Interactive Ubiquitous Visualization of Construction Progress Monitoring with D4AR (4 Dimensional Augmented Reality) Models

使用 D4AR（4 维增强现实）模型实现施工进度监控的交互式无处不在的可视化

• 批准号: 1063559
• 负责人: Feniosky Pena-Mora
• 金额: $ 31.27万
• 依托单位: Columbia University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2010
• 资助国家: 美国
• 起止时间: 2010-04-29 至 2012-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1063559&HistoricalAwards=false
• 关键词: Interactive; Ubiquitous; Visualization; Construction; Progress

Systematic progress monitoring through rigorous identification, processing, and visualization of progress discrepancies between planned and actual performances, is a necessity for successful project management. It provides an early warning to project managers and allows them to diagnose performance problems and to take corrective actions in a timely manner to reduce the impacts of progress discrepancies. However, systematic implementation of progress monitoring has significant challenges because: 1) Current practice is time-consuming since it needs extensive data extraction; 2) Quality of the manually collected data maybe low and subjective; 3) Existing methods are also non-systematic; 4) Reports can be visually complex and thus not effectively represent multivariable progress information (i.e., schedule, cost, and performance) in a holistic manner nor intuitively reflect information pertaining to spatial aspects of the construction progress and their associated complexities; 5) Furthermore, current reporting methods increase the time required to describe and explain progress situation in coordination meetings and in turn could delay decision making. Thus, with current mechanism, it may be not be easy to understand the progress situation clearly and quickly. The proposed research will therefore focus on addressing these challenges through an interactive visualization of construction progress monitoring using 4 Dimensional Augmented Reality (D4AR) models. In the research, progress deviations (e.g., schedule, cost and performance) would be visualized through superimposition of planned models onto site photographs using different visualization techniques such as traffic light and weather metaphors, fish eye, and x-ray. Within the proposed framework, a project manager would be able to: 1) use the planned model as a baseline for progress monitoring; 2) collect progress data from site photographs during construction; 3) automatically analyze images and quantify built components; 4) automatically measure discrepancies between planned and built progress; and 5) visually represent progress deviations through superimposition of planned models with on-site photographs combining interactive visual imageries. In such an environment, project manager would be able to walk through or browse through a rich visual dataset of augmented progress photographs , analyze progress, make control decisions and effectively communicate with project participants. To achieve these objectives, the proposed research methodology will consist of a formal cycle of data collection, formalization, testing, validation, and refinement. This will involve integrating research, education and practice at every stage of the project. Through this integration, students will gain valuable experiences by having industry partners involved in the process. In order to reach educators, practitioners, researchers, and students who were not involved in the proposed activity, broad dissemination of the project results will be actively sought. Ultimately, in order to make the research truly successful, involvement from a diverse group of students and practitioners, in terms of economic, social, cultural, racial, and gender backgrounds, will be a priority. This diversity will provide the research project insights from different point of views, providing new perspectives for all involved.

通过严格的识别、处理和可视化计划和实际绩效之间的进度差异来进行系统的进度监控，是成功的项目管理的必要条件。它为项目经理提供了早期预警，并允许他们诊断性能问题，并及时采取纠正措施，以减少进度差异的影响。然而，系统地实施进度监测面临重大挑战，因为：1)目前的做法是耗时的，因为它需要大量的数据提取；2)人工采集的数据质量可能较低且主观；3)现有方法也不系统；4)报告可能在视觉上很复杂，因此不能以整体的方式有效地表示多变量进度信息（即进度、成本和性能），也不能直观地反映有关施工进度的空间方面及其相关复杂性的信息；5)此外，目前的报告方法增加了在协调会议中描述和解释进展情况所需的时间，从而可能推迟决策。因此，在目前的机制下，可能不容易清楚和快速地了解进展情况。因此，拟议的研究将侧重于通过使用4维增强现实（D4AR）模型进行施工进度监测的交互式可视化来解决这些挑战。在研究中，进度偏差（如进度、成本和性能）将通过使用不同的可视化技术，如交通灯和天气隐喻、鱼眼和x射线，将计划模型叠加到现场照片上，从而可视化。在建议的框架内，项目经理将能够：1)使用计划的模型作为进度监测的基线；2)从施工现场照片中收集进度数据；3)自动分析图像和量化构建组件；4)自动测量计划进度和建成进度之间的差异；5)将规划模型与现场照片叠加，结合交互式视觉图像，直观地表示进度偏差。在这样的环境中，项目经理将能够浏览或浏览丰富的可视化数据集，增强进度照片，分析进度，做出控制决策，并与项目参与者进行有效沟通。为了实现这些目标，建议的研究方法将包括数据收集、形式化、测试、验证和改进的正式周期。这将包括在项目的每个阶段整合研究、教育和实践。通过这种整合，学生将通过让行业合作伙伴参与到这个过程中来获得宝贵的经验。为了接触到未参与拟议活动的教育工作者、从业人员、研究人员和学生，将积极寻求广泛传播项目结果。最终，为了使研究真正成功，在经济、社会、文化、种族和性别背景方面，来自不同群体的学生和实践者的参与将是一个优先事项。这种多样性将从不同的角度提供研究项目的见解，为所有相关人员提供新的视角。