Remote assistance framework for maintenance through virtual & augmented reality

• Autor(a) principal: Félix, Ivo Miguel dos Santos
• Data de Publicação: 2023
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Texto Completo: http://hdl.handle.net/10773/41848
• topic: Remote assistance; Augmented reality; Virtual reality; Mixed reality; Computer supported mixed reality; Cooperative work; Virtual replica

The increasing job specialization and globalized economy often produce situations for which an expert is required on-site but co-location is either not possible or convenient due to time or cost constraints. Connecting the expert with a local technician remotely is a viable option but remote assistance for maintenance tasks presents several challenges that are not satisfied efficiently by traditional videoconference approaches, as spatial information is not easily conveyed through these mediums. Virtual and augmented reality technologies, on the other hand, are well suited to that task as the information added as virtual objects is spatially referenced. Making use of these advantages, in an attempt to address the challenges of the proliferation of remote work, a remote assistance framework for maintenance tasks making use of virtual and augmented reality was developed, with support for: asynchronous assistance; independent user views through 3D model reconstruction of the local technician’s environment; communication through non-verbal cues using annotations; virtual replica annotations. The combination of these characteristics is uncommon in current research and provides a high degree of autonomy between users, with the potential for a more efficient use of the remote expert’s time. After development, the system was evaluated with a user study targeting a building maintenance use case. The system performed better than traditional methods in task failure rate, while completion time was better only for the tasks that did not require text-insertion. Furthermore, it showed promise as a training tool, and garnered positive feedback from users.