3D Feature Extraction and Modelling for SLAM

Paper submitted to the 39th International Symposium on Robotics ISR 2008, Seoul, South Korea, October 15-17, 2008.

Model of Competencies for Decomposition of Human Behavior: Application to Control System of Robots

Humans and machines have shared the same physical space for many years. To share the same space, we want the robots to behave like human beings. This will facilitate their social integration, their interaction with humans and create an intelligent behavior. To achieve this goal, we need to understand how human behavior is generated, analyze tasks running our nerves and how they relate to them. Then and only then can we implement these mechanisms in robotic beings. In this study, we propose a model of competencies based on human neuroregulator system for analysis and decomposition of behavior into functional modules. Using this model allow separate and locate the tasks to be implemented in a robot that displays human-like behavior. As an example, we show the application of model to the autonomous movement behavior on unfamiliar environments and its implementation in various simulated and real robots with different physical configurations and physical devices of different nature. The main result of this study has been to build a model of competencies that is being used to build robotic systems capable of displaying behaviors similar to humans and consider the specific characteristics of robots.

La enseñanza de robótica en el Máster en Automática y Robótica mediante un robot modular

Este trabajo muestra cómo se realiza la enseñanza de robótica mediante un robot modular y los resultados educativos obtenidos en el Máster Universitario en Automática y Robótica de la Escuela Politécnica Superior de la Universidad de Alicante. En el artículo se describen los resultados obtenidos con el uso de este robot modular tanto en competencias genéricas como específicas, en las enseñanzas de electrónica, control y programación del Máster. En este artículo se exponen los objetivos de aprendizaje para cada uno de ellos, su aplicación a la enseñanza y los resultados educativos obtenidos. En los resultados del estudio, cabe destacar que el alumno ha mostrado mayor interés y ha fomentado su aprendizaje autónomo. Para ello, el robot modular se construyó con herramientas para fomentar este tipo de enseñanza y aprendizaje, tales como comunicaciones interactivas para monitorizar, cambiar y adaptar diversos parámetros de control y potencia del robot.

Desarrollo de un robot modular para la enseñanza en el Máster en Automática y Robótica

Este trabajo presenta el diseño, construcción y programación de un robot modular para el desarrollo tanto de competencias genéricas como específicas, en las enseñanzas de electrónica, control y programación del Master de Automática y Robótica de la Escuela Politécnica Superior de la Universidad de Alicante. En este trabajo se exponen los diferentes módulos propuestos, así como los objetivos de aprendizaje para cada uno de ellos. Uno de los factores más importantes a destacar en el presente estudio es el posible desarrollo de la creatividad y el aprendizaje autónomo. Para ello, se desarrollará especialmente un módulo de comunicación por bluetooth que servirá para monitorizar, cambiar y adaptar on-line diversos parámetros de control y potencia del robot. Además, dicha herramienta se ha introducido como parte de la metodología en las asignaturas del Máster de Electromecánica y Sistemas de Control Automático. En esta memoria se mostrarán los distintos resultados obtenidos durante y en la finalización de este trabajo.

Contributions to 3D Data Registration and Representation

Nowadays, new computers generation provides a high performance that enables to build computationally expensive computer vision applications applied to mobile robotics. Building a map of the environment is a common task of a robot and is an essential part to allow the robots to move through these environments. Traditionally, mobile robots used a combination of several sensors from different technologies. Lasers, sonars and contact sensors have been typically used in any mobile robotic architecture, however color cameras are an important sensor due to we want the robots to use the same information that humans to sense and move through the different environments. Color cameras are cheap and flexible but a lot of work need to be done to give robots enough visual understanding of the scenes. Computer vision algorithms are computational complex problems but nowadays robots have access to different and powerful architectures that can be used for mobile robotics purposes. The advent of low-cost RGB-D sensors like Microsoft Kinect which provide 3D colored point clouds at high frame rates made the computer vision even more relevant in the mobile robotics field. The combination of visual and 3D data allows the systems to use both computer vision and 3D processing and therefore to be aware of more details of the surrounding environment. The research described in this thesis was motivated by the need of scene mapping. Being aware of the surrounding environment is a key feature in many mobile robotics applications from simple robotic navigation to complex surveillance applications. In addition, the acquisition of a 3D model of the scenes is useful in many areas as video games scene modeling where well-known places are reconstructed and added to game systems or advertising where once you get the 3D model of one room the system can add furniture pieces using augmented reality techniques. In this thesis we perform an experimental study of the state-of-the-art registration methods to find which one fits better to our scene mapping purposes. Different methods are tested and analyzed on different scene distributions of visual and geometry appearance. In addition, this thesis proposes two methods for 3d data compression and representation of 3D maps. Our 3D representation proposal is based on the use of Growing Neural Gas (GNG) method. This Self-Organizing Maps (SOMs) has been successfully used for clustering, pattern recognition and topology representation of various kind of data. Until now, Self-Organizing Maps have been primarily computed offline and their application in 3D data has mainly focused on free noise models without considering time constraints. Self-organising neural models have the ability to provide a good representation of the input space. In particular, the Growing Neural Gas (GNG) is a suitable model because of its flexibility, rapid adaptation and excellent quality of representation. However, this type of learning is time consuming, specially for high-dimensional input data. Since real applications often work under time constraints, it is necessary to adapt the learning process in order to complete it in a predefined time. This thesis proposes a hardware implementation leveraging the computing power of modern GPUs which takes advantage of a new paradigm coined as General-Purpose Computing on Graphics Processing Units (GPGPU). Our proposed geometrical 3D compression method seeks to reduce the 3D information using plane detection as basic structure to compress the data. This is due to our target environments are man-made and therefore there are a lot of points that belong to a plane surface. Our proposed method is able to get good compression results in those man-made scenarios. The detected and compressed planes can be also used in other applications as surface reconstruction or plane-based registration algorithms. Finally, we have also demonstrated the goodness of the GPU technologies getting a high performance implementation of a CAD/CAM common technique called Virtual Digitizing.

Where Are We After Five Editions?: Robot Vision Challenge, a Competition that Evaluates Solutions for the Visual Place Classification Problem

This article describes the Robot Vision challenge, a competition that evaluates solutions for the visual place classification problem. Since its origin, this challenge has been proposed as a common benchmark where worldwide proposals are measured using a common overall score. Each new edition of the competition introduced novelties, both for the type of input data and subobjectives of the challenge. All the techniques used by the participants have been gathered up and published to make it accessible for future developments. The legacy of the Robot Vision challenge includes data sets, benchmarking techniques, and a wide experience in the place classification research that is reflected in this article.