Who maintains the control of the scale for the longest seems to be heavily dependent on who gets the first cube on it, which makes placing it on in auton an ideal strategy (plus its double the points while in auton). Do you plan on trying this? How do you plan on lining up?
Vision seems like a poor option for the scale, since there isn’t any retro tape on it. The outriggers are angled in and aren’t necessarily a good judge of where you are. The easiest way I can think of is a downward facing ultrasonic to detect the 0.7" bump for the cables (to judge distance), and use a gyro to turn 90 degrees, and another ultrasonic to detect the wall of the field behind you, though this seems a bit complex for a single function.
The game-specific info string tells you which side of the scale is yours. Beyond that, dead reckoning is easily accurate enough. Drive, turn, drive (a different distance depending on which side it’s on), turn, place. You have a 3ft target. A gyro and encoders is easily enough to do this, even without motion profiling or path following.
A gyro and some encoders on the drive seem like the easiest option here. Start with your bot in front of driverstation 1 or 3, have it drive forward ~25ft while using the gyro to ensure your heading is constant, and then rotate 90deg (left or right depending where you start).
One possible autonomous strategy is a double move, which may be possible depending on when field randomization is set.
Score on switch
Pickup cube on field
Try to score on scale
Even if only one and two are complete, you’re in position to take the scale. I am not sure if your camera can determine field orientation, the ability to detect and pickup a cube. Our team is looking at using a XV11 Lidar to pickup cubes in auto.