Pasar al contenido principal

COMPORTAMIENTOS INTELIGENTES MULTITAREA DE ENJAMBRES ADAPTATIVOS

RTI2018-096219-B-I00

•
Nombre agencia financiadora Agencia Estatal de Investigación
Acrónimo agencia financiadora AEI
Programa Programa Estatal de I+D+i Orientado a los Retos de la Sociedad
Convocatoria Proyectos I+D+i Retos de Investigación
Año convocatoria 2018
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD DE ALICANTE

Resultados relacionados

Resultados totales (Incluyendo duplicados): 12
Encontrada(s) 1 página(s)

RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

UAV + BIM: Incorporation of Photogrammetric Techniques in Architectural Projects with Building Information Modeling Versus Classical Work Processes

  • Rizo-Maestre, Carlos
  • González Avilés, Ángel Benigno
  • Galiano-Garrigós, Antonio
  • Andújar-Montoya, María Dolores
  • Puchol García, Juan Antonio
  • Andújar-Montoya, María Dolores|||/items/996fa4b2-4c2b-4ec8-8f1d-080dcbbcf481
  • Galiano-Garrigós, Antonio|||/items/e3920700-d391-4957-b84c-7615cbeaf96c
  • González Avilés, Ángel Benigno|||/items/3da0f460-8d19-45e7-8456-3cf1b393b4f6
  • Puchol García, Juan Antonio|||/items/6f49553c-9062-47de-b2b7-2f867a93c739
  • Rizo-Maestre, Carlos|||/items/7945a54f-dc26-4b42-8c83-229c5580ef2b
The current computer technology facilitates the processing of large volumes of information in architectural design teams, in parallel with recent advances in-flight automation in unmanned aerial vehicles (UAVs) along with lower costs, facilitates their use to capture aerial photographs and obtain orthophotographs and 3D models of relief and terrain textures. With these technologies, 3D models can be produced that allow different geometric configurations of the distribution of construction elements on the ground to be analyzed. This article presents the process of implementation in a terrain integrated into the early stages of architectural design. A methodology is proposed that covers the detailed capture of terrain, the relationship with the architectural design environment, and its implementation on the plot. As a novelty, an inverse perspective to the remaining disciplines is presented, from the inside of the object to the outside. The proposed methodology for the use of UAVs integrates terrain capture, generation of the 3D mesh, superimposition of environmental realities and architectural design using building information modeling (BIM) technologies. In addition, it represents the beginning of a line of research on the implementation of the plot and the layout of foundations using UAVs. The results obtained in the study carried out in three different projects comparing traditional technologies with the integration of UAVs + BIM show a clear improvement in the second option. The use of new technologies applied to the execution and control of work not only improves accuracy but also reduces errors and saves time, which undoubtedly indicates significant savings in costs and deviations in the project.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Movement Patterns in Students Diagnosed with ADHD, Objective Measurement in a Natural Learning Environment

  • Sempere-Tortosa, Mireia
  • Fernández-Carrasco, Francisco
  • Navarro Soria, Ignasi
  • Rizo-Maestre, Carlos
  • Fernández-Carrasco, Francisco|||/items/5400c73a-85c4-4777-9fc7-6367e00ead71
  • Navarro Soria, Ignasi|||/items/1c7c4096-530f-4487-8b9c-e472938c5df6
  • Rizo-Maestre, Carlos|||/items/7945a54f-dc26-4b42-8c83-229c5580ef2b
  • Sempere-Tortosa, Mireia|||/items/830e6185-0ff6-433c-8558-be6aff5ecae5
Attention deficit hyperactivity disorder is the most common neuropsychological disorder in childhood and adolescence, affecting the basic psychological processes involved in learning, social adaptation and affective adjustment. From previous research, the disorder is linked to problems in different areas of development, with deficiencies in psychological processes leading to the development of the most common characteristics of the disorder such as inattention, excess of activity and lack of inhibitory control. As for the diagnosis, in spite of being a very frequent disorder, there are multiple controversies about which tools are the most suitable for evaluation. One of the most widespread tools in the professional field is behavior inventories such as the Strengths and Difficulties Questionnaires for Parents and Teachers or the ADHD Rating Scale-V. The main disadvantage of these assessment tools is that they do not provide an objective observation. For this reason, there are different studies focused on recording objective measures of the subjects’ movement, since hyperkinesia is one of the most characteristic symptoms of this disorder. In this sense, we have developed an application that, using a Kinect device, is capable of measuring the movement of the different parts of the body of up to six subjects in the classroom, being a natural context for the student. The main objective of this work is twofold, on the one hand, to investigate whether there are correlations between excessive movement and high scores in the inventories for the diagnosis of ADHD, Rating Scale-V and Strengths and Difficulties Questionnaire (SDQ) and, on the other hand, to determine which sections of the body present the most significant mobility in subjects diagnosed with ADHD. Results show that the control group, composed of neurotypical subjects, presents less kinaesthetic activity than the clinical group diagnosed with ADHD. This indicates that the experimental group presents one of the main characteristics of the disorder. In addition, results also show that practically all the measured body parts present significant differences, being higher in the clinical group, highlighting the head as the joint with the highest effect size.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Navegación Visual Autónoma en Robótica de Enjambre mediante Control Distribuido: Un Enfoque Híbrido con Aprendizaje por Refuerzo Multiagente

