On measuring the accuracy of SLAM algorithms

By Rainer Kümmerle, Bastian Steder, C. Dornhege, Michael Ruhnke, Giorgio Grisetti, Cyrill Stachniss, A. Kleiner. 2009

358 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Control and Systems Engineering

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 Control and Systems Engineering

Citations in the corpus

Simitate: A Hybrid Imitation Learning Benchmark (2019, positive): “This metric is inspired by suggestions for the accuracy of SLAM systems [50], [15].”
Generic 3D Representation via Pose Estimation and Matching (2016, background): “The angular evaluation metric is the standard overall angular error [33,20], defined as the angle between the predicted pose vector and the ground truth vector in the plane defined by their cross product.”
Switchable constraints for robust pose graph SLAM (2012, methodology): “3 reveal that the value of ξ = Ξij indeed influences the quality of the optimization result which is measured by the RPE metric.”
Evaluating Egomotion and Structure-from-Motion Approaches Using the TUM RGB-D Benchmark (2012, background): “of SLAM systems [45], [46] which is especially useful if only sparse, relative relations are available as ground truth.”
Double window optimisation for constant time visual SLAM (2011, background): “[10] and define a relative error in terms of the relative differences ∆Tij := TiT j between two absolute poses.”
Real-time loop closure in 2D LIDAR SLAM (2016, methodology): “TABLE III QUANTITATIVE COMPARISON OF ERROR WITH [9] The most significant differences between all data sets is the frequency and quality of the laser scans as well as the availability and quality of odometry.”
On behalf of: Multimedia Archives (2012, neutral): “…dozen to a few hundreds ; examples include 20 (Fairfield et al., 2006; Tomono, 2007), 30–40 (Stachniss et al., 2005), 20–160 (Welle et al., 2010), 50 (Kuemmerle et al., 2009), up to 100 (Grisetti et al., 2005), 100–200 (Marks et al., 2009), 200 (Schroeter and Gross, 2008), 250 (Koenig et al.,…”
Are we ready for autonomous driving? The KITTI vision benchmark suite (2012, neutral): “Thus often only qualitative results are presented, with the notable exception of laser-based SLAM [28].”
Multiple‐Robot Simultaneous Localization and Mapping: A Review (2016, methodology): “For instance, (Kümmerle et al., 2009) propose a method that relies on the relative geometric relations between the poses of the robot.”
Towards a benchmark for RGB-D SLAM evaluation (2011, methodology): “For evaluating visual SLAM algorithms on our dataset, we propose a metric similar to the one introduced by [10].”
A benchmark for the evaluation of RGB-D SLAM systems (2012, background): “All sequences and movies are available on our website [1].”
A Tutorial on Quantitative Trajectory Evaluation for Visual(-Inertial) Odometry (2018, background): “However, as recognized by several researchers [5], [6], [7], ATE is sensitive to the time when the error occurs.”