IR tracking

[nzj1]

I was wondering if instead of tracking two IRs on the field if you could track one IR with two or three sensors on the robot and be just as effective?

[mtrawls]

Originally Posted by nzj1

I was wondering if instead of tracking two IRs on the field if you could track one IR with two or three sensors on the robot and be just as effective?

I'm not sure if I understand what you are saying. The code at kevin.org/frc only tracks one IR beacon (either type-0 or type-1), and it uses four sensors, two on each servo. Is this what you mean by "tracking two IRs"? Also, you could use just "two or three" sensors -- but since they give us four, and code to use all four to boot, why not just use all four? If you don't entirely understand how the code operates, I recommend you look at it again, and also search around here in the programming forums, as there have been several threads about how the code works.

[Al Skierkiewicz]

Originally Posted by nzj1

I was wondering if instead of tracking two IRs on the field if you could track one IR with two or three sensors on the robot and be just as effective?

If you only track one beacon you can determine that you are located on a circle whose center is the beacon. With two sensors you can tell the distance to the center of the circle. Unfortunately you can't tell where on the circle you are without another reference of some sort. You would need to know where you started and how far and in what direction you travled since the start to be accurate.

[Oode]

Originally Posted by Al Skierkiewicz

If you only track one beacon you can determine that you are located on a circle whose center is the beacon. With two sensors you can tell the distance to the center of the circle. Unfortunately you can't tell where on the circle you are without another reference of some sort. You would need to know where you started and how far and in what direction you travled since the start to be accurate.

Or......
you can use the idea of a differential measurment.

TAke a snapshot of the 2 angles, go a known distance, then take another snapshot. That gives the third data point, and a "fix" of where you are.

Hmmmmm. interesting.