Recent Advances in SLAM for Degraded Environments: A Review
Wenda Wang, Qiuzhao Zhang, Yongfeng Hu, Michal Gallay, Wen Zheng, Jianye Guo
Abstract
Simultaneous localization and mapping (SLAM) is a technology that relies on self-carried sensors, such as light detection and ranging (LiDAR), inertial measurement units (IMU), and cameras, to perform autonomous navigation positioning and mapping in unknown environments, characterized by non-contact operation, global coverage, and high precision. SLAM in optimal conditions is already a well-established field. However, there is still a significant need to develop SLAM techniques that can effectively operate in degraded scenarios, such as during sensor failures or perception degradation. This review article focuses on the recent advancement of SLAM, particularly in degraded environments including GNSS-denied, visually-degraded, and feature-correspondence degradation settings. This article first introduces the development of SLAM. Secondly, it details the progress of SLAM in single degraded environments, such as GNSS-denied, visually-degraded, and feature-correspondence degradation settings. Following this, the progress of SLAM in complex degraded environments such as mines, tunnels, and indoor-degraded environments is introduced. Finally, this article summarizes the advancements of SLAM technology in complex degraded environments and discusses potential future development trends. This detailed review article is of great significance for autonomous exploration and mapping in degraded environments.