Skip to main content

Resultados relacionados

Resultados totales (Incluyendo duplicados): 47
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Calibration of an outdoor distributed camera network with a 3D point cloud

  • Ortega, Agustín
  • Silva, Manuel
  • Teniente, Ernesto H.
  • Ferreira, Ricardo
  • Bernardino, Alexandre
  • Gaspar, José
  • Andrade-Cetto, Juan
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution license., Outdoor camera networks are becoming ubiquitous in critical urban areas of the largest cities around the world. Although current applications of camera networks are mostly tailored to video surveillance, recent research projects are exploiting their use to aid robotic systems in people-assisting tasks. Such systems require precise calibration of the internal and external parameters of the distributed camera network. Despite the fact that camera calibration has been an extensively studied topic, the development of practical methods for user-assisted calibration that minimize user intervention time and maximize precision still pose significant challenges. These camera systems have non-overlapping fields of view, are subject to environmental stress, and are likely to suffer frequent recalibration. In this paper, we propose the use of a 3D map covering the area to support the calibration process and develop an automated method that allows quick and precise calibration of a large camera network. We present two cases of study of the proposed calibration method: one is the calibration of the Barcelona Robot Lab camera network, which also includes direct mappings (homographies) between image coordinates and world points in the ground plane (walking areas) to support person and robot detection and localization algorithms. The second case consist of improving the GPS positioning of geo-tagged images taken with a mobile device in the Facultat de Matemàtiques i Estadística (FME) patio at the Universitat Politècnica de Catalunya (UPC)., Agustín Ortega and Ernesto H. Teniente acknowledge support from the Mexican Council of Science and Technology. This work received support from the Agència de Gestiò d’Ajuts Universitaris i de Recerca of The Generalitat of Catalonia (2012 CTP00013), the Spanish Ministry of Economy and Competitiveness, project PAU+: Perception and Action in Robotics Problems with Large State Spaces (DPI-2011-27510), the EU project, Cargo-ANTs: Cargo Handling by Automated Next Generation Transportation Systems for Ports and Terminals (FP7-SST-2013-RTD-1 605598), the Portuguese Science Foundation FCT project DCCAL: Discrete Cameras Calibration using Properties of Natural Scenes (PTDC/EEACRO/105413/2008), and the Portuguese Quadro de Referéncia Estratégico Nacional project HDA: High Definition Analytics (QREN I&D em Co-Promoção 13750)., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Dense entropy decrease estimation for mobile robot exploration

  • Vallvé, Joan
  • Andrade-Cetto, Juan
Presentado al ICRA 2014 celebrado en Hong Kong del 31 de mayo al 7 de junio., We propose a method for the computation of entropy decrease in C-space. These estimates are then used to evaluate candidate exploratory trajectories in the context of autonomous mobile robot mapping. The method evaluates both map and path entropy reduction and uses such estimates to compute trajectories that maximize coverage whilst min- imizing localization uncertainty, hence reducing map error. Very efficient kernel convolution mechanisms are used to evaluate entropy reduction at each sensor ray, and for each possible robot position and orientation, taking frontiers and obstacles into account. In contrast to most other exploration methods that evaluate entropy reduction at a small number of discrete robot configurations, we do it densely for the entire C-space. The computation of such dense entropy reduction maps opens the window to new exploratory strategies. In this paper we present two such strategies. In the first one we drive exploration through a gradient descent on the entropy decrease field. The second strategy chooses maximal entropy reduction configurations as candidate exploration goals, and plans paths to them using RRT*. Both methods use PoseSLAM as their estimation backbone, and are tested and compared with classical frontier-based exploration in simulations using common publicly available datasets., This work has been supported by the Spanish Ministry of Economy and Competitiveness under Project DPI-2011-27510 and by the EU Project CargoAnts FP7-605598., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Localization in highly dynamic environments using dual-timescale NDT-MCL

  • Valencia, Rafael
  • Saarinen, Jari
  • Andreasson, Henrik
  • Vallvé, Joan
  • Andrade-Cetto, Juan
  • Lilienthal, Achim J.
Industrial environments are rarely static and often their configuration is continuously changing due to the material transfer flow. This is a major challenge for infrastructure free localization systems. In this paper we address this challenge by introducing a localization approach that uses a dual- timescale approach. The proposed approach - Dual-Timescale Normal Distributions Transform Monte Carlo Localization (DT- NDT-MCL) - is a particle filter based localization method, which simultaneously keeps track of the pose using an apriori known static map and a short-term map. The short-term map is continuously updated and uses Normal Distributions Transform Occupancy maps to maintain the current state of the environment. A key novelty of this approach is that it does not have to select an entire timescale map but rather use the best timescale locally. The approach has real-time performance and is evaluated using three datasets with increasing levels of dynamics. We compare our approach against previously pro- posed NDT-MCL and commonly used SLAM algorithms and show that DT-NDT-MCL outperforms competing algorithms with regards to accuracy in all three test cases., This work has been supported by the European Commission under contract number FP7-ICT-600877 (SPENCER) and by the Spanish Ministry of Economy and Competitiveness under Project DPI-2011-27510 and the EU Project CargoAnts FP7-605598., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Uncertainty analysis of the DLT-Lines calibration algorithm for cameras with radial distortion

  • Galego, Ricardo
  • Ortega, Agustín
  • Ferreira, Ricardo
  • Bernardino, Alexandre
  • Andrade-Cetto, Juan
  • Gaspar, José
3D metric data of environmental structures is nowadays present in many information sources (maps, GIS) and can be easily acquired with modern depth sensing technology (RGBD, laser). This wealth of information can be readily used for single view calibration of 2D cameras with radial distortion, provided that image structures can be matched with the 3D data. In this paper we present an analysis of the level of accuracy that can be obtained when such calibration is performed with the 2D-3D DLT-Lines algorithm. The analysis propagates uncertainty in the detection of features at the image level to camera pose, and from there to 3D reconstruction. The analytic error propagation expressions are derived using first order uncertainty models, and are validated with Monte Carlo simulations in a virtual indoor environment. The method is general and can be applied to other calibration methods, as long as explicit or implicit expressions can be derived for the transformation from image coordinates to 3D reconstruction. We present results with real data for two applications: i) the 3D reconstruction of an outdoors building for which 3D information is given by a map, observed by a mobile phone camera; and ii) the uncertainty in the localization at the floor plane of points observed by a fixed camera calibrated by a robot equipped with an RGBD camera navigating in a typical indoor environment., This work has been partially supported by the FCT project [UID/EEA/50009/2013], by the MINECO project DPI2011-27510, and by the EU Projects POETICON++EU-FP7-ICT-288382 and Cargo-ANTs FP7-SST-605598., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Terrain classification in complex three-dimensional outdoor environments

  • Santamaria-Navarro, Àngel
  • Teniente, Ernesto H.
  • Morta, Martí
  • Andrade-Cetto, Juan
This paper presents two techniques to detect and classify navigable terrain in complex three-dimensional (3D) environments. The first method is a low level on-line mechanism aimed at detecting obstacles and holes at a fast frame rate using a time-of-flight camera as the main sensor. The second technique is a high-level offline classification mechanism that learns traversable regions from larger 3D point clouds acquired with a laser range scanner. We approach the problem using Gaussian processes as a regression tool, in which the terrain parameters are learned, and also for classification, using samples from traversed areas to build the traversable terrain class. The two methods are compared against unsupervised classification, and sample trajectories are generated in the classified areas using a nonholonomic path planner. We show results of both the low-level and the high-level terrain classification approaches in simulations and in real-time navigation experiments using a Segway RMP400 robot., This work has been partially funded by the Spanish Ministry of Economy and Competitiveness under project PAU+ DPI2011-27510 and by the EU projects ARCAS FP7-287617 and Cargo-ANTs FP7-605598, Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Potential information fields for mobile robot exploration

