Hello,
I’m new to unreal en c++ i use chatgpt and unreal engine 5.7 to help me with coding.
I built a modular event-driven AR reticle system for Unreal Engine.
The architecture is split into clear runtime layers:
ARTraceComponent
→ realtime spatial acquisition
CoreRuntimeSubsystem
→ runtime spatial candidate database/cache
ARReticleComponent
→ candidate evaluation + UI state broadcasting
WBP_Reticle
→ pure UI presentation layer
The reticle itself is a static 2D screen-space UI element, not a 3D world object.
The system continuously:
-
Scans the current AR scene
-
Registers visible wall candidates
-
Maintains a live runtime cache
-
Evaluates whether a valid wall candidate currently exists
-
Broadcasts UI state through delegates
Behavior:
-
Valid visible wall candidate → GREEN reticle
-
No valid candidate → RED reticle
Key architectural improvements:
-
Removed UI rendering logic from C++
-
Moved all visual behavior to Blueprint/UI
-
Added temporal cache invalidation using
LastSeenTimestamp -
Prevented stale AR candidates from remaining permanently active
-
Introduced modular delegate-driven communication
The result is a scalable AR interaction system where:
-
C++ handles spatial reasoning
-
Blueprint handles presentation
-
UI can evolve independently without changing runtime logic
This now supports future extensions like:
-
dynamic reticle text
-
selection prompts
-
progress indicators
-
accessibility overlays
-
interaction buttons
-
animated feedback systems
Give me some feedback, if this is usefull or am i thinking way to complex for a simple task