  • Díez, Álvaro
  • Díez, Álvaro|||/items/ab81fc32-cd4b-45d6-a9c9-993edc6b95c2
La mayoría de los entornos del mundo real se caracterizan por ser desconocidos, dinámicos y cambiantes, presentando problemas de alta complejidad. Los agentes destinados a solucionar estos problemas generalmente carecen de información o conocimiento a priori sobre estos entornos. Como respuesta a estas dificultades, surgieron los sistemas basados en el comportamiento, inspirados en procesos y mecanismos observados en diversos sistemas biológicos existentes en la naturaleza. En los últimos años, estos sistemas han sido ampliamente utilizados en el campo de la robótica, dando lugar a la robótica basada en el comportamiento. Esta permite acometer tareas cuya resolución puede resultar extremadamente complicada o inviable para un ser humano, ofreciendo la posibilidad de utilizar agentes capaces de resolver dichas tareas de manera autónoma, gracias a su capacidad para desplegar un comportamiento inteligente, reactivo y adaptativo. En este contexto, la presente tesis doctoral se centra en el diseño y desarrollo de un método innovador para el control distribuido de un sistema robótico de enjambre, compuesto por agentes reactivos de bajo coste. El enfoque propuesto está orientado a aplicaciones de navegación autónoma visual en entornos complejos, dinámicos y cambiantes del mundo real. La investigación aborda el desafío de generar comportamientos colectivos flexibles, robustos y escalables, que permitan a los robots del enjambre adaptarse en tiempo real a diversas situaciones y cambios en las condiciones del entorno. La principal aportación de esta tesis es la definición de un modelo híbrido que combina aprendizaje por refuerzo multiagente y neuroevolución para el diseño automático de controladores de enjambre. Este enfoque integra un modelo microscópico, basado en una arquitectura profunda reactiva para la navegación individual, y un modelo macroscópico que gobierna el comportamiento colectivo emergente. La arquitectura propuesta prioriza la eficiencia computacional y las capacidades en tiempo real, aspectos cruciales para su aplicación en robótica de bajo coste y sistemas artificiales de enjambre. La validez del enfoque propuesto se demuestra a través de una serie de experimentos rigurosos, tanto en entornos simulados como reales. A nivel microscópico, se valida la eficacia de los comportamientos reactivos individuales utilizando un robot de bajo coste, evidenciando la viabilidad del sistema de navegación autónoma en escenarios dinámicos y cambiantes. A nivel macroscópico, se evalúan los controladores de navegación colectiva desarrollados, analizando su robustez, flexibilidad y escalabilidad en diversas condiciones. Los resultados obtenidos confirman la capacidad del sistema para generar comportamientos de enjambre efectivos y adaptativos, adecuados para la implementación en entornos desafiantes del mundo real. La eficiencia del enfoque, basado en hardware de bajo coste y navegación reactiva sin mapas, destaca su potencial para aplicaciones prácticas en diversos ámbitos de la robótica, como exploración de áreas de desastre, monitorización ambiental o agricultura de precisión.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Objective Analysis of Movement in Subjects with ADHD. Multidisciplinary Control Tool for Students in the Classroom

