How To Use
APIs
Debugging
Basic Concepts
Advanced Concepts
TCATNavigationQueryFilter(PathFinding)(한글)
The Tactical Crowd AI Toolkit (TCAT) enables the creation of large-scale AI systems that make informed movement and navigation decisions using GPU-based influence maps. Designed for performance, it scales to hundreds -- if not thousands -- of simultaneous units, making it ideal for crowd-level decision making where traditional query-based approaches struggle.
You can download it from the link below.
Tactical Crowd AI Toolkit: GPU Influence Maps
TCAT significantly outperforms Unreal Engine’s traditional AI system, EQS, especially in large-scale scenarios.
To demonstrate this, we compare performance under the following setup.
■ = AI attempting to approach the cone ▲
■ = AI blocking the approach of incoming enemies ■
<aside> 💡
Test Environment
As shown below, when the number of AI agents increases significantly, TCAT clearly outperforms EQS in both execution time and memory usage.
So why does TCAT outperform EQS? The key reason is that TCAT is built on a GPU-based influence map architecture.
For a detailed explanation of GPU-based influence map, please refer to the link below.
Youtube Tutorial (to be inserted later)
<aside> 💡
</aside>