Using Condition Nodes for Decision Logic

In this lesson, you’ll learn how to inject dynamic decision-making into your Behavior Trees by using Condition Nodes to branch execution based on real-time game states.

The Role of Condition Nodes

Up until now, your Behavior Tree Selector and Sequence nodes have handled the flow of execution, while Action nodes performed the tasks. However, a tree that simply runs top-to-bottom lacks intelligence. Condition Nodes act as the gatekeepers of your tree; they evaluate specific variables—like distance to a target or current health levels—and return Success only if the criteria are met. If they return Failure, the tree immediately bypasses that branch, allowing your AI to pivot to a different behavior.

Implementing a Condition Node

To keep your architecture modular, define a base Node class that returns a NodeState. A condition node doesn't "do" anything in the world; it merely inspects the environment and reports back to its parent Selector.

public class IsTargetInRange : Node {
    private Transform _agent;
    private Transform _target;
    private float _range;

    public override NodeState Evaluate() {
        float distance = Vector3.Distance(_agent.position, _target.position);
        // If in range, return Success so the parent node proceeds to an action
        return (distance <= _range) ? NodeState.SUCCESS : NodeState.FAILURE;
    }
}

Building the Decision Path

By placing this IsTargetInRange node as the first child of a Selector, you create a branching path. If the condition returns SUCCESS, the Selector knows it is "safe" to execute the following Action node (e.g., AttackAction). If it returns FAILURE, the Selector ignores the attack branch and moves to the next child, perhaps an IsLowHealth condition leading to a FleeAction.

This structure decouples your "decision logic" (the criteria for acting) from your "execution logic" (the act itself). By nesting these conditions within your Sequence nodes, you ensure that complex behaviors only trigger when the exact prerequisites defined in your game state are satisfied.

Key Takeaway: Condition Nodes serve as the brain’s sensors, transforming raw variables into boolean-like gates that direct the AI's execution path through the Behavior Tree.