3D Map Building Research Articles

Robot@Home, a robotic dataset for semantic mapping of home environments

This paper presents the Robot-at-Home dataset (Robot@Home), a collection of raw and processed sensory data from domestic settings aimed at serving as a benchmark for semantic mapping algorithms through the categorization of objects and/or rooms. The dataset contains 87,000+ time-stamped observations gathered by a mobile robot endowed with a rig of four RGB-D cameras and a 2D laser scanner. Raw observations have been processed to produce different outcomes also distributed with the dataset, including 3D reconstructions and 2D geometric maps of the inspected rooms, both annotated with the ground truth categories of the surveyed rooms and objects. The proposed dataset is particularly suited as a testbed for object and/or room categorization systems, but it can be also exploited for a variety of tasks, including robot localization, 3D map building, SLAM, and object segmentation. Robot@Home is publicly available for the research community at http://mapir.isa.uma.es/work/robot-at-home-dataset .

Corrigendum to “A Novel Improved Probability-Guided RANSAC Algorithm for Robot 3D Map Building”

Corrigendum to “A Novel Improved Probability-Guided RANSAC Algorithm for Robot 3D Map Building”

3D Map Building Using Mobile Robot with Scan Device

3D Map Building Using Mobile Robot with Scan Device

A Novel Improved Probability-Guided RANSAC Algorithm for Robot 3D Map Building

This paper presents a novel improved RANSAC algorithm based on probability and DS evidence theory to deal with the robust pose estimation in robot 3D map building. In this proposed RANSAC algorithm, a parameter model is estimated by using a random sampling test set. Based on this estimated model, all points are tested to evaluate the fitness of current parameter model and their probabilities are updated by using a total probability formula during the iterations. The maximum size of inlier set containing the test point is taken into account to get a more reliable evaluation for test points by using DS evidence theory. Furthermore, the theories of forgetting are utilized to filter out the unstable inliers and improve the stability of the proposed algorithm. In order to boost a high performance, an inverse mapping sampling strategy is adopted based on the updated probabilities of points. Both the simulations and real experimental results demonstrate the feasibility and effectiveness of the proposed algorithm.

Passive Sensor Integration for Vehicle Self-Localization in Urban Traffic Environment.

This research proposes an accurate vehicular positioning system which can achieve lane-level performance in urban canyons. Multiple passive sensors, which include Global Navigation Satellite System (GNSS) receivers, onboard cameras and inertial sensors, are integrated in the proposed system. As the main source for the localization, the GNSS technique suffers from Non-Line-Of-Sight (NLOS) propagation and multipath effects in urban canyons. This paper proposes to employ a novel GNSS positioning technique in the integration. The employed GNSS technique reduces the multipath and NLOS effects by using the 3D building map. In addition, the inertial sensor can describe the vehicle motion, but has a drift problem as time increases. This paper develops vision-based lane detection, which is firstly used for controlling the drift of the inertial sensor. Moreover, the lane keeping and changing behaviors are extracted from the lane detection function, and further reduce the lateral positioning error in the proposed localization system. We evaluate the integrated localization system in the challenging city urban scenario. The experiments demonstrate the proposed method has sub-meter accuracy with respect to mean positioning error.

2차원 레이저 거리계를 이용한 수직/수평 다각평면 기반의 위치인식 및 3차원 지도제작

2차원 레이저 거리계를 이용한 수직/수평 다각평면 기반의 위치인식 및 3차원 지도제작

Implementing Kinect Sensor for Building 3D Maps of Indoor Environments

pioneer 3-dx is the robot used in this projects that is programmed by using the Advanced Robotics Interface for Applications (ARIA) that program with C++ package ( Visual C++.Net ), and ARNetworking software is used for setup Wireless TCP/IP Ethernet-to-Serial connection between robot and PC.

Indoor partial 3D map building based on Kinect depth sensor

Indoor partial 3D map building based on Kinect depth sensor

Digitalización del entorno a partir de un LIDAR HDL-64E

En este trabajo se propone un sistema de reconstrucción de mapas para un vehículo autónomo equipado con tecnología LIDAR, capaz de obtener más de un millón de puntos por segundo. Dentro del artículo proponemos la construcción rápida de Mapas Locales (ML) 2D que nos servirán para la navegación local autónoma, y la construcción de Mapas Globales (MG) 2D y 3D que modelizán el ambiente recorrido por el vehículo. Durante el proceso de construcción del Mapa Global se calcula la localización del vehículo con respecto a su posición inicial.DOI: http://dx.doi.org/10.5377/nexo.v25i1.795 Nexo Revista Científica,Vol. 25, No. 01, pp.28-37/Junio 2012

