Papers by Dennis Barrios Aranibar
Demystifying Educational Robotics with FOCORE: from Very Low Cost Software and Hardware Technologies to the Development of New Methodologies and Curriculum for Continuing Teacher Education and Teaching of Brazilian Basic Education Students
2020 Latin American Robotics Symposium (LARS), 2020 Brazilian Symposium on Robotics (SBR) and 2020 Workshop on Robotics in Education (WRE)
This article is of informative proposal about the first research network with Latin American scop... more This article is of informative proposal about the first research network with Latin American scope in the area of Robotics in Education (RE). We came up with a heterogeneous and multidisciplinary group, with the participation of universities and federal institutes in Brazil and a Peruvian university. The network aims to Study and Develop Methodologies for Implementing Educational Robotics in Public Education Systems. The proposal emphasizes the analysis of policies and experiences in the Continuing Education of Teachers as well as the development and improvement of software, hardware, curricula, and teaching methodologies in the Robotics in Education area. During the manuscript, some of the state of the art is given on public policies aiming the use of Educational Robotics in the Brazilian education scenario, with a critical reflection on the contributions of RE, either as an active methodology for different purposes or in the teaching-learning process in public schools. Next, based on the initial study, we will be showing developed and improved tools for RE, including software and hardware. Our studies also seek the scientific foundations to propose improvements in curricula and course proposals for continuing education and learning of RE in schools.
A new approach for supervised learning based influence value reinforcement learning
Proceedings of the 2nd International Conference on Machine Learning and Soft Computing
The neural self-organization, is an innate feature of the mammal's brains, and is necessary f... more The neural self-organization, is an innate feature of the mammal's brains, and is necessary for its operation. The most known neuronal models that use this characteristic are the self-organized maps (SOM) and the adaptive resonance theory (ART), but those models, did not take the neuron as a processing unit, as the biological counterpart. On the other hand, the influence value learning paradigm [1], used in multi-agent environments, proof that agents can communicate with each other [2]; and they can self-organize to assign tasks; without any interference. Motivated by this missing feature in artificial networks, and with the influence value reinforcement learning algorithm; a new approach to supervised learning was modeled using the neuron as an agent learning by reinforcement.
An Approach for Representing Maps in SLAM Based on Grid Maps and Sparse Matrices
2019 Latin American Robotics Symposium (LARS), 2019 Brazilian Symposium on Robotics (SBR) and 2019 Workshop on Robotics in Education (WRE)
Graphs and matrices are widely used to represent maps in solutions for Simultaneous Localization ... more Graphs and matrices are widely used to represent maps in solutions for Simultaneous Localization and Mapping (SLAM). Space representation is crucial for map creation, finding the shortest path, planning a trajectory or route, tracking landmarks, etc. The choice of the data structure can present disadvantages such as more memory consumption when using matrices, or, the need of a transforming function to represent a map when graphs (as structure) are applied as structure. This paper propose a representing map using a storing structure that allows dynamic growth and optimization of memory consumption in SLAM. The so-called MPTE-SLAM structure, proposed here, suggests the usage of a sparse matrix of occupancy matrices to represent a map for SLAM. The goal of this approach is to reduce the use of memory (if compared with traditional metric maps also) also having a compact representation suitable for path planning and other tasks for autonomous robots; without the need of a transforming function (if compared with graph based solutions that return a map). This approach allows the addition and updating of values of the sub-matrices quickly and with dynamic growth on the overall map within the structure in the same way graphs permit; also it is suitable for a real-time implementation of solutions for SLAM. For implementation and testing, a non-holonomic mobile robot in an indoor environment was used. The final results showed that the MPTE-SLAM structure uses less memory than metric maps when optimal sub-matrices sizes are applied. Through experimentation an optimal size for submatrices (10x10) was determined; this size yielded a memory consumption less than 4.3 GB in all test experiments; also, it was observed that this approach also has dynamic growth in the structure plus the no-loss of its metric nature which gives it an advantage over graph-based representations.
Optimal Selection and Identification of Defects in Chestnuts Processing, through Computer Vision, Taking Advantage of its Inherent Characteristics
2019 24th IEEE International Conference on Emerging Technologies and Factory Automation (ETFA)
In the agro-industry automation, computer vision has become very important to the product selecti... more In the agro-industry automation, computer vision has become very important to the product selection and classification process. The problem becomes more challenging when it is necessary to detect defects or diseases in the product images. In literature, it was observed that when the fruit or vegetable image is treated as only one problem, efficiency is lower than when dividing it into sub-problems considering regions with similar appearance. Thus, in this paper, the target is to automate the detection and identification of visual defects in Brazil nuts by dividing the problem into two sub-problems (pulp and epidermis defects recognition) and by using color, shape and texture descriptors. First, the original image is segmented into two regions (one dark and one light). Then, First Order Descriptor, is applied to detect the presence or absence of defects in each region through the texture descriptor. Next, color, size and texture descriptors are used to the identification of each defect. This approach improves results obtained in previous research (Álvarez-Valera et al. [1]). We obtained an efficiency rate of 98.03 % with a processing time of 75 ms at worst and 51 at the best for every 3 images processed, unlike the previous attempt that had an efficiency rate of 91.79 % with a processing time of 130 ms. Finally, this approach can be applied in different types of products with other characteristics, since its inherent characteristics allows us to divide the original problem in two or more sub-problems.