  • Vallvé, Joan
  • Andrade-Cetto, Juan
We present a decision theoretic approach to mobile robot exploration. The method evaluates the reduction of joint path and map entropy and computes a potential information field in robot configuration space using these joint entropy reduction estimates. The exploration trajectory is computed descending on the gradient of this field. The technique uses Pose SLAM as its estimation backbone. Very efficient kernel convolution mechanisms are used to evaluate entropy reduction for each sensor ray, and for each possible robot orientation, taking frontiers and obstacles into account. In the end, the computation of this field on the entire configuration space is shown to be very efficient. The approach is tested in simulations in a pair of publicly available datasets comparing favorably both in quality of estimates and in execution time against an RRT∗-based search for the nearest frontier and also against a locally optimal exploration strategy., This work has been supported by the Spanish Ministry of Economy and Competitiveness under Project DPI-2011-27510 and by the EU Project CargoANTs FP7-605598., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Chromatic shadow detection and tracking for moving foreground segmentation

  • Huerta, Iván
  • Holte, Michael B.
  • Moeslund, Thomas B.
  • Gonzàlez, Jordi
Advanced segmentation techniques in the surveillance domain deal with shadows to avoid distortions when detecting moving objects. Most approaches for shadow detection are still typically restricted to penumbra shadows and cannot cope well with umbra shadows. Consequently, umbra shadow regions are usually detected as part of moving objects, thus affecting the performance of the final detection. In this paper we address the detection of both penumbra and umbra shadow regions. First, a novel bottom-up approach is presented based on gradient and colour models, which successfully discriminates between chromatic moving cast shadow regions and those regions detected as moving objects. In essence, those regions corresponding to potential shadows are detected based on edge partitioning and colour statistics. Subsequently (i) temporal similarities between textures and (ii) spatial similarities between chrominance angle and brightness distortions are analysed for each potential shadow region for detecting the umbra shadow regions. Our second contribution refines even further the segmentation results: a tracking-based top-down approach increases the performance of our bottom-up chromatic shadow detection algorithm by properly correcting non-detected shadows. To do so, a combination of motion filters in a data association framework exploits the temporal consistency between objects and shadows to increase the shadow detection rate. Experimental results exceed current state-of-the-art in shadow accuracy for multiple well-known surveillance image databases which contain different shadowed materials and illumination conditions., This work has been partially funded by the European project CargoANTs (FP7-SST-2013-605598), by the Spanish CICYT project DPI2013-42458-P and by the CICYT TIN2012-39051., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Multimodal feedback fusion of laser, image and temporal information

  • Huerta, Iván
  • Ferrer, Gonzalo
  • Herrero, Fernando
  • Prati, Andrea
  • Sanfeliu, Alberto
Trabajo presentado a la 8th International Conference on Distributed Smart Cameras (ICDSC) celebrada en Venecia (Italia) del 4 al 7 de noviembre de 2014., In the present paper, we propose a highly accurate and robust people detector, which works well under highly variant and uncertain conditions, such as occlusions, false positives and false detections. These adverse conditions, which initially motivated this research, occur when a robotic platform navigates in an urban environment, and although the scope is originally within the robotics field, the authors believe that our contributions can be extended to other fields. To this end, we propose a multimodal information fusion consisting of laser and monocular camera information. Laser information is modelled using a set of weak classifiers (Adaboost) to detect people. Camera information is processed by using HOG descriptors to classify person/non person based on a linear SVM. A multi-hypothesis tracker trails the position and velocity of each of the targets, providing temporal information to the fusion, allowing recovery of detections even when the laser segmentation fails. Experimental results show that our feedback-based system outperforms previous state-of-the-art methods in performance and accuracy, and that near real-time detection performance can be achieved., This work has been partially funded by the European project CargoANTs (FP7-SST-2013- 605598) and by the Spanish CICYT project DPI2013-42458-P., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Active pose SLAM with RRT*

  • Vallvé, Joan
  • Andrade-Cetto, Juan
Trabajo presentado al ICRA celebrado en Seattle (US) del 26 al 30 de mayo de 2015., We propose a novel method for robotic exploration that evaluates paths that minimize both the joint path and map entropy per meter traveled. The method uses Pose SLAM to update the path estimate, and grows an RRT* tree to generate the set of candidate paths. This action selection mechanism contrasts with previous appoaches in which the action set was built heuristically from a sparse set of candidate actions. The technique favorably compares agains the classical frontier- based exploration and other Active Pose SLAM methods in simulations in a common publicly available dataset., This work has been supported by the Spanish Ministry of Economy and Competitiveness under project PAU+ DPI-2011-27510 and by the EU Project CargoAnts FP7-605598., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Real time people detection combining appearance and depth image spaces using boosted random ferns

  • Vaquero, Victor
  • Villamizar, Michael
  • Sanfeliu, Alberto
Trabajo presentado a la 2nd Iberian Robotics Conference (ROBOT-2015)., This paper presents a robust and real-time method for people detection in urban and crowed environments. Unlike other conventional methods which either focus on single features or compute multiple and independent classifiers specialized in a particular feature space, the proposed approach creates a synergic combination of appearance and depth cues in a unique classifier. The core of our method is a Boosted Random Ferns classifier that selects automatically the most discriminative local binary features for both the appearance and depth image spaces. Based on this classifier, a fast and robust people detector which maintains high detection rates in spite of environmental changes is created. The proposed method has been validated in a challenging RGB-D database of people in urban scenarios and has shown that outperforms state-of-the-art approaches in spite of the difficult environment conditions. As a result, this method is of special interest for real-time robotic applications where people detection is a key matter, such as human-robot interaction or safe navigation of mobile robots for example., This work has been partially funded by the EU project CargoANTs FP7-SST-2013-605598 and the Spanish CICYT project DPI2013-42458-P., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Low cost, robust and real time system for detecting and tracking moving objects to automate cargo handling in port terminals

  • Vaquero, Victor
  • Repiso, Ely
  • Sanfeliu, Alberto
  • Vissers, John
  • Kwakkernaat, Maurice
Trabajo presentado a la 2nd Iberian Robotics Conference (ROBOT-2015)., The presented paper addresses the problem of detecting and tracking moving objects for autonomous cargo handling in port terminals using a perception system which input data is a single layer laser scanner. A computationally low cost and robust Detection and Tracking Moving Objects (DATMO) algorithm is presented to be used in autonomous guided vehicles and autonomous trucks for efficient transportation of cargo in ports. The method first detects moving objects and then tracks them, taking into account that in port terminals the structure of the environment is formed by containers and that the moving objects can be trucks, AGV, cars, straddle carriers and people among others. Two approaches of the DATMO system have been tested, the first one is oriented to detect moving obstacles and focused on tracking and filtering those detections; and the second one is focused on keepking targets when no detections are provided. The system has been evaluated with real data obtained in the CTT port terminal in Hengelo, the Netherlands. Both methods have been tested in the dataset with good results in tracking moving objects., This work has been partially funded by the EU project CargoANTs FP7-SST-2013-605598 and the Spanish CICYT project DPI2013-42458-P., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Tesis de maestría (MasterThesis)

