The structural condition of bridges in Germany is often affected by aging and increased traffic loads. However, retrofitting or renewing all bridges is not always feasible due to limited resources. Indirect or drive‐by monitoring of bridges with sensors on passing vehicles has been explored as a cost‐effective and scalable alternative to direct monitoring. In this study, we propose a novel High‐Speed Monitoring (HSM) method for determining natural frequencies, based on accelerometers mounted on train axles, developed for speeds under normal operating conditions. In combination with Bayesian optimisation, a resonance curve can be generated using only a few bridge passages, from which the dominant bending natural frequency of a bridge can be determined with an accuracy comparable to direct monitoring. Measurements on an ICE‐TD passaging bridges travelling at high speeds of up to 200 km/h have shown that the natural frequencies can be determined with the same accuracy as direct measurements. Thus, in this study we were able to show that the method is not only accurate, cost‐effective and scalable, but can also be applied at high speeds without affecting operations.
