south pole problems

[Wayneo]

Have had a odd problem several times landing on duna, tylo, moho if i land a rover near the south pole the controls don't work properly its sluggish and trying to drive straight it keeps turning, if i revert to a quick save in orbit and land anywhere else it works fine, revert to the same quick save in orbit and land at the south pole same problems again.

[swjr-swis]

2 hours ago, Wayneo said:

Have had a odd problem several times landing on duna, tylo, moho if i land a rover near the south pole the controls don't work properly its sluggish and trying to drive straight it keeps turning

If you go to the North Poles, you'll notice the same issues. Unfortunately there isn't much to do about this: due to the way the navball tracks a 'heading', the poles are singularities - places where the heading is undefined. Which means that the closer you get to the poles, the more and faster the navball heading swings around. Which is what affects the camera and your controls.

[Wayneo]

No I'm not talking about the nav ball rotating about the pole, No the problem dose not happen at the north pole, The rover is just uncontrollable near the south pole if you keep trying to drive it eventual all the wheels start pointing in different directions.and I'm not talking about just right at the pole but 100km odd square area.

Edited November 5, 2016 by Wayneo

[Wayneo]

This person seams to have the same south pole north pole problem also

[sal_vager]

Hi @Wayneo, can you please provide your save with the affected craft, you can put it on dropbox, see here for details.

 

[Wayneo]

It’s not a rover design specific problem its a rover near the south pole problem. Build a rover, launch it, debug screen it into orbit around kerbin with infinite fuel,quick save,land anywhere near the south pole it will not drive, revert to quick save, land at the north pole of anywhere else and it works just fine, try a few other planets/moons same problem on the ones i have tried at the south pole.

 

https://www.dropbox.com/sh/6fimynzsylmn5uq/AAAIlYXqix6ROzJKt_Kd9X6wa?dl=0

[swjr-swis]

I have tried one of my own rovers at a couple south poles now... I don't see this issue you mention, except when the rover is directly on top of the exact south pole or driving right over it. Wheels steering reacts exactly as one would expect.

 

2 hours ago, Wayneo said:

It’s not a rover design specific problem its a rover near the south pole problem.

It is, actually, a rover design problem. If you build your rover with the probe core oriented like a rocket (forward = up into the sky), like your rover1 is, you will have problems with the wheels and controls at one point or another, and mixed with the heading singularity at the south pole it is made especially obvious.

If you keep the probe core oriented like a plane (forward = towards the horizon, parallel to the surface), those problems go away.

 

[Wayneo]

No sorry your wrong again, I have had this problem using the probododyne rovermate, and tried a probe core orientated in every direction, still had the same problem AND NOT JUST ON TOP OF THE SOUTH POLE BUT WITHIN HUNDREDS OF KM OF IT. Have been playing this since 0.24, been to the poles before with similar designs of rover and never had the problem until 1.2

[swjr-swis]

1 hour ago, Wayneo said:

I have had this problem using the probododyne rovermate

This is not surprising: the rovermate is, for reasons unknown, oriented 'upwards' instead of forward, which due to the effects I explained before on wheel control/steering, makes it inadequate for rover control. Maybe whomever designed it that way was thinking more of VTOL craft than rovers, but the name makes it a bit misleading in that case.

I have nothing invested in giving you bad advice. I even wasted some time downloading and testing both your save and rover, and one of my own rovers, on several south poles, which confirmed what I told you. My rover, with the probe core properly oriented, works fine, even in the vicinity of the south poles. You are of course free to ignore that.

[Wayneo]

If the rover with the core pointing up works everywhere except the south pole, its not a rover design problem its a game problem, and its a game problem that has not existed till now. Having to stick you probe core on its side to solve the problem is not a fix for the problem its just a workaround.

 

Don't worry i don't think you deliberately give bad advice.

[T-Red]

I have the same problem. I don't know if it's south pole specific, though. I landed a small rover near the south pole of the Mun to complete a contract. The rover was, indeed, hardly controllable, while it behaved perfectly fine at KSC. In my case it steered somewhat to the right when I pressed forward, and when countered by steering to the left, the rover came to a dead stop, sometimes even reversing and going in circles. And yes, it was based on "Rovemate", too.

[T-Red]

39 minutes ago, swjr-swis said:

the rovermate is, for reasons unknown, oriented 'upwards' instead of forward

It doesn't matter as long as the wheels are oriented properly. The wheels work very simple: they rotate forward/backward when you press W or S respectively, and steer left/right when you press A/D. The probe's orientation doesn't matter. And the "Rovemate" doesn't even have reaction wheels that could interfere with controls.

[Wayneo]

Yes had the exact same problem as you described it with two "Rovemate" rovers, except on duna near the south pole trying to get to the pyramid.

Edited November 6, 2016 by Wayneo

[swjr-swis]

38 minutes ago, T-Red said:

The wheels work very simple: they rotate forward/backward when you press W or S respectively, and steer left/right when you press A/D. The probe's orientation doesn't matter.

