The SPmap: a probabilistic framework for simultaneous localization and map building

By José A. Castellanos, José M. M. Montiel, José Neira, Juan D. Tardós. 1999

This article describes a rigorous and complete framework for the simultaneous localization and map building problem for mobile robots: the symmetries and perturbation map (SPmap), which is based on a general probabilistic representation of uncertain geometric information. We present a complete experiment with a LabMate/sup TM/ mobile robot navigating in a human-made indoor environment and equipped with a rotating 2D laser rangefinder. Experiments validate the appropriateness of our approach and provide a real measurement of the precision of the algorithms.

Subfields: Probabilistic & Bayesian Methods, Robotics & Control

How researchers see this paper

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

In Probabilistic & Bayesian Methods

In Robotics & Control

Citations in the corpus

Square Root SAM: Simultaneous Localization and Mapping via Square Root Information Smoothing (2006, methodology): “Below we assume familiarity with EKF-based approaches to SLAM (Smith et al. 1990; Leonard et al. 1992; Castellanos et al. 1999; Dissanayake et al. 2001).”
A Probabilistic On-Line Mapping Algorithm for Teams of Mobile Robots (2001, methodology): “The SLAM family of approaches (Dissanayake, Durrant-Whyte, and Bailey 2000; Leonard, Durrant-Whyte, and Cox 1992; Moutarlier and Chatila 1989b; Smith, Self, and Cheeseman 1990), to date the only approach that implements the full posterior calculation as specified in eq.”
Map building with mobile robots in dynamic environments (2003, background): “In the literature, the mobile robot mapping problem is often referred to as the simultaneous localization and mapping problem (SLAM) [5] [7] [13] [17] [14] [10] [19].”
Robotic mapping: a survey (2003, background): “In the following years, a number of researchers developed this approach further [13, 14, 30, 26, 27, 56, 73], most notably a group of researchers located at the University of Sydney in Australia.”
Factor Graphs for Robot Perception (2017, background): “Many attempts were made to extending the filtering approach to cope with larger-scale environments [26, 52, 130, 91, 184, 188, 92], but filtering itself was shown to be inconsistent [105] when applied to the inherently nonlinear SLAM problem.”
Mobile Robot Localisation and Mapping in Extensive Outdoor Environments (2002, methodology): “The underlying estimation process used in [24] is a stochastic SLAM variant called the symmetries and perturbations map (SP map) [25].”
Outdoor SLAM using visual appearance and laser ranging (2006, methodology): “They subsequently extract planar features from the scans and maintain parameterized estimates in a probabilistic manner using the SP formulation [ 2 ].”
A Tutorial on Graph-Based SLAM (2010, methodology): “Popular techniques like Kalman and information ﬁlters [28], [3], particle ﬁlters [22], [12], [9], or information ﬁlters [7], [31] fall into this category.”
Hierarchical SLAM: real-time accurate mapping of large environments (2005, methodology): “At each step, data association is robustly solved using the joint compatibility test [22], and the local map is updated with standard stochastic mapping techniques [23].”
Coordinated multi-robot exploration (2005, methodology): “If they use the recursive update rule for occupancy grids to combine and the information originally contained in is integrated twice in the resulting map, which is not admissible.”
Robust Mapping and Localization in Indoor Environments Using Sonar Data (2002, methodology): “Given that the full covariance matrix must be maintained to assure consistency, updating a full stochastic map of n features is well known to be O(n(2)) (Castellanos et al. 1999; Guivant and Nebot 2001).”
Probabilistic Algorithms and the Interactive Museum Tour-Guide Robot Minerva (2000, methodology): “One popular family of approaches, known as SLAM algorithms [22, 23, 84, 85, 134], employs Kalman lters [72, 92] for concurrently estimating robot poses and maps.”