We should know our robot position (pose) at any given time given our odometry and vision. There should be a function to "snap" to areas of interest like the speaker. We know the pose of the speaker, we know our alliance color, and we know our position. That should be enough information to turn our robot to the desired angle which will point us at the object of interest. Once the object comes into view, we should be able to use vision to reduce the error ~0
We should know our robot position (pose) at any given time given our odometry and vision. There should be a function to "snap" to areas of interest like the speaker. We know the pose of the speaker, we know our alliance color, and we know our position. That should be enough information to turn our robot to the desired angle which will point us at the object of interest. Once the object comes into view, we should be able to use vision to reduce the error ~0