Marie-José Aldon

Line Extraction in 2D Range Images for Mobile Robotics

This paper presents a geometrical feature detection framework for use with conventional 2D laser rangefinders. This framework is composed of three main procedures: data pre-processing, breakpoint detection and line extraction. In data pre-processing, low-level data organization and processing are discussed, with emphasis to sensor bias compensation. Breakpoint detection allows to determine sequences of measurements which are not interrupted by scanning surface changing. Two breakpoint detectors are investigated, one based on adaptive thresholding, and the other on Kalman filtering. Implementation and tuning of both detectors are also investigated. Line extraction is performed to each continuous scan sequence in a range image by applying line kernels. We have investigated two classic kernels, commonly used in mobile robots, and our Split-and-Merge Fuzzy (SMF) line extractor. SMF employs fuzzy clustering in a split-and-merge framework without the need to guess the number of clusters. Qualitative and quantitative comparisons using simulated and real images illustrate the main characteristics of the framework when using different methods for breakpoint and line detection. These comparisons illustrate the characteristics of each estimator, which can be exploited according to the platform computing power and the application accuracy requirements.

A split-and-merge segmentation algorithm for line extraction in 2D range images

This paper presents a segmentation method for line extraction in 2D range images. It uses a prototype-based fuzzy clustering algorithm in a split-and-merge framework. The split-and-merge structure allows one to use the fuzzy clustering algorithm without any previous knowledge on the number of prototypes. This algorithm aims to be used in mobile robots navigation systems for dynamic map building. Simulation results show its good performance compared to some classical approaches.

Optimal mobile robot pose estimation using geometrical maps

We propose a weighted least-squares (WLS) algorithm for optimal pose estimation of mobile robots using geometrical maps as environment models. Pose estimation is achieved from feature correspondences in a nonlinear framework without linearization. The proposed WLS approach yields optimal estimates in the least-squares sense, is applicable to heterogeneous geometrical features decomposed in points and lines, and has an O(N) computation time.

Robustified estimation algorithms for mobile robot localization based on geometrical environment maps

Abstract This paper presents an improved weighted least-squares algorithm used for optimal 2D pose estimation of mobile robots navigating in real environments represented by geometrical maps. Following this map representation paradigm, feature matching is an important step in pose estimation. In this process, false feature matches may be accepted as reliable. Thus, in order to provide reliable pose estimation even in the presence of a certain level of false matches, robust M-estimators are derived. We further apply some concepts of outlier rejection for deriving a robust Kalman filter-based pose estimator. Extensive comparisons of the proposed robust methods with classic Kalman filtering-based approaches were carried out in real environments.

Color Constrained ICP for Registration of Large Unstructured 3D Color Data Sets

In this paper, we address the problem of pair-wise registration of large unstructured 3D/color datasets. Our purpose is to improve the classical ICP (Iterative closest point) algorithm by using color information, in order to deal with large datasets and with objects for which the geometric information is not significant enough. After a brief presentation of classical ICP (iterative closest point) algorithm and of the research works developed to improve its performance, we propose a new strategy to improve the selection of points. Color information is used to reduce the search space during the matching step. Experimental results obtained with real range images show that the algorithm provides an accurate estimation of the rigid transformation

Rough histograms for robust statistics

Applied statistics are widely used in pattern recognition and other computing applications to find the most likely value of a parameter. The use of classical empirical statistics is based upon assumption about normality of underlying density distribution of data. When the data is corrupted by contaminated noise, then classical tools are usually not robust enough and the estimation of the mode is biased. In this article, we propose to estimate the main mode of a distribution by means of a rough histogram and we show that this estimation is robust to contamination.

An optimal pose estimator for map-based mobile robot dynamic localization: experimental comparison with the EKF

Theoretical solutions based on the matching of 2D range measurements with a map of the environment have been proposed to solve the robot localization problem. However most of them have not been experimented with in real conditions: the robot was stopped or it moved slowly during range data acquisition, and the environment was supposed to be static. We propose and evaluate a dynamic localization method based on feature matching. Experiments carried out in real cluttered indoor environments including people and unknown obstacles show the good performance of the proposed algorithm against the classical solution based on Kalman filtering.

Possibility Theory and Rough Histograms for Motion Estimation in a Video Sequence

This article proposes to use both theories of possibility and rough histograms to deal with estimation of the movement between two images in a video sequence. A fuzzy modeling of data and a reasoning based on imprecise statistics allow us to partly cope with the constraints associated to classical movement estimation methods such as correlation or optical flow based-methods. The theoretical aspect of our method will be explained in details, and its properties will be shown. An illustrative example will also be presented.

Motion estimation by iterative 2D features matching in range images

This article describes an iterative algorithm for relative motion estimation from 2D range images. The matching is achieved in the geometric feature domain represented by straight lines and ellipsoidal clusters. Using infinite length straight lines instead of line segments as features, the algorithm attempts to achieve robustness to partial occlusions. The motion estimates and the feature correspondence measures are determined in order to minimize a cost function. The performance of this algorithm was evaluated with experiments carried out in real cluttered indoor environments.

A decoupled approach for simultaneous stochastic mapping and mobile robot localization

This paper introduces a decoupled approach of concurrent mapping and localization for mobile robots. Its theoretical aspects rely on recent techniques for correct uncertainty handling using stochastic models: covariance intersection and unscented transform. Further, stochastic constraints are considered as a way to minimize map incoherence with respect to the real environment. Experimental results obtained from multisensory data acquired in a large real environment illustrate the performance of the proposed method.