about this during a day and came to a conclusion that it would be easier for me to just add a simple programming interface to TCA itself than to directly implement your proposition. It is easier to implement all these triggers, timers and simple logic inside of TCA than to build a second system of controls to delegate the task. Besides, not everyone would want to use Smart Parts, but the idea of programmable autopilot is undoubtedly an appealing one

[qm3ster]

I thought about this during a day and came to a conclusion that it would be easier for me to just add a simple programming interface to TCA itself than to directly implement your proposition. It is easier to implement all these triggers, timers and simple logic inside of TCA than to build a second system of controls to delegate the task. Besides, not everyone would want to use Smart Parts, but the idea of programmable autopilot is undoubtedly an appealing one

Orite, orite, so from the most stable auto differential thrust with no configurable settings you want to maek a whole autopilot with a lot of configurability? :v

My idea was that the complete CoM-multiengine stabilisation and control core would be one layer, and the high level autopilots would live on top of that abstraction. That requires much less collaboration. But in retrospect, it might end up just like mechjeb differential thrust, since mechjeb for instance would not account for the dV that gets spent on rotation and the reduced TWR from not burning all at 100% for suicide burns, etc.

[allista]

Orite, orite, so from the most stable auto differential thrust with no configurable settings you want to maek a whole autopilot with a lot of configurability? :v

Not quite so. The stable auto-differential-thrust part is still the essence of it. But on top I've already built a lot of autopilot functionality and high-level configurability. Try the beta if you're interested

My idea was that the complete CoM-multiengine stabilisation and control core would be one layer, and the high level autopilots would live on top of that abstraction. That requires much less collaboration. But in retrospect, it might end up just like mechjeb differential thrust, since mechjeb for instance would not account for the dV that gets spent on rotation and the reduced TWR from not burning all at 100% for suicide burns, etc.

I understand the idea and that indeed would be great, but someone should develop and maintain that core, including stable API for everyone else; and modders would need to link against it, and there would be multi-version problems like with KAE, and so on. And that is an entirely different project from what we have now. The one that I couldn't pull out myself in my free time.

[smjjames]

When I try to have it add a MechJeb waypoint as a waypoint, I get spammed with this exception and it shows the waypoint as being n/a.

Edit: Also, I can't put down any new waypoints if there is a MJ waypoint active.

NullReferenceException: Object reference not set to an instance of an object
  at ThrottleControlledAvionics.WayPoint.UpdateCoordinates (.CelestialBody body) [0x00000] in <filename unknown>:0 

  at ThrottleControlledAvionics.ThrottleControlledAvionics.WaypointOverlay () [0x00000] in <filename unknown>:0 

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at RenderingManager.OnGUI () [0x00000] in <filename unknown>:0 

Edited October 13, 2015 by smjjames

[allista]

When I try to have it add a MechJeb waypoint as a waypoint, I get spammed with this exception and it shows the waypoint as being n/a.

Edit: Also, I can't put down any new waypoints if there is a MJ waypoint active.

NullReferenceException: Object reference not set to an instance of an object
  at ThrottleControlledAvionics.WayPoint.UpdateCoordinates (.CelestialBody body) [0x00000] in <filename unknown>:0 

  at ThrottleControlledAvionics.ThrottleControlledAvionics.WaypointOverlay () [0x00000] in <filename unknown>:0 

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at (wrapper delegate-invoke) Callback:invoke_void__this__ ()

  at RenderingManager.OnGUI () [0x00000] in <filename unknown>:0 

http://img.photobucket.com/albums/v214/smjjames/Kolonial%20sci_screenshot_2015-10-13--14-23-49_zpstp8jyjuo.png

Thanks! Don't know if I can get the MJ waypoints to work with TCA, but I'll fix the NRE.

[Rory Yammomoto]

This is a great mod, but,if I may ask, it needs to be able to handle asymmetrical rockets going into orbit. I say this because I have a Space shuttle that works excellently with the TCA enabled until it has to perform a horisontal gravity turn, where the thing decides to stop working and leaving me to the balancing. Could you please add a way to balance a spacecraft not hovering in the air, but flying?

