Book Image

Unreal Engine 4 AI Programming Essentials

By : Jie Feng, Peter Newton
Book Image

Unreal Engine 4 AI Programming Essentials

By: Jie Feng, Peter Newton

Overview of this book

Unreal Engine is a powerful game development engine that provides rich functionalities to create 2D and 3D games. Developers have the opportunity to build cross-platform mobile and desktop games from scratch. This book will show you how to apply artificial intelligence (AI) techniques to your Unreal project using blueprints as your scripting language. You will start with an introduction to AI, and learn how it is applied to gaming. Then you'll jump right in and create a simple AI bot and apply basic behaviors to allow it to move randomly. As you progress, you'll find out how to implement randomness and probability traits. Using NavMesh, you will impart navigation components such as character movement, MoveTo nodes, settings, and world objects, and implement Behavior Trees. At the end of the book, you will troubleshoot any issues that might crop up while building the game.

Related Content you might be interested in

Current Title:

Small book image
Unreal Engine 4 AI Programming Essentials
Jie Feng | Peter Newton
Mar, 2016 | 188 pages
No titles found
Table of Contents (16 chapters)
Unreal Engine 4 AI Programming Essentials
Unreal Engine 4 AI Programming Essentials
Credits
Credits
About the Authors
About the Authors
About the Reviewer
About the Reviewer
www.PacktPub.com
www.PacktPub.com
Preface
Preface
Free Chapter
1
Introduction to Game AI
Introduction to Game AI
Game Artificial Intelligence
How AI affects the gaming experience
Techniques and practices of game AI
Unreal Engine 4 tools
Summary
2
Creating Basic AI
Creating Basic AI
Goal
Setting up the project
Using our new AIController class
Reviewing the current progress
Adding the challenge
Traces
The Enemy logic
Adding the Enemy AI
Summary
3
Adding Randomness and Probability
Adding Randomness and Probability
Introducing probability
RandomStream in Unreal Engine 4
The plan
Adding Wander
Fleeing and attacking
Summary
4
Introducing Movement
Introducing Movement
Overview
Path Finding
The A* algorithm
Navigation Mesh
The movement component
Let's start!
Waypoints
Navigation
Summary
5
Giving AI Choices
Giving AI Choices
Behavior Tree in AIController
Creating Behavior Tree
Blackboard
Designing Behavior Tree
Summary
6
How Does Our AI Sense?
How Does Our AI Sense?
Overview
AI Sense
AI Perception components
State machines
Pawn detection
State transition
Resetting the state
Simulating and playing
Summary
7
More Advanced Movement
More Advanced Movement
Setting up the agents
Viewing the agent
Following the agent
Steering behavior: Flocking
Controlling behavior through UMG
Summary
8
Creating Patrol, Chase, and Attack AI
Creating Patrol, Chase, and Attack AI
Creating a Blackboard
Mid-range attack
Controllers
Waypoints
BT Composites, Task, Decorator, and Service
Creating the logic
Summary
9
What Have We Learned?
What Have We Learned?
Creating basic AI
Adding randomness and probability
Introducing movement
Giving our AI choice
How does our AI sense?
More advanced movement
Creating patrol, chase, and attack AI
Summary
Index
Index

Blackboard

This stores the memory for an individual agent or agents. This works hand in hand with the Behavior Tree. It allows easy and direct access to variables from any of your Task nodes. Imagine that your Behavior Tree is your EventGraph, and Blackboard is your variable that you use in EventGraph. With the Instance Synced function, you can replicate the variable to every instance of the Blackboard class within the world. Now, let's start creating our Blackboard:

  1. We need to create a new Blackboard (which is under AI assets when you right-click within Content Browser) and name it DogBrain. This will store our State as Enum DogState for the Behavior Tree branches, Mailman as ThirdPersonCharacter if we have one found, Foodbowl as itself if we have found, and lastly our latest search Locations as a vector for EQS. You may see the elements we created as follows:

  2. Let's save this all. We have to create Behavior Tree now that we have the bare components that make up our Behavior Tree. Let's go...

Previous Section
End of Section 4
Next Section