Beginner

Introduction to AI in Unity

Unity provides a rich set of built-in tools and packages for creating intelligent game characters, from simple navigation to advanced machine learning.

Unity's AI Ecosystem

Unity is one of the most popular game engines, powering over half of all mobile games and a large share of indie and AA titles. Its AI ecosystem includes built-in systems, official packages, and a thriving asset store of third-party tools.

Built-in AI Features

Feature	Description	Use Case
NavMesh	Navigation mesh system for pathfinding and agent movement	Character navigation in 3D worlds
NavMeshAgent	Component for automatic pathfinding and obstacle avoidance	NPCs that walk around obstacles
NavMeshObstacle	Dynamic obstacles that carve the NavMesh at runtime	Moving barriers, destructible walls
OffMeshLink	Connections between disconnected NavMesh surfaces	Jumping gaps, ladders, teleporters

Official AI Packages

ML-Agents Toolkit: Unity's open-source package for training intelligent agents using reinforcement learning, imitation learning, and neuroevolution. Integrates with PyTorch.
Unity Sentis: Run neural network models (ONNX format) directly inside Unity for inference at runtime. No external dependencies required.
AI Navigation: The modernized NavMesh package with improved APIs, NavMesh surfaces, and modifiers for fine-grained control.

Popular Third-Party AI Assets

Behavior Designer: Visual behavior tree editor with a large library of pre-built tasks. The most popular BT solution on the Asset Store.
NodeCanvas: Combined behavior tree and FSM editor with a visual scripting interface.
A* Pathfinding Project: High-performance pathfinding with grid, navmesh, and point graph support. More flexible than Unity's built-in NavMesh.
RAIN AI: Complete AI framework with behavior trees, navigation, sensors, and animation integration.

Setting Up Your AI Project

C# - Basic AI Controller in Unity

using UnityEngine;
using UnityEngine.AI;

public class SimpleAIController : MonoBehaviour
{
    public Transform target;
    public float detectionRange = 10f;
    private NavMeshAgent agent;

    void Start()
    {
        agent = GetComponent<NavMeshAgent>();
    }

    void Update()
    {
        float distance = Vector3.Distance(
            transform.position, target.position
        );

        if (distance <= detectionRange)
        {
            agent.SetDestination(target.position);
        }
        else
        {
            agent.ResetPath();
        }
    }
}

✅

Key takeaway: Unity offers a comprehensive AI toolkit from built-in NavMesh navigation to the ML-Agents package for deep reinforcement learning. Start with the built-in tools and add complexity as your project demands.

Next → ML-Agents