Simultaneous Localization and Mapping

Also known as SLAM

Overview

SLAM estimates a robot’s changing position and a consistent map at the same time, enabling navigation where GPS or prior maps are unavailable. Systems combine camera, LiDAR, radar, sonar, and inertial measurements using feature or scan matching, sensor-fusion filters, and factor-graph optimization with loop closure to limit accumulated drift. Dense mapping increasingly uses neural fields, semantic scene graphs, and 3D Gaussian splatting, evaluated on indoor RGB-D benchmarks as well as outdoor driving and field-robot datasets. A central challenge is balancing accuracy, dense and useful maps, and real-time resource limits while remaining reliable in dynamic, low-texture, or changing environments.

Latest papers

CardsList

215 papers16 papers in the last four weeks, down 20% on the four weeks before. 0.1% of all new papers.

  1. 2Fast-2Lamaa: Large-Scale Lidar-Inertial Localization and Mapping with Continuous Distance Fields

    Oct 7, 2024Cedric Le Gentil, Raphael Falque, Daniil Lisus +1Simultaneous Localization and MappingInertial Odometry

  2. Generation of Uncertainty-Aware High-Level Spatial Concepts in Factorized 3D Scene Graphs via Graph Neural Networks

    Sep 18, 2024Jose Andres Millan-Romera, Muhammad Shaheer, Miguel Fernandez-Cortizas +3Simultaneous Localization and Mapping3D SGG

  3. A Survey on Reinforcement Learning Applications in SLAM

    Aug 26, 2024Mohammad Dehghani Tezerjani, Mohammad Khoshnazar, Mohammadhamed Tangestanizadeh +2Simultaneous Localization and MappingRL for Robotics

  4. NGD-SLAM: Towards Real-Time Dynamic SLAM without GPU

    May 12, 2024Yuhao Zhang, Mihai Bujanca, Mikel LujánSimultaneous Localization and MappingDynamic SLAM