Sensor fusion of raw GPS measurements for autonomous vehicle localization

  • Tirindelli, Pier
Master in Artificial Intelligence, Facultat d’Informàtica de Barcelona (FIB), Facultat de Matemàtiques (UB), Escola Tècnica Superior d’Enginyeria (URV), Universitat Politècnica de Catalunya (UPC), Universitat de Barcelona (UB), Universitat Rovira i Virgili (URV)., This dissertation outlines the benefits that sensor fusion brings in the mobile robotics problem of incrementally constructing a consistent map of an unknown environment with a moving robot, while simultaneously determining its location within this map. The work of this Master’s Thesis focuses on outdoor localization of autonomous vehicles, and aims to use raw GPS pseudorange data to establish geometric constraints to be added in the localization problem. This localization problem will solve for the optimal vehicle localization state, satisfying these pseudorange constraints, but also other geometric constraints imposed by other sensor measurements.We developed a software able to establish geometric constraints for a localization problem from raw GPS measurements. Then we integrated it in Wolf, a software framework for managing SLAM, enriching its sensor fusion capabilities. In the end we tested the sensor fusion between raw GPS and odometry., This project has been developed in the context of the European project Cargo-ANTS, devoted to investigate and develop novel techniques for autonomous navigation of wheeled vehicles in port terminal areas., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Tesis de maestría (MasterThesis)

Robust multi-hypothesis tracker fusing diverse sensor information

  • Repiso, Ely
Master’s degree in Automatic Control and Robotics. Escola Tècnica Superior d’Enginyeria Industrial de Barcelona (ETSEIB-UPC)., This Master’s Thesis presents a developed tracking algorithm which fuses different object detections. These object detections are obtained from different detectors. These detectors are: a laser leg detector and a tag vision detector. The fusion has the objective to obtain more robust object detections; moreover, it has the capability to distinguish specific people. In the future, it is intend to add new detectors, such as: a skeleton person detector and a radar detector. Adding new detectors is direct and easy, since the tracker has a general and flexible implementation which allows the change of the type and number of detectors. The Robust multi-hypothesis tracker fusing diverse sensor information uses as the back- bone of the theory the multi-hypothesis tracker approach proposed by Reid [28]. Then, the algorithm has been modified to add some new improvements. For work with social robots in the algorithm was added two functions in the probabilities of the hypothesis which allows us to control the confirmation and the deletion of the tracks. Also, the algorithm was improved by adding another function that lets we use the people’s velocities orientations to improve the association between tracks and detections in crossing situations. Furthermore, the algorithm was extended with the fusion of different detections. For work with autonomous vehicles in the European Project Cargo ANTs1 the algorithm was adapted for tracking locally (without map) and global (with a map); was improved to use the track velocity and a specific detector to distinguish between dynamic and static objects and was adapted to group different objects of the same type using distance. The objects can have dynamic or static type. Finally, the tests of the tracker algorithm have been carried out in the fields of social robots and autonomous vehicles (Cargo ANTs project). The experiments have been carried out with Tibi and Dabo robots, illustrate that our tracking approach can robustly and efficiently track multiple people without changes on their id ’s and without losing them. Also, the experiments show that the tracker can distinguish a specific person by using the fusion; even in situations which has several people or objects that cross with or occlude the specific person. The experiments have been carried out in the Cargo ANTs project show that the tracker is able to follow only the single real dynamic object, and that the tracker is able to group different detections obtained inside the same object., The European Project Cargo ANTs., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Otros (Other)

Characterization of a radar for autonomous vehicle location systems

  • Flores Abad, Ismael
The goal of this project was to set-up a radar sensor fully operational in a mobile robotics research laboratory. In order to achieve this goal, several steps were done: study of radar principles, data sheet device evaluation, sensor purchasing, C++ development for real-time data acquisition, ROS integration and data visualization. At the end, the device was successfully used to carry out research experiments in the context of autonomous driving vehicles., The work is under the scope of the following projects: Cargo-ANTS: Cargo handling by Automated Next generation Transportation Systems for ports and terminals., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Contribución de congreso (ConferenceOutput)

Observability analysis and optimal sensor placement in stereo radar odometry

  • Corominas Murtra, Andreu
  • Vallvé, Joan
  • Solà, Joan
  • Flores Abad, Ismael
  • Andrade-Cetto, Juan
Trabajo presentado a la IEEE Conference on Roboticand Automation (ICRA), celebrada en Estocolmo (Suecia) del 16 al 21 de mayo de 2016., Localization is the key perceptual process closing the loop of autonomous navigation, allowing self-driving vehi- cles to operate in a deliberate way. To ensure robust localization, autonomous vehicles have to implement redundant estimation processes, ideally independent in terms of the underlying physics behind sensing principles. This paper presents a stereo radar odometry system, which can be used as such a redundant system, complementary to other odometry estimation processes, providing robustness for long-term operability. The presented work is novel with respect to previously published methods in that it contains: (i) a detailed formulation of the Doppler error and its associated uncertainty; (ii) an observability analysis that gives the minimal conditions to infer a 2D twist from radar readings; and (iii) a numerical analysis for optimal vehicle sensor placement. Experimental results are also detailed that validate the theoretical insights., This work was supported by the EU Project CargoAnts: Cargo Handling by Automated Next Generation Transportation Systems for Ports and Terminals (FP7-SST-2013-RTD-1-605598)., Peer Reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Artículo científico (JournalArticle)

Robust and Real-Time Detection and Tracking of Moving Objects with Minimum 2D LiDAR Information to Advance Autonomous Cargo Handling in Ports

  • Vaquero, Victor
  • Repiso, Ely
  • Sanfeliu, Alberto
Detecting and tracking moving objects (DATMO) is an essential component for autonomous driving and transportation. In this paper, we present a computationally low-cost and robust DATMO system which uses as input only 2D laser rangefinder (LRF) information. Due to its low requirements both in sensor needs and computation, our DATMO algorithm is meant to be used in current Autonomous Guided Vehicles (AGVs) to improve their reliability for the cargo transportation tasks at port terminals, advancing towards the next generation of fully autonomous transportation vehicles. Our method follows a Detection plus Tracking paradigm. In the detection step we exploit the minimum information of 2D-LRFs by segmenting the elements of the scene in a model-free way and performing a fast object matching to pair segmented elements from two different scans. In this way, we easily recognize dynamic objects and thus reduce consistently by between two and five times the computational burden of the adjacent tracking method. We track the final dynamic objects with an improved Multiple-Hypothesis Tracking (MHT), to which special functions for filtering, confirming, holding, and deleting targets have been included. The full system is evaluated in simulated and real scenarios producing solid results. Specifically, a simulated port environment has been developed to gather realistic data of common autonomous transportation situations such as observing an intersection, joining vehicle platoons, and perceiving overtaking maneuvers. We use different sensor configurations to demonstrate the robustness and adaptability of our approach. We additionally evaluate our system with real data collected in a port terminal the Netherlands. We show that it is able to accomplish the vehicle following task successfully, obtaining a total system recall of more than 98% while running faster than 30 Hz, This work has been partly funded by the EU project CargoANTs FP7-SST-2013-605598, the Spanish MINECO projects DPI2016-78957-R (AEI/FEDER EU) and Unidad de Excelencia Maria de Maeztu 2016 (MDM-2016-0656)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publication Tesis doctoral (DoctoralThesis)