  • Sempere-Tortosa, Mireia
  • Fernández-Carrasco, Francisco
  • Mora Lizán, Francisco José
  • Rizo-Maestre, Carlos
  • Fernández-Carrasco, Francisco|||/items/5400c73a-85c4-4777-9fc7-6367e00ead71
  • Mora Lizán, Francisco José|||/items/c7620d9a-4458-4bf0-a8b9-6feba0924f10
  • Rizo-Maestre, Carlos|||/items/7945a54f-dc26-4b42-8c83-229c5580ef2b
  • Sempere-Tortosa, Mireia|||/items/830e6185-0ff6-433c-8558-be6aff5ecae5
The term Attention Deficit Hyperactivity Disorder (ADHD) has a long history of problems behind it. The origin of all these problems lies in the lack of agreement in the assessment procedures and evaluation instruments. The diagnosis is clinical and is determined by the observation and information provided by parents and teachers. So, this is highly subjective and leads to disparate results. Therefore, on the one hand the inaccuracy of the diagnosis of ADHD, which has been based on subjective criteria, together with the fact that hyperactivity is one of the main symptoms of this disorder, implies that several studies (with limitations) have been carried out to record objective measures of movement in subjects in at least the last ten years. In order to solve some of this derived problems and limitations of previous studies, a computer program has been developed to objectively record the amount of movement of subjects. The main objective of this study is threefold: first to register the amount of movement of both experimental group and control group, then to compare them with the movement registered by observers and finally to determine the validity of the software developed as a tool to support the diagnosis of ADHD. Results show that there are significant differences in the amount of objective movement between a clinical group of subjects with ADHD and a control group, obtaining a higher average of movement the experimental group. In addition, results also demonstrate that the developed software is a valid tool for the evaluation of movement that solves the limitations of previous studies. The proposed tool is developed from different aspects to give it a multidisciplinary character.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Radon Gas in the City of Alicante. High Risk of Low Indoor Air Quality in Poorly Ventilated Buildings

  • Rizo-Maestre, Carlos
  • Echarri-Iribarren, Víctor
  • Echarri-Iribarren, Víctor|||/items/7a3a9348-e61e-4195-b1f5-bbda963d3d05
  • Rizo-Maestre, Carlos|||/items/7945a54f-dc26-4b42-8c83-229c5580ef2b
In December 2019, Spain considered for the first time the presence of radon to the Technical Building Code (Basic Document HS 6: Radon Exposure Protection), although it only mentions minimum presences and the need for ventilation. This research shows that in buried structures or in places with little ventilation, even in soils with a low probability of granite, a high content of radon gas can be found. The city of Alicante has been used as a measurement location for different architectural sites; here, the level of 100 Bq/m3 is the first threshold where the gas must be monitored, and the level of 300 Bq/m3 is the maximum threshold above which corrective ventilation measures must be taken. The research conducted during the years 2015 and 2016 shows that it is necessary to account for also the areas considered to be “low presence of radon gas” to achieve healthy constructions. The renewal of air in the different places will be tested for the presence of radon, i.e., the greater the accumulation is, the less ventilation and the greater the risk of accumulation of radon gas. This study is located in the city of Alicante, where the seven civil constructions are located: two Civil War shelters, the Santa Barbara Castle, the Ereta Powder Keg, the Luceros-Marq and Serra Grossa railway tunnels and the Británica underground deposits. Radon gas is currently a concern for major health and medical agencies because it is considered to be a chemical element that is very harmful to people. The World Health Organization is one of the organisations that has the objective of studying and researching this element, to develop solutions. Radon gas is normally found in a gaseous state and is highly radioactive. It is present in many terrains and it is mostly found in those with granite; although the presence of this element is very low, there is always a minimum presence. In the past, in nongranite soils, the dose of radon was considered to be so low that it was insignificant. Therefore, in this research, the aim is to consider the high presence of radon gas in nongranite soils as long as the conditions for its accumulation are present.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Learning a Swarm Foraging Behavior with Microscopic Fuzzy Controllers Using Deep Reinforcement Learning

  • Aznar Gregori, Fidel
  • Pujol, Mar
  • Rizo, Ramón
  • Aznar Gregori, Fidel|||/items/bfc4a10c-2436-455c-b46d-a905a778a632
  • Pujol, Mar|||/items/041a4307-ba25-407c-9780-971b7f92890d
  • Rizo, Ramón|||/items/a500bfe2-013b-4f8f-8131-a7054ca58f19
This article presents a macroscopic swarm foraging behavior obtained using deep reinforcement learning. The selected behavior is a complex task in which a group of simple agents must be directed towards an object to move it to a target position without the use of special gripping mechanisms, using only their own bodies. Our system has been designed to use and combine basic fuzzy behaviors to control obstacle avoidance and the low-level rendezvous processes needed for the foraging task. We use a realistically modeled swarm based on differential robots equipped with light detection and ranging (LiDAR) sensors. It is important to highlight that the obtained macroscopic behavior, in contrast to that of end-to-end systems, combines existing microscopic tasks, which allows us to apply these learning techniques even with the dimensionality and complexity of the problem in a realistic robotic swarm system. The presented behavior is capable of correctly developing the macroscopic foraging task in a robust and scalable way, even in situations that have not been seen in the training phase. An exhaustive analysis of the obtained behavior is carried out, where both the movement of the swarm while performing the task and the swarm scalability are analyzed.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