Sensors
Many authors have been working on approaches that can be applied to social robots to allow a more... more Many authors have been working on approaches that can be applied to social robots to allow a more realistic/comfortable relationship between humans and robots in the same space. This paper proposes a new navigation strategy for social environments by recognizing and considering the social conventions of people and groups. To achieve that, we proposed the application of Delaunay triangulation for connecting people as vertices of a triangle network. Then, we defined a complete asymmetric Gaussian function (for individuals and groups) to decide zones where the robot must avoid passing. Furthermore, a feature generalization scheme called socialization feature was proposed to incorporate perception information that can be used to change the variance of the Gaussian function. Simulation results have been presented to demonstrate that the proposed approach can modify the path according to the perception of the robot compared to a standard A* algorithm.
A Survey of Ontologies for Simultaneous Localization and Mapping in Mobile Robots
ACM Computing Surveys
Autonomous robots are playing important roles in academic, technological, and scientific activiti... more Autonomous robots are playing important roles in academic, technological, and scientific activities. Thus, their behavior is getting more complex, particularly, in tasks related to mapping an environment and localizing themselves. These tasks comprise the Simultaneous Localization and Mapping (SLAM) problem. Representation of knowledge related to the SLAM problem with a standard, flexible, and well-defined model, provides the base to develop efficient and interoperable solutions. As many existing works demonstrate, Semantic Web seems to be a clear approach, since they have formulated ontologies, as the base data model to represent such knowledge. In this article, we survey the most popular and recent SLAM ontologies with our aim being threefold: (i) propose a classification of SLAM ontologies according to the main knowledge needed to model the SLAM problem; (ii) identify existing ontologies for classifying, comparing, and contrasting them, in order to conceptualize SLAM domain for m...
Journal of Intelligent & Robotic Systems, 2021
With advances in science and technology, several innovative researches have been developed trying... more With advances in science and technology, several innovative researches have been developed trying to figure out the main problems related to children's learning. It is known that issues such as frustration and inattention, between others, affect student learning. In this fashion, robotics is an important resource that can be used towards helping to solve these issues, empowering our students in order to push their learning up. In this case, robotic tools are generally used considering two different paradigms: as the main focus and as a secondary focus. Actually, these paradigms define the way that Educational Robotics is implemented in schools. Most of the approaches have implemented it as the main focus, which is teaching Robotics. Nevertheless, there are quite a few works that implement robotics as a secondary focus, which is currently assisting the learning process in several disciplines. The main contribution of this work is a complete three steps methodology for Robotics in Education to guide projects in order to either use it alone or to teach robotics with others topics. Our experiments show the importance of devising a study plan and evaluation method because the process is iterative and could improve the final results. As a novelty, here we have joined and extended our previous works by proposing a new set of methods with guidelines and strategies for applying the educational robotics standard curriculum for kids, named EDUROSC-Kids. We propose several tools that have been developed to organize the learning topics of Robotics for children, including the desired outcomes during the learning process. As said our current approach is divided in three steps (or phases): setting up the environment, defining the project, and performing evaluation. The proposed curriculum organizes robotics contents into five disciplines: Robotics and Society, Mechanics, Electronics, Programming, and Control Theory. Also, it considers a set of topics for each discipline and defines the level of knowledge that is recommended to achieve each group of children based on Bloom's Nomenclature. The contribution on this paper is a crucial step towards linking the general learning process with Educational Robotics approaches. Our methodology is validated by presenting practical experiences with application of EDUROSC-kids and the proposed method with a rubric guidelines into groups of children.
Fast Iterative and Fast Iterative Square Methods for Path Planning in Mobile Robots
2018 Latin American Robotic Symposium, 2018 Brazilian Symposium on Robotics (SBR) and 2018 Workshop on Robotics in Education (WRE), 2018
Autonomous mobile robots face many challenges, and path planning is one of the first to be solved... more Autonomous mobile robots face many challenges, and path planning is one of the first to be solved. To solve path planning, many solutions have been proposed, being the deterministic methods the ones that produced the best results. One of the these methods inspired by the concept of Potential Fields is the Fast Marching Method. This method belongs to a family of methods that solve the Eikonal equation, called Fast Methods. However, the only method applied to solve path planning is Fast Marching Method, then we propose to use the Fast Iterative Method instead due its scalability feature. Also we proposed an improvement to Fast Iterative Method called Fast Iterative Method Square, this comes from the need to have methods that generates safe paths. This proposal seeks to demostrate that the Fast Iterative Method and Fast Iterative Method Square are applicable to solve path planning and that they also present a better performance than the Fast Marching Method in all the comparisons made. In addition, Fast Iterative Method allows us to handle maps of large dimensions due to its highly parallel feature.
