Very simple Supervised Learning program. The robot tries a position at random. Gathers feedback from a human. Adds good positions to a list. Moves through the good positions.