모바일 로봇의 경사 주행 시 3차원 지도작성 알고리즘

This paper proposes a 3D map-building algorithm using one LRF (Laser Range Finder) while a mobile robot is navigating on the slanted surface. There are several researches on 3D map buildings using the LRF. However most of them are performing the map building only on the flat surface. While a mobile robot is moving on the slanted surface, the view angle of LRF is dynamically changing, which makes it very difficult to build the 3D map using encoder data. To cope with this dynamic change of the view angle in build 3D map, IMU and balance filters are fused to correct the unstable encoder data in this research. Through the real navigation experiments, it is verified that the fusion of multiple sensors are properly performed to correct the slope angle of the slanted surface. The effectiveness of the balance filter are also checked through the hill climbing navigations.

Mobile robot 3D map building and path planning based on multi-sensor data fusion

3D map building and path planning serve as two essential tasks for mobile robot to work within a complex outdoor environment. An elevation map built from 3D laser points is utilised to extract terrain feature while ground surface character is acquired from visual information by matching characteristics vectors. By projecting the units of the elevation map into the image, the fusion of terrain feature and ground surface character is achieved using statistical method. Then the units which fuse multi-sensor information are evaluated and given traversable weights to extract constraints for autonomous path planning in outdoor scene. After clustering the units into several regions, a hierarchical path planning strategy uses A* and Probabilistic Roadmap Method (PRM) to plan region paths and unit paths, respectively. The PRM is improved by choosing the units with the largest traversable weights in the regions to ensure a uniform distribution instead of random distribution. Experiment results show the validity and practicability of the proposed approaches.

Attention-driven monocular scene reconstruction for obstacle detection, robot navigation and map building

In this paper, we present a feature-based approach for monocular scene reconstruction based on Extended Kalman Filters (EKF). Our method processes a sequence of images taken by a single camera mounted frontally on a mobile robot. Using a combination of various techniques, we are able to produce a precise reconstruction that is free from outliers and can therefore be used for reliable obstacle detection and 3D map building. Furthermore, we present an attention-driven method that focuses the feature selection to image areas where the obstacle situation is unclear and where a more detailed scene reconstruction is necessary. In extensive real-world field tests we show that the presented approach is able to detect obstacles that are not seen by other sensors, such as laser range finders. Furthermore, we show that visual obstacle detection combined with a laser range finder can increase the detection rate of obstacles considerably, allowing the autonomous use of mobile robots in complex public and home environments.

Wide-area augmented reality using camera tracking and mapping in multiple regions

We show how a system for video-rate parallel camera tracking and 3D map-building can be readily extended to allow one or more cameras to work in several maps, separately or simultaneously. The ability to handle several thousand features per map at video-rate, and for the cameras to switch automatically between maps, allows spatially localized AR workcells to be constructed and used with very little intervention from the user of a wearable vision system. The user can explore an environment in a natural way, acquiring local maps in real-time. When revisiting those areas the camera will select the correct local map from store and continue tracking and structural acquisition, while the user views relevant AR constructs registered to that map. The method is shown working in a progressively larger environments, from desktop to large building.

Mobile Robot Self Localization and 3D Map Building using a 3D PMD-Camera for tele-robotic applications

AbstractThis work purposes using a 3D PMD camera as a sensor for mobile robot applications. The use of image sensors for environment detection is substantial and common usage in mobile robot exploration. Laser scanners, 3D stereo vision or a combination of them are be used generally. The PMD camera is a novel 3D measuring system, which is based on a time-of-flight principle. The main feature is an array sensor, which can measure the distance to the target without scanning. This relatively new technology offers a cost-effective alternative to the mentioned sensors. In the following we present methods for self-localization and 3D map building by remote sensor data acquisition. This allows the teleoperative control of a mobile robot in an unknown environment. To enhance the mobile robot tasks, a CCD and the PMD camera are registered in order to create the three dimensional mapping. The CCD texture data will be overlaid with the PMD depth data to increase the accuracy of the lateral resolution. The visual input from the CCD camera not only delivers high resolution texture data, it can also apply for object recognition in clutter environment.

Path planning and traversable area marking for stereo vision-based 3D map building

This paper deals with the path-planning of a robot during the exploration and 3D map building in the unknown environment. The robot uses a stereocamera for significant features detection, then matches them in both images and computes their 3D coordinates. This way the robot incrementally creates the 3D map of its surroundings and tries to explore as much area as possible. This paper describes the way of marking the traversable path in the map created so far and the path planning to the next explored goal point. As the map uses points and triangles to represent shape of the surface, the traversable area marking and the path planning is also done in the map which is represented in the same way.