UAV Deployment Using Two Levels of Stigmergy for Unstructured Environments

  • Aznar Gregori, Fidel
  • Pujol, Mar
  • Rizo, Ramón
  • Aznar Gregori, Fidel|||/items/bfc4a10c-2436-455c-b46d-a905a778a632
  • Pujol, Mar|||/items/041a4307-ba25-407c-9780-971b7f92890d
  • Rizo, Ramón|||/items/a500bfe2-013b-4f8f-8131-a7054ca58f19
This article will present two swarming behaviors for deployment in unstructured environments using unmanned aerial vehicles (UAVs). These behaviors will use stigmergy for communication. We found that there are currently few realistic deployment approaches that use stigmergy, due mainly to the difficulty of building transmitters and receivers for this type of communication. In this paper, we will provide the microscopic design of two behaviors with different technological and information requirements. We will compare them and also investigate how the number of agents influences the deployment. In this work, these behaviors will be exhaustively analyzed, taking into account different take-off time interval strategies, the number of collisions, and the time and energy required by the swarm. Numerous simulations will be conducted using unstructured maps generated at random, which will enable the establishment of the general functioning of the behaviors independently of the map used. Finally, we will show how both behaviors are capable of achieving the required deployment task in terms of covering time and energy consumed by the swarm. We will discuss how, depending on the type of map used, this task can be performed at a lower cost without using a more informed (but expensive) robotic swarm.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Deep Reinforcement Learning for Swarm Navigation Using Different Evolution Strategies

  • Díez, Álvaro
  • Aznar Gregori, Fidel
  • Aznar Gregori, Fidel|||/items/bfc4a10c-2436-455c-b46d-a905a778a632
  • Díez, Álvaro|||/items/ab81fc32-cd4b-45d6-a9c9-993edc6b95c2
At present, swarm robotics represents a very interesting option for solving complex problems in dynamic and changing environments. To do so, the agents that form a robot swarm must be able to deploy one or more collective behaviors. In order to learn the latter, in the last few years, different automatic design methods have become very popular. In this paper, we propose a hybrid method to automatically design an autonomous navigation behavior for swarm robotics. The idea is to combine multi-agent reinforcement learning and neuro-evolution. Furthermore, we present and compare two evolutionary algorithms whose aim is to implement the learning process of the aforementioned behavior. In particular, they are the well-known cross-entropy method and covariance matrix adaptation evolution strategy. Finally, we include some experiments in which we demonstrate that the covariance matrix adaptation evolution strategy is more appropriate than the cross-entropy method, when performing a vision-based collision-free exploration task in a simulated indoor corridor.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Modelado de un sistema multiagente para coordinación de dispositivos IoT

  • Nieto Pérez, Mario
En este trabajo se analizarán los distintos modelos/arquitecturas de agentes con el fin de aplicarlos al Internet de las cosas. La idea es proponer un modelo que permita resolver problemas de recogida de datos (sensores), planificación y consecución de objetivos, y actuación en dispositivos IoT.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Study of image sensors for enhanced face recognition at a distance in the Smart City context

  • Llauradó, José M.
  • Pujol, Francisco A.
  • Tomás, David
  • Visvizi, Anna
  • Pujol, Mar
  • Pujol, Francisco A.|||/items/f6155231-4b70-412a-a6af-2bea51ac458a
  • Pujol, Mar|||/items/041a4307-ba25-407c-9780-971b7f92890d
  • Tomás, David|||/items/cf031fc3-3c30-454b-a6df-3762c29cc296
Smart monitoring and surveillance systems have become one of the fundamental areas in the context of security applications in Smart Cities. In particular, video surveillance for Human Activity Recognition (HAR) applied to the recognition of potential offenders and to the detection and prevention of violent acts is a challenging task that is still undergoing. This paper presents a method based on deep learning for face recognition at a distance for security applications. Due to the absence of available datasets on face recognition at a distance, a methodology to generate a reliable dataset that relates the distance of the individuals from the camera, the focal length of the image sensors and the size in pixels of the target face is introduced. To generate the extended dataset, the Georgia Tech Face and Quality Dataset for Distance Faces databases were chosen. Our method is then tested and applied to a set of commercial image sensors for surveillance cameras using this dataset. The system achieves an average accuracy above 99% for several sensors and allows to calculate the maximum distance for a sensor to get the required accuracy in the recognition, which could be crucial in security applications in smart cities.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

