Data-driven behavior tree system for AI decision making. More...

#include "../ECS_Entity.h"
#include "../vector.h"
#include <vector>
#include <string>
#include <cstdint>
#include <map>
#include <algorithm>
#include <set>

Include dependency graph for BehaviorTree.h:

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes
struct	BTNode
	Represents a single node in a behavior tree. More...

struct	BTValidationMessage
	Validation message for behavior tree structure checking. More...

struct	BehaviorTreeAsset

struct	SubGraphCallStack

class	BehaviorTreeManager

Enumerations
enum class	BTNodeType : uint8_t { Selector = 0 , Sequence , Condition , Action , Inverter , Repeater , Root , OnEvent , SubGraph = 8 }
	Behavior tree node types. More...

enum class	BTStatus : uint8_t { Idle = 0 , Running = 1 , Success = 2 , Failure = 3 , Aborted = 4 }
	Behavior tree node execution status. More...

enum class	BTConditionType : uint8_t { TargetVisible = 0 , TargetInRange , HealthBelow , HasMoveGoal , CanAttack , HeardNoise , IsWaitTimerExpired , HasNavigableDestination , HasValidPath , HasReachedDestination , HasTarget = TargetVisible , IsTargetInAttackRange = TargetInRange }
	Built-in condition types for behavior trees. More...

enum class	BTActionType : uint8_t { SetMoveGoalToLastKnownTargetPos = 0 , SetMoveGoalToTarget , SetMoveGoalToPatrolPoint , MoveToGoal , AttackIfClose , PatrolPickNextPoint , ClearTarget , Idle , WaitRandomTime , ChooseRandomNavigablePoint , RequestPathfinding , FollowPath , SendMessage , MoveTo = MoveToGoal , AttackMelee = AttackIfClose }
	Built-in action types for behavior trees. More...

Functions
BTStatus	ExecuteBTNode (const BTNode &node, EntityID entity, AIBlackboard_data &blackboard, const BehaviorTreeAsset &tree)

BTStatus	ExecuteBTCondition (BTConditionType condType, float param, EntityID entity, const AIBlackboard_data &blackboard)

BTStatus	ExecuteBTAction (BTActionType actionType, float param1, float param2, EntityID entity, AIBlackboard_data &blackboard)

void	TickEventRoots (class EventQueue &eventQueue, const BehaviorTreeAsset &tree, EntityID entity, AIBlackboard_data &blackboard)

Detailed Description

Data-driven behavior tree system for AI decision making.

Author: Nicolas Chereau

Date: 2025

Version: 2.0

This file implements a behavior tree system for creating complex AI behaviors. Behavior trees are hierarchical structures that make decisions based on conditions and execute actions based on those decisions.

Key features:

Composite nodes: Selector (OR), Sequence (AND)
Decorator nodes: Inverter, Repeater
Condition nodes: State checking (health, target, etc.)
Action nodes: Behaviors (move, attack, patrol, etc.)
JSON-based tree definitions
Per-entity tree execution

Note: Behavior Tree purpose: Data-driven behavior tree system for AI decision making.

Definition in file BehaviorTree.h.

Enumeration Type Documentation

◆ BTActionType

enum class BTActionType : uint8_t

strong

Built-in action types for behavior trees.

Predefined actions that can be executed during tree execution.

Enumerator
SetMoveGoalToLastKnownTargetPos	Move to last seen target position.
SetMoveGoalToTarget	Move towards current target.
SetMoveGoalToPatrolPoint	Move to next patrol waypoint.
MoveToGoal	Execute movement to goal.
AttackIfClose	Attack if in range.
PatrolPickNextPoint	Select next patrol point.
ClearTarget	Clear current target.
Idle	Do nothing.
WaitRandomTime	Initialize random timer (param1=min, param2=max)
ChooseRandomNavigablePoint	Choose navigable point (param1=searchRadius, param2=maxAttempts)
RequestPathfinding	Request pathfinding to moveGoal via MoveIntent.
FollowPath	Follow the path (check progression)
SendMessage	Emit event to EventQueue (param1=EventType enum)
MoveTo	Alias for MoveTo action.
AttackMelee	Alias for melee attack.

Examples: /home/runner/work/Olympe-Engine/Olympe-Engine/Source/AI/BehaviorTree.h.

Definition at line 111 of file BehaviorTree.h.

◆ BTConditionType

enum class BTConditionType : uint8_t

strong

Built-in condition types for behavior trees.

Predefined conditions that can be checked during tree execution.

Enumerator
TargetVisible	Can see target entity.
TargetInRange	Target within specified range.
HealthBelow	Health below threshold.
HasMoveGoal	Movement goal is set.
CanAttack	Attack is available.
HeardNoise	Detected noise.
IsWaitTimerExpired	Wait timer expired?
HasNavigableDestination	Navigable destination chosen?
HasValidPath	Valid path calculated?
HasReachedDestination	Reached destination?
HasTarget	Alias for HasTarget condition.
IsTargetInAttackRange	Alias for range check.

Examples: /home/runner/work/Olympe-Engine/Olympe-Engine/Source/AI/BehaviorTree.h.

Definition at line 87 of file BehaviorTree.h.

◆ BTNodeType

enum class BTNodeType : uint8_t

strong

Behavior tree node types.

Defines the different types of nodes that can exist in a behavior tree.

Enumerator
Selector	OR node - succeeds if any child succeeds.
Sequence	AND node - succeeds if all children succeed.
Condition	Leaf node - checks a condition.
Action	Leaf node - performs an action.
Inverter	Decorator - inverts child result.
Repeater	Decorator - repeats child N times.
Root	Phase 38b: Root node - entry point of behavior tree (green, fixed position)
OnEvent	Phase 38b: OnEvent root - event-driven entry point (orange, event-triggered)
SubGraph	Phase 39: SubGraph - external graph reference (recursive BT or ATS)

Examples: /home/runner/work/Olympe-Engine/Olympe-Engine/Source/AI/BehaviorTree.h.

Definition at line 53 of file BehaviorTree.h.

◆ BTStatus

enum class BTStatus : uint8_t

strong

Behavior tree node execution status.

Represents the current state of a behavior tree node.

Enumerator
Idle	Node waiting for execution (not yet started)
Running	Node is currently executing.
Success	Node completed successfully.
Failure	Node failed.
Aborted	Node execution interrupted (e.g., entity destroyed)