904 research outputs found
Augmented Reality and Deep Learning towards the Management of Secondary Building Assets
No full textThe retrieval of as-is information for existing buildings is a prerequisite for effectively operating facilities, through the creation or updating of Building/Asset Information Models (BIM/AIM), or Digital Twins. At present, many studies focus on the capture of geometry for the modelling of primary components, overlooking the fact that many recurring actions need to be conducted on assets inside buildings. Furthermore, highly accurate survey techniques like laser scanning need long offsite processing for object recognition. Performing such process on site would dramatically impact efficiency and also prevent the need to revisit the site in the case of insufficient/incomplete data. In this paper, an Augmented Reality (AR) system is proposed enabling inventory, information retrieval and information update directly on-site. It would reduce post-processing work and avoid loss of information and unreliability of data. The system has a Head-Mounted Display (HMD) AR interface that lets the technician interact handsfree with the real world and digital information contained in the BIM/AIM. A trained Deep Learning Neural Network operates the automatic recognition of objects in the field of view of the user and their placement into the digital BIM. In this paper, two uses cases are described: one is the inventory of small assets inside buildings to populate a BIM/AIM, and the second is the retrieval of relevant information from the AIM to support maintenance operations. Partial development and feasibility tests of the first use case applied to fire extinguishers, have been carried out to assess the feasibility and value of this system
Double Bubbles in Spaces of Constant Curvature
No full text(Statement of Responsibility) by Joseph A. Corneli(Thesis) Thesis (B.A.) -- New College of Florida, 2002RESTRICTED TO NCF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE(Bibliography) Includes bibliographical references.This bibliographic record is available under the Creative Commons CC0 public domain dedication. The New College of Florida, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.Faculty Sponsor: McDonald, Patric
Development of a Twin Model for Real-time Detection of Fall Hazards
No full textThe Architecture, Engineering and Construction (AEC) industry is still one of the most hazardous industries in the world. Researchers impute this trend to many factors such as the separation between the phases of safety planning and project execution, implicit safety issues and, most of all, the dynamic and complex nature of construction projects. Several studies show that the AEC industry could greatly benefit of latest advances in Information and Communication Technologies (ICTs) to develop tools contributing to safety management.
A digital twin of the construction site, which is automatically instantiated and updated by real-time collected data, can run fast forward simulations in order to pro-actively support activities and forecast dangerous scenarios. In this paper, the twin model of the Digital Construction Capability Centre (DC3) at the Polytechnic University of Marche (UNIVPM) is developed and run as a mock-up, thanks to the adoption of a serious game engine. This mock-up is able to mirror all the relevant features of a job site during the execution of works from a safety-wide perspective. In such a scenario, virtual avatars randomly explore the construction site in order to detect accessible, unprotected and risky workspaces at height, while warning the safety inspector in case additional safety measures are needed
Sovereign debt and reserves with liquidity and productivity crises
No full textDuring the recent financial crisis, emerging economies have kept
accumulating both sovereign reserves and debt. To account for this
empirical fact, we model the optimal portfolio choice of a sovereign that is subject to liquidity and productivity shocks. We
determine the equilibrium level of debt and its cost by solving a contracting game between sovereign and international lenders. Although
raising debt increases the sovereign exposure to liquidity and productivity crises, the simultaneous accumulation of reserves can
mitigate the negative effects of such crises. This mechanism rationalizes the complementarity between debt and reserves
Digital Twin for a resilient management of the built environment
No full textResilient management of the built environment is a strategic objective, made even more evident by recent events, not only in terms of managing chronic stresses such as social, economic and financial stresses, but especially emergency situations. Human settlements can be considered complex systems and when disaster strikes, the degree of complexity is amplified exponentially. Thus, traditional management systems, based on classical paradigms, may be completely ineffective. Therefore, new management approaches are needed to implement safety, resilience and sustainability. In this context, this paper illustrates a research where Digital Twin (DT) is exploited as an adaptive system for the built environment, as a support to optimize post-disaster reconstruction processes with a focus on reactive security management, in order to implement resilience in a smart city perspective. Following this vision, the research project aims to realise a first extended framework enabling the implementation of digital twins in the built environment, then defining some DT demonstrators in the laboratory
Artificial Intelligence assisted Building Digitization using Mixed Reality
No full textIl Facility Management in edifici complessi richiede una grande quantità di informazioni che possono essere archiviate in un modello funzionale dell’edificio. Un modello funzionale è una rappresentazione strutturata dell'edificio che include informazioni cruciali per funzioni specifiche come la sicurezza, le azioni di ristrutturazione o il funzionamento e la manutenzione. Il rilevamento di questo tipo di dati, come le proprietà tecniche dei componenti dell'edificio, è un processo costoso. Per questo motivo, è necessario uno strumento avanzato il rilievo ingegneristico. Oggi molti studi si concentrano ancora sull'acquisizione della geometria, trascurando il fatto che molte azioni ricorrenti sono condotte su componenti all'interno degli edifici. Molti sistemi proposti sfruttano tecniche di rilevamento altamente accurate, come la scansione laser o la fotogrammetria, ma che richiedono lunghi sforzi di post-elaborazione per interpretare i dati raccolti. Inoltre, queste operazioni non vengono eseguite sul posto, portando a imprecisioni causate da un’interpretazione errata dei dati. In queste circostanze, la possibilità di eseguire la maggior parte delle operazioni in loco renderebbe sicuramente il processo più efficiente e ridurrebbe gli errori.
