Approximate LVLH #818
Approximate LVLH #818
Conversation
| camera.transform = Transforms.eastNorthUpToFixedFrame(cartesian, ellipsoid, update3DTransform); | ||
| var failed = false; | ||
|
|
||
| var deltaTime = time.addSeconds(1.0); |
There was a problem hiding this comment.
I think the reason you are seeing odd behavior in vehicle.czml and BAMS is because this value is too big. There's no reason this can't be a small value because interpolation will always point you in the right direction to the next point. The smaller the better. I changed it to 0.01 to test and I think you'll find it's much nicer that way.
There was a problem hiding this comment.
Why not use the derivative of the Lagange (or whatever) polynomial as the velocity, rather than finite differencing? I believe this is what Components does. CC @rcpage3
There was a problem hiding this comment.
Yes, I suggested this. If there is a way to get a derivative out of the Lagrange interpolation, that is definitely preferable, especially since it eliminates the need to perform the frame transformation twice.
|
@mramato This is ready for another review. |
|
After a bunch of offline discussion, it was decided that while this code is not optimal, it's probably the best we can do in the near term. If the finite difference being used in this change proves sub-par, we'll revisit the issue. |
Makes an approximate LVLH reference frame the default when the camera is following an object.
@mramato @emackey This looks good when following a satellite, but not when following a vehicle. Try following a vehicle in Vehicle.czml or BAMS_C4ISR.czml. Let me know what you think.