Industrial AR Mobile Application Development
A technician at an oil refinery needs to replace a valve. Equipment documentation is a 400-page PDF, pipeline schematic in AutoCAD on the server, replacement instructions in a laminated folder in the office. An AR app on the tablet points directly at the needed valve, overlay shows bolt torque specifications and step-by-step instructions over real equipment. This is not the future—this is working PTC Vuforia, Scope AR, Augmentir. We build similar systems for clients needing custom solutions for their ERP and production specifics.
Technical Complexity: Industrial Equipment Recognition
Standard flat-image tracking from AR catalogs doesn't work for factories: equipment is three-dimensional, metallic (reflections), often in poor lighting and hazy conditions. Approaches:
Marker-based. ArUco or QR markers affixed to equipment—reliable, independent of object appearance. Use ARImageTrackingConfiguration (ARKit) or AugmentedImageDatabase (ARCore). Downside: markers must be physically placed, can get dirty.
Markerless object recognition. ARKit provides ARObjectScanningConfiguration for scanning 3D objects into .arobject format and subsequent recognition via ARObjectAnchor. Works for compact equipment (pump, valve block) with sufficient texture. For large objects (turbine, tank)—train model via Create ML Object Detection or use Vuforia Model Targets SDK with custom app integration.
Spatial anchors for large objects. Azure Spatial Anchors or ARCore Cloud Anchors: position AR scene relative to object once, save anchor to cloud, on next app launch position recovers. Works across app restarts and device switches.
Integration with Production Systems
Industrial AR without integration is expensive toy. Real value:
- SAP PM / Maximo: on equipment scan, app fetches current maintenance work order, repair history, needed parts
- IoT/SCADA: overlay with real sensor readings (temperature, pressure, vibration) over physical equipment via OPC-UA or MQTT
- Equipment digital passport: user sees AR layer with ERP data right above the machine
API integrations via REST or GraphQL, authentication via corporate SSO (SAML 2.0, Azure AD). Offline mode critical: production floor may lack stable Wi-Fi, all instructions cached locally via Core Data or Room.
UX Specifics for Industrial Use
Engineer in protective gloves with tablet—interface must work via large buttons, voice commands (SFSpeechRecognizer / Android Speech API), without removing gloves. Screen brightness in direct sun—use UIScreen.brightness = 1.0 programmatically and High Brightness Mode on supported devices. At explosion-hazard sites—only ATEX/IECEx certified devices, limiting hardware to specialized tablets.
Development Workflow
Audit equipment and production processes → choose recognition method (markers / object tracking / spatial anchors) → prototype on one equipment type → ERP/SCADA integration → develop authoring tool (so technologists create new AR instructions without programmers) → pilot in one shop → scale.
Timeline: pilot with marker tracking and 10–20 AR instructions—8–12 weeks. Full system with ERP integration and authoring tool—5–9 months. Cost individual.