[allista]

This is a great mod, but,if I may ask, it needs to be able to handle asymmetrical rockets going into orbit. I say this because I have a Space shuttle that works excellently with the TCA enabled until it has to perform a horisontal gravity turn, where the thing decides to stop working and leaving me to the balancing. Could you please add a way to balance a spacecraft not hovering in the air, but flying?

In my experience it works fine in your scenario if you setup engine roles correctly. The only inconvenience is the need to switch roles when the booster separates, but it's doable.

On the other hand, beta version has engine profiles that may be automatically switched on staging, so you may try it out, or wait until we publish it as stable.

Edited October 16, 2015 by allista

[allista]

Here's the

next BETA version.

Big changes to Radar, CPS, HSC; improvements of AutoLander, PN, GUI; bugfixes. Added FlightStabilizer module.

* Fixed VTOLAssistant that caused Autopilot and Navigation shutdown on ground/building collision.

* Reimplemented Radar using horizontal swipes of 3 spherecasts each frame in the velocity direction; this allows to detect edges of buildings/mountains and latice-like structures and take horizontal or vertical evasive actions apropriately.

* Reworked CPS collision prediction and maneuver calculations; in my tests works much better.

* Increased range of squad actions by explicitly unpacking loaded but still packed vessels.

* Added FlightStabilizer module that catches a ship that rotates without control (by either user, SAS or autopilots) and levels it, then enables SAS.

* Moved TCA version info, TCA hotkey button and Autotune toggle to Advanced Settings. Titlebar now displays the name of the active vessel. Also TCA window now may be moved by titlebar ONLY.

* Improved squad landing: first the leader finds a suitable landing site and starts landing, then it sends the Land command to the squad. Thus the squad has more chances to land compactly.

* Improved autolander's search algorithm by excluding sites that were already checked.

* Fixed residual near-zero thrust limit of MANUAL engines.

* Added system of multiple CourseCorrections that could be added by any module and are executed solely by HSC, which thus may use maneuver engines to correct course faster with less tilting.

* Fixed manual landing from high altitude that caused ground collision in the end.

* Fixed profile activation when OnPlanet status changes.

* Tried to fix NRE when using MJ waypoints as TCA waypoints.

Note: this is a debug version which will show CPS and AutoLander state vectors.

As usual, I'm asking you for help:

I've created a google spreadsheet with the tests that came to mind. You are free to add other tests, comment and mark existing tests as done (with some screenshots or videos, maybe).

Edited October 19, 2015 by allista

[smjjames]

Nice!

Also, nice idea with adding the craft name to the title bar of the TCA window.

[smjjames]

Does anybody know what would cause the lateral engines of other craft with TCA on to go all, well, wierd, they act like they are elastically attached rather than rigidly attached to the craft when I leave the physics loading range and then return?

I noticed it when I left the squad landed to check if a landing spot was actually working and when I returned, they were all like that. The main vertical engines appear to be fine though. I initially thought that the bug was due to the known issue mentioned here, but I checked and none of the engines on the landed craft are active.

https://www.youtube.com/watch?v=jdUqtxJRETo&feature=youtu.be

Going to do some more investigation into this, actually, I should investigate on my branchoff KSP.

Edit: Strange, I can't get it to reproduce on the branchoff KSP, I'll see if the quick craft that I made in the branchoff reproduces it on my main.

Edit2: The hell? When I got past 22km, the landed craft that I had, which still had the maneuver engines firing for some reason, abruptly despawned on me. Apparently the game thought it was still in flight or something.

Edit3: Okay..... now the stockbased craft isn't reproducing it on the main save either. Is it specific to the part?.....

Edit4: This is looking entirely like it's specifically that part, I've asked in the KWrocketry thread, which is what that part is from.

Edit5: Okay, TCA is involved somehow, maybe, I dunno, arg.............

