Overview of AI

• This is based on Introduction to AI with Blueprint
  • You will be doing this for your next assignment.
  • It is really good and I feel the results are worth it.
    • You end up with a NPC who can chase you, look for food, run to random locations or hide
  • He does a nice job integrating some theory as well.
  • There is a missing bit, but I will cover that in my assignment notes.
  • The framework appears to already have the project completed, so you will probably want to delete the AI parts and recreate them. Again, this is discussed in my notes.
  • This goes very fast and is very in depth so I thought a preview might be in order.
• Some Ideas.
  • An Intelligent Agent or IA is an autonomous entity that acts
  • In this case, he says that there are three components
    • Sense or discover what is in the environment.
    • Think or figure out what to do next based on the input.
    • Act or take an action based on the action.
• For sensing, we will use two different items.
  • Navigation Meshes
    • This is a resonably modern concept.
    • Build a mesh of convex polygons that "covers" the areas where we can navigate.
    • In (A navigation mesh for dynamic environments† ) they present this navigateion mesh for "Zelda" A navigation mesh for dynamic environments†
    • 
    • I grabbed this one from Gamedev.net.
    • 
    • 
    • Navigation within a polygon is simple since it is convex, just move directly there.
    • Navigation outside of polygons, form a graph and do a shortest path computation (Algorithms or Discrete II)
  • Building a nav mesh is easy,
    • just drag a Nav Mesh Bounds Volume into the scene
    • There are some parameters you might need to adjust
    • But this will generate the mesh for you.
    • He cautions about large nav meshes.
    • You can stitch mutliplies together if you wish.
      • There are functions like GetRandomReachablePointInRadius
        • He employs this to make the NPC wander about in the scene.
    • The next item is the AI controller
      • Remember a character is a pawn that moves around.
      • These can be controled by a controller.
      • You will probably have a controler for each different type of NPC in your scene.
      • He uses the controler for two things.
        • To run the AI, which controls the character (NPC)
        • To implment fnctions which communicate between the character (NPC) and the AI.
    • The AI is implmented as a Behavior Tree.
      • A blackboard
        • Think of this as a variable storage location.
        • In the blackboard he declares the variables that control the AI
        • 
        • Just like we need to transfer data to the GUI, we need to send data to the blackboard as well.
        • Here is a script in the AI_Player Controller that does this.
        • 
        • And this is called by a function in player (NPC)
        • 
      • The tree
        • The tree is constructed from a series of nodes
        • The end nodes are called tasks.
        • There are some pre-built tasks.
          • Move to
          • Wait
          • 
          • Notice the move_to is accessing a value stored in the blackboard.
        • Or you can build a task.
          • Find navigable location inteacts with the nav-mesh to generate a location the player can move to.
          • 
          • Here is the blueprint
          • 
          • Notice here it is setting a value in the blackboard.
        • These tasks are combined in a sequence
          • This is a special node that attempts to execute ALL of the tasks associated with it.
          • 
        • Another type of node is a Selector
          • This type of node stops executing as soon as the first child executes successfully.
          • 
        • There is another type of node called Parallel, but it is not discussed in the tutorial. (But a presentation on this might be great)
        • These nodes can have a Decorator
          • This is a conditional that effects if the node should fire or not.
          • They interact with variables from the blackboard.
          • And determine if a node should fire or not.
          • 
          • Here, we check to see if hunger has reaced 100%, and if so, search for food. If not, do something else.
        • A Node can also have a Service
          • This is an action associated with a composite node.
          • It is a function that gets executed when the node is executed.
          • 
          • Here, every time the taks fires, it also runs the service.
          • This is just a blueprint that increments the "Hunger" variable.
          • 
    • The final portion discussed is the Environmental Query System
      • This is a cool experimental portion of Unreal.
      • It gives the NPC "Senses"
      • In this system, you describe a series of queries.
      • In this case, we are asking for the closest food sources within 1000 units
      • 
      • This gets tied to the BTree
      • 
      • And executed when the player is trying to find food.
      •