Collaborative Research: A Novel Video-Assisted Integrated Approach for Enhancing Bridge Health Monitoring

摘要合作项目0510507（M. Feng，加州/ Ivine）0510655（Chin-An Tang，韦恩Statye Univ）在桥梁结构的长期健康监测中，对评估桥梁整体状况至关重要的系统识别通常基于测量的振动响应或系统输出。 在实现综合健康监测系统的道路上仍然存在两个主要障碍。 首先，输入的知识（即，交通激励）是未知的或有限的，使得难以准确评估桥梁结构的状态。 其次，必须开发有效的程序来处理和解释来自传感器的大量数据。 本研究的重点是开发一种新的，视频辅助的方法，以加强桥梁健康监测。 重要的交通信息特征的输入激励，如车辆到达时间，速度和类型，建议从数字视频中提取。 由于摄像机系统在许多现有桥梁中很容易获得（作为监控摄像机），因此该提案代表了一种低成本的方法，可以显着提高我们对交通输入的理解。 用于处理和分析同步交通信息和桥梁振动数据的新算法将执行集成系统识别并改善健康监测这是两所大学之间的联合研究，包括成功完成该项目所需的专业知识。 来自华盛顿州立大学的Tan教授和Yin教授将负责（1）交通激励模型的开发和验证以及基于统计的桥梁损伤诊断，（2）视频模式识别的车辆分类和（3）基于网络的数据共享和演示用于教育目的的监测项目。 UCI的Feng教授将带头（1）在全仪表化的桥梁上安装防风雨摄像机和现场数据收集，（2）在芯片上开发实时算法，允许传输处理后的交通数据，以显着减少数据传输量，以及（3）基于系统输出和从视频中获得的部分输入信息开发系统识别算法。

Abstract Collaborative Projects0510507 (M. Feng, Univ California/ Ivine)0510655 (Chin-An Tang, Wayne Statye Univ)In the long-term health monitoring of bridge structures, system identification that is essential for assessing the global condition of the bridge is often based on measured vibration response or the system output. Two major obstacles remain in the path to realization of an integrated health monitoring system. First, knowledge of the input (i.e., traffic excitation) is unknown or limited, rendering an accurate assessment of the state of the bridge structures difficult. Second, efficient procedure must be developed for processing and interpreting the vast amount of data from sensors. This research focuses on developing a novel, video-assisted approach for enhancing bridge health monitoring. Important traffic information characterizing the input excitation, such as the vehicle arrival times, speeds and types, is proposed to be extracted from digital video. Since camera systems are readily available in many existing bridges (as surveillance cameras), this proposal represents a low-cost approach to significantly enhance our understanding of the traffic input. Novel algorithms for processing and analyzing synchronized traffic information and bridge vibration data will perform integrated system identification and improve the health monitoring This is a joint research between two universities encompassing the expertise necessary for the successful completion of this project. Professor Tan and Professor Yin from WSU will be in charge of (1) development and verification of the traffic excitation model and the bridge damage diagnosis on a statistical basis, (2) vehicle classification from the video pattern recognition and (3) web-based data sharing and presentation of the monitoring project for educational purpose. Professor Feng at UCI will take the lead in (1) installation of weather-proof video cameras and field data collection at a fully instrumented bridge, (2) development of real-time algorithms on a chip allowing transmission of processed traffic data to significantly reduce the amount of data transmission and (3) development of system identification algorithms based on system output and partial input information obtained from the video.