Lidar-Based Scene Understanding for Autonomous Driving Using Deep Learning

  • Vaquero, Victor
Tesis llevada a cabo para conseguir el grado de Doctor por la Universidad Politécnica de Cataluña.--2020-02-25, [EN] With over 1.35 million fatalities related to traffic accidents worldwide, autonomous driving was foreseen at the beginning of this century as a feasible solution to improve security in our roads. Nevertheless, it is meant to disrupt our transportation paradigm, allowing to reduce congestion, pollution, and costs, while increasing the accessibility, efficiency, and reliability of the transportation for both people and goods. Although some advances have gradually been transferred into commercial vehicles in the way of Advanced Driving Assistance Systems (ADAS) such as adaptive cruise control, blind spot detection or automatic parking, however, the technology is far from mature. A full understanding of the scene is actually needed so that allowing the vehicles to be aware of the surroundings, knowing the existing elements of the scene, as well as their motion, intentions and interactions. In this PhD dissertation, we explore new approaches for understanding driving scenes from 3D LiDAR point clouds by using Deep Learning methods. To this end, in Part I we analyze the scene from a static perspective using independent frames to detect the neighboring vehicles. Next, in Part II we develop new ways for understanding the dynamics of the scene. Finally, in Part III we apply all the developed methods to accomplish higher level challenges such as segmenting moving obstacles while obtaining their rigid motion vector over the ground. More specifically, in Chapter 2 we develop a 3D vehicle detection pipeline based on a multi-branch deep-learning architecture and propose a Front (FR-V) and a Bird’s Eye view (BE-V) as 2D representations of the 3D point cloud to serve as input for training our models. Later on, in Chapter 3 we apply and further test this method on two real uses-cases, for pre-filtering moving obstacles while creating maps to better localize ourselves on subsequent days, as well as for vehicle tracking. From the dynamic perspective, in Chapter 4 we learn from the 3D point cloud a novel dynamic feature that resembles optical flow from RGB images. For that, we develop a new approach to leverage RGB optical flow as pseudo ground truth for training purposes but allowing the use of only 3D LiDAR data at inference time. Additionally, in Chapter 5 we explore the benefits of combining classification and regression learning problems to face the optical flow estimation task in a joint coarse-and-fine manner. Lastly, in Chapter 6 we gather the previous methods and demonstrate that with these independent tasks we can guide the learning of higher challenging problems such as segmentation and motion estimation of moving vehicles from our own moving perspective., [ES] Con más de 1,35 millones de muertes por accidentes de tráfico en el mundo, a principios de siglo se predijo que la conducción autónoma sería una solución viable para mejorar la seguridad en nuestras carreteras. Además la conducción autónoma está destinada a cambiar nuestros paradigmas de transporte, permitiendo reducir la congestión del tráfico, la contaminación y el coste, a la vez que aumentando la accesibilidad, la eficiencia y confiabilidad del transporte tanto de personas como de mercancías. Aunque algunos avances, como el control de crucero adaptativo, la detección de puntos ciegos o el estacionamiento automático, se han transferido gradualmente a vehículos comerciales en la forma de los Sistemas Avanzados de Asistencia a la Conducción (ADAS), la tecnología aún no ha alcanzado el suficiente grado de madurez. Se necesita una comprensión completa de la escena para que los vehículos puedan entender el entorno, detectando los elementos presentes, así como su movimiento, intenciones e interacciones. En la presente tesis doctoral, exploramos nuevos enfoques para comprender escenarios de conducción utilizando nubes de puntos en 3D capturadas con sensores LiDAR, para lo cual empleamos métodos de aprendizaje profundo. Con este fin, en la Parte I analizamos la escena desde una perspectiva estática para detectar vehículos. A continuación, en la Parte II, desarrollamos nuevas formas de entender las dinámicas del entorno. Finalmente, en la Parte III aplicamos los métodos previamente desarrollados para lograr desafíos de nivel superior, como segmentar obstáculos dinámicos a la vez que estimamos su vector de movimiento sobre el suelo. Específicamente, en el Capítulo 2 detectamos vehículos en 3D creando una arquitectura de aprendizaje profundo de dos ramas y proponemos una vista frontal (FR-V) y una vista de pájaro (BE-V) como representaciones 2D de la nube de puntos 3D que sirven como entrada para entrenar nuestros modelos. Más adelante, en el Capítulo 3 aplicamos y probamos aún más este método en dos casos de uso reales, tanto para filtrar obstáculos en movimiento previamente a la creación de mapas sobre los que poder localizarnos mejor en los días posteriores, como para el seguimiento de vehículos. Desde la perspectiva dinámica, en el Capítulo 4 aprendemos de la nube de puntos en 3D una característica dinámica novedosa que se asemeja al flujo óptico sobre imágenes RGB. Para ello, desarrollamos un nuevo enfoque que aprovecha el flujo óptico RGB como pseudo muestras reales para entrenamiento, usando solo information 3D durante la inferencia. Además, en el Capítulo 5 exploramos los beneficios de combinar los aprendizajes de problemas de clasificación y regresión para la tarea de estimación de flujo óptico de manera conjunta. Por último, en el Capítulo 6 reunimos los métodos anteriores y demostramos que con estas tareas independientes podemos guiar el aprendizaje de problemas de más alto nivel, como la segmentación y estimación del movimiento de vehículos desde nuestra propia perspectiva, This work has been partially supported by the European projects: FP7-SST-2013-RTD-1-605598 Cargo-ANTS: Cargo handling by Automated Next generation Transportation Systems for ports and terminals. FLAG-ERA-JTC-2016 ROBOCOM++: Rethinking robotics for the robot companion of the future. As well as by the Spanish Ministry of Education and Science under the research project: DPI2016-78957-R ColRobTransp: Colaboración robots-humanos para el transporte de productos en zonas urbanas
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Uncertainty analysis of the DLT-Lines calibration algorithm for cameras with radial distortion

  • Galego, Ricardo
  • Ortega, Agustin
  • Ferreira, Ricardo
  • Bernardino, Alexandre
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
  • Gaspar, José
3D metric data of environmental structures is nowadays present in many information sources (maps, GIS) and can be easily acquired with modern depth sensing technology (RGBD, laser). This wealth of information can be readily used for single view calibration of 2D cameras with radial distortion, provided that image structures can be matched with the 3D data. In this paper we present an analysis of the level of accuracy that can be obtained when such calibration is performed with the 2D-3D DLT-Lines algorithm. The analysis propagates uncertainty in the detection of features at the image level to camera pose, and from there to 3D reconstruction. The analytic error propagation expressions are derived using first order uncertainty models, and are validated with Monte Carlo simulations in a virtual indoor environment. The method is general and can be applied to other calibration methods, as long as explicit or implicit expressions can be derived for the transformation from image coordinates to 3D reconstruction. We present results with real data for two applications: i) the 3D reconstruction of an outdoors building for which 3D information is given by a map, observed by a mobile phone camera: and ii) the uncertainty in the localization at the floor plane of points observed by a fixed camera calibrated by a robot equipped with an RGBD camera navigating in a typical indoor environment. (C) 2015 Elsevier Inc. All rights reserved., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Tesis de maestría (MasterThesis)

Sensor Fusion of Raw GPS Measurements for Autonomous Vehicle Localization

  • Tirindelli, Pier
We developed a software able to establish geometric constraints for a localization problem from raw GPS measurements. Then we integrated it in Wolf, a software framework for managing SLAM, enriching its sensor fusion capabilities. In the end we tested the sensor fusion between raw GPS and odometry.
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Multimodal feedback fusion of laser, image and temporal information

