bitbrain/beehave
Beehave
Beehave is a behavior tree AI framework and plugin for the Godot engine. It serves as an agent logic orchestrator for designing adaptive non-player character behaviors using hierarchical behavior trees.
The system features a visual debugger that provides runtime visualization of the execution flow and internal state of active trees. This allows for AI execution analysis and the identification of logic errors through a state-based runtime interface.
The framework utilizes a composite-decorator pattern and a tick-based polling cycle to execute tree-based logic. It integrates directly with the Godot scene hierarchy, mapping visual node graphs to scene objects to control entity actions. Additionally, it includes monitoring tools to track resource usage and execution efficiency.
Features
- Behavior Trees - Provides a complete framework for implementing adaptive agent logic using hierarchical behavior trees.
- NPC Integration - Provides a system for attaching behavior trees to Godot scene nodes to control non-player character actions.
- Stateful Agent Orchestration - Serves as a state-based orchestration framework for managing complex non-player character behavior transitions.
- Game Engine AI Plugins - Integrates directly with the Godot game engine to allow behavior tree composition within its scene hierarchy.
- NPC Logic Composition - Enables the design and attachment of behavior trees to game objects to orchestrate NPC actions.
- State-Based Execution Logic - Evaluates a hierarchical node structure to determine the next AI action based on success, failure, or running states.
- Visual Editors - Provides an interactive visual editor for inspecting and modifying behavior trees during runtime.
- AI Application Debugging - Provides tools for visualizing the execution flow and internal state of behavior trees to debug AI logic.
- State Visualizations - Provides real-time visual indicators of the active path and internal state of the behavior tree within the editor.
- Logic-to-Graph Mapping - Links internal behavior tree logic to a graphical interface for designing agent behaviors without coding.
- Behavior Controller Bindings - Attaches behavior tree controllers directly to scene objects to map logic execution to entity movements.
- Composite-Decorator Nodes - Utilizes a composite-decorator pattern to organize branching and execution modifiers within behavior trees.
- AI Agent Execution Monitors - Captures real-time execution events and logic flow to analyze agent decision-making processes.
- Behavior Tree Visualizers - Ships a runtime interface for visualizing the execution flow and internal state of active behavior trees.
- Plugins and Scripts - Behavior tree system for NPC AI.
