Recent Advances in UAV-Based SLAM: A Survey

With the rapid development of unmanned aerial vehicle (UAV) technologies, simultaneous localization and mapping (SLAM) has emerged as a key enabling paradigm for autonomous navigation and environmental perception. This paper presents a comprehensive survey of recent trends in UAV-based SLAM. First, we review the fundamental components of UAV-based SLAM systems, including commonly used onboard sensors and front-end odometry methods such as visual odometry, visual-inertial odometry, and LiDAR-inertial odometry, which provide reliable ego-motion estimation. Next, we summarize back-end methodologies that enhance estimation accuracy and global consistency, covering pose graph optimization, 3D reconstruction techniques, filter-based SLAM, fusion-based multi-UAV SLAM, and emerging blockchain-enabled frameworks for secure and distributed mapping. Furthermore, representative application scenarios are discussed, including a newly proposed category of collaborative UAV-SLAM platforms, termed the UAV-plus system. We also identify several remaining research gaps and highlight future research directions, including observability challenges, operation in imperfect and dynamic environments, and the miniaturization of UAV-based SLAM systems.