Creation of a fine controlled action for a robot

A common problem facing roboticists is the creation of fine controlled actions for a robot that must be interspersed with a baseline motion. We define a fine controlled action to be one in which small errors can make the difference in the success or failure of the action. A baseline motion is one that is executed repeatedly over time, such as walking straight or remaining idle. We examine the considerations that can affect the success of a fine controlled action that transitions between baseline motions. We introduce a general technique for implementing a fine controlled action that transitions from and to a baseline motion using the example of pushing an elevator button. We implement this technique in the area of robotic soccer. Our results demonstrate that this technique can successfully create fine controlled actions for a robot.