The control logic expects the controlling part to be oriented to face forwards towards the horizon. This is a fact. Doing it any other way will inevitably lead to control/steering 'errors' that are only errors due to an incorrect design. Those errors happen everywhere, not just the south pole. Wheels will appear to work, but if you observe closely you will notice there is always one or more wheels either turning or accelerating or both in a wrong direction.

Proof 1: load the savefile from the dropbox link. Load the 'rover 1' from that save and launch onto the pad. Drive it off the pad. PAY ATTENTION. When you steer left while driving forward, the wheels make the rover turn RIGHT. When you steer right, it goes left. This is already happening at the KSC.

Proof 2: back in the VAB, load the 'rover 1' again. Attach two launch clamps to the sides of the reaction wheel to lift the rover off the ground. Select the probe core with the rotate tool and rotate it 45 degrees as if it's climbing a hill. Launch and without staging, steer left/right and watch the wheels. Notice that now, front and back wheels no longer steer in unison to be able to make a turn. The incorrect probe core orientation confuses the control logic when on an incline, and the steering input to the wheels becomes 'incorrect'. This is not a game error, this is a design error. Were the probe core oriented correctly, this would not happen (try it to convince yourself). We haven't even left the launch pad yet, this is all happening at the KSC.

Proof 3: put a rover with the probe core correctly oriented forward to the horizon on the south pole, any south pole, and drive it around the same area you saw these problems. None of this weird behaviour you're seeing with your current rover will happen (*).

 

4 hours ago, Wayneo said:

Don't worry i don't think you deliberately give bad advice.

Feel free to believe what you will. It's not my problem and I've already wasted too much of my free time on this trying to help. I have no interest in arguing. Good luck getting your rovers to work reliably, both of you.

 

(* with the sole exception of standing/driving over the exact pole where the singularity will flip the camera around or may even put the camera underground, which is a game engine limitation - when on the exact pole, by definition, all directions point north, 'heading' no longer works, and the camera view is apparently clamped to some sort of default to prevent a code exception).

[Wayneo]

Proof 1: Invert the steering on the wheels. this is not what is coursing the problem at the south pole you can invert the steering revers the motors all you want it will not fix the south pole problem.

 

Proof 2: You just created a design error to prove a design error, original rover drives fine everywhere(invert the steering or reverse the motors,should have done that when building it but forgot) but at the south pole its not drivable, most probably because something in the game is not compensating for the rover being at the south pole just like the nav ball and camera angle swing round to compensate for the change in direction, but something in the game code is not swinging the probe cores direction around like it should, and always has up until now.

 

Proof 3: It is not a fix for the problem just a workaround. The stock Rover+Skycrane also suffers from the same south pole problems though luckily who ever designed it remembered to invert the steering.

 

Thank you swjr-swis for understand this is not your problem and you should no longer keep trying to helping.

Edited November 6, 2016 by Wayneo

[Wayneo]

Tried a rover made using a ML1 Lander Can a kerbal and no probe core and it is undrivable too.

[T-Red]

No, I think that swjr-swis's explanation makes sense. I did indeed forget that the direction wheels rotate and steer is automatically defined by where they are put respectively to the probe core. It would have been nice to have wheels just follow direct input commands (WSAD), bypassing probe core logic. I think that I can make a workaround by placing "Clamp-o-Tron Jr" on the rover in the correct direction and setting the rover to "control from here".

Edited November 6, 2016 by T-Red

[Wayneo]

Yes it would be nice if it WSAD automatically, but that would make assumptions about witch end you considered the front, the fact is you have controls to invert steering and reverse motors so when building it you can set it up anyway you want, or if you make a rover with two front ends revers the steering and motors in flight so you can drag a broken wheel instead of trying to push a broken wheel, or just change the steering if you forget to when building it, but all that is a completely different problem from having a rover that you have designed and tested that drives the way you want it to until you land it near the south pole. A problem that even the stock vehicles in the game have not ben designed for.

 

Attaching a docking port to the end and controlling it from there dose work, but it is just a workaround.

 

If you can drive a vehicle off the launch pad you should be able to drive it all the way up over the north pole and down over the south pole back round to the Launchpad, not hit a point near the south pole where the controls stop working.

 

As you found out yourself if you design a rover and test it at the lunch complex and you have all the controls all set up correctly, it should not suddenly go spinning in circles just because you landed it near the south pole of the mun.

[Wayneo]

Actually WSAD could be improved buy when you change the name and designation of a craft from say a porb to a rover the nave ball switching it’s orientation from pointing at the sky to pointing at the horizon. That way you could change it yourself, when you land your probe that doubles as a rover, you could change the designation and switch the orientation, then when you finished driving and want to go back into orbit change the designation back to a probe and switch the orientation back.

[Wayneo]

On ‎5‎/‎11‎/‎2016 at 10:30 PM, sal_vager said:

Hi @Wayneo, can you please provide your save with the affected craft, you can put it on dropbox, see here for details.

 

 

Here is a better save that demonstrates the problem, there are some rovers near the south pole of kerbin that are undrivable.

