SBIR Phase I: Robotic Inspection and Data Analytics to Localize and Visualize the Structural Defects of Civil Infrastructure

SBIR 第一阶段：机器人检查和数据分析，以本地化和可视化民用基础设施的结构缺陷

• 批准号: 1915721
• 负责人: Diar Sanakov
• 金额: $ 22.5万
• 依托单位: InnovBot LLC
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-07-15 至 2020-12-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1915721&HistoricalAwards=false
• 关键词: SBIR; Phase; Robotic; Inspection; Data

The broader impact/commercial potential of this Small Business Innovation Research (SBIR) Phase I project is that the proposed innovative research has the potential to be developed into a self-contained robotic inspection tool with vertical mobility that carries an RGB-D camera and ground penetrating radar (GPR) to detect and characterize both surface flaws and subsurface defects. The software algorithms and functions will be integrated into this wall-climbing robot to automate the data collection and analysis process, especially at critical locations that are difficult to access by human operators. The use of the robotic inspection tool will allow the evaluation and condition health monitoring of human-built concrete structures to be performed significantly faster, more thoroughly and at a lower cost by eliminating the need for scaffolding and blocking traffic. It will also improve inspection safety and speed which leads to more frequent and on-demand inspections, thus making the national infrastructure (bridges, tunnels, dams, buildings) more secure. This Small Business Innovation Research (SBIR) Phase I project focuses on developing innovative methods and software algorithms for 3D GPR imaging of subsurface defects, vision-based accurate positioning and surface flaw detection, characterization and mapping. The software functions will be integrated into this wall-climbing robot to evaluate the performance and validate the feasibility of the innovation. The intellectual merit of this project includes the 3D GPR imaging method that combines robot control and vision-based accurate positioning with GPR signal processing to locate the subsurface defects and embedment (rebar, pipes, fractures, voids, delamination, etc.) in concrete structures that will revolutionize the way GPR data is collected, interpreted and displayed. This method enables the GPR-Rover to scan the surface in arbitrary and irregular trajectory rather than move along grid lines to locate subsurface targets and discover the areas of delamination. The proposed robotic visual inspection and machine learning algorithm is novel because it can not only detect and characterize surface flaws but also precisely register them on 3D map for better localization and visualization.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

这个小型企业创新研究（SBIR）第一阶段项目的更广泛的影响/商业潜力是，拟议的创新研究有可能开发成一个独立的机器人检测工具，具有垂直移动性，携带RGB-D相机和探地雷达（GPR），以检测和表征表面缺陷和地下缺陷。软件算法和功能将集成到这个爬壁机器人中，以自动化数据收集和分析过程，特别是在人类操作员难以访问的关键位置。机器人检测工具的使用将使人类建造的混凝土结构的评估和状态健康监测能够更快，更彻底，以更低的成本进行，消除了对脚手架和阻塞交通的需求。它还将提高检查的安全性和速度，导致更频繁和按需检查，从而使国家基础设施（桥梁，隧道，水坝，建筑物）更加安全。该小型企业创新研究（SBIR）第一阶段项目的重点是开发创新方法和软件算法，用于地下缺陷的3D GPR成像，基于视觉的精确定位和表面缺陷检测，表征和绘图。将软件功能集成到该爬壁机器人中，以评估其性能并验证创新的可行性。该项目的智力价值包括3D GPR成像方法，该方法将机器人控制和基于视觉的精确定位与GPR信号处理相结合，以定位地下缺陷和埋置（钢筋，管道，裂缝，空隙，分层等）。这将彻底改变探地雷达数据的收集、解释和显示方式。该方法使GPR-Rover能够以任意和不规则的轨迹扫描表面，而不是沿着网格线移动，以定位地下目标并发现分层区域。提出的机器人视觉检测和机器学习算法是新颖的，因为它不仅可以检测和表征表面缺陷，还可以在3D地图上精确地将它们注册，以更好地定位和可视化。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Subsurface Pipes Detection Using DNN-based Back Projection on GPR Data

使用基于 DNN 的探地雷达数据反投影进行地下管道检测

• 发表时间: 2021
• 期刊: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision
• 作者: Feng, Jinglun;Yang, Liang;Wang, Haiyan;Tian, Yingli;Xiao, Jizhong.
• 通讯作者: Xiao, Jizhong.

Deep Neural Network based Visual Inspection with 3D Metric Measurement of Concrete Defects using Wall-climbing Robot

• DOI: 10.1109/iros40897.2019.8968195
• 发表时间: 2019-11
• 期刊: 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
• 作者: Liang Yang;Bing Li;Guoyong Yang;Yong Chang;Zhaoming Liu;Biao Jiang;Jizhong Xiao

GPR-based Subsurface Object Detection and Reconstruction Using Random Motion and DepthNet

使用随机运动和深度网络进行基于探地雷达的地下物体检测和重建

• 发表时间: 2020
• 期刊: 2020 IEEE International Conference on Robotics and Automation
• 作者: Feng, Jinglun;Yang, Liang;Wang, Haiyan;Song, Yifeng;Xiao, Jizhong.
• 通讯作者: Xiao, Jizhong.

Automated wall‐climbing robot for concrete construction inspection

• DOI: 10.1002/rob.22119
• 发表时间: 2022-09
• 期刊: Journal of Field Robotics
• 影响因子: 8.3
• 作者: Liang Yang;Bing Li;Jinglun Feng;Guoyong Yang;Yong Chang;Bo Jiang;Jizhong Xiao