NRI: Large: Collaborative Research: Fast and Accurate Infrastructure Modeling and Inspection with Low-Flying Robots

NRI：大型：协作研究：使用低空飞行机器人进行快速准确的基础设施建模和检查

• 批准号: 1328930
• 负责人: Sanjiv Singh
• 金额: $ 198.88万
• 依托单位: Carnegie-Mellon University
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2013
• 资助国家: 美国
• 起止时间: 2013-09-15 至 2018-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1328930&HistoricalAwards=false
• 关键词: NRI; Large; Collaborative; Research; Fast

The goal of this project is to transform the efficiency, fidelity, and safety of current critical infrastructure inspection methods by combining human judgment with machine intelligence through the development of an autonomous robotic inspection assistant. The proposed work utilizes small aerial robots, coupled with three-dimensional imaging and the state-of-the-art in planning, modeling, and analysis to develop safe and efficient, high-precision assessment of structures. The key themes of the proposed work are: (1) rapid infrastructure modeling and analysis of large complex structures via a small autonomous aerial robot with 3D mapping capabilities; (2) immersive inspection and structural assessment to combine shape and appearance into an integrated representation amenable to structural health evaluation by an inspector; and (3) adaptive aerial vehicle motion plans that seek to learn from the experience of human inspectors and facilitate as autonomous inspection assistants. The proposed work is exploring the role of humans in the entire cycle from deployment of flying robots to registering data to the assessment. This project brings together members of participating communities and is developing curriculum to engage undergraduate and graduate students from robotics and civil engineering in the proposed research.

该项目的目标是通过开发自主机器人检查助手，将人类判断与机器智能相结合，从而改变当前关键基础设施检查方法的效率，保真度和安全性。拟议的工作利用小型空中机器人，再加上三维成像和最先进的规划，建模和分析，以开发安全，高效，高精度的结构评估。建议的工作的关键主题是：（1）快速基础设施建模和分析的大型复杂结构通过一个小型自主航空机器人与三维测绘能力;（2）沉浸式检查和结构评估，结合联合收割机的形状和外观到一个综合的代表服从结构健康评估的检查员;以及（3）自适应飞行器运动计划，其寻求从人类检查员的经验中学习并且作为自主检查助理而便利。拟议的工作是探索人类在从部署飞行机器人到登记数据到评估的整个周期中的作用。 该项目汇集了参与社区的成员，并正在开发课程，以吸引机器人和土木工程领域的本科生和研究生参与拟议的研究。