Introduction to Behavior Trees

🛈 Note: Demo project includes a tutorial that provides an introduction to behavior trees through illustrative examples.

Behavior Trees (BT) are hierarchical structures used to model and control the behavior of agents in a game (e.g., characters, enemies, entities). They are designed to make it easier to create complex and highly modular behaviors for your games.

Behavior Trees are composed of tasks that represent specific actions or decision-making rules. Tasks can be broadly categorized into two main types: control tasks and leaf tasks. Control tasks determine the execution flow within the tree. They include Sequence, Selector, and Invert. Leaf tasks represent specific actions to perform, like moving or attacking, or conditions that need to be checked. The BTTask class provides the foundation for various building blocks of the Behavior Trees. Such tasks can share data using the Blackboard.

🛈 Note: To create your own actions, extend the BTAction class.

A Behavior Tree is usually processed each frame. It is traversed from top to bottom, with the control tasks determining the control flow. Each task has a _tick method, which performs the task’s work and returns a status indicating its progress: SUCCESS, FAILURE, or RUNNING. SUCCESS and FAILURE indicate the outcome of finished work, while RUNNING status is returned when a task requires more than one tick to complete its job. These statuses determine how the tree progresses, with the RUNNING status usually meaning that the tree will continue execution during the next frame.

There are four types of tasks:

  • Actions are leaf tasks that perform the actual work.

  • Conditions are leaf tasks that conduct various checks.

  • Composites can have one or more child tasks, and dictate the execution flow of their children.

  • Decorators can only have a single child and they change how their child task operates.

Sequence is one of the core composite tasks. It executes its child tasks sequentially, from first to last, until one of them returns FAILURE, or all of them result in SUCCESS. In other words, if any child task results in FAILURE, the Sequence execution will be aborted, and the Sequence itself will return FAILURE.

Selector is another essential composite task. It executes its child tasks sequentially, from first to last, until one of them returns SUCCESS or all of them result in FAILURE. In other words, when a child task results in FAILURE, it moves on to the next one until it finds the one that returns SUCCESS. Once a child task results in SUCCESS, the Selector stops and also returns SUCCESS. The purpose of the Selector is to find a child that succeeds.

Behavior Trees handle conditional logic using condition tasks. These tasks check for specific conditions and return either SUCCESS or FAILURE based on the state of the agent or its environment (e.g., “IsLowOnHealth”, “IsTargetInSight”). Conditions can be used together with Sequence and Selector to craft your decision-making logic.

🛈 Note: To create your own conditions, extend the BTCondition class.

Check out the BTTask class documentation, which provides the foundation for various building blocks of Behavior Trees.