Last season we purchased a Google Coral for artificial intelligence.  Combined with vision processing we have a camera pointing towards our intake - notes are identified in real time with location information.  As the driver approaches the note and deploys the intake the robot automatically adjusts the robot angle to increase the rate we are able to pick up notes.

The current generation of brushless motors benefit from built in encoders.  Our intake and elevator always start in the closed position.  PID control is used to quickly move them to their deployed positions.  

When the intake is deployed there is a risk that it can be damaged by another robot hitting it.  Our goal is to be able to quickly determine whether a note was intaked regardless of the view the driver has.  We explored the idea of multiple sensors but rejected them as they added extra wiring and complexity.  Modern motor controllers allow us to easily access the amperage the motor is drawing.  When a note is intaked there is a very repeatable amperage spike allowing us to reduce the need for extra sensors and complexitiry.  Fewer failure points was a key goal this year.