What’s this?
This is a repository of scientific papers whose experimental results were carried out using the MRPT library and tools. Along with each paper, you’ll also find the corresponding datasets, the source code used for the experiments and/or instructions/scripts to reproduce the results.
If you want to add a paper to this list, just contact us and we’ll include it.

See also: A direct search in Google Scholar for publications mentioning MRPT.

“A tutorial on SE(3) transformation parameterizations and on-manifold optimization”, J.L. Blanco, Technical report, 2010. (PDF, Bibtex). Updated: 14/AUG/2012.
Abstract: An arbitrary rigid transformation in SE(3) can be separated into two parts, namely, a translation and a rigid rotation. This technical report reviews, under a unifying viewpoint, three common alternatives to representing the rotation part: sets of three (yaw-pitch-roll) Euler angles, orthogonal rotation matrices from SO(3) and quaternions. It will be described: (i) the equivalence between these representations and the formulas for transforming one to each other (in all cases considering the translational and rotational parts as a whole), (ii) how to compose poses in each representation and (iii) how the uncertainty of the poses (when modeled as Gaussian distributions) is affected by these transformations and compositions. Some brief notes are also given about the Jacobians required to implement least-squares optimization on manifolds, an very promising approach in recent SLAM literature. The text reflects which MRPT C++ library functions implement each of the described algorithms. All the implementations have been thoroughly validated by means of unit testing. read more

Contributions to Localization, Mapping and Navigation in Mobile Robotics, PhD Thesis, Jose-Luis Blanco-Claraco, November 13th, 2009.
Downloads: PDF (12.6Mb) – Citation (Bibtex) – Slides (17.5 Mb) – Slides+videos (196 Mb)
Abstract: This thesis focuses on the problem of enabling mobile robots to autonomously build world models of their environments and to employ them as a reference to self–localization and navigation. For mobile robots to become truly autonomous and useful, they must be able of reliably moving towards the locations required by their tasks. This simple requirement gives raise to a myriad of problems that has populated research in the mobile robotics community for decades. Among these issues, two of the most relevant are: (i) secure autonomous navigation while avoiding collisions and (ii) the employment of an adequate world model for robot self-referencing within the environment and also for locating places of interest. In spite of presenting contributions regarding mobile robot navigation, the main focus of this thesis is on the latter problem, usually referred to as Simultaneous Localization and Mapping (SLAM). One of the most interesting contributions of this thesis is a novel approach to extend SLAM to large-scale scenarios by means of a seamless integration of geometric and topological map building in a probabilistic framework that estimates the hybrid metric-topological (HMT) state space of the robot path. The proposed framework unifies in an elegant manner the research areas of topological mapping, reasoning on topological maps and metric SLAM, providing also a natural integration of SLAM and the “robot awakening” problem. Other contributions presented in this thesis cover a wide variety of topics, such as optimal estimation in particle filters, a new probabilistic observation model for laser scanners based on consensus theory, a novel measure of the uncertainty in grid mapping, an efficient method for range-only SLAM, a grounded method for partitioning large maps into submaps, a multi-hypotheses approach to grid map matching, and a mathematical framework for extending simple obstacle avoidance methods to realistic robots.
PhD Awards: read more

“An alternative to the Mahalanobis distance for determining optimal correspondences in data association”. J.L. Blanco, J. Gonzalez-Jimenez, J.A. Fernandez-Madrigal. IEEE Transactions on Robotics (T-RO), vol. 28, no.4, 980-986, 2012. (Bibtex, Draft PDF)

DOI: 10.1109/TRO.2012.2193706

Abstract: The most common criteria for determining data association rely on minimizing the squared Mahalanobis distance (SMD) between observations and predictions. We hold that the SMD is just a heuristic, while the alternative matching likelihood (ML) is the optimal statistic to be maximized. Thorough experiments undoubtedly confirm this idea, with false positive reductions of up to 16%. read more

“An Optimal Filtering Algorithm for Non-Parametric Observation Models in Robot Localization”, ICRA 2008, Pasadena (California, USA), May 19-23, 2008. (PDF) – (Slides)

Abstract: The lack of a parameterized observation model in robot localization using occupancy grids requires the application of sampling-based methods, or particle filters. This work addresses the problem of optimal Bayesian filtering for dynamic systems with observation models that cannot be approximated properly as any parameterized distribution, which includes localization and SLAM with occupancy grids. By integrating ideas from previous works on adaptive sample size, auxiliary particle filters, and rejection sampling, we derive a new particle filter algorithm that enables the usage of the optimal proposal distribution to estimate the true posterior density of a non-parametric dynamic system. Our solution avoids approximations adopted in previous approaches at the cost of a higher computational burden. We present simulations and experimental results for a real robot showing the suitability of the method for localization.

