CAREER: Hybrid 3D Unstructured Workspace Modeling: A Critical Component in Developing an Automated Construction Site

该学院早期职业发展(Career)计划奖项的目标是创建一种创新的3D工作空间可视化方法，使自动化建筑设备能够安全高效地处理用于机器人现场安装和连接的建筑材料。机器人建筑材料搬运需要工作空间的快速可视化和高度准确的位置数据，以确保目标对象和机器人之间安全可靠的物理接触。这些都是目前的传感器技术难以实现的。为了应对这一挑战，将通过一种混合方法来研究目标聚焦的距离数据收集和建模方法，该方法将轻型3D激光扫描仪与激光测距仪和红外相机等二次和三次传感设备相结合。可安装在机器人或设备上的混合系统将显著减少数据收集时间，同时提高数据准确性。将针对不同形状的建筑材料部件开发自动目标检测和基于模型的拟合匹配算法。这种混合方法将使用真实世界的机器人测试台进行测试，其中包括内布拉斯加州大学林肯分校的一个多关节手臂机器人，以及位于韩国大学自动化建筑系统中的一个螺栓连接机器人和一个机器人起重机。如果成功，这项研究的结果将通过为材料、设备和操作员提供一个全面感知和监控的环境，促进更智能和自动化的建筑工地和制造流程的发展。这项研究也将通过减少大量的建筑垃圾对环境产生积极的影响，这将极大地造福于建筑业和整个社会。最后，该项目有强有力的教育计划，将培养未来的工程和科学领导者，促进工业从业者的终身学习，并扩大代表不足的群体对研究的参与。

The objective of this Faculty Early Career Development (CAREER) program award is to create an innovative 3D workspace visualization method, which will enable automated construction equipment to safely and efficiently handle construction materials for robotic field erection and connection. Robotic construction material handling requires both rapid visualization of the workspace and highly accurate position data for safe and secure physical contact between a target object and a robot. These are difficult to achieve with current sensor technologies. To meet this challenge, a target-focused range data collection and modeling method will be researched through a hybrid approach that integrates a light-weight 3D laser scanner with secondary and tertiary sensory devices such as a laser rangefinder and an infrared camera. A hybrid system mountable on a robot or piece of equipment will significantly reduce data collection time while enhancing data accuracy. Automatic target detection and model-based fitting and matching algorithms will be developed for various shapes of construction material components. The hybrid approach will be tested using real-world robotic test beds, including a multi-joint arm robot at the University of Nebraska-Lincoln and a bolting robot and a robot crane located within Korea University's automated building construction system. If successful, the results of this research will enhance the development of more intelligent and automated construction job sites and manufacturing processes by providing a fully sensed, monitored environment for materials, equipment, and operators. This research will also positively impact the environment by reducing the enormous amount of construction waste, which will greatly benefit the construction industry and society in general. Finally, this project has strong educational plans that will nurture tomorrow's leaders in engineering and science, promote lifelong learning for industrial practitioners, and broaden participation of underrepresented groups in research.