Abstract — This paper presents the extension and experimental validation of the widely used EKF 1-based SLAM 2 algorithm to 3D space. It uses planar features extracted probabilistically from dense three-dimensional point clouds generated by a rotating 2D laser scanner. These features are represented in compliance with the Symmetries and Perturbation model (SPmodel) in a stochastic map. As the robot moves, this map is updated incrementally while its pose is tracked by using an Extended Kalman Filter. After showing how three-dimensional data can be generated, the probabilistic feature extraction method is described, capable of robustly extracting (infinite) planes from structured environments. The SLAM algorithm is then used to track a robot moving through an indoor environment and its capabilities in terms of 3D reconstruction are analyzed.

Abstract — This paper presents an approach to generating compact 3D maps of urban environments using mobile robots and laser range finders. Our algorithm extracts planar information from 3D point cloud maps. The planar representation is very efficient for representing building structures in urban environments when a high level of detail is not required. We also present preliminary results on 3D geometric mapping with incomplete data. Based on previously known models and incomplete data, our system is able to estimate parts of buildings which have never been seen before. As validation we present experimental results using a Segway RMP vehicle in two environments, both approximately the size of a city block. I.

Objects within human environments are usually found on flat surfaces that are orthogonal to gravity, such as floors, tables, and shelves. We first present a new assistive robot that is explicitly designed to take advantage of this common structure in order to retrieve unmodeled, everyday objects for people with motor impairments. This compact, statically stable mobile manipulator has a novel kinematic and sensory configuration that facilitates autonomy and humanrobot interaction within indoor human environments. Second, we present a behavior system that enables this robot to fetch objects selected with a laser pointer from the floor and tables. The robot can approach an object selected with the laser pointer interface, detect if the object is on an elevated surface, raise or lower its arm and sensors to this surface, and visually and tacitly grasp the object. Once the object is acquired, the robot can place the object on a laser designated surface above the floor, follow the laser pointer on the floor, or deliver the object to a seated person selected with the laser pointer. Within this paper we present initial results for object acquisition and delivery to a seated, ablebodied individual. For this test, the robot succeeded in 6 out of 7 trials (86%). 1.

We describe a novel method for discovering and incorporating higher level map structure in a real-time visual SLAM system. Previous approaches use sparse maps, populated by isolated features such as 3-D points or edgelets. Although this facilitates efficient localisation, it yields very limited scene representation and ignores the inherent redundancy amongst features resulting from physical structure in the scene. In this work, higher level structure, in the form of lines and surfaces, is discovered concurrently with SLAM operation and then incorporated into the map in a rigorous manner, attempting to maintain important cross covariance information and allow consistent update of the feature parameters. This is achieved by using a bottom-up process, in which subsets of low level features are ‘folded in ’ to a parameterisation of an associated higher level feature, thus collapsing the state space as well as building structure into the map. We demonstrate and analyse the effects of the approach for the cases of line and plane discovery, both in simulation and within a real-time system operating with a hand-held camera in an office environment.

Abstract — Standardized middleware for autonomous robot control has proven itself to enable faster deployment of robots, to make robot control code more interchangeable, and experiments easier to replicate. Unfortunately, the support provided by current middleware is in most cases limited to what current robots do: navigation. However, as we tackle more ambitious service robot applications more comprehensive middleware support is needed. We increasingly need the middleware to support ubiquitous sensing infrastructures, robot manipulation tasks, and cognitive capabilities. In this paper we describe and discuss current extensions of the Player/Stage/Gazebo (P/S/G) middleware, one of the most widespread used robot middleware, of which we are active developers, that satisfy these requirements. I.

This paper addresses the question of generating large-scale 3d models from a set of 2d laser scans. We investigate two configurations: a rotating laser scanner and a setup of two orthogonally mounted laser range finders. Both systems are used to combine data from several viewpoints into a map using global registration. A feature-based and a raw-data based method for pose estimation and global registration are presented. These approaches are applied to data from indoor and outdoor scenes. For data acquisition the Biba-robot and a Smart-car were used. With additional information from a panoramic camera a textured 3d model of a part of EPFL campus was created. The precision of the resulting 3d model is evaluated by comparing it to an orthophoto of the environment. 1

Real-time 3D localization and mapping is eventually needed in many service robotic applications. Toward a light and practical SLAM algorithm, we focus on feature extraction via segmentation of range images. Using horizontal and vertical traces of the range matrix, 2D observed polygons are considered for calculation of a one-dimensional measure of direction, called Bearing Angle (BA). BA is the incident angle between the laser beam and edges of the observed polygon by the scanner in the selected direction. Based on this measure, two different approaches to range image segmentation, region- and edge-based, are proposed and evaluated through a set of standard analysis. It is experimentally shown that for navigation applications, edge based approaches are more efficient. Extensive tests on real robots prove BA-based segmentation is successful for SLAM. 1.

Abstract — This paper concerns context and feature-sensitive re-sampling of workspace surfaces represented by 3D point clouds. We interpret a point cloud as the outcome of repetitive and non-uniform sampling of the surfaces in the workspace. The nature of this sampling may not be ideal for all applications, representations and downstream processing. For example it might be preferable to have a high point density around sharp edges or near marked changes in texture. Additionally such preferences might be dependent on the semantic classification of the surface in question. This paper addresses this issue and provides a framework which given a raw point cloud as input, produces a new point cloud by re-sampling from the underlying workspace surfaces. Moreover it does this in a manner which can be biased by local low-level geometric or appearance properties and higher level (semantic) classification of the surface. We are in no way prescriptive about what justifies a biasing in the re-sampling scheme — this is left up to the user who may encapsulate what constitutes “interesting ” into one or more “policies ” which are used to modulate the default re-sampling behavior. I.

Abstract — The fast and accurate computation of surface normals from a point cloud is a critical step for many 3D robotics and automotive problems, including terrain estimation, mapping, navigation, object segmentation, and object recognition. To obtain the tangent plane to the surface at a point, the traditional approach applies total least squares to its small neighborhood. However, least squares becomes computationally very expensive when applied to the millions of measurements per second that current range sensors can generate. We reformulate the traditional least squares solution to allow the fast computation of surface normals, and propose a new approach that obtains the normals by calculating the derivatives of the surface from a spherical range image. Furthermore, we show that the traditional least squares problem is very sensitive to range noise and must be normalized to obtain accurate results. Experimental results with synthetic and real data demonstrate that our proposed method is not only more efficient by up to two orders of magnitude, but provides better accuracy than the traditional least squares for practical neighborhood sizes. I.

We present experimental results with uncertainty analysis method for a proposed distance measurement protocol for mid- range terrestrial laser scanners (TLS). Three-dimensional (3D) imaging systems evaluation is performed in the environmentally controlled laboratory at NRC-IIT with the use of specially designed reference test object (RTO), test procedures and mathematical methods. The procedure is based on the evaluation of point-to-point distance measurements. The distance measurements characterizing a system under test (SUT) are taken between comparative points of reference (CPOR) located on the RTO plane surfaces. Selected CPORs are the intersections of particular lines of measurements with estimated RTO plane surfaces. The virtual lines of measurement are determined by using a spherical RTO. Uncertainty analysis for fitted planes, virtual lines of measurements and CPORs is performed with our software utilities especially created for this purpose. Data on commercial products are provided for the sake of illustrating the proposed protocol. 1.