Bibtex:

@INPROCEEDINGS{blanco2008ofa,
     author = {Blanco, J.L. and Gonzalez, J. and Fern\'andez-Madrigal, J.A.},
      pages = {461--466},
      month = {May},
      title = {An Optimal Filtering Algorithm for Non-Parametric Observation Models in Robot Localization},
  booktitle = {IEEE International Conference on Robotics and Automation (ICRA'08)},
       year = {2008},
   location = {Pasadena, CA (USA)}
}

2. Source code

The “optimal sampling” method described in this paper is implemented in the class mrpt::slam::CMonteCarloLocalization2D, which is the main class used in the ready-to-use application pf-localization.

In particular, the optimal sampling algorithm for (global) localization is implemented in the method prediction_and_update_pfAuxiliaryPFOptimal.

3. Reproducing the experiments

First, grab the configuration and data files of the experiments from icra2008optsamp.zip and decompress it.

Then, execute:

$ pf-localization icra08_demo_opt.ini # for running the method proposed in this paper, or $ pf-localization icra08_demo_std.ini # for using the "standard" proposal distribution. read more

“Stereo vision-specific models for Particle Filter-based SLAM”, Robotic and Autonomous Systems – PDF.

Abstract:This work addresses the SLAM problem for stereo vision systems under the unified formulation of particle filter methods. In contrast to most existing approaches to visual SLAM, the present method does not rely on restrictive smooth camera motion models, but on computing incremental 6D pose differences from the image flow through a probabilistic visual odometry method. Moreover, our observation model, which considers both the 3D positions and the SIFT descriptors of the landmarks, avoids explicit data association between the observations and the map by marginalizing the observation likelihood over all the possible associations. We have experimentally validated our research with two experiments in indoor scenarios. read more

“Efficient Probabilistic Range-Only SLAM”, IROS 2008 – PDF – Slides PPT
Abstract: This work addresses Range-Only SLAM (RO-SLAM) as the Bayesian inference problem of sequentially tracking a vehicle while estimating the location of a set of beacons without any prior information. The only assumptions are the availability of odometry and a range sensor able of identifying the different beacons. We propose exploiting the conditional independence between each beacon distribution within a Rao-Blackwellized Particle Filter (RBPF) for maintaining independent Sum of Gaussians (SOGs) for each map element. It is shown then that a proper probabilistic observation model can be derived for online operation with no need for delayed initializations. We provide a rigorous statistical comparison of this proposal with previous work of the authors where a Monte-Carlo approximation was employed instead for the conditional densities. As verified experimentally, this new proposal represents a significant improvement in accuracy, computation time, and robustness against outliers. read more

Subjective Local Maps for Hybrid Metric-Topological SLAM, Robotics and Autonomous Systems, 2009 – (PDF).

Abstract: Hybrid maps where local metric sub-maps are kept in the nodes of a graph-based topological structure are gaining relevance as the focus of robot Simultaneous Localization and Mapping (SLAM) shifts towards spatial scalability and long-term operation. In this paper we examine the applicability of spectral graph partitioning techniques to automatically generate metric sub-maps by establishing groups in the sequence of observations gathered by the robot. One of the main aims of this work is to provide a probabilistically grounded interpretation of such a partitioning technique in the context of generating these local maps. We also discuss how to apply it to different kinds of sensory data (stereo images and laser range scans) and how to consider them simultaneously. An important feature of our approach is that it implicitly takes into account the intrinsic characteristics of the sensors, such as the sensor field of view, to perform the partitioning instead of applying heuristics supplied by a human as in other works, and thus the robot builds “subjective” local maps. The ideas presented here are supported by experimental results from a real mobile robot as well as simulations for statistical analysis. We discuss the effects of considering different combinations of sensors in the resulting clustering of the environment.