Micromachines, 2021
Nowadays, mobile robots are playing an important role in different areas of science, industry, ac... more Nowadays, mobile robots are playing an important role in different areas of science, industry, academia and even in everyday life. In this sense, their abilities and behaviours become increasingly complex. In particular, in indoor environments, such as hospitals, schools, banks and museums, where the robot coincides with people and other robots, its movement and navigation must be programmed and adapted to robot–robot and human–robot interactions. However, existing approaches are focused either on multi-robot navigation (robot–robot interaction) or social navigation with human presence (human–robot interaction), neglecting the integration of both approaches. Proxemic interaction is recently being used in this domain of research, to improve Human–Robot Interaction (HRI). In this context, we propose an autonomous navigation approach for mobile robots in indoor environments, based on the principles of proxemic theory, integrated with classical navigation algorithms, such as ORCA, Socia...
Local Coordination Diagrams for Collision Avoidance in Multi-Robot Path Planning
2019 Latin American Robotics Symposium (LARS), 2019 Brazilian Symposium on Robotics (SBR) and 2019 Workshop on Robotics in Education (WRE), 2019
Autonomous robots are intended to help humans performing a lot of different tasks in a safer and ... more Autonomous robots are intended to help humans performing a lot of different tasks in a safer and more efficient way. Some of those tasks must be solved by a group of autonomous robots. Also, when the task can be solved only by one robot, for cost constraints, it is cheaper (for development and maintenance) to implement solutions including a group of simple robots. Solutions including multiple robots have to solve group problems like communication and coordination; also, common problems of autonomous robots like the widely studied problem of Path Planning must be rethought. In this case, finding a collision-free path is not enough because each robot also has to avoid collisions with other robots (by coordinating their movements). In this scenario, the path planning problem turns into the multi-robot motion planning (MRMP) problem. There are two approaches for solving the MRMP problem: coupled and decoupled. This work is focused on the decoupled approach because it has the potential to solve MRMP not only in a centralized way but, also in a concurrent or distributed way. In this sense, a new parallelizable algorithm, called Local Coordination Diagrams - LCD, is presented in this paper. Experimental results show that our approach can be applied efficiently to a large number of robots.
The shucking of the Argopecten Purpuratus (scallop) is a process where the viscera and the kidney... more The shucking of the Argopecten Purpuratus (scallop) is a process where the viscera and the kidney are removed. The use of computer vision is an alternative to automate this process, however the extraction of the kidney is the most difficult task because in many cases the scallop's stem covers the kidney avoiding the kidney extraction. Due to this circumstance the whole shucking process fails. This paper proposes a method of computer vision to determine the movement direction of a finger mechatronic, which move the stem in the opposite direction of the kidney improving its visibility. In order to achieve this improvement we use image segmentation, logical operations on images, geometric and mathematical calculations, etc. The proposed method provides the efficiency 99.38% for removing Argopecten Purpuratus's kidney.
Sistema de Custos para o Centro de Diagnóstico por Imagens do Hospital Universitário Onofre Lopes
Neste artigo apresentamos um sistema computacional de custos desenvolvido especialmente para o Ce... more Neste artigo apresentamos um sistema computacional de custos desenvolvido especialmente para o Centro de Diagnostico por Imagens (CDI) do Hospital Universitario Onofre Lopes (HUOL). Este sistema de custos possibilita determinar os custos e receitas totais de cada procedimento realizado no CDI do HUOL, assim como, alguns dados estatisticos (custos medios por procedimento de acordo a um certo perfil, desvio padrao, etc.). Com isto pretende-se fornecer aos administradores do CDI do HUOL informacoes que lhes auxiliem na tomada de suas decisoes gerenciais.
Proceedings of the 35th Annual ACM Symposium on Applied Computing, 2020
Autonomous robots are playing important roles in academic, technological, and scientific activiti... more Autonomous robots are playing important roles in academic, technological, and scientific activities. Thus, their behavior is getting more complex. The main tasks of autonomous robots include mapping an environment and localize themselves. These tasks comprise the Simultaneous Localization and Mapping (SLAM) problem. Representation of the SLAM knowledge (e.g., robot characteristics, environment information, mapping and location information), with a standard and well-defined model, provides the base to develop efficient and interoperable solutions. However, as far as we know, there is not a common classification of such knowledge. Many existing works based on Semantic Web, have formulated ontologies to model information related to only some SLAM aspects, without a standard arrangement. In this paper, we propose a categorization of the knowledge managed in SLAM, based on existing ontologies and SLAM principles. We also classify recent and popular ontologies according to our proposed ca...