  • Huerta Casado, Iván
  • Ferrer Mínguez, Gonzalo
  • Herrero Cotarelo, Fernando|||0000-0003-4116-0265
  • Prati, Andrea
  • Sanfeliu Cortés, Alberto|||0000-0003-3868-9678
In the present paper, we propose a highly accurate and robust people detector, which works well under highly variant and uncertain conditions, such as occlusions, false positives and false detections. These adverse conditions, which initially motivated this research, occur when a robotic platform navigates in an urban environment, and although the scope is originally within the robotics field, the authors believe that our contributions can be extended to other fields. To this end, we propose a multimodal information fusion consisting of laser and monocular camera information. Laser information is modelled using a set of weak classifiers (Adaboost) to detect people. Camera information is processed by using HOG descriptors to classify person/non person based on a linear SVM. A multi-hypothesis tracker trails the position and velocity of each of the targets, providing temporal information to the fusion, allowing recovery of detections even when the laser segmentation fails. Experimental results show that our feedback-based system outperforms previous state-of-the-art methods in performance and accuracy, and that near real-time detection performance can be achieved., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Low cost, robust and real time system for detecting and tracking moving objects to automate cargo handling in port terminals

  • Vaquero Gómez, Víctor|||0000-0003-2516-2551
  • Repiso Polo, Ely|||0000-0001-6540-8347
  • Sanfeliu Cortés, Alberto|||0000-0003-3868-9678
  • Vissers, John
  • Kwakkernaat, Maurice
The presented paper addresses the problem of detecting and tracking moving objects for autonomous cargo handling in port terminals using a perception system which input data is a single layer laser scanner. A computationally low cost and robust Detection and Tracking Moving Objects (DATMO) algorithm is presented to be used in autonomous guided vehicles and autonomous trucks for efficient transportation of cargo in ports. The method first detects moving objects and then tracks them, taking into account that in port terminals the structure of the environment is formed by containers and that the moving objects can be trucks, AGV, cars, straddle carriers and people among others. Two approaches of the DATMO system have been tested, the first one is oriented to detect moving obstacles and focused on tracking and filtering those detections; and the second one is focused on keepking targets when no detections are provided. The system has been evaluated with real data obtained in the CTT port terminal in Hengelo, the Netherlands. Both methods have been tested in the dataset with good results in tracking moving objects., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Localization in highly dynamic environments using dual-timescale NDT-MCL

  • Valencia Carreño, Rafael
  • Saarinen, Jari
  • Andreasson, Henrik
  • Vallvé Navarro, Joan|||0000-0002-1029-356X
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
  • Lilienthal, Achim
Industrial environments are rarely static and often their configuration is continuously changing due to the material transfer flow. This is a major challenge for infrastructure free localization systems. In this paper we address this challenge by introducing a localization approach that uses a dual- timescale approach. The proposed approach - Dual-Timescale Normal Distributions Transform Monte Carlo Localization (DT- NDT-MCL) - is a particle filter based localization method, which simultaneously keeps track of the pose using an apriori known static map and a short-term map. The short-term map is continuously updated and uses Normal Distributions Transform Occupancy maps to maintain the current state of the environment. A key novelty of this approach is that it does not have to select an entire timescale map but rather use the best timescale locally. The approach has real-time performance and is evaluated using three datasets with increasing levels of dynamics. We compare our approach against previously pro- posed NDT-MCL and commonly used SLAM algorithms and show that DT-NDT-MCL outperforms competing algorithms with regards to accuracy in all three test cases.
UPCommons. Portal del coneixement obert de la UPC
Publication Tesis de maestría (MasterThesis)

Robust multi-hypothesis tracker fusing diverse sensor information

  • Repiso Polo, Ely|||0000-0001-6540-8347
This Master’s Thesis presents a developed tracking algorithm which fuses different object detections. These object detections are obtained from different detectors. These detectors are: a laser leg detector and a tag vision detector. The fusion has the objective to obtain more robust object detections; moreover, it has the capability to distinguish specific people. In the future, it is intend to add new detectors, such as: a skeleton person detector and a radar detector. Adding new detectors is direct and easy, since the tracker has a general and flexible implementation which allows the change of the type and number of detectors. The Robust multi-hypothesis tracker fusing diverse sensor information uses as the back- bone of the theory the multi-hypothesis tracker approach proposed by Reid [28]. Then, the algorithm has been modified to add some new improvements. For work with social robots in the algorithm was added two functions in the probabilities of the hypothesis which allows us to control the confirmation and the deletion of the tracks. Also, the algorithm was improved by adding another function that lets we use the people’s velocities orientations to improve the association between tracks and detections in crossing situations. Furthermore, the algorithm was extended with the fusion of different detections. For work with autonomous vehicles in the European Project Cargo ANTs1 the algorithm was adapted for tracking locally (without map) and global (with a map); was improved to use the track velocity and a specific detector to distinguish between dynamic and static objects and was adapted to group different objects of the same type using distance. The objects can have dynamic or static type. Finally, the tests of the tracker algorithm have been carried out in the fields of social robots and autonomous vehicles (Cargo ANTs project). The experiments have been carried out with Tibi and Dabo robots, illustrate that our tracking approach can robustly and efficiently track multiple people without changes on their id ’s and without losing them. Also, the experiments show that the tracker can distinguish a specific person by using the fusion; even in situations which has several people or objects that cross with or occlude the specific person. The experiments have been carried out in the Cargo ANTs project show that the tracker is able to follow only the single real dynamic object, and that the tracker is able to group different detections obtained inside the same object.
UPCommons. Portal del coneixement obert de la UPC
Publication Proyecto fin de carrera (BachelorThesis)

Characterization of a radar for autonomous vehicle location systems

UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Real time people detection combining appearance and depth image spaces using boosted random ferns

  • Vaquero Gómez, Víctor|||0000-0003-2516-2551
  • Villamizar Vergel, Michael Alejandro
  • Sanfeliu Cortés, Alberto|||0000-0003-3868-9678
This paper presents a robust and real-time method for people detection in urban and crowed environments. Unlike other conventional methods which either focus on single features or compute multiple and independent classifiers specialized in a particular feature space, the proposed approach creates a synergic combination of appearance and depth cues in a unique classifier. The core of our method is a Boosted Random Ferns classifier that selects automatically the most discriminative local binary features for both the appearance and depth image spaces. Based on this classifier, a fast and robust people detector which maintains high detection rates in spite of environmental changes is created. The proposed method has been validated in a challenging RGB-D database of people in urban scenarios and has shown that outperforms state-of-the-art approaches in spite of the difficult environment conditions. As a result, this method is of special interest for real-time robotic applications where people detection is a key matter, such as human-robot interaction or safe navigation of mobile robots for example., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Terrain classification in complex three-dimensional outdoor environments

  • Santamaria Navarro, Àngel|||0000-0002-6328-1423
  • Teniente Avilés, Ernesto Homar
  • Morta Garriga, Martí
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
This paper presents two techniques to detect and classify navigable terrain in complex three-dimensional (3D) environments. The first method is a low level on-line mechanism aimed at detecting obstacles and holes at a fast frame rate using a time-of-flight camera as the main sensor. The second technique is a high-level offline classification mechanism that learns traversable regions from larger 3D point clouds acquired with a laser range scanner. We approach the problem using Gaussian processes as a regression tool, in which the terrain parameters are learned, and also for classification, using samples from traversed areas to build the traversable terrain class. The two methods are compared against unsupervised classification, and sample trajectories are generated in the classified areas using a nonholonomic path planner. We show results of both the low-level and the high-level terrain classification approaches in simulations and in real-time navigation experiments using a Segway RMP400 robot. (C) 2014 Wiley Periodicals, Inc., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Observability analysis and optimal sensor placement in stereo radar odometry

