Underwater environments are tough for mapping and navigation systems. To solve this, the researchers introduce AQUA-SLAM, a new system that combines data from three sources—a sonar-like device (DVL), a stereo camera, and motion sensors (IMU)—to help underwater vehicles know where they are and build maps more accurately. They tested AQUA-SLAM in a controlled tank and out at sea in the North Sea, and it outperformed other leading systems in accuracy and reliability. They’re also making the system open-source, so others can use and improve it.
Abstract: Underwater environments pose significant challenges for visual Simultaneous Localization and Mapping (SLAM) systems due to limited visibility, inadequate illumination, and sporadic loss of structural features in images. Addressing these challenges, this paper introduces a novel, tightly-coupled Acoustic-Visual-Inertial SLAM approach, termed AQUA-SLAM, to fuse a Doppler Velocity Log (DVL), a stereo camera, and an Inertial Measurement Unit (IMU) within a graph optimization framework. Moreover, we propose an efficient sensor calibration technique, encompassing multi-sensor extrinsic calibration (among the DVL, camera and IMU) and DVL transducer misalignment calibration, with a fast linear approximation procedure for real-time online execution. The proposed methods are extensively evaluated in a tank environment with ground truth, and validated for offshore applications in the North Sea. The results demonstrate that our method surpasses current state-ofthe-art underwater and visual-inertial SLAM systems in terms of localization accuracy and robustness. The proposed system will be made open-source for the community.
Authors: Shida Xu, S., Zhang, K., Wang, S.
Journal: arXiv
To read more, click here!