ABSTRACT
The drive-by inspection methods have garnered significant attention in recent years, owing to their advantages in mobility, cost-effectiveness, and efficiency. However, in practical applications, even simple extraction of bridge frequencies from an ordinary/commercial vehicle proves to be challenging; components associated with road roughness and vehicle dynamics typically dominate the vehicle vibration response. Current research typically attempts to alleviate these influences by minimizing variability (e.g., driving in the same lane, using the same speed, etc.), and by employing custom vehicles, though such practices are often difficult to implement. Recent advancements have pivoted towards “crowd-sensing” or “fleet monitoring” approaches, which undoubtedly introduce greater variability, thus potentially making the aforementioned strategies less preferable. In light of this, this paper presents a novel coherence-based approach applied to crowd-sensing drive-by monitoring systems which leverage data from multiple vehicles. The fundamental idea is to identify bridge frequencies as common vibrational components across a variety of vehicles driving over it. It encourages the introduction of variability in drive-by measurements to filter bridge frequencies. The proposed method is explored through numerical experiments, incorporating Monte Carlo methods. The results highlight the method’s efficacy in eliminating the influences of road roughness, environmental noise, and vehicle parameter variations, with accurate identification of bridge frequencies. From the perspective of the method, this paper also illustrates the advantages of crowd-sensing drive-by monitoring and its practical feasibility in engineering problems.
