AI Basics

Labgames Studio - 蓝图 - 2023/01/22

Everything you need to get your project moving! Create your own artificial intelligence now and bring your game to life! More details about what is included in this package can be found below.

  • 支持的平台
  • 支持的引擎版本
    4.26 - 4.27, 5.0 - 5.1
  • 下载类型

Thanks for checking in!


This package contains everything you need to build your own AI. The whole system works with a behavior tree which executes different blueprints during runtime. The package contains an example character which introduces the features of the AI in the demo map. Please note that this character is in no way a finished AI and that this package is designed for you to extend the AI on your own. Please note that this package assumes that you already have a fundamental knowledge on how AI in Unreal Engine works.

If you are looking for a ready-made AI that is easier for you to use, I can recommend Basic Enemy AI.

Included AI abilities:

  • Patrolling
  • Random Roam
  • Idle Mode
  • Noise detection [Sensing]
  • Visual detection [Sensing]

Furthermore, I offer a detailed documentation where it is recorded how the Blueprints, which are included this project work. With this I want to make sure that everyone who buys this product will be able to work with it later and will not spend unnecessary time trying to understand my code without any help. If you still have questions despite the documentation you can of course contact me at:, I will try to answer your question as soon as possible.

Here you will find a more detailed list of each feature and a short explanation:


Patrol Mode:

In patrol mode you can pass an array of positions (Vector) to the AI. The AI will then run through all the positions with a small pause at each point. Once the AI has worked its way through the entire array it will reset itself and restart at the first array element.

[At the moment the individual positions are inserted directly via the behavior tree. This solution is quick and easy, but ensures that every AI has the same points and there is no room for diversity. This can be solved by passing the points for each AI itself via a function. How you design this function is up to you.]

Random Roam

In Random Roam mode, the AI calculates random points within a radius you specify. Then the AI moves to the just calculated point and waits there for either a random or fixed time. After the time expires, a new point is calculated and the AI repeats the entire process until it is interrupted.

[The radius and the settings for determining the waiting time are also set directly in the behavior tree, here you also have to create your own function to be able to access the desired values and pass them to the behavior tree.]

Idle Mode

In idle mode, the AI waits at the place where it was placed in the first frames until a sense event is triggerd or it is triggered in another way. After the event, such as following a player, is over, the AI returns to its original location.

Noise Detection [Sensing]

As soon as the AI detects a noise that has been played in combination with a "Report Noise Event" node, the AI moves to the source of the noise and examines the environment around the said position for a few seconds.

[Here too you have to independently pass your own settings if desired to align the system to you.]

Visual Detection [Sensing]

As soon as the AI perceives a character other than itself or instances of itself, it immediately commences the persecution. All other activities are interrupted. Tracking the target has the highest priority in the standard design.



Number of Blueprints: 25

Network Replicated: (No)

Supported Development Platforms:

Windows: (Yes)

Mac: (Not Tested)

Documentation: Here!

Important/Additional Notes: Please make sure to have some basic knowledge of AI in Unreal Engine to be able to properly work with this product. No real physical product.