Torque forces or Quaternion transforms for hovercraft rotations?

There are so many ways to do rotations and I am not sure which way to go.
I’m making a hovercraft thing which I need to have its rotation behaving in various ways that need combining.
It needs to:

  • Steer left and right relative to local orientation and with some ‘force buildup’ to the steering, to feel more dynamic.
  • Banking when turning based on amount of ‘turn velocity’. Maybe cheated?
  • Keep aligned with ground but with ideal up vector and overshoot when trying to reach this. I want overshoot only left and right though, not front and back.
  • This alignment will be different when close to ground and slopes, than when the ship is free flying when it needs to gradually align to ‘world up’,

And I want a nice a smooth feel with small overshoots though not too big to feel ‘floppy’.

I do have most of this behaviour sort of working but using an unholy mix of rigidbody AddRelativeTorque and direct transform.rotation manipulation.
But I now want to ‘unify’ my behaviour model.

Which way to go? Forces or Quaternions?
Do people mix these much?

Cheers
Fred

Forces will give you a much more natural feel & will help you model inertia easily.

You may try spring joints. Have an empty game object have your logic script with a child game object containing the mesh. Then use rigidbodies and a spring joint. That’s how I’d do it. Also as stated, forces will make it feel more natural and modern. Direct rotation by quaternions give it an older, more 90s feel.

For the banking, you can just change the z axis rotation (assuming z is forward) to simulate banking on the mesh only, while the logic controller doesn’t actually bank.