The Graph SLAM Algorithm with Applications to Large-Scale Mapping of Urban Structures

By Sebastian Thrun, Michael Montemerlo. Int. J. Robotics Res., 2006

784 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Environmental Engineering

View Paper: https://www.semanticscholar.org/paper/afe451172f1f0ad4b3555274afaaccc56f3a3921

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Environmental Engineering

Citations in the corpus

Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age (2016, neutral): “[125], [192], [241].”
An Overview to Visual Odometry and Visual SLAM: Applications to Mobile Robotics (2015, methodology): “GraphSLAM GraphSLAM [111] is a full-SLAM problem which means that a posterior (robot poses and map) is calculated over the entire path x 1 : t along with the map, instead of just the current pose calculation in online-SLAM [109] (such as the EKF-SLAM).”
Visual Place Recognition: A Survey (2016, background): “Thrun and Leonard [164] identify three key SLAM paradigms: extended Kalman filters (EKF) [37], [38], [165]–[167] and Rao- Blackwellized particle filters [168], as well as the pose graph approach discussed above [160], [161], [169]–[171].”
Keyframe-based visual–inertial odometry using nonlinear optimization (2015, background): “Figure 3 shows the respective graph representation inspired by (Thrun and Montemerlo, 2006): it displays measurements as edges with square boxes and estimated quantities as round nodes.”
Efficient Sparse Pose Adjustment for 2D mapping (2010, methodology): “The combination of an SBA/GraphSLAM optimizer with efficient methods for solving the linear subproblem has the following advantages.”
Geometry-Aware Learning of Maps for Camera Localization (2017, methodology): “In visual SLAM, 3D landmarks (with feature descriptors of the corresponding image patches) are often defined as maps in both Bayesian filtering approaches [14, 41, 53] and key-frame based approaches [34, 42, 49].”
Map-Based Precision Vehicle Localization in Urban Environments (2007, methodology): “Particles are projected forward through time via the velocity outputs from a tightly-coupled inertial navigation system, which relies on wheel odometry, an IMU and a GPS system for determining vehicle velocity.”
A Tutorial on Graph-Based SLAM (2010, methodology): “GraphSLAM [32] applies variable elimination techniques to reduce the dimensionality of the optimization problem.”
Mapping a Suburb With a Single Camera Using a Biologically Inspired SLAM System (2008, neutral): “arcs in algorithms such as GraphSLAM [46].”
Persistent Navigation and Mapping using a Biologically Inspired SLAM System (2010, methodology): “Graphical SLAM techniques represent the robot and landmark locations as nodes in a graph (Lu and Milios, 1997 Duckett et al., 2002 Folkesson and Christensen, 2004 Thrun and Montemerlo, 2006).”
LiDARsim: Realistic LiDAR Simulation by Leveraging the Real World (2020, methodology): “Multiple LiDAR sweeps are then associated to a common coordinate system (the map frame) using offline Graph-SLAM [37] with multi-sensor fusion leveraging wheel-odometry, IMU, LiDAR and GPS.”
A Survey of Motion Planning Algorithms from the Perspective of Autonomous UAV Guidance (2010, background): “[116] shows what part SLAM played in their DARPA Grand Challenge entry.”
Trajectory Optimization using Reinforcement Learning for Map Exploration (2008, background): “Further, these innovations have led to robots that can map large-scale, outdoor environments with a variety of different sensors and to the creation of commodity open-source community around mapping software (Haehnel et al. 2003; Thrun and Montemerlo 2005; Martinez-Cantin and Castellanos 2005; Marti”
One-Shot Reinforcement Learning for Robot Navigation with Interactive Replay (2017, methodology): “The space is then discretized into a pose graph [20] consisting of sensory snapshots taken every ∆pos meters, which are connected by edges representing temporal adjacency and loop closures.”