資源
香港專上院校所提供之論文/研究刊物
院校 | 題目 | 類型 | 日期 | 作者 | 摘要 | 網頁 |
---|---|---|---|---|---|---|
HKUST | Transfer learning enhanced AR spatial registration for facility maintenance management | Journal | 02/2020 | Chen, K., Yang, J., Cheng, J.C.P., Chen, W., and Li, C.T. | Augmented reality (AR), which requires a spatial registration technique, has proved to greatly improve the efficiency of facility maintenance management (FMM) activities. Being one of the most promising techniques for indoor localization, Wi-Fi fingerprinting has been widely used for AR spatial registration. However, localization accuracy of Wi-Fi fingerprinting decreases over time due to dynamics of environmental factors. Readings from different mobile devices can also affect the accuracy negatively. In this paper, a transfer learning technique named transferable CNN-LSTM is proposed for improving the robustness of Wi-Fi fingerprinting while implementing AR in FMM activities. Convolutional neural network (CNN), embedded with long short term memory (LSTM) networks, is utilized to predict the location of unlabeled fingerprints. Multiple kernel variant of maximum mean discrepancy (MK-MMD) is adopted to reduce the distribution difference between the source domain and the target domain, so that the location of the newly collected unlabeled fingerprints can be predicted accurately. As shown in the experimental validation, the transferable CNN-LSTM can achieve an accuracy of 97.1% in short-term (without significant environmental changes) spatial registration, 87.8% in long-term (with significant environmental changes) spatial registration, and around 90% in multi-device spatial registration, indicating a higher accuracy and better robustness over other conventional approaches. | 連結 |
HKUST | Developing efficient mechanisms for BIM-to-AR/VR data transfer | Journal | 06/2020 | Chen, K., Chen, W., Wang, Q., and Cheng, J.C.P. | Augmented reality/virtual reality (AR/VR) has been increasingly adopted to enhance visualization of building information modeling (BIM) models. However, there is a lack of mechanisms for efficient data transfer from BIM to AR/VR. On one hand, most semantic information is lost while importing BIM models into AR/VR engines. On the other hand, huge and complicated BIM models can increase the time for model transfer, increase the computation work load while rendering, and reduce the fluency when using AR/VR applications. Therefore, this paper aims to develop efficient mechanisms for BIM-to-AR/VR data transfer to better utilize the information of BIM. In this paper, an ontology-based approach is proposed to transfer semantic information of BIM. Building components in geometric models are classified according to their features and simplified with different polygon reduction methods. As shown in the experimental validation, the proposed mechanisms have the capability to efficiently transfer semantic information of BIM to AR/VR, greatly reduce the number of triangles for geometric models while maximizing the consistency of the overall shape, and improve the framerate in corresponding AR/VR applications. | N.A. |
HKUST | A BIM-based location aware AR collaborative framework for facility maintenance management | Journal | 07/2019 | Chen, K., Chen, W., Li, C.T., and Cheng, J.C.P. | Facility maintenance management (FMM) accounts for a large amount of the total cost of facilities’ lifecycle, illustrating the importance of improving FMM efficiency. Many mechanical facilities, like ventilation ducts above ceilings, are normally hidden, indicating the necessity of applying certain technology that can enable users to visualize and update the information of hidden facilities. Real-time location information is also needed so that users can be aware of their current location and the surrounding facility can be displayed accordingly. Therefore, this paper aims to develop location aware augmented reality (AR) framework for FMM, with building information modeling (BIM) as the data source, AR for the interaction between users and facilities, and Wi-Fi fingerprinting for providing real-time location information. The developed framework has the following features: (1) a proposed softmax-based weighted K nearest neighbour (S-WKNN) algorithm is used for Wi-Fi fingerprinting to obtain the current location of users; (2) a room identification method, based on BIM, the obtained location, and ray casting algorithm, is proposed to identify which room the user is currently in; (3) according to the obtained location and the identified room, users can visualize and interact with their surrounding facilities through the AR devices; and (4) users in a remote location can visualize site situation and interact with site facilities in real time through video streaming and the shared database. At the end of the paper, an experiment was designed to evaluate the effectiveness of the developed system. As shown by the experiment, the developed AR collaborative system can reduce the completion time of the designed task by around 65% compared with traditional 2D drawing-based method, and can provide a localization accuracy of around 1m |
