Skip to content

splibiplay/xrcore-training-toolkit

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 

Repository files navigation

XRCore Training Toolkit

Unity Category AI License: MIT

XRCore Training Toolkit is a guided training layer for Unity XR built on top of XRCore — AI Agent Framework for Unity XR.

Where XRCore focuses on the agent architecture (perception → events → reasoning → behaviour), XRCore Training Toolkit adds what many real projects are missing: structured training scenarios, steps, validators, and clear user feedback.


Demo

▶ Watch full demo on YouTube

Look, grab and place a tool in XR,
guided step-by-step with instructions, highlighting and audio.

Why XRCore Training Toolkit?

In many XR training projects, the pattern is always the same:

  • Demo scenes are built for one very specific flow.
  • Training logic ("look", "grab", "place", "press the button"…) is mixed with interaction scripts.
  • Every new training flow means duplicating scenes, prefabs and ad-hoc logic.

XRCore Training Toolkit proposes a different approach:

  • Separate the XR engine (XRCore) from the training logic.
  • Define trainings as data scenarios (XRTrainingScenario + XRTrainingStep).
  • Express validation rules as reusable validators (IXRTrainingValidator).
  • Connect everything to a feedback layer (UI, highlight, audio) that makes the training understandable in seconds.

This lets you reuse the same XR agent engine (XRCore) across multiple projects and build new XR courses by changing scenarios and validators, not by rewriting low-level interaction code.


Relationship with XRCore

XRCore Training Toolkit is explicitly designed to work together with XRCore SDK:

  • XRCore provides:
    • XRCoreEventBus (event bus).
    • Perception providers (detection, vision, etc.).
    • Agents (XRGuideAgent) and reasoners (rule engine, state machine, LLM…).
  • XRCore Training Toolkit adds:
    • XRTrainingScenario and XRTrainingStep to define training flows.
    • XRTrainingScenarioRunner to orchestrate step progression.
    • IXRTrainingValidator and ready-to-use validators (look, grab, place).
    • A feedback system (instruction UI, progress, step state, audio, highlight).
    • An industrial-style demo ready to record in seconds.

Requirement: for the full experience you should also import XRCore into the Unity project. The toolkit can run a limited demo without XRCore, but it is designed as a natural companion to the core framework.


Key Features

  • Training scenarios as data

    • XRTrainingScenario: ScriptableObject defining the sequence of steps.
    • XRTrainingStep: title, instruction, validators, validation mode (Any / All).
  • Modular validation system (IXRTrainingValidator)

    • ObjectFocusValidator: succeeds when the user looks at a target object.
    • ObjectGrabValidator: succeeds when the object has been grabbed.
    • ObjectPlaceValidator: succeeds when the object is placed in a target zone.
    • Easy to extend with custom validators (e.g. press a button, operate a lever, complete a checklist).
  • Integration with XRCore events (no hard coupling)

    • An internal bridge listens to XRCoreEventBus events via reflection when XRCore is present.
    • If XRCore is missing, the toolkit still runs the basic demo and logs clear warnings.
  • User feedback layer

    • XRTrainingInstructionPanel: shows step title, instruction and progress.
    • XRTrainingProgressUI: renders step progress with text and dots.
    • XRTrainingStepIndicator: shows step state (INACTIVE / ACTIVE / COMPLETED / ERROR).
    • XRTrainingHighlightBehaviour: highlights the relevant object.
    • XRTrainingAudioFeedback: plays success / error audio cues.
  • XR-ready UI

    • Prefabs XRTrainingUIPanel and XRTrainingUIPanel_XRDemo using World Space canvases.
    • XRTrainingUIFollowCamera keeps the panel comfortably in front of the XR camera.
  • Industrial demo ready to record

    • Example scene where the user looks at, grabs, and places a tool.
    • Bootstrap logic that assembles camera, environment, UI and training logic automatically.

Requirements

  • Unity 2022+ or Unity 6.
  • Project configured for XR (OpenXR, XR Interaction Toolkit, etc.).
  • XRCore SDK installed for full integration.

Installation

  1. Install XRCore SDK (from the Asset Store or the xrcore-sdk repository).
  2. Import the XRCore Training Toolkit package into your Unity project.
  3. Open the demo scene:
    • Assets/XRCoreTrainingToolkit/Samples/IndustrialTrainingDemo/Scene/Demo_Training_Basic.unity
  4. Press Play and follow the on-screen training instructions.

Example Interaction Flow

User looks at the tool
        ↓
A perception / interaction event is generated
        ↓
XRCoreEventBus (in XRCore) publishes the event
        ↓
XRTrainingScenarioRunner advances the training step
        ↓
The feedback layer updates UI / highlight / audio


Use Cases

XRCore Training Toolkit is designed for:

  • Industrial training simulations (step-by-step procedures).
  • XR onboarding for new users or employees.
  • Real-time task guidance in industrial or maintenance environments.
  • XR product demos where you need to show a workflow quickly.
  • XR training prototypes before investing in heavy content production.

Additional Documentation

Toolkit-specific documentation (Quick Start, Scenario Guide, Validators Guide, Demo Setup, Asset Store Checklist) ships inside the Unity package at:

Assets/XRCoreTrainingToolkit/Documentation/

For a deeper understanding of the underlying agent architecture, see the XRCore docs:


License

XRCore Training Toolkit is distributed under the MIT License.

See the LICENSE file for full details.

About

XRCore Training Toolkit for Unity XR

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors