Rover does pirouette when directed to move forward?

[SpacedInvader]

Trying to build my first rover in 1.1 and running into what seems to be a bug in the wheels, though it could just be me. I've got a pretty standard 6 wheel rover design, which seems to handle well, except that when I tell it to start moving forward, it will spin around in place 180°-360° before settling and moving forward. This might be manageable if it were predictable, but it seems pretty random and then on top of that, attempt to correct the direction of the rover result in a repeat of the event, usually with crashy results. Is this a known issue (searching hasn't really found the answer), and if so, how can I fix it?

[Majorjim!]

Wheels and the way they work now is totally broken no matter what anyone says. The probe core or seat or pod position now has an effect on how the wheels turn and drive. Unless the control point is between them some wheels may drive in different directions or turn to different degrees. It makes zero sense and is much worse than the behaviour of wheels before unity 5. 

 Can you post a pic of the rover?

Edited May 5, 2016 by Majorjim

[SpacedInvader]

1 hour ago, Majorjim said:

Wheels and the way they work now is totally broken no matter what anyone says. The probe core or seat or pod position now has an effect on how the wheels turn and drive. Unless the control point is between them some wheels may drive in different directions or turn to different degrees. It makes zero sense and is much worse than the behaviour of wheels before unity 5. 

 Can you post a pic of the rover?

I'm out and about, but I'll upload a screenshot a little later. 

[OutlawFarmer]

I haven't played 1.1 yet, but maybe try and making so that only the front wheels turn, it could be that the game gets confused with directional control on all wheels

[Snark]

Moving to Gameplay Questions.

[SpacedInvader]

Ok, here is the offender:

 

I've noticed that it always turned to the left when driving forward (the batteries are on the back), and also that some of the wheels turn less than others depending on what direction you are turning.

 

EDIT: A couple more shots from above showing wheel positions with right and left turns held:

Right

 

Left

 

Looks like only two opposite corner wheels are turning depending on the direction. I tried taking off steering inversion, and then only the two wheels opposite of the desired direction of turn work. I should mention that the middle wheels have steering off, so they are not supposed to turn.

 

EDIT2: Apparently it had to do with the orientation of the probe core... I tried redesigning it with the core rotated 90° so that it was stacked with the cubic struts and the spinning seems to have disappeared. There is still some funny wheel movement, but it doesn't seem to be affecting the direction of movement as much as I had thought.

Edited May 6, 2016 by SpacedInvader

[purpletarget]

Yes, it's all to do with the orientation of your control point. In your original case, the probe core pointing to the sky won't allow the wheels to behave as expected. In your edit 2 example, the wheels will be slightly better in that your probe is facing the horizon, however... it's still upside down (navball, blue should be on top). Flip your core around right side up, and you'll probably get more desirable results.