Not that I doubt your conclusions necessarily, but isn't this what sensor fusion is for? You can cast it as sensor "selection", which is fine, but given two sensors that show 10/1 accuracy (variance 100:1), and the estimates are consistent, I don't know why you'd expect it to have divergent results. (Am I understanding the problem here?). Your pos/alt is still measurable but with big old error bars until the precise sensors kick in.
You can only fuse sensors that are online. In the recent crashes the radar wasn't. The point is that you cannot swap out a high accuracy sensor, not when doing suicide burns with zero margin.
Worth noting too that your original, pre-LPS[1] position/orientation/trajectory is coming from other sensors with their own error bars, namely your IMU and whatever information the ground can glean from radio signals.
If your LPS accuracy is better than your IMU accuracy, I don't see why it wouldn't make sense to start using it once it's available.
[1] gotta call it something and GPS doesn't really fit
