DECIDE

The increasing complexity of modern hardware systems requires ever more precise positioning and handling of components. As a result, training demands are high, and even minor mistakes can lead to significant damage, delays, or additional costs. Current assembly manuals are often text-heavy static, and difficult to adapt to variations between hardware versions, making them inefficient to use during real-world operations.

This project is driven by the need to provide workers with clear, intuitive, and real-time guidance that directly corresponds to the specific components they are handling, reducing cognitive load and improving overall efficiency.

The project introduces an XR-based guidance system that replaces traditional static manuals with interactive, spatially aligned instructions. By leveraging Deep Learning models, the system is able to detect and recognize components in real time, enabling context-aware assistance.

Augmented overlays provide step-by-step assembly instructions directly on physical components, including part identification, correct orientation, and visual references to virtual models for validation.

LIST contributes its expertise in XR technologies and AI to develop a robust, adaptive solution that bridges the gap between digital instructions and physical assembly tasks.

The proposed solution aims to significantly improve assembly efficiency by reducing errors and accelerating task execution. It also simplifies onboarding by enabling new technicians to quickly understand and perform complex procedures with minimal training.

Beyond immediate productivity gains, the approach is scalable across different hardware configurations and use cases, making it applicable to a wide range of industrial assembly scenarios