Vidal Moreno

Clustering and line detection in laser range measurements

This article presents two algorithms that extract information from laser range data. They are designed to work sequentially. The first method (dcc) separates the data into clusters by means of a convolution operation, using a high-pass filter. The second one (reholt) performs line detection in each of the clusters previously discovered. The reliability of the algorithms devised is tested on the experimental data collected both indoors and outdoors. When compared with other methods found in the literature, the ones proposed here prove to achieve higher performance.

A general method for C-space evaluation and its application to articulated robots

A general method to obtain the obstacle representation in the robot configuration space is presented. The method is based on a solid mathematical formalism that has as a key element a mathematical expression to evaluate the C-obstacle representation. The use of the proposed method and the choice of a suitable coordinate system, in the workspace as well as in the configuration space, lead to the fact that the obstacle representation in the configuration space could be seen as a convolution of two functions that describe the robot and the obstacles, respectively. The computational load can then be reduced drastically if convolution properties are applied. The method is directly applicable to mobile and articulated robots without any kind of adaptation or restriction of the shape of the elements that constitute the robot and the shape of the obstacles. Hence, its utilization is quite suitable to motion planning and control problems.

Mathematical formalism for the fast evaluation of the configuration space

In this paper a mathematical formalism for the configuration space evaluation of robots is presented, that tries to optimise the computational time. The use of the proposed method and the choice of a suitable coordinate system, in the workspace, as well as in the configuration space, lead to the fact that the obstacle representation in the configuration space could be seen as a convolution of two functions that describe the robot and the obstacles respectively. The method has been applied to mobile and articulated robots in the two dimensional plane, but its application to the most popular robots can be easily done. Additionally, the computational load is independent of the shape and number of obstacles and of the robot shape. The mathematical tool that has been used is the FFT (fast Fourier transform) since it eases the parallel implementation of the resulting algorithms reducing significantly the computational load.

An experience of a CORBA based architecture for computer integrated manufacturing

The main limitations of robotic workcell software design can be exposed in terms of rigidity and low levels of reusability, so temporal and economic costs related with software development are very high. We present a software architecture design with a layered structure. This architecture includes a middleware layer, which guarantees that decentralized control can be performed, which let us have a higher degree of flexibility. In order to prove the validity of this design, a set of components is being implemented in an industrial-like robotic workcell.

Robotics in Education

Robotics in Education (R-in-E) tries to strengthen the learning skills of future engineers and scientists by means of robot-based projects. Both in school and in college, presenting robots in the classroom will give students a more interesting (and fun) vision of science and engineering, and they will be able to observe directly the practical application of theoretical concepts in the fields of mathematics and technology. R-in-E is inclusive in nature, which can lead students to orient their university studies in fields related to the STEM subjects. From the point of view of University, robot-based projects put together, since their inception, key competences related to mechatronics. This will help students deepen their knowledge, and it will also help engineers and scientists of the 21st century to finish their studies. One could also point out that R-in-E furthers and strengthens both in curricular and in extracurricular activities values such as creativity, innovation, support, cooperation and teamwork. Clearly, these values must be promoted in our society. “R-in-E” puts together actors from disparate ecosystems: on the one hand, experts and researchers in education, teachers and professors, cultural entertainers in public and private institutions, and on the

A Real-time Door Detection System for Domestic Robotic Navigation

Human beings use doors to access rooms and corridors, to know where they are, to know where they have to go, etc. Similarly, it would be quite useful for robots to be able to detect doors in order to accomplish more complex and flexible navigation tasks. Such a goal is even more desirable when domestic environments are taken into account. Moreover, if the human-robot interaction is considered, the use of this semantic information can be broadly used. In this paper we present a solid and complete door detection system which fuses data from an end-user camera and a laser rangefinder. By using both Haar-like features and the Integral Image, the computation time is significantly reduced when compared to other methods found in the literature. Extensive tests in real-world environments have been performed in order to prove the efficiency, robustness and real-time ability of our system.

A real-time indoor localization approach integrated with a Geographic Information System (GIS)

Nowadays, accurate maps from mostly anywhere in the world can be obtained for free, with the exception of indoor spaces. However, the evidence seems to suggest that in the next few years indoor maps will be more and more available for anyone. Thus, profiting from the idea of easily obtainable indoor maps, we present a novel approach for real-time mobile robot localization that focuses on spatial reasoning at a high abstraction level. In order to manage and query existing indoor spatial models, we rely on the power of Geographic Information Systems (GIS) and spatial databases. Moreover, to extract symbolic information from the environment, we have developed a door detection system that fuses 2D laser and vision data. We have integrated these two ideas into an extended Kalman filter localization framework. Our proposal has been implemented and tested through autonomous navigation missions in real-world scenarios. Extensive experimental results are provided, which show robustness and accuracy concerning both door detection and localization. Indoor spaces, obtained from standard like AUTOCAD, are exploited by Spatial Database.We propose a real-time door detection system that fuses laser and vision data.The robot performs EKF localization, fusing our door detection system with the GIS.Our data modeling increases interoperability, scalability and spatial abstraction.The approach for robotic autonomous missions can be quite useful for rescue missions.

C-Space Evaluation for Mobile Robots at Large Workspaces

In this work we present a hierarchical method to evaluate the configuration space of a mobile robot working in large environments. It is based on convolution of discrete functions representing the workspace and the robot to obtain a function that represents the C-space. An important key in the work is that the representation of discrete functions is made by hierarchical data structures, so the information is highly compact, and high dimensions or/and resolutions workspaces can be considered. It is important to highlight that the hierarchical data structures no only have been used for the representation, but for the evaluation. So, intermediate arrays to represent the discrete information are not used in the proposed method. Also, as the method is hierarchical, calculations at high resolutions are performed only in those regions where it is necessary. This makes the method suitable when the robot moves in large environments where other methods will fail due to the memory needs.

Transnational lifelong education course in robotic systems

Robotics constitutes a multidisciplinary area, congregating knowledge from different scientific domains. The learning of robotic systems requires the acquisition of multidisciplinary scientific bases, and high integration and synthesis abilities, which is not an easy task. This paper describes the implementation of a lifelong course that aims to provide a global insight on robotics field, introducing the concepts and technologies for different domain applications, namely industrial robotics, autonomous mobile robotics and robotics applied in medicine. This is accomplished in an international framework where individual knowledge and experiences will be confronted in a multidisciplinary level and intercultural environment.

A new approach to robotics teaching at computer science engineering degree based on action-research

Due to its multidisciplinary nature, the progressive introduction of Robotics within the Engineering requires a pedagogy that should be capable of providing the key elements (how it is taught, what is taught and how to learn) in line with European convergence processes that are inherent in European Higher Education Area (EHEA). This paper provides a new approach to Robotics teaching that is specifically aimed to studies of Computer Engineering. In order to achieve this improvement process we propose the change in the educational practice through reflection and teaching experience, that is, using action-research in the classroom. We will show the underlying approach by describing the performed steps: problem detection, the plan formulation to solve the problem, the plan implementation in the classroom and the evaluation of the obtained results during the academic years 2007--08 through 2011--12.