Map-Based Localization
Overview
Map-based localization estimates position and orientation by matching camera, LiDAR, radar, radio, or inertial observations to a reference map. Maps range from point clouds and 3D models to floor plans, road networks, satellite imagery, and compact semantic or topological representations; methods use retrieval and feature matching, geometric registration, particle filters, or pose-graph optimization. The field covers both global relocalization and continuous tracking, with benchmarks spanning indoor buildings, roads, forests, and aerial or GPS-denied settings. A central challenge is balancing accurate, reliable localization against compact maps and efficient computation, especially when scenes change, look alike, or provide incomplete observations.
Latest papers
106 papers16 papers in the last four weeks, up 14% on the four weeks before. 0.1% of all new papers.