  • Corominas Murtra, Andreu|||0000-0001-8111-2351
  • Vallvé Navarro, Joan|||0000-0002-1029-356X
  • Solà Ortega, Joan|||0000-0002-2933-3381
  • Flores Abad, Ismael
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Localization is the key perceptual process closing the loop of autonomous navigation, allowing self-driving vehicles to operate in a deliberate way. To ensure robust localization, autonomous vehicles have to implement redundant estimation processes, ideally independent in terms of the underlying physics behind sensing principles. This paper presents a stereo radar odometry system, which can be used as such a redundant system, complementary to other odometry estimation processes, providing robustness for long-term operability. The presented work is novel with respect to previously published methods in that it contains: (i) a detailed formulation of the Doppler error and its associated uncertainty; (ii) an observability analysis that gives the minimal conditions to infer a 2D twist from radar readings; and (iii) a numerical analysis for optimal vehicle sensor placement. Experimental results are also detailed that validate the theoretical insights., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Calibration of an outdoor distributed camera network with a 3D point cloud

  • Ortega, Agustin
  • Silva, Manuel
  • Teniente Avilés, Ernesto Homar
  • Ferreira, Ricardo
  • Bernardino Bernardino Malheiro, Alexandre José
  • Gaspar, José
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
Outdoor camera networks are becoming ubiquitous in critical urban areas of the largest cities around the world. Although current applications of camera networks are mostly tailored to video surveillance, recent research projects are exploiting their use to aid robotic systems in people-assisting tasks. Such systems require precise calibration of the internal and external parameters of the distributed camera network. Despite the fact that camera calibration has been an extensively studied topic, the development of practical methods for user-assisted calibration that minimize user intervention time and maximize precision still pose significant challenges. These camera systems have non-overlapping fields of view, are subject to environmental stress, and are likely to suffer frequent recalibration. In this paper, we propose the use of a 3D map covering the area to support the calibration process and develop an automated method that allows quick and precise calibration of a large camera network. We present two cases of study of the proposed calibration method: one is the calibration of the Barcelona Robot Lab camera network, which also includes direct mappings (homographies) between image coordinates and world points in the ground plane (walking areas) to support person and robot detection and localization algorithms. The second case consist of improving the GPS positioning of geo-tagged images taken with a mobile device in the Facultat de Matematiques i Estadistica (FME) patio at the Universitat Politecnica de Catalunya (UPC)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Potential information fields for mobile robot exploration

  • Vallvé Navarro, Joan|||0000-0002-1029-356X
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
We present a decision theoretic approach to mobile robot exploration. The method evaluates the reduction of joint path and map entropy and computes a potential information field in robot configuration space using these joint entropy reduction estimates. The exploration trajectory is computed descending on the gradient of this field. The technique uses Pose SLAM as its estimation backbone. Very efficient kernel convolution mechanisms are used to evaluate entropy reduction for each sensor ray, and for each possible robot orientation, taking frontiers and obstacles into account. In the end, the computation of this field on the entire configuration space is shown to be very efficient. The approach is tested in simulations in a pair of publicly available datasets comparing favorably both in quality of estimates and in execution time against an RRT*-based search for the nearest frontier and also against a locally optimal exploration strategy. (C) 2014 Elsevier B.V. All rights reserved., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Active pose SLAM with RRT*

  • Vallvé Navarro, Joan|||0000-0002-1029-356X
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
We propose a novel method for robotic exploration that evaluates paths that minimize both the joint path and map entropy per meter traveled. The method uses Pose SLAM to update the path estimate, and grows an RRT* tree to generate the set of candidate paths. This action selection mechanism contrasts with previous appoaches in which the action set was built heuristically from a sparse set of candidate actions. The technique favorably compares agains the classical frontier- based exploration and other Active Pose SLAM methods in simulations in a common publicly available dataset., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Artículo científico (JournalArticle)

Chromatic shadow detection and tracking for moving foreground segmentation

  • Huerta Casado, Iván
  • Holte, Michael B.
  • Moeslund, Thomas
  • Gonzàlez, Jordi
Advanced segmentation techniques in the surveillance domain deal with shadows to avoid distortions when detecting moving objects. Most approaches for shadow detection are still typically restricted to penumbra shadows and cannot cope well with umbra shadows. Consequently, umbra shadow regions are usually detected as part of moving objects, thus affecting the performance of the final detection. In this paper we address the detection of both penumbra and umbra shadow regions. First, a novel bottom-up approach is presented based on gradient and colour models, which successfully discriminates between chromatic moving cast shadow regions and those regions detected as moving objects. In essence, those regions corresponding to potential shadows are detected based on edge partitioning and colour statistics. Subsequently (i) temporal similarities between textures and (ii) spatial similarities between chrominance angle and brightness distortions are analysed for each potential shadow region for detecting the umbra shadow regions. Our second contribution refines even further the segmentation results: a tracking-based top-down approach increases the performance of our bottom-up chromatic shadow detection algorithm by properly correcting non-detected shadows. To do so, a combination of motion filters in a data association framework exploits the temporal consistency between objects and shadows to increase the shadow detection rate. Experimental results exceed current state-of-the-art in shadow accuracy for multiple well-known surveillance image databases which contain different shadowed materials and illumination conditions., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publication Contribución de congreso (ConferenceOutput)

Dense entropy decrease estimation for mobile robot exploration

  • Vallvé Navarro, Joan|||0000-0002-1029-356X
  • Andrade-Cetto, Juan|||0000-0002-6354-8941
We propose a method for the computation of entropy decrease in C-space. These estimates are then used to evaluate candidate exploratory trajectories in the context of autonomous mobile robot mapping. The method evaluates both map and path entropy reduction and uses such estimates to compute trajectories that maximize coverage whilst min- imizing localization uncertainty, hence reducing map error. Very efficient kernel convolution mechanisms are used to evaluate entropy reduction at each sensor ray, and for each possible robot position and orientation, taking frontiers and obstacles into account. In contrast to most other exploration methods that evaluate entropy reduction at a small number of discrete robot configurations, we do it densely for the entire C-space. The computation of such dense entropy reduction maps opens the window to new exploratory strategies. In this paper we present two such strategies. In the first one we drive exploration through a gradient descent on the entropy decrease field. The second strategy chooses maximal entropy reduction configurations as candidate exploration goals, and plans paths to them using RRT*. Both methods use PoseSLAM as their estimation backbone, and are tested and compared with classical frontier-based exploration in simulations using common publicly available datasets., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Proyecto fin de carrera (BachelorThesis)

Characterization of a radar for autonomous vehicle location systems

Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Multimodal feedback fusion of laser, image and temporal information

  • Huerta Casado, Iván
  • Ferrer Mínguez, Gonzalo
  • Herrero Cotarelo, Fernando
  • Prati, Andrea
  • Sanfeliu Cortés, Alberto
In the present paper, we propose a highly accurate and robust people detector, which works well under highly variant and uncertain conditions, such as occlusions, false positives and false detections. These adverse conditions, which initially motivated this research, occur when a robotic platform navigates in an urban environment, and although the scope is originally within the robotics field, the authors believe that our contributions can be extended to other fields. To this end, we propose a multimodal information fusion consisting of laser and monocular camera information. Laser information is modelled using a set of weak classifiers (Adaboost) to detect people. Camera information is processed by using HOG descriptors to classify person/non person based on a linear SVM. A multi-hypothesis tracker trails the position and velocity of each of the targets, providing temporal information to the fusion, allowing recovery of detections even when the laser segmentation fails. Experimental results show that our feedback-based system outperforms previous state-of-the-art methods in performance and accuracy, and that near real-time detection performance can be achieved., Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Potential information fields for mobile robot exploration

