Your browser does not support JavaScript!

Home    Search  

Results - Details

Search command : Author="Στεφανίδης"  And Author="Κωνσταντίνος"

Current Record: 7 of 72

Back to Results Previous page
Next page
Add to Basket
[Add to Basket]
Identifier 000456479
Title Development of an Augmented Reality (AR) library for real-time visualization of defect detection, facilitating on-the-fly reconfiguration of manufacturing production lines
Alternative Title Ανάπτυξη μιας βιβλιοθήκης επαυξημένης πραγματικότητας (AR) για την οπτικοποίηση της ανίχνευσης ελαττωμάτων σε πραγματικό χρόνο, διευκολύνοντας την εν κινήσει αναδιαμόρφωση των γραμμών παραγωγής
Author Γαυγιωτάκη, Δέσποινα Γ.
Thesis advisor Στεφανίδης, Κωνσταντίνος
Reviewer Παπαγιαννάκης, Γεώργιος
Αντόνα, Μαργαρίτα
Abstract Augmented Reality (AR) involves overlaying computer-generated virtual elements onto the user's realworld environment, thereby enhancing the user's sensory experience by merging digital information with the physical world. AR is increasingly being utilized in diverse domains, including education, entertainment, healthcare, retail and tourism, among others, due to its ability to enhance user engagement, improve learning, increase efficiency, and provide immersive experiences. AR also plays a vital role in Industry 4.0, which emphasizes the integration of advanced technologies into manufacturing processes. As a fundamental component of human-machine interaction, AR facilitates real-time information and guidance provision to Production Line Operators (PLOs) in smart factories, thus enhancing their situational awareness, decision-making capability and operational efficiency in reducing errors and defects. As technological advancements occur at an unprecedented rate, the need for systems that can adapt to new environments, hardware configurations, and software protocols is paramount. Adaptive systems are critical in modern technological environments due to the everchanging nature of the industry and the complex demands placed by users on systems. In this challenging industrial context, adaptive AR systems that use Context Awareness (CA) and gesture-based interaction can play a critical role in enhancing efficiency and productivity. Context awareness refers to the ability of a system, device or application to perceive, understand, and utilize relevant contextual information, such as user location, time of day, activity and environment. In the industrial sector, context-aware systems can continuously monitor and analyse various aspects of the manufacturing process, in real-time, in order to optimize performance and minimize downtime. Context awareness can further enhance safety by detecting potential risks or hazards and alerting workers or adjusting the behaviour of machines accordingly. Furthermore, the ability of context-aware adaptive systems to analyse and comprehend the user’s context enables them to adapt their behaviour and content to better suit their needs and preferences, leading to a more efficient, effective and satisfactory user experience. Gesture-based interaction constitutes the prominent interaction modality for AR systems, which in the context of manufacturing environments can also enhance the user experience and improve safety, by reducing the need for manual input or physical contact with machines. More specifically, the need for shop floor operators to maintain unobstructed use of their hands, which stems from the fact that their work tasks necessitate a combination of manual dexterity, precise tool manipulation and meticulous product inspection, leads to the need of adopting mid-air hand gesture interaction. However, the majority of research approaches regarding the integration of mid-air gestures in AR applications utilize handheld devices, which limits the user’s manual dexterity to a singular hand, thereby restricting their ability to engage in concurrent manual tasks. The integration of context awareness and gesture-based interaction in adaptive AR systems can thus enable manufacturers to achieve more efficient and safe operations, reduce costs and downtime, and improve the quality and consistency of their products. Although in recent years the research on smart systems in the manufacturing sector is gaining momentum, aiming at enhancing manufacturing performance, there is still limited investigation into the application of these technological approaches for the pursuit of a zero-defect context. This work introduces a novel human-centric approach for supporting shop floor operators, utilizing a contextually-aware adaptive and rapidly reconfigurable gesture-based interactive AR system. The main objective of the proposed approach is to optimize the displayed Graphical User Interface (GUI) at runtime, while assisting the operators in the completion of everyday shopfloor tasks. In this regard, the system is modelled using a comprehensive ontological model that captures all the pertinent relationships, encompassing the operator, production lines, tasks, various GUI components and component types, available gestures from a predefined vocabulary, and possible operator reactions. The proposed system exhibits the capacity to dynamically adapt the GUI and allow the operators to naturally interact and reconfigure effectively the manufacturing process, through ontology reasoning, leveraging contextual factors such as the production line the operator is working on, the task they are performing, and their level of expertise. The presented approach is versatile and can be used across various manufacturing sites and production lines. To demonstrate the versatility and generality of the proposed work, we implemented our computational approach in three distinct scenarios, namely, a microchip production line, an antenna production line, and a lift production line. To evaluate the effectiveness of the developed system, we conducted cognitive walkthroughs with experts, as well as evaluations with twenty shop floor representative end-users. Despite minor variations, the core scenario in all cases involved the system alerting the user of a detected defective product, followed by user investigation of the cause of the defect and corrective actions based on the system’s recommendations to prevent further defects. The study’s results demonstrate that the system is intuitive and user-friendly, facilitating operator engagement and situational awareness. Additionally, they highlight that the system’s instantaneous feedback capabilities can enhance operator attentiveness, resulting in improved operational outcomes.
Language English
Issue date 2023-07-21
Collection   School/Department--School of Sciences and Engineering--Department of Computer Science--Post-graduate theses
  Type of Work--Post-graduate theses
Permanent Link https://elocus.lib.uoc.gr//dlib/4/6/6/metadata-dlib-1687329542-664238-4559.tkl Bookmark and Share
Views 597

Digital Documents
No preview available

No permission to view document.
It won't be available until: 2024-07-21