Inverse Kinematics and Interoperability Standards for Visualization of Construction Activities at the Operations Level of Detail

该项目将研究如何根据执行该操作的资源（如设备、工匠）的运动学属性和基础设施的几何形状自动生成施工操作的3D动画。这将通过设计一种分析技术来实现，该分析技术将铰接资源视为可以作为开放运动链进行分析的连杆系统。这将允许这样一个资源的实现（例如挖掘机的铲斗，泥瓦匠的手臂等）被认为是运动链的末端执行器，可以使用基于逆运动学的目标导向技术来操纵和控制。末端执行器本身的目标，即在操作过程中资源执行器应该在关键实例中的位置，将自动从设施的可互操作的3D产品模型中提取出来。将设计和实现一种简单的、软件可编写的、面向对象的语言来定义铰接的资源。离散事件仿真模型（或任何其他外部软件进程）将能够在3D虚拟世界中实例化用这种语言定义的资源，并指导它们使用高级的、类似于构造工作的术语来执行操作。这将使任何长度和复杂性的施工操作的自动3D动画。由此产生的能力将显著改善离散事件仿真模型的验证、确认和沟通，并使其更加可信，从而用于作战计划和决策。允许在执行计划的工作之前就其进行适当的交流，将改善实际外地业务的业绩。除了交流未来可能发生的事情（通过可视化模拟操作）之外，还可以在虚拟世界中自动重新创建过去发生的事情和当前正在发生的事情（从实时数据中）。在制造业、航空业、采矿业和造船业等对可视化操作的需求与建筑业一样迫切的其他领域，也会产生这种效益。该技术将使教育工作者能够有效地向建筑和其他领域的学生教授运营规划、分析和设计。研究得出的工具将用于加强密歇根大学目前教授的建筑课程，并将向其他机构的教育工作者公开提供。这项工作还将通过为研究人员提供有效的技术来研究建筑和其他领域的操作、安全和教育问题，从而对研究基础设施产生重大影响。总而言之，该项目的社会效益是：1)通过合理规划和设计施工作业，可以降低施工生命周期成本；2)参与项目人员的职业发展；3)未来建筑工程师的有效教育和培训。

This project will investigate the requirements that will enable the 3D animation of a construction operation to be automatically generated from the kinematic properties of the resources (e.g. equipment, craftsmen) that perform that operation, and the geometry of the infrastructure. This will be achieved by designing an analysis technique that considers an articulated resource to be a system of linkages that can be analyzed as an open kinematic chain. This will allow the implement of such a resource (e.g. excavator's bucket, mason's arm, etc.) to be considered as the kinematic chain's end-effector that can be manipulated and controlled using goal-oriented techniques based on inverse kinematics. The goals of the end-effectors themselves, i.e. positions where a resource's implement should be at key instances during an operation will be automatically extracted from interoperable 3D product models of facilities.A simple, software-authorable, object-oriented language to define articulated resources will be designed and implemented. A discrete-event simulation model (or any other external software process) will be able to instantiate resources defined in this language inside 3D virtual worlds, and instruct them to perform operations using a high-level, construction work-like terminology. This will enable the automated 3D animation of construction operations of any length and complexity. The resulting capability will significantly improve the verification, validation, and communication of discrete-event simulation models, and make them more credible and thus used in operations planning and decision-making. Performance of actual field operations will be improved by allowing proper communication of the planned work prior to its execution. In addition to communicating what may happen in the future (by visualizing simulated operations), it will be possible to automatically re-create in virtual worlds what happened in the past, and what is currently happening (from real-time data). Such benefits will also accrue in other fields such as manufacturing, aviation, mining, and ship-building where the need to visualize operations is as acute as in construction. The enabled technology will allow educators to effectively teach operations planning, analysis, and design to students in construction and other domains. The tools resulting from the research will be used to enhance the construction courses being currently taught at the University of Michigan, and will be made publicly available to educators at other institutions. This work will also significantly impact the infrastructure for research by providing an effective technology for investigators to study operations, safety, and educational issues in construction and other fields. In summary, the societal benefits of the project are: 1) the reductions in construction life-cycle costs that will be possible through proper planning and design of construction operations; 2) the career development of the personnel participating in the project; and 3) the effective education and training of future construction engineers.