13.3 Tutorial 2

This tutorial follows on from the previous and for this one we will be designing and implementing a basic decision tree. The idea here is that we can have multiple agents all processing the same tree and end up with dynamic behaviour. A decision tree is a flowchart-like structure in which each internal node represents a “test” on an attribute (e.g. whether a coin flip comes up heads or tails), each branch represents the outcome of the test, and each leaf node represents a class label. For our case we will implement some functions which represent those “test”’s that were mentioned and our leaf nodes will be our basic action like KICK. This tutorial contains:

• Design and Implement Decision Tree

• Implement functions required by our tree

  • IsClosestTeam - Am I the closest teammate to the ball
  • IsinPossesion - Is atleast one member of my team closer to the ball than any opponent
  • IsClosest - Am I the closest agent to the ball

• Demonstrate the tree in action with RoboViz