連結 |
HKUST | Analysis and Evaluation of Low Carbon Building Features Using Building Information Modeling | FYP | 06/2018 | CHAN, Yin Yee TSANG, Chun Kit |
Building sector contributes to more than 30% of the global greenhouse gas emissions, which is the major source of greenhouse gas emissions. In Hong Kong, a high-rise and high-density city, about 60% of carbon emissions and 90% of energy expenditure come from buildings. Mitigating the environmental impacts caused by the building sector can be achieved by low carbon buildings. However, previous studies on carbon emissions from buildings mainly adopted manual processes and only a few studies applied computational fluid dynamics (CFD) into the analysis and calculated the carbon emissions using the CFD results. Therefore, the comparison between buildings with different features is laborious. Building information modelling (BIM) enables comprehensive and accurate analysis of low carbon building features by collaborating with various simulation systems. By incorporating CFD into the analysis and evaluation of the carbon footprint of different Hong Kong public housing standard blocks using BIM, the research of low carbon building is extended. Revit models of three common Hong Kong public housing blocks are created, and the embodied carbon is quantified by using the material schedules and the corresponding carbon emission factors of different construction materials. The operational carbon is quantified by using the energy simulation results and the CFD results. By considering the total carbon emissions throughout the life-cycle of the buildings, it is found that the harmony block has the lowest carbon emissions among studied public housing standard blocks. When considered the effect of natural ventilation, the energy consumption of the buildings can be reduced up to 17%. | N.A. |
HKUST | Integrating Building Information Modeling and Internet of Things for Building Facility Management | FYP | 06/2019 | CHAN, Sum Chau DWIVEDY, Sampriti |
In Hong Kong’s Smart City Blueprint, promoting ‘Green and Intelligent Buildings, and Energy Efficiency’ is one of the most important initiatives. HKUST, as the leading university in Hong Kong, has been working for years to build a better, smarter and greener campus. Keeping in line with HKUST’s “Sustainable Smart Campus as a Living Lab (SSC)” initiative, this project seeks to enable the Facilities Management Office to make better decisions with respect to balancing the trade-off between human thermal comfort and energy costs. This can be done by optimizing the operational controls of the existing heating, ventilation and air-conditioning systems (HVAC) to the occupancy level of the facility. The research was divided into two case studies, one that focuses on occupancy prediction with the use of machine learning and the other seeks to demonstrate how building information modelling (BIM) and Internet of Things (IoT) can be used to visualize the tradeoff between user thermal comfort and energy costs. This project also discusses a flowchart to integrate the various technologies being suggested. and identifies certain software tools that can be used to assist in the integration process, for instance Autodesk’s Forge. A web-based graphical user interface for an integrated smart facility management system was also constructed in order to provide a direction for future works on this topic. |
N.A. |
HKUST | Developing a Context-Aware Building Information Modeling Framework for Construction Monitoring and Management | FYP | 06/2017 | CHAN, Kei Yiu LI, Chun Ting |
With the global popularization of smartphones, which are equipped with various electronic sensors and hardware, the smartphones can collect useful information, such as location, light intensity, speed from the surroundings almost everywhere and anytime. The instant availability of the useful information has led to the formulation of a novel concept called context-awareness, which is developing computer programs to perform specific functions based on the acquired information. Location-awareness, which focuses only on collecting location information, is one of the future trends for building information modelling (BIM) development. The primary purpose of this project is to incorporate the idea of location-awareness to BIM in construction management and monitoring. To achieve this purpose, this project is objected to accomplish three main objectives, which are locating and analyzing the user current indoor position, acquiring and transferring the information in from BIM models to local devices and establishing the location-aware BIM framework on a viable and convenient platform. Thus, the location-aware BIM framework is developed as a mobile application named as “HKUST Library Helper”. The mobile application is not only equipped with Wi-Fi fingerprinting technology to support indoor localization, but also it is designed to provide different useful functions such as identifying rooms based on user position or by touch, extracting room information and creating and retrieving special notes and tasks for different rooms. | N.A. |