Visual Sensing for Inspection

Private industries and government spend enormous time and cost in managing infrastructure and energy facilities. Predicting locations and degrees of degradations in any facility is a challenge even latest condition monitoring technologies. A key reason that limits our ability to predict degradation is related to our reliance on data from discrete sensor locations. As the system start degrading, the effectiveness of digital twins (DT) can be enhanced significantly if sensor data can be collected at additional locations. Such additional locations are different at different times due to changes in operating conditions and loads and differential degradation at different locations. However, flexibility to collect sensor data at additional locations does not exist. Moreover, much of the sensor data for assessing structural degradation is limited to acceleration data (vibrations) resulting from operating machinery, flow induced vibrations, or other environmental conditions. Collection of additional data (i.e., curvatures, strains, velocities, and even displacements) is either not reliable or is economically impractical.

Our inspection solution leverages various visual sensors (i.e., high resolution cameras, laser scanner, high-speed cameras, and thermal scanners) mounted on an unmanned ground vehicle (UGV) to capture geometrical deviations, structural vibrations, and thermal displacements, which then can be integrated with our digital twin (DT) solution. Such contactless visual sensing will allow extraction of a highly valuable and diverse set of information (geometrical difference, accelerations, curvatures, displacements, rotations, etc.) at any desired location.

The collected data will significantly enhance the quality and quantity of information available on the condition of facilities and infrastructure systems. The reduction in uncertainty associated with recorded data and availability of more diverse data will enable substantial advancement in the development of diagnosis and prognosis functionalities associated with DTs.