An Approach to Improve Simultaneous Localization and Mapping in Human Populated Environments
2021 Latin American Robotics Symposium (LARS), 2021 Brazilian Symposium on Robotics (SBR), and 2021 Workshop on Robotics in Education (WRE), 2021
One task that autonomous mobile robots have to perform in indoor spaces is to construct the map o... more One task that autonomous mobile robots have to perform in indoor spaces is to construct the map of their environment and report their location and orientation. This process is called Simultaneous Localization and Mapping (SLAM). To do so, robots extract data through their sensors. However, in dynamic indoor environments, moving objects induce the SLAM process to collapse or diverge. Moving objects should not be taken into account to generate the map and the occlusions that they generate should be solved. In this work, we propose a robust and flexible approach for SLAM algorithms to perform better in human populated environments; by integrating a filtering scheme that manages moving and static objects. To illustrate the suitability of our approach, we implement Gmapping, as the classical SLAM algorithm, and RANSAC as the filter. Nevertheless, any other SLAM algorithm and filter can be implemented. The simulation tests have been carried out using three museum environments, which the robot can face in real life. Through the results obtained, it is possible to conclude that the proposed approach is efficient in managing the sensor data, filtering the outliers, and thus removing dynamic objects from the map.
EduRoSC-Prof: Continuous Education Method for Teacher Formation in Educational Robotics for K-12 Teaching
2021 Latin American Robotics Symposium (LARS), 2021 Brazilian Symposium on Robotics (SBR), and 2021 Workshop on Robotics in Education (WRE), 2021
This paper proposes a methodological approach to continuum formation in Educational Robotics incl... more This paper proposes a methodological approach to continuum formation in Educational Robotics including a course approach and a standard curriculum to be used for training teachers of the K-12 system. This paper extends the EduRoSC-Kids as proposed previously, in order for providing teachers with a broader knowledge about the basic knowledge in Robotics and in Educational Robotics. The curriculum takes into account only the strict necessary contents for the teachers training considering that they are of cross disciplines as Physical Education, Biology, Mathematics, Portuguese, Geography, History, Arts, Sociology and Philosophy. These knowledge have proven to be enough, and further used in order to improve their students learning skills. The complete methodology has been successfully applied in the basic public educational system at Natal, in the Brazil Northeast, and in Arequipa, Perú, and has has proven to be a useful tool to help in the learning process.
CLEI electronic journal, 2016
As joint invited editors, we are proud to present the April 2016 and August 2016 issues of CLEIej... more As joint invited editors, we are proud to present the April 2016 and August 2016 issues of CLEIej, which include a number of revised and reviewed versions of the best papers presented at CLEI 2015 in Arequipa, Peru in October 2015. Authors were asked to prepare extended papers with new contributions with respect to the conference versions; a total of 16 papers were finally accepted and are now published in these two issues.
Reinforcement learning-based path planning for autonomous robots
The Brazil-nuts classification is a process where the brazil-nuts that do not present damages are... more The Brazil-nuts classification is a process where the brazil-nuts that do not present damages are classified according to its size: Large, Medium, Small and Tiny, prior to its exportation. The current method used for this process is a manual one, presenting several deficiencies, because it is subjective, slow and imprecise. In this sense, this study proposes the application of computational vision to automate this process, considering that there is a direct relation between the weight and size of a Brazil-nut, using a conversion factor and the Brazil-Nut's areas in order to estimate the weights and infer the type. The segmentation was obtained using the YCrCb colour space with a dynamic threshold for binarization, since the background of the images change by external factors such as illumination.The experimental results show that the performance achieved by this approach is 99.7%.
Argopecten Purpuratus Codification Based on Determination of Weight by Conversion and Adjustment Factors
2011 30th International Conference of the Chilean Computer Science Society, 2011
The Codification of Argopecten Purpuratus is a process, where the Stem and Coral are classified b... more The Codification of Argopecten Purpuratus is a process, where the Stem and Coral are classified by their weight in different codes. This process is done manually, therefore is linked to the subjectivity and the fatigue of people involved in the work. The use of computer vision is an alternative to automate this process. The present work proposes a method to classify the Argopecten Purpuratus based on determination of weights by conversion and adjustment factors. These factors use the area of the whole scallop and of the coral to make the estimation. Results of experiments show that the computer vision system achieved an overall acccuracy of 98%.
Estratégias baseadas em aprendizado para coordenação de uma frota de robôs em tarefas cooperativas
… work document. Departamento de Eng. de …, 2005
Uploads
Papers by Dennis Barrios Aranibar