  • Vallvé Navarro, Joan
  • Andrade-Cetto, Juan
We present a decision theoretic approach to mobile robot exploration. The method evaluates the reduction of joint path and map entropy and computes a potential information field in robot configuration space using these joint entropy reduction estimates. The exploration trajectory is computed descending on the gradient of this field. The technique uses Pose SLAM as its estimation backbone. Very efficient kernel convolution mechanisms are used to evaluate entropy reduction for each sensor ray, and for each possible robot orientation, taking frontiers and obstacles into account. In the end, the computation of this field on the entire configuration space is shown to be very efficient. The approach is tested in simulations in a pair of publicly available datasets comparing favorably both in quality of estimates and in execution time against an RRT*-based search for the nearest frontier and also against a locally optimal exploration strategy. (C) 2014 Elsevier B.V. All rights reserved., Peer Reviewed, Preprint
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Real time people detection combining appearance and depth image spaces using boosted random ferns

  • Vaquero Gómez, Víctor
  • Villamizar Vergel, Michael Alejandro
  • Sanfeliu Cortés, Alberto
This paper presents a robust and real-time method for people detection in urban and crowed environments. Unlike other conventional methods which either focus on single features or compute multiple and independent classifiers specialized in a particular feature space, the proposed approach creates a synergic combination of appearance and depth cues in a unique classifier. The core of our method is a Boosted Random Ferns classifier that selects automatically the most discriminative local binary features for both the appearance and depth image spaces. Based on this classifier, a fast and robust people detector which maintains high detection rates in spite of environmental changes is created. The proposed method has been validated in a challenging RGB-D database of people in urban scenarios and has shown that outperforms state-of-the-art approaches in spite of the difficult environment conditions. As a result, this method is of special interest for real-time robotic applications where people detection is a key matter, such as human-robot interaction or safe navigation of mobile robots for example., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Uncertainty analysis of the DLT-Lines calibration algorithm for cameras with radial distortion

  • Galego, Ricardo
  • Ortega, Agustin
  • Ferreira, Ricardo
  • Bernardino, Alexandre
  • Andrade-Cetto, Juan
  • Gaspar, José
3D metric data of environmental structures is nowadays present in many information sources (maps, GIS) and can be easily acquired with modern depth sensing technology (RGBD, laser). This wealth of information can be readily used for single view calibration of 2D cameras with radial distortion, provided that image structures can be matched with the 3D data. In this paper we present an analysis of the level of accuracy that can be obtained when such calibration is performed with the 2D-3D DLT-Lines algorithm. The analysis propagates uncertainty in the detection of features at the image level to camera pose, and from there to 3D reconstruction. The analytic error propagation expressions are derived using first order uncertainty models, and are validated with Monte Carlo simulations in a virtual indoor environment. The method is general and can be applied to other calibration methods, as long as explicit or implicit expressions can be derived for the transformation from image coordinates to 3D reconstruction. We present results with real data for two applications: i) the 3D reconstruction of an outdoors building for which 3D information is given by a map, observed by a mobile phone camera: and ii) the uncertainty in the localization at the floor plane of points observed by a fixed camera calibrated by a robot equipped with an RGBD camera navigating in a typical indoor environment. (C) 2015 Elsevier Inc. All rights reserved., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Calibration of an outdoor distributed camera network with a 3D point cloud

  • Ortega, Agustin
  • Silva, Manuel
  • Teniente Avilés, Ernesto Homar
  • Ferreira, Ricardo
  • Bernardino Bernardino Malheiro, Alexandre José
  • Gaspar, José
  • Andrade-Cetto, Juan
Outdoor camera networks are becoming ubiquitous in critical urban areas of the largest cities around the world. Although current applications of camera networks are mostly tailored to video surveillance, recent research projects are exploiting their use to aid robotic systems in people-assisting tasks. Such systems require precise calibration of the internal and external parameters of the distributed camera network. Despite the fact that camera calibration has been an extensively studied topic, the development of practical methods for user-assisted calibration that minimize user intervention time and maximize precision still pose significant challenges. These camera systems have non-overlapping fields of view, are subject to environmental stress, and are likely to suffer frequent recalibration. In this paper, we propose the use of a 3D map covering the area to support the calibration process and develop an automated method that allows quick and precise calibration of a large camera network. We present two cases of study of the proposed calibration method: one is the calibration of the Barcelona Robot Lab camera network, which also includes direct mappings (homographies) between image coordinates and world points in the ground plane (walking areas) to support person and robot detection and localization algorithms. The second case consist of improving the GPS positioning of geo-tagged images taken with a mobile device in the Facultat de Matematiques i Estadistica (FME) patio at the Universitat Politecnica de Catalunya (UPC)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Observability analysis and optimal sensor placement in stereo radar odometry

  • Corominas Murtra, Andreu
  • Vallvé Navarro, Joan
  • Solà Ortega, Joan
  • Flores Abad, Ismael
  • Andrade-Cetto, Juan
© 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Localization is the key perceptual process closing the loop of autonomous navigation, allowing self-driving vehicles to operate in a deliberate way. To ensure robust localization, autonomous vehicles have to implement redundant estimation processes, ideally independent in terms of the underlying physics behind sensing principles. This paper presents a stereo radar odometry system, which can be used as such a redundant system, complementary to other odometry estimation processes, providing robustness for long-term operability. The presented work is novel with respect to previously published methods in that it contains: (i) a detailed formulation of the Doppler error and its associated uncertainty; (ii) an observability analysis that gives the minimal conditions to infer a 2D twist from radar readings; and (iii) a numerical analysis for optimal vehicle sensor placement. Experimental results are also detailed that validate the theoretical insights., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Active pose SLAM with RRT*

  • Vallvé Navarro, Joan
  • Andrade-Cetto, Juan
We propose a novel method for robotic exploration that evaluates paths that minimize both the joint path and map entropy per meter traveled. The method uses Pose SLAM to update the path estimate, and grows an RRT* tree to generate the set of candidate paths. This action selection mechanism contrasts with previous appoaches in which the action set was built heuristically from a sparse set of candidate actions. The technique favorably compares agains the classical frontier- based exploration and other Active Pose SLAM methods in simulations in a common publicly available dataset., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Low cost, robust and real time system for detecting and tracking moving objects to automate cargo handling in port terminals

  • Vaquero Gómez, Víctor
  • Repiso Polo, Ely
  • Sanfeliu Cortés, Alberto
  • Vissers, John
  • Kwakkernaat, Maurice
The presented paper addresses the problem of detecting and tracking moving objects for autonomous cargo handling in port terminals using a perception system which input data is a single layer laser scanner. A computationally low cost and robust Detection and Tracking Moving Objects (DATMO) algorithm is presented to be used in autonomous guided vehicles and autonomous trucks for efficient transportation of cargo in ports. The method first detects moving objects and then tracks them, taking into account that in port terminals the structure of the environment is formed by containers and that the moving objects can be trucks, AGV, cars, straddle carriers and people among others. Two approaches of the DATMO system have been tested, the first one is oriented to detect moving obstacles and focused on tracking and filtering those detections; and the second one is focused on keepking targets when no detections are provided. The system has been evaluated with real data obtained in the CTT port terminal in Hengelo, the Netherlands. Both methods have been tested in the dataset with good results in tracking moving objects., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Dense entropy decrease estimation for mobile robot exploration

