Augmented Reality (AR) using object detection and displaying manuals for transformer components.
Reference number | |
Coordinator | AugmentedRealm AB |
Funding from Vinnova | SEK 576 373 |
Project duration | November 2023 - August 2024 |
Status | Completed |
Venture | Advanced digitalization - Enabling technologies |
Call | AI for advanced digitalization, 2 |
Important results from the project
Project has explored improving power transformer production through augmented reality (AR). An advanced AR application was developed that combines real-time object detection with the display of PDF instructions and schematics, with the goal of reducing manual consultation and increasing productivity. The project implements: - AI-powered chat functionality for PDF content. - Multiple object detection and location. - Video to the AR world based on which object is viewed. - Flexible web portal for configuring PDFs and videos for each detected object.
Expected long term effects
The project resulted in an AR application that integrates the display of PDF files with real-time object detection. An innovation was the implementation of an AI chat function for the PDF content, which increases user-friendliness. The system includes a web portal for configuring PDFs and videos. Although the project primarily resulted in an advanced demonstration, it has clearly demonstrated the potential of AR technology. The results have laid the foundation for continued development, an area that is now being discussed is video recording of silent knowledge.
Approach and implementation
We have completed 3 trips to hitachi - Factory display of focus area, where the Tap changers in Hitachi Energy Ludvika were chosen as the focus area. - A review and recording of material on how to assemble the component, the films were used in the final demo in AR. - Demonstration of AR application for operators. Continuously during the project we have shown demonstrations of what is being developed. We have taken in views on things that should be changed. For example, voice recognition is included as a way to interact and to be able to talk to the documents.