El framework ROS. Implementación y experimentación de herramientas para el diseño y desarrollo de comportamientos en robots tipo Turtlebot

  • Sánchez Gea, Sandra
En este trabajo se utilizan diferentes herramientas proporcionadas por el framework ROS en el robot Turtlebot, trabajando tanto en la versión de simulador como experimentando con los robots reales. En primer lugar, se ha llevado a cabo un estudio del estado del arte actual del framework ROS, de los campos de aplicación actuales de ROS, y el concepto de robot móvil. A continuación, se ha explicado el funcionamiento de la arquitectura de la red en la que se basan las conexiones en ROS. Tras esto, se ha realizado una descripción de los componentes que conforman al robot Turtlebot utilizado en este proyecto. A continuación, se ha llevado a cabo la puesta en marcha del robot Turtlebot real y en simulación. Donde se han habilitado los sensores y actuadores que conforman al robot y se han realizado unas pruebas iniciales para verificar el correcto funcionamiento de éstos. Después de la puesta en marcha, se han realizado diferentes experimentos. En primer lugar, en cuanto a ROS, se ha visto cómo crear un espacio de trabajo y cómo crear unos paquetes básicos de un nodo publicador y un suscriptor. En cuanto a la simulación, se ha visto cómo insertar unos mundos creados en el simulador Gazebo junto con el robot Turtlebot. A continuación, se ha creado un paquete donde se ha utilizado uno de los mundos creados, el cual simula una carretera. Este paquete realiza un procesamiento de las imágenes capturadas por el Turtlebot donde se detectan las líneas de la carretera y a partir de éstas, se calculan las velocidades lineales y angulares que se publican para mover al robot a través de esta carretera de una forma autónoma. También se ha creado un paquete donde se han insertado varios robots Turtlebots dentro del mundo especificado en el simulador Gazebo, el cual ayudará a desarrollar futuros algoritmos para crear comportamientos multirobots. En cuanto al robot real, se ha conseguido realizar una conexión entre los dos PC, el del Turtlebot y el ordenador personal, en la que ambos se encuentran dentro de la misma red de nodos de ROS, donde se consigue una comunicación transversal en la que desde los dos PC se pueden ejecutar nodos de forma bidireccional. También se han construido diversos mapas a través de los datos capturados a partir de los ficheros bags y en tiempo real, del robot en simulación y del Turtlebot real. También se ha conseguido mover al robot Turtlebot de forma autónoma solo indicando el objetivo a alcanzar a partir de estos mapas conocidos. Por último, se han realizado distintas pruebas con la herramienta ROS Toolbox de Matlab, en las que Matlab se ha conectado al mismo ROS Master en el que se encontraba también conectados los nodos del Turtlebot, real y simulado. A través de Matlab se han realizado las mismas pruebas realizadas con las otras herramientas de ROS, creación de un suscriptor y un publicador, también, la construcción de un mapa a partir de una rosbag.
RUA. Repositorio Institucional de la Universidad de Alicante
Publicación Otros (Other)

Obtaining emergent behaviors for swarm robotics singling with deep reinforcement learning

  • Arques Corrales, Pilar
  • Aznar Gregori, Fidel
  • Pujol, Mar
  • Rizo, Ramón
  • Arques Corrales, Pilar|||/items/7ad758de-c67a-4e5b-9dca-6c55ae6098d1
  • Aznar Gregori, Fidel|||/items/bfc4a10c-2436-455c-b46d-a905a778a632
  • Pujol, Mar|||/items/041a4307-ba25-407c-9780-971b7f92890d
  • Rizo, Ramón|||/items/a500bfe2-013b-4f8f-8131-a7054ca58f19
Isolating (singling) an individual from a group can be essential for protection, rescue or capture tasks. In this paper a system with multiple shepherds who must coordinate the sheep to achieve a specific singleness is proposed. We present a realistically modeled system that will be finally tested in a real robotic system. We want to encourage the adaptability of the system and provide different solutions by promoting the emergence of the swarm. In this line we will focus on the use of reinforcement learning, avoiding a manual design of the behavior in order to not restrict the resulting behaviors and to facilitate their adaptation. A detailed MDP model will be specified as well as the keys to reduce its dimensionality and facilitate its training. We will check the results of the obtained singling policy with respect to a greedy policy and focus on evaluating different behavioral strategies that can solve the problem in different ways. In addition, one of the obtained policies will be analyzed in detail to check both its robustness and its scalability with respect to the number of shepherds and sheep. This policy will be finally tested on a physical robotic swarm.