Högskolan i Skövde

In the context of Industry 4.0, the prominence of robotics has grown significantly, leading to a pressing need for advanced monitoring techniques. This thesis explores the potential role of Mixed Reality (MR) in robot condition monitoring through an exhaustive literature review of 138 selected studies. The investigation showed prevalent methods in robot condition monitoring, such as Fault Detection and Diagnosis, Machine Learning Techniques, Signal-based Monitoring, Model-based Monitoring, and Real-time Monitoring. MR, while not yet abundant in this context, is emerging as a promising tool, especially for real-time data visualization, remote maintenance, and integration with other technologies. By visually representing data and predictions directly on the robot, MR can speed up the diagnostic process, improve safety, and promote remote collaboration. However, challenges such as integration with legacy systems, effective data management, and hardware limitations were identified. The research also observed trends, benefits, and challenges in the broader application of MR in industrial settings. While MR offers significant advantages, including enhanced visualization, improved efficiency, and cost savings, its full integration into the world of robot condition monitoring necessitates further research and iterative refinement. In essence, this thesis presents a balanced overview of the potential and challenges of MR in robot condition monitoring, setting the stage for future exploration in this burgeoning domain.