AI-Powered Aircraft Inspection

XR inspection system that spots panel defects in real time and auto-generates the maintenance documentation that normally follows by hand. Domain: Aerospace maintenance.

The challenge

V-22 maintainers triage wiring, chafing, and thermal-damage defects manually, generating documentation overhead that slows inspection throughput.

The approach

HoloLens 2 XR application detects panel-level anomalies (clamps, ties, defects) in real time, then an N8N + PostgreSQL agent queries tech manuals via RAG and auto-generates standardized defect reports into live maintenance logs.

About this project

This project demonstration shows an AI inspection system that identifies clamps and ties on a simulated aircraft panel in real-time.

Role

Lead developer. Built the HoloLens 2 detection application and the RAG-backed reporting agent that writes standardized defect reports into the maintenance log.

Outcome

-25% documentation overhead

Technology and methods

  • HoloLens 2
  • Azure Custom Vision
  • N8N
  • PostgreSQL
  • Claude API
  • RAG
  • NVIDIA Jetson

Related work