Investigation of vibration serviceability of a footbridge using computer vision-based methods

Abstract In current standards and codes for vibration serviceability assessment, the serviceability limits are generally defined by acceleration related criteria or indices such as root-mean-square (RMS) or peak value of acceleration time histories. The accelerations are collected by using conventional accelerometers, which have several drawbacks such as traffic closure, setup time, and labor force to deal with the cable wiring work. It may not be convenient or practical to conduct such monitoring, especially for certain field applications. This study proposes to assess the vibration serviceability of a footbridge using computer vision-based methods. The proposed vision-based approach shows great advantages such as non-contact, long distance, low cost, time saving, and ease of use. The proposed approach is validated by a series of experiments on a footbridge under various types of pedestrian loading including walking, jumping and running with different paces (beat per minute). Displacement and velocity are first estimated from the image sequence and then converted to acceleration for the assessment of vibration serviceability. The feasibility of the proposed approach is verified by the comparative analysis between the results of serviceability assessments using the proposed approach and the conventional accelerometers. Consideration and recommendations of the process of converting displacement/velocity to acceleration are also discussed. The proposed approach provides a promising and efficient alternative for the vibration serviceability assessment of footbridges combined with the current standards and codes. As a result, it provides another approach for the serviceability assessment using different data type such as displacement and velocity.