  • Vallvé Navarro, Joan
  • Andrade-Cetto, Juan
We propose a method for the computation of entropy decrease in C-space. These estimates are then used to evaluate candidate exploratory trajectories in the context of autonomous mobile robot mapping. The method evaluates both map and path entropy reduction and uses such estimates to compute trajectories that maximize coverage whilst min- imizing localization uncertainty, hence reducing map error. Very efficient kernel convolution mechanisms are used to evaluate entropy reduction at each sensor ray, and for each possible robot position and orientation, taking frontiers and obstacles into account. In contrast to most other exploration methods that evaluate entropy reduction at a small number of discrete robot configurations, we do it densely for the entire C-space. The computation of such dense entropy reduction maps opens the window to new exploratory strategies. In this paper we present two such strategies. In the first one we drive exploration through a gradient descent on the entropy decrease field. The second strategy chooses maximal entropy reduction configurations as candidate exploration goals, and plans paths to them using RRT*. Both methods use PoseSLAM as their estimation backbone, and are tested and compared with classical frontier-based exploration in simulations using common publicly available datasets., Peer Reviewed, Postprint (author’s final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Tesis de maestría (MasterThesis)

Sensor Fusion of Raw GPS Measurements for Autonomous Vehicle Localization

  • Tirindelli, Pier
We developed a software able to establish geometric constraints for a localization problem from raw GPS measurements. Then we integrated it in Wolf, a software framework for managing SLAM, enriching its sensor fusion capabilities. In the end we tested the sensor fusion between raw GPS and odometry.
Recercat. Dipósit de la Recerca de Catalunya
Publication Contribución de congreso (ConferenceOutput)

Localization in highly dynamic environments using dual-timescale NDT-MCL

  • Valencia Carreño, Rafael
  • Saarinen, Jari
  • Andreasson, Henrik
  • Vallvé Navarro, Joan
  • Andrade-Cetto, Juan
  • Lilienthal, Achim
Industrial environments are rarely static and often their configuration is continuously changing due to the material transfer flow. This is a major challenge for infrastructure free localization systems. In this paper we address this challenge by introducing a localization approach that uses a dual- timescale approach. The proposed approach - Dual-Timescale Normal Distributions Transform Monte Carlo Localization (DT- NDT-MCL) - is a particle filter based localization method, which simultaneously keeps track of the pose using an apriori known static map and a short-term map. The short-term map is continuously updated and uses Normal Distributions Transform Occupancy maps to maintain the current state of the environment. A key novelty of this approach is that it does not have to select an entire timescale map but rather use the best timescale locally. The approach has real-time performance and is evaluated using three datasets with increasing levels of dynamics. We compare our approach against previously pro- posed NDT-MCL and commonly used SLAM algorithms and show that DT-NDT-MCL outperforms competing algorithms with regards to accuracy in all three test cases., Postprint (author’s final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Tesis de maestría (MasterThesis)

Robust multi-hypothesis tracker fusing diverse sensor information

  • Repiso Polo, Ely
This Master’s Thesis presents a developed tracking algorithm which fuses different object detections. These object detections are obtained from different detectors. These detectors are: a laser leg detector and a tag vision detector. The fusion has the objective to obtain more robust object detections; moreover, it has the capability to distinguish specific people. In the future, it is intend to add new detectors, such as: a skeleton person detector and a radar detector. Adding new detectors is direct and easy, since the tracker has a general and flexible implementation which allows the change of the type and number of detectors. The Robust multi-hypothesis tracker fusing diverse sensor information uses as the back- bone of the theory the multi-hypothesis tracker approach proposed by Reid [28]. Then, the algorithm has been modified to add some new improvements. For work with social robots in the algorithm was added two functions in the probabilities of the hypothesis which allows us to control the confirmation and the deletion of the tracks. Also, the algorithm was improved by adding another function that lets we use the people’s velocities orientations to improve the association between tracks and detections in crossing situations. Furthermore, the algorithm was extended with the fusion of different detections. For work with autonomous vehicles in the European Project Cargo ANTs1 the algorithm was adapted for tracking locally (without map) and global (with a map); was improved to use the track velocity and a specific detector to distinguish between dynamic and static objects and was adapted to group different objects of the same type using distance. The objects can have dynamic or static type. Finally, the tests of the tracker algorithm have been carried out in the fields of social robots and autonomous vehicles (Cargo ANTs project). The experiments have been carried out with Tibi and Dabo robots, illustrate that our tracking approach can robustly and efficiently track multiple people without changes on their id ’s and without losing them. Also, the experiments show that the tracker can distinguish a specific person by using the fusion; even in situations which has several people or objects that cross with or occlude the specific person. The experiments have been carried out in the Cargo ANTs project show that the tracker is able to follow only the single real dynamic object, and that the tracker is able to group different detections obtained inside the same object.
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Terrain classification in complex three-dimensional outdoor environments

  • Santamaria Navarro, Àngel
  • Teniente Avilés, Ernesto Homar
  • Morta Garriga, Martí
  • Andrade-Cetto, Juan
This paper presents two techniques to detect and classify navigable terrain in complex three-dimensional (3D) environments. The first method is a low level on-line mechanism aimed at detecting obstacles and holes at a fast frame rate using a time-of-flight camera as the main sensor. The second technique is a high-level offline classification mechanism that learns traversable regions from larger 3D point clouds acquired with a laser range scanner. We approach the problem using Gaussian processes as a regression tool, in which the terrain parameters are learned, and also for classification, using samples from traversed areas to build the traversable terrain class. The two methods are compared against unsupervised classification, and sample trajectories are generated in the classified areas using a nonholonomic path planner. We show results of both the low-level and the high-level terrain classification approaches in simulations and in real-time navigation experiments using a Segway RMP400 robot. (C) 2014 Wiley Periodicals, Inc., Peer Reviewed, Postprint (author’s final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publication Artículo científico (JournalArticle)

Chromatic shadow detection and tracking for moving foreground segmentation

  • Huerta Casado, Iván
  • Holte, Michael B.
  • Moeslund, Thomas
  • Gonzàlez, Jordi
Advanced segmentation techniques in the surveillance domain deal with shadows to avoid distortions when detecting moving objects. Most approaches for shadow detection are still typically restricted to penumbra shadows and cannot cope well with umbra shadows. Consequently, umbra shadow regions are usually detected as part of moving objects, thus affecting the performance of the final detection. In this paper we address the detection of both penumbra and umbra shadow regions. First, a novel bottom-up approach is presented based on gradient and colour models, which successfully discriminates between chromatic moving cast shadow regions and those regions detected as moving objects. In essence, those regions corresponding to potential shadows are detected based on edge partitioning and colour statistics. Subsequently (i) temporal similarities between textures and (ii) spatial similarities between chrominance angle and brightness distortions are analysed for each potential shadow region for detecting the umbra shadow regions. Our second contribution refines even further the segmentation results: a tracking-based top-down approach increases the performance of our bottom-up chromatic shadow detection algorithm by properly correcting non-detected shadows. To do so, a combination of motion filters in a data association framework exploits the temporal consistency between objects and shadows to increase the shadow detection rate. Experimental results exceed current state-of-the-art in shadow accuracy for multiple well-known surveillance image databases which contain different shadowed materials and illumination conditions., Peer Reviewed, Postprint (author's final draft)