Sonar based underwater vehicle dataset

In particular, we teleoperated the BlueROV and recorded readings from various sensors at $10$--$100$~Hz, including a high-resolution multibeam imaging sonar, an inertial measurement unit, and a Doppler velocity log (which also provides depth measurements).

In addition, the underwater motion capture system provided the ground-truth trajectories of the BlueROV at $53$~Hz, which is the key enabler to effectively quantify the performance of underwater simultaneous localization and mapping (SLAM) algorithms.The dataset contains three folders and a configuration file. Each folder contains a Rosbag dataset, which are circle, line and L-shape trajectory data respectively. Configuration file is the transformation matrix between each sensor.