Wang Zhuo, Cheng Qin, Mu Xiaokai
Science and Technology on Underwater Vehicle Laboratory, Harbin Engineering University, Harbin 150001, China.
Qingdao Innovation and Development Center, Harbin Engineering University, Qingdao 266000, China.
Sensors (Basel). 2024 Mar 18;24(6):1937. doi: 10.3390/s24061937.
Accurate and robust simultaneous localization and mapping (SLAM) systems are crucial for autonomous underwater vehicles (AUVs) to perform missions in unknown environments. However, directly applying deep learning-based SLAM methods to underwater environments poses challenges due to weak textures, image degradation, and the inability to accurately annotate keypoints. In this paper, a robust deep-learning visual SLAM system is proposed. First, a feature generator named UWNet is designed to address weak texture and image degradation problems and extract more accurate keypoint features and their descriptors. Further, the idea of knowledge distillation is introduced based on an improved underwater imaging physical model to train the network in a self-supervised manner. Finally, UWNet is integrated into the ORB-SLAM3 to replace the traditional feature extractor. The extracted local and global features are respectively utilized in the feature tracking and closed-loop detection modules. Experimental results on public datasets and self-collected pool datasets verify that the proposed system maintains high accuracy and robustness in complex scenarios.
准确且稳健的同步定位与地图构建（SLAM）系统对于自主水下航行器（AUV）在未知环境中执行任务至关重要。然而，由于纹理薄弱、图像退化以及无法准确标注关键点，直接将基于深度学习的SLAM方法应用于水下环境会带来挑战。本文提出了一种稳健的深度学习视觉SLAM系统。首先，设计了一个名为UWNet的特征生成器，以解决纹理薄弱和图像退化问题，并提取更准确的关键点特征及其描述符。此外，基于改进的水下成像物理模型引入知识蒸馏的思想，以自监督方式训练网络。最后，将UWNet集成到ORB-SLAM3中以取代传统的特征提取器。提取的局部和全局特征分别用于特征跟踪和闭环检测模块。在公共数据集和自行采集的水池数据集上的实验结果验证了所提出的系统在复杂场景中保持了高精度和稳健性。
