The Behaviour Foundation

Page Goals:
- Give overview of Behaviour Foundation
- Describe Key Advantages and Features
- Brief Example of Advanced Behaviour, simply explaining how to use VisAI



VisAI uses a unique behaviour structure called The Behaviour Foundation. The Behaviour Foundation consists of 3 core pieces. Behaviour, Decision, Action.

Allow click and scroll to Behaviour/Decision/Action overview here



Behaviour

A Behaviour is a set of Goals. This determines the AI's overall behaviour, hence the name. When designing Behaviour, think of what the AI does.

Goal

A Goal is a set of Plans. A goal is the driving force of the AI. When designing Goals, think of what the AI would want.

Plan

A Plan is a set of Actions. Plans are how the AI *plans* to complete a specific goal, hence the name. When thinking of plans, think of how the AI could get what it wants.

Action

An Action is the specific logic that tells the AI what to do. It comes in the form of modular Behaviour Tree Tasks. When thinking of actions, think of the steps an AI would take to complete it's plan.

Example: B/G/P/A

As you can see, this structure allows you to create virtually any type of AI you can imagine. It's also incredibly performance efficient when optimized.

B/G/P Design Chart

This chart outlines the general B/G/P structure of an AI. You can create your own, detailed Behaviour Charts by using a program such as SimpleMind.

Now that you've seen an overview of what a B/G/P structure looks like, let's take a look at a working B/G/P structure in the behaviour tree.

Example: B/G/P Behaviour Tree Structure

As you can see, this AI has the following B/G/P structure outlined:
Behaviour: Team Deathmatch"
Goal: "Search for Enemy"
Plans: "Investigate Sound" and "Look for Lost Enemy".




Action Sequence Design Chart

What you see above is the overall layout of an Action Sequence. The Actions within the Sequence depend on the steps it takes to complete the Plan

Example: Action Sequences

Here you can see a sequence of Actions for the Plan "Look for Lost Enemy"
The Actions tell the AI how to complete the plan
The "Search" action creates variety in the plan by using "Percent Chance" to explore.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License