Bibtex:

@article{blanco2009slm,
  title={{Subjective local maps for hybrid metric-topological SLAM}},
  author={Blanco, JL and Gonz{\'a}lez, J. and Fern{\'a}ndez-Madrigal, J.A.},
  journal={Robotics and Autonomous Systems},
  volume={57},
  number={1},
  pages={64--74},
  year={2009},
  publisher={Elsevier}
}

2. Source code

The main method described in the paper, partitioning a map using a spectral graph-cut technique, is implemented in the class mrpt::slam::CIncrementalMapPartitioner. The generic algorithm for partitioning any graph given its adjacency matrix is implemented in mrpt::math::CGraphPartitioner.

There is a ready-to-use application, map-partition, which takes a map as input (a “.simplemap” file), and processes it to build the adjacency matrix and generate the partitions. It also shows graphically the original and the rearranged weight matrix (Ω in the paper).

3. The experiments

3.1. Statistical results

The statistical experiments reported in the paper have been obtained by invoking 150 times the following MRPT programs:

  $ simul-landmarks
  $ kf-slam ./partition_EKF-SLAM_test.ini

The first program simul-landmarks, takes its parameters from “config.ini” and generates a synthetic dataset “out.rawlog”.

Next, the application kf-slam executes the EKF filter and generates a file “OUT_KF-SLAM/ERRORS.txt” with an analysis of the information losses for each partitioning algorithm (refer to the paper for details).

The contents of the configuration files is:

config.ini

<br />; Parameters for the Landmark-Map Rawlog simulator.<br />[Params] nSteps=250 ; Number of steps to simulate<br />circularPath=0 ; Circular/Square paths<br />squarePathLength=10000 ; "Steps" or length of the square path (integer)<br />; If file is empty, generate them at random from parameters below:<br />; file contains one row X Y Z per landmark.<br />loadLandmarksFromFile=<br />odometryNoiseXY_std=0.01 ; In meters (1 sigma)<br />odometryNoisePhi_std_deg=0.1 ; In degs (1 sigma)<br />minSensorDistance=0<br />maxSensorDistance=5<br />fieldOfView_deg=180<br />stdRange=0.01 ; The "sigma" of the sensor in range<br />stdYaw_deg=0.1 ; The "sigma" of the sensor in yaw (degrees)<br />stdPitch_deg=0.0 ; The "sigma" of the sensor in pitch (degrees)<br />outFile=simLandmarkPath.rawlog<br />outDir=OUT <br /><br />[RANDOMSET_1]<br />; Number of landmarks if there is no file<br />nLandmarks=60<br />min_x=0<br />max_x=55<br />min_y=-2<br />max_y=2<br />min_z=0<br />max_z=0<br />  read more

Paper: J.L. Blanco, J. Gonzalez-Jimenez, J.A. Fernandez-Madrigal, “A Robust, Multi-Hypothesis Approach to Matching Occupancy Grid Maps”, Robotica, 2013 (DOI, draft PDF).

Abstract:This article presents a new approach to matching occupancy grid maps by means of finding correspondences between a set of sparse features detected in the maps. The problem is stated here as a special instance of generic image registration. To cope with the uncertainty and ambiguity that arise from matching grid maps, we introduce a modified RANSAC algorithm which searches for a dynamic number of internally consistent subsets of feature pairings from which to compute hypotheses about the translation and rotation between the maps. By providing a (possibly multi-modal) probability distribution of the relative pose of the maps, our method can be seamlessly integrated into large-scale mapping frameworks for mobile robots. This article provides a benchmarking of different detectors and descriptors, along extensive experimental results that illustrate the robustness of the algorithm with a 97% success ratio in loop-closure detection for ~1700 matchings between local maps obtained from four publicly available datasets. read more

Technical report: “Derivation and Implementation of a Full 6D EKF-based Solution to Range-Bearing SLAM”, Jose-Luis Blanco, Perception and Mobile Robots Research Group, University of Malaga, Spain. (new version with corrected Jacobians, soon!, old version: PDF)
This document describes the theory behind the application kf-slam. read more