This prototype is initially provided to students as a 3D template containing a player-controlled character and an agent which does nothing. The goal is for students to implement the agent’s behavior specified by the following decision tree:
flowchart LR
id1(( )) --- A
A{"Player in sight?"}
A -- True --> B{"Random decision\n(4 sec.)"}
A -- False --> C["Patrol"]
B -- True --> D["Attack"]
B -- False --> E["Stand still\nin fear!"]
This behavior consists of moving between waypoints, unless the player character gets too close, in which case it can either go after the player (using kinematic seek) or get confused and stop to think for a while; the selected behavior should hold for a few seconds.
This behavior should be implemented with decision trees module of the libgameai library. After students implement the required behavior, the complete prototype is then provided, so that students can compare it to their solution.
This prototype uses the DecisionTrees module of the libgameai library (MIT License).