A sound exposure level study near the Camden Sound area, Browse Basin

A common approach to environmental risk assessment when assessing the level of risk to marine animals during marine seismic surveys is to compare either measured or modelled estimates of underwater sound levels to a known sound level threshold criteria. Over the years, a number of measures have been used to calculate and characterise underwater sound thresholds, including: sound pressure level (SPL), a root mean square (RMS) calculation, and—more recently—sound exposure level (SEL). SEL was suggested in 2007 by an expert panel of researchers as an appropriate measure to reduce the risk of temporary threshold shift (TTS) occurring in marine animals. Petroleum Geo-Services (PGS) commissioned an SEL study in preparation for a large MultiClient 3D seismic survey planned for the environmentally sensitive Camden Sound area, Western Australia, in late 2010. The Centre for Marine Science and Technology (CMST) at Curtin University built a comprehensive bathymetric model of the Camden Sound area and used two seabed geoacoustic models, with sand/limestone—as worst case, most reflective—and silt—as best case, least reflective. Acoustic propagation modelling over the frequency range of 8–1 kHz was used to build 3D maps of sound levels throughout all areas relevant to the seismic survey. The CMST SEL modelling results were integrated with other airgun array and SEL modelling pursued by PGS, and represent part of a broader environmental management strategy by PGS. Discussion is also given to other variants of SEL modelling.

A novel 3D sensory system for robot-assisted mapping of cluttered urban search and rescue environments

In this paper, the first application of utilizing a unique 3D sensor for sequential 3D map building in unknown cluttered urban search and rescue (USAR) environments is proposed. The sensor utilizes a digital fringe projection and phase shifting technique to provide real-time 2D and 3D sensory information of the environment. The proposed sensor is unique over current technologies in that high-resolution 3D information of rubble filled environments can be acquired from the single sensor at a speed of 30 frames per second (fps). Furthermore, we propose the development of a novel robust and reliable landmark identification technique that utilizes both 2D and 3D depth images taken by the sensor for 3D mapping. Preliminary experiments show the potential of the real-time 3D sensory system and landmark identification scheme for robotic 3D mapping in unknown cluttered USAR-like environments.

Integrating Route and Survey Learning in Complex Virtual Environments: Using a 3D Map

This study aimed to explore how a three-dimensional (3D) map affects the acquisition of survey knowledge in complex virtual environments (VEs). Subject navigated in virtual and real subway stations, where the virtual station was augmented with either a 3D floor map, a 3D building map, or neither. After subject's navigation in the station, they made the judgment of spatial relative directions between objects. The analyses of response time showed that participants responded faster when they were assisted by the 3D floor or building map than without a map aid in virtual stations. There was no significant difference between survey knowledge acquired in the virtual station augmented with the 3D floor or building map and in the real station. Participants with the 3D floor map performed best in spatial judgments. Implications of these findings to a 3D map design were discussed.

Spatial learning in a virtual multilevel building: Evaluating three exocentric view aids

The present study explores how the design of the exocentric view aid affects the acquisition of survey knowledge in virtual environments. The exocentric view was provided by either a 3D floor map, a 3D building map or the elevation of viewpoint in air. Participants navigated a virtual multilevel building and their survey knowledge was measured by the judgment of spatial relative direction. The results showed that (1) the accuracy of spatial judgment along the horizontal direction and response time were improved for participants with the exocentric view aid; (2) the accuracy of spatial judgment along the vertical direction was worst in the condition with a 3D floor map; (3) in general participants with a 3D building map performed best. The data suggested that the large scale of an exocentric view aid and the increased number of exocentric perspective through which the spatial layout is observed can facilitate the acquisition of survey knowledge in a virtual building. Potential applications of the findings include the design of a 3D map for navigation in both real and virtual buildings.

PVHL-based Local 3D Map Building

D map building is an important task of autonomous land vehicle(ALV) for obstacle detection, path planning. This paper analyzes the defects of the traditional 3D map building method, and then proposes a novel method based on projection of virtual height line(PVHL). The method projects virtual height line(VHL) into the stereo vision system by transforming the world coordinate of each point on VHL into its image coordinate, and generates two projection lines(PLs) in two images. On the two PLs, only the correspondence that is the projection of the intersection point of VHL and the surface of the scene has the maximum similarity metric. Then finding the height of the scene point is equivalent to proving whether the real correspondence of VHL has the maximum similarity metric or not. Experiments show that the proposed method is effective. And compared with that of the traditional, the results show that the proposed avoids unnecessary computation time and has better performance than the traditional, and can build the interested region only.