Silvio Da Col
Mixed Reality Application for Inspection and Validation in Industrial Environments: Human Performance and Brain-Computer Interface Advantages over Gestures.
Rel. Andrea Sanna. Politecnico di Torino, Corso di laurea magistrale in Automotive Engineering (Ingegneria Dell'Autoveicolo), 2021
|
PDF (Tesi_di_laurea)
- Tesi
Licenza: Creative Commons Attribution Non-commercial No Derivatives. Download (4MB) | Preview |
Abstract: |
Shortening product development cycles demand increasingly efficient methods and tools for the planning of complex production systems. Recently, augmented reality (AR) technologies have been introduced in manufacturing planning functions. Working in augmented environments, users usually select virtual objects with hand gestures that are associated with arm fatigue. In the first study, a mixed reality (MR) application for inspection and validation of a production line has been developed. The application allows to show the virtual environment in Augmented/Mixed Reality, and it allows to import three-dimensional (3D) CAD with information and structure in a specific position. The application makes the user interacting with the scene through a user-friendly user interface (UI) and changing the position of 3D objects in the space. The application also allows taking a measurement between two points in the space. The Measurement Tool has been validated, and the absolute average error for dimensions that are lower than 100 cm is 3.59%, while it is 1.30% for dimensions that are higher than 100 cm. In the second study, a steady-state visual evoked potentials (SSVEP) brain-computer interface (BCI) for “hologram” selection in AR is proposed. The usefulness of the BCI was demonstrated with one experiment in dense and dynamic tasks, a NASA TLX test, and a usability test. On the one hand, the BCI is 2.5 seconds slower than the hand gestures in the static tasks, while the time of selection for the two interfaces is comparable in the dynamic environments. On the other hand, the BCI is more precise, with close to 100% accuracy for all tasks. In addition, the BCI resulted in having a lower overall workload (38.52) compared to hand gestures (52.40) and final usability of 77.8 on the System Usability Scale. The results indicate the potential of a BCI in dense and dynamic environments, demonstrating a possible application for AR technologies in industrial settings. |
---|---|
Relators: | Andrea Sanna |
Academic year: | 2021/22 |
Publication type: | Electronic |
Number of Pages: | 93 |
Subjects: | |
Corso di laurea: | Corso di laurea magistrale in Automotive Engineering (Ingegneria Dell'Autoveicolo) |
Classe di laurea: | New organization > Master science > LM-33 - MECHANICAL ENGINEERING |
Ente in cotutela: | UNIVERSITY OF WINDSOR (CANADA) |
Aziende collaboratrici: | University of Windsor |
URI: | http://webthesis.biblio.polito.it/id/eprint/20123 |
Modify record (reserved for operators) |