Why use two different pose estimators?

[Akshaj-Katkuri]

In drive.java lines 71-76, there are 2 pose estimators initialized, but they are referenced the same way. Additionally, in lines 321-336, both methods seem to be doing the same thing, just using the two different pose estimators.

[PRODOFFICAL]

In Vision.java lines 148–160, it looks like the if (cameraIndex < 2) is written twice, once for consumer and once for reefConsumer. I’m super curious to why you guys did it this way instead of just combining them.

[Luddo183]

I’m super curious to why you guys did it this way instead of just combining them.

A quick fix during the middle of a competition. We used to have a back camera on our robot, and it threw off the poses close to the reef. We just made another estimator that we specifically used for alignment a quick fix, since our auto-align was failing and we were having issues scoring. That's why the cameraIndex < 2 line exists for reefConsumer, to stop the back camera results from being added. I think during our second competition, our poses got weird again, so we tried isolating the back camera using the cameraIndex < 2 line on consumer. The camera mount ended up getting sheared off of our bot, so we just removed the back camera entirely. The code got left behind, and since it never broke or caused an issue, we didn't bother changing it.