Questa ricerca propone un sistema di digitalizzazione che sfrutta la collaborazione uomo-macchina evitando fasi di post-elaborazione del dato. A questo scopo, vengono sfruttate le potenzialità della Mixed Reality quali la sua capacità di interagire con il mondo reale, creando un ambiente ideale per la collaborazione uomo-macchina. La capacità della Mixed Reality di sovrapporre i dati digitali all'ambiente reale rende possibile il controllo dei dati direttamente in sito. Per il processo di riconoscimento degli oggetti il sistema proposto in questa ricerca si avvale di rete neurale. La rete neurale YOLO (You Only Look Once) è stata scelta per la sua velocità e funzionalità di rilevamento multiplo, ideale per applicazioni in tempo reale. Il sistema è stato sviluppato e le sue prestazioni sono state valutate per il rilevamento di componenti del sistema antincendio. Il primo set di allenamenti è stato testato ed ha raggiunto sempre più dell'85% del fattore F1. Quindi l'intero sistema è stato testato in sito per dimostrare la sua fattibilità in uno scenario del mondo reale.Facility Management in complex buildings requires a large amount of information that can be stored in a functional building model. A functional building model is a structured representation of the building including information crucial for specific functions such as safety, refurbishment actions or operation and maintenance. Surveying this kind of data, such as technical properties of building components, is a costly process. For this reason, an advanced tool for engineering surveys is needed. Nowadays many studies still focus on capturing
geometry, overlooking the fact that many recurring actions are conducted on assets inside buildings. Many systems proposed exploit highly accurate survey techniques, like laser scanning or photogrammetry, but they need long postprocessing efforts to interpret data collected. Moreover, these operations are not pursued on site leading to inaccuracies for the incorrect interpretation of
data. Under these circumstances, the possibility of performing the majority of operation on-site would definitely make the process more efficient and it would reduce errors. This research proposes a system for digitization exploiting manmachine intelligence collaboration without post-processing. To this aim, Mixed Reality with its capability of interacting with real world is applied giving an environment for man-machine collaboration. The capability of Mixed Reality of overlapping digital data to the real environment makes possible checking data directly on site. For the object recognition process the system proposed in this research make use of Neural Network. YOLO (You Only Look Once) Neural Networks has been chosen for its speed and multiple detection features, ideal for real-time applications. The system has been developed and its performance evaluated for the detection of fire protection system components. First single Neural Network have been tested reaching always more than 85%of F1 factor. Then the whole embedded system proposed has been tested on site to prove its feasibility in a real-world scenario
Development of digital twin models supporting ambient assisted living
No full textWorld population aging requires finding solutions to
improve independent living options. Ambient Assisted
Living (AAL) is making step forward developing services
supporting the elderly, but the implementation of
predictive environments is still far away.
Besides, the emerging Digital Twin (DT) concept has
begun to shape the first cognitive environments that
integrate users into assessments, improving efficiency,
prevention, and prediction of likely events through realtime
AI computing.
This paper aims to provide a prototype of a Cognitive
Building framework based on DT models that develop
high-level knowledge to achieve real-time Scenario
Awareness and offer appropriate AAL services once
anomalies are detected
Circular economy in the built environment management supported by Digital Twin. A review.
No full textManaging the built environment according to the principles of the Circular Economy (CE), with the aim of limiting energy consumption and environmental pollution, is a highly topical issue. In this context, digitization can play a crucial role in promoting virtuous scenarios.
However, the review of the literature reveals a distinct lack of in-depth investigation of the intersection between promotion of strategies related to the CE of the built environment using a digital approach, as Digital Twin (DT). In fact, in a sector characterized by low permeability to the introduction of new technologies, introducing digitization with the aim of greater efficiency is not easy. In recent years the DT approach has been gaining ground, namely, the development of digital copies that make it possible to investigate or simulate scenarios in real time.
This brief review aims to examine the existing studies and to identify challenges and further research needs. The results will ensure a deeper understanding of the topic, facilitating its development and promoting circular building management practices
Augmented Reality Application Supporting On-Site Secondary Building Assets Management
No full textSecondary building assets management requires a large amount of information related to them. Nevertheless, building assets surveysare cost and time demanding, especially because they need long post-processing efforts in order to systematize collected data.Furthermore, with the recent transition towards the BIM methodology for building management also modeling building objectsboth in their geometric features and in their related information is a long process and error-prone task. Under these circumstancesthe possibility of performing the majority of operation on-site would definitely make the process more efficient and it would reduceerrors. Augmented Reality (AR) with its capability of overlapping digital data to the real scene is the right tool to support operatorson-site.The proposed system has the aim of reducing the time of secondary building assets survey and provide a tool for the automaticenrichment of BIM models. An AR device (Hololens) with an embedded computer and a neural compute stick constitute the portableon-site system for the automatic recognition of assets objects, removing the necessity of reworking data off site. A trained DeepLearning Neural Network inside the neural compute stick performs the recognition providing the operator with objects features andposition. The AR application inside the Hololens operates as an interface between the user and the digital information just created.Finally, data is stored in a NoSQL database linked to the BIM model so as to be available for further operations. The visuallysupporting information provided by the AR tool, the possibility of working on data directly on site and the portability of the systemrepresent means for increasing efficiency in survey operations. First tests have been conducted so as to prove the feasibility of thesystem and its use on site without further equipment
- …