 

Out on vall there is a string of 5 stock Rover+Skycrane rovers, on these rovers when you try to drive straight buy pressing W they turn to the right, trying to correct the turn buy pressing A makes then stop and go backwards. Number 1 the featheriest from the pole is drivable but dose slew to the right, each rove getting closer to the pole gets progressively worse.

https://www.dropbox.com/sh/2w6rb7bx5h72pjr/AADu2uRtBTnP1k4ckBrWRD8sa?dl=0

Edited November 7, 2016 by Wayneo

[Wayneo]

On ‎6‎/‎11‎/‎2016 at 3:31 PM, swjr-swis said:

The control logic expects the controlling part to be oriented to face forwards towards the horizon. This is a fact. Doing it any other way will inevitably lead to control/steering 'errors' that are only errors due to an incorrect design. Those errors happen everywhere, not just the south pole. Wheels will appear to work, but if you observe closely you will notice there is always one or more wheels either turning or accelerating or both in a wrong direction.

 

I tried a rover designed the way you recommend, as i suspected a probe core on its side just shifts the south pole problem by 90°

I places rovers every 90° around the equator of vall and the one at 0° 24’ 58’’ S 13° 1’ 6’’ W is undrivable feel free to place one of your rover designs there.

 

Actually i need to correct that its not a fixed point it seams to move as the moon rotates, like its a fixed area but the moons surface is rotating over it

Edited November 7, 2016 by Wayneo

[swjr-swis]

1 hour ago, Wayneo said:

I tried a rover designed the way you recommend, as i suspected a probe core on its side just shifts the south pole problem by 90°

I places rovers every 90° around the equator of vall and the one at 0° 24’ 58’’ S 13° 1’ 6’’ W is undrivable feel free to place one of your rover designs there.

Using the exact same rover I linked before, the one I also used before to test on the poles, launched as is and edited to the Vall surface at the coordinates you mention above, I see no anomalies at all in its steering or controls. Just in case, I also went and tested it around 90, 180, and -90 degrees longitude on the equator... no problems at all in steering at any of those locations, just as the poles before. (*)

Please try that rover in the same situations. I'm pretty sure you will not see any control problems with that one, in any of the locations.

BUT.

I have a sneak suspicion of what might be causing your issue, and this may in fact be considered a game bug: from your save file and your rover craft file, I take it that you are building your rovers in the VAB instead of in the SPH, true? And is it also true that you are building them with the wheels pointed at the VAB floor, instead of pointed to the VAB north wall (which would be the default 'down' direction for craft built in the VAB)?

An additional test I did by building a rover from scratch in the VAB showed control/steering anomalies when placed in certain situations/locations. It would appear that the control logic is somehow also affected by the orientation differences of VAB and SPH, leading to similar problems as when the probe core is not oriented in the same plane as the wheels.

You should be able to test this by building your rover in the SPH. Don't simply reload the one you already built in the VAB: I tried that too, and it did not solve the issue. Build a new rover in the SPH, with the probe core oriented the way it loads by default (pointing forward to the door, or 'on the side' as you say), and the wheels pointing at the SPH floor. Don't use the rovermate - as said before, that one is oriented wrong for rover use (something I consider a bug).

Then try your new rover on the poles and the equator. It should not show any of the issues you have experienced so far.

(*: irrelevant to the issue discussed here, but worth a mention - some of those coordinates place the rover on a steep incline several km up. I had an absolute blast driving my rover down the incline at top speed and doing the 'yeeehaw' thing every time a dip in the surface made the rover experience several seconds of airtime. Driving on Vall is fun! )

Edited November 7, 2016 by swjr-swis
irrelevant note on driving fun

[Wayneo]

I did build it in the SPH

It could be any where in the 360 of the equator it shows up as a much smaller problem area because instead of turning the rover to the right or left it wants to turn the rover down when you press W or S

Edited November 7, 2016 by Wayneo

[swjr-swis]

5 minutes ago, Wayneo said:

I did build it in the SPH

Well, I simply cannot reproduce the issue you describe with rovers built the way the game logic expects them; controls and steering work in all locations.

Have you tried the rover I linked? Does it show the same issues at the poles and/or the equator, anywhere? If that one also shows the problems you describe, then your KSP install may not be entirely default/stock. Maybe one of the game files got corrupted somewhere, in which case a reinstall of KSP might be in order. It's not your savefile, because I tested with that (and my own rover) and that worked fine.

[Wayneo]

Should also not that building a rover that doubles as a lander in the SPH means the controls are all messed up when flying it, having to point at anti-normal to do a retrograde burn and control radial manually with QE, as apposed to building a rover in VAB where the markers are not needed for driving but you can track retrograde all the way down.

 

The following shows its a game problem, it only happens in the southen hemisphere, if it was probe core alinment issue the same problem would happen going towards the north pole but in the oposit direction.

Out on vall there is a string of 5 stock Rover+Skycrane rovers, on these rovers when you try to drive straight buy pressing W they turn to the right, trying to correct the turn buy pressing A makes then stop and go backwards. Number 1 the featheriest from the pole is drivable but dose slew to the right, each rove getting closer to the pole gets progressively worse.

https://www.dropbox.com/sh/2w6rb7bx5h72pjr/AADu2uRtBTnP1k4ckBrWRD8sa?dl=0

 

Edited November 7, 2016 by Wayneo