Edited October 23, 2015 by smjjames

[allista]

Does anybody know what would cause the lateral engines of other craft with TCA on to go all, well, wierd, they act like they are elastically attached rather than rigidly attached to the craft when I leave the physics loading range and then return?

I noticed it when I left the squad landed to check if a landing spot was actually working and when I returned, they were all like that. The main vertical engines appear to be fine though. I initially thought that the bug was due to the known issue mentioned here, but I checked and none of the engines on the landed craft are active.

https://www.youtube.com/watch?v=jdUqtxJRETo&feature=youtu.be

Going to do some more investigation into this, actually, I should investigate on my branchoff KSP.

Edit: Strange, I can't get it to reproduce on the branchoff KSP, I'll see if the quick craft that I made in the branchoff reproduces it on my main.

Edit2: The hell? When I got past 22km, the landed craft that I had, which still had the maneuver engines firing for some reason, abruptly despawned on me. Apparently the game thought it was still in flight or something.

Edit3: Okay..... now the stockbased craft isn't reproducing it on the main save either. Is it specific to the part?.....

Edit4: This is looking entirely like it's specifically that part, I've asked in the KWrocketry thread, which is what that part is from.

Edit5: Okay, TCA is involved somehow, maybe, I dunno, arg.............

Never saw anything like it

But the part with despawning vessel is unnerving...

[allista]

Here's the

next BETA version.

* Improved HSC by using information about ManualEngines thrust directions, allowing it to turn around vertical axis to use them to change course or brake speed.

* Improved navigation by more accurately predicting arrival time and adding a CourseCorrection.

* Fixed unpacking problem.

* Improved followers logic.

* Improved Radar maneuvering logic and obstacle detection.

* Fixed NRE in AutoLander. Improved algorithm by adding UnevennesThreshold: if the flattest node is more uneven, do not try to move to it, but increase the altitude.

* Improved Anchor by predicting position at LookAheadTime.

* Fixed Navigation.Paused status when LandedOrSplashed.

* Grayed out Navigation buttons when Navigation is Paused.

* Fixed vertical overshooting caused by sudden changes in relative vertical speed.

* Disabled CPS if LandedOrSplashed.

* Increasing pack/unpack distance of TCA-enabled vessels by default.

Note: this is a debug version which will show CPS and AutoLander state vectors.

As usual, I'm asking you for help:

I've created a google spreadsheet with the tests that came to mind. You are free to add other tests, comment and mark existing tests as done (with some screenshots or videos, maybe).

[smjjames]

What's the green debug line, out of curiosity?

Also, that strange bug is still not fixed. Maneuver engines being on while landed appears fixed at least.

Edited October 23, 2015 by smjjames

[allista]

What's the green debug line, out of curiosity?

Also, that strange bug is still not fixed. Maneuver engines being on while landed appears fixed at least.

Is it reproducible with stock parts, or with KWR only? Also, there are no exceptions in the log when this happens, are there?

[smjjames]

Is it reproducible with stock parts, or with KWR only? Also, there are no exceptions in the log when this happens, are there?

I tried reproducing it with stock parts and I couldn't and as far as exceptions, I don't think so, will have to look again.

I did try to reproduce it on other craft and with the same engine and the honeybadger control module (the combo SAS+RCS part), although I've only tried it on one other craft which was smaller and there was a bit of flexing of the whole craft, so, it was inconclusive as I couldn't tell for sure if it was actually loose or the whole craft was just flexing due to being really unbalanced on one end.

I'll have to try to look more into it later.

Also, that uncontrolled ascent with follower craft that I mentioned in the google sheets is still there. Only seems to happen in non-atmospheric conditions.

edit: It only seems to happen with that specific craft design, so.... maybe theres something with the craft....

Edit2: Umkay, wierd, it seems to have disappeared with me fiddling around with the craft a little....

Edit3: The bug seems to have disappeared for sure. I think we're going to have to blame this strange one on the kraken.

Edited October 24, 2015 by smjjames

[allista]

What's the green debug line, out of curiosity?

It's the 'turn the nose to' direction; I've separated it from the needed horizontal velocity to be able to correct course with lateral engines.