This project proposed an unmanned aerial vehicle (UAV)-based method to measure the change in magnetic field as a result of the movement of a smart rock to monitor the maximum depth of bridge scour. A high resolution 3-axis magnetometer and global positioning system (GPS) unit were installed on the UAV to obtain the magnetic field intensity and its corresponding coordinate. The optimization algorithm for the localization of smart rock was demonstrated and adopted. The impact factors on accurate smart rock positioning were experimentally investigated. Utilizing finite element method, the influence of steel reinforcing bars on the distribution of ambient magnetic field was demonstrated, which agree well with experimental results. The UAV-based smart rock technology was implemented at the Roubidoux Creek Bridge and tested for six times since January 24, 2018. The overall smart rock positioning error, compared with ground truth (survey) data, is less than 0.5 m, even during an flood event. The performance of proposed method was compared favorably with the conventional crane-based positioning method.