Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.107
Ángela Sánchez-Pérez, J. M. Torres, Carlos A. Jara, J. Pomares, Gabriel J. García, A. Úbeda
espanolEn este articulo se presenta un esquema de control mioelectrico bidimensional orientado al control de la orientacion de protesis roboticas. Tras preprocesar la senal EMG (rectificado y suavizado) obtenida a partir de cuatro canales, se han aplicado cuatro clasificadores distintos con el fin de discriminar entre movimientos de flexion, extension, abduccion y aduccion de muneca. Once participantes han evaluado el esquema y los resultados obtenidos muestran una gran robustez superando el 70% de acierto en todos los casos y alcanzando, en muchos casos, el 80%. Ademas, los resultados obtenidos sugieren que no hay diferencias entre genero y estado fisico de los participantes. EnglishIn this paper, we present a bidimensional myoelectric control scheme oriented to the control of the orientation of a robotic hand prosthesis. After preprocessing the EMG signal (rectification and linear envelope) obtained from four channels, four different classifiers have been applied to discriminate between flexion, extension, adduction and abduction wrist movements. Eleven participants evaluated the control scheme showing results over a 70% accuracy, most of them surpassing an 80%. Moreover, the obtained results suggest that there are no differences in gender or physical condition.
{"title":"Evaluación de un esquema de control mioeléctrico bidimensional para prótesis robóticas","authors":"Ángela Sánchez-Pérez, J. M. Torres, Carlos A. Jara, J. Pomares, Gabriel J. García, A. Úbeda","doi":"10.17979/SPUDC.9788497497169.107","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.107","url":null,"abstract":"espanolEn este articulo se presenta un esquema de control mioelectrico bidimensional orientado al control de la orientacion de protesis roboticas. Tras preprocesar la senal EMG (rectificado y suavizado) obtenida a partir de cuatro canales, se han aplicado cuatro clasificadores distintos con el fin de discriminar entre movimientos de flexion, extension, abduccion y aduccion de muneca. Once participantes han evaluado el esquema y los resultados obtenidos muestran una gran robustez superando el 70% de acierto en todos los casos y alcanzando, en muchos casos, el 80%. Ademas, los resultados obtenidos sugieren que no hay diferencias entre genero y estado fisico de los participantes. EnglishIn this paper, we present a bidimensional myoelectric control scheme oriented to the control of the orientation of a robotic hand prosthesis. After preprocessing the EMG signal (rectification and linear envelope) obtained from four channels, four different classifiers have been applied to discriminate between flexion, extension, adduction and abduction wrist movements. Eleven participants evaluated the control scheme showing results over a 70% accuracy, most of them surpassing an 80%. Moreover, the obtained results suggest that there are no differences in gender or physical condition.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132113696","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.001
Juan Almenara Ahijón, Matilde Santos, M. Tomás-Rodríguez
Among the current deployment of renewable energy systems, floating wind turbines are a promising resource. They take advantage of the wind at deep seas, where it reaches higher and more constant speed. However, being located in deep waters they are subject to heavy loads, caused mainly by waves and wind. To reduce the vibrations in the structure, passive control devices can be used, which come from civil engineering. But its design and tuning are not a simple task due to the diverse goals that coexist in the application of these passive control systems. In this work the use of different methods of signal analysis for the optimization of TMD devices (Tuned Mass Damper) is explored through genetic algorithms. Various error criteria have been applied for the optimization in order to obtain a greater reduction of vibrations in the various elements of the floating turbine.
{"title":"Optimización heurística con criterios de error de control TMD en turbinas marinas flotantes","authors":"Juan Almenara Ahijón, Matilde Santos, M. Tomás-Rodríguez","doi":"10.17979/SPUDC.9788497497169.001","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.001","url":null,"abstract":"Among the current deployment of renewable energy systems, floating wind turbines are a promising resource. They take advantage of the wind at deep seas, where it reaches higher and more constant speed. However, being located in deep waters they are subject to heavy loads, caused mainly by waves and wind. To reduce the vibrations in the structure, passive control devices can be used, which come from civil engineering. But its design and tuning are not a simple task due to the diverse goals that coexist in the application of these passive control systems. In this work the use of different methods of signal analysis for the optimization of TMD devices (Tuned Mass Damper) is explored through genetic algorithms. Various error criteria have been applied for the optimization in order to obtain a greater reduction of vibrations in the various elements of the floating turbine.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116205542","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.421
Pablo Otálora Berenguel, L. Roca, Javier Bonilla, Jose Luis Alfonso Sanchez
espanolEste trabajo presenta un sistema de control adaptativo para un recuperador de gases como parte de una planta termosolar hibrida. La variable que se desea controlar en este sistema es la temperatura de salida de las sales, que presenta una dinamica fuertemente no lineal afectada por los flujos masicos de sales y gases. Por ello, el proposito de este articulo ha sido obtener los modelos necesarios para elaborar una estrategia de control con un buen desempeno en todos los puntos de operacion, sin necesidad de hacer uso de modelos no lineales, y utilizando esquemas de ajuste de ganancia por tabla e interpolacion de parametros. El algoritmo de control ha sido evaluado por completo en simulacion obteniendo resultados satisfactorios. EnglishThis paper presents a non-linear control approach for a heat recovery system in a hybrid power plant. The salt temperature is the controlled variable, which presents a highly non-linear dynamics, affected by the salt and gas mass ows. Therefore, the aim of this paper is to obtain the process models to elaborate a control strategy with an adequate performance in the desired operating point, without making use of non-linear models, and employing gain scheduling and parameter interpolation control schemes. The control algorithms have been completely evaluated in simulation obtaining satisfactory results.
{"title":"Control de un recuperador de sales en una planta termosolar híbrida","authors":"Pablo Otálora Berenguel, L. Roca, Javier Bonilla, Jose Luis Alfonso Sanchez","doi":"10.17979/SPUDC.9788497497169.421","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.421","url":null,"abstract":"espanolEste trabajo presenta un sistema de control adaptativo para un recuperador de gases como parte de una planta termosolar hibrida. La variable que se desea controlar en este sistema es la temperatura de salida de las sales, que presenta una dinamica fuertemente no lineal afectada por los flujos masicos de sales y gases. Por ello, el proposito de este articulo ha sido obtener los modelos necesarios para elaborar una estrategia de control con un buen desempeno en todos los puntos de operacion, sin necesidad de hacer uso de modelos no lineales, y utilizando esquemas de ajuste de ganancia por tabla e interpolacion de parametros. El algoritmo de control ha sido evaluado por completo en simulacion obteniendo resultados satisfactorios. EnglishThis paper presents a non-linear control approach for a heat recovery system in a hybrid power plant. The salt temperature is the controlled variable, which presents a highly non-linear dynamics, affected by the salt and gas mass ows. Therefore, the aim of this paper is to obtain the process models to elaborate a control strategy with an adequate performance in the desired operating point, without making use of non-linear models, and employing gain scheduling and parameter interpolation control schemes. The control algorithms have been completely evaluated in simulation obtaining satisfactory results.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123824069","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.756
E. D. Oña, Bartek Łukawski, A. Jardón, Carlos Balaguer
espanolA dia de hoy, el desarrollo de sistemas roboticos de asistencia en procedimientos de rehabilitacion ha sido elevado. Sin embargo, existen ciertas aspectos que limitan un uso mas extendido. Una posible causa puede ser que las mejoras en funcion motora no son traspasadas directamente al desarrollo de actividades de la vida diaria. Esto puede provenir de que las estrategias utilizadas no promueven el entrenamiento de otros factores esenciales en la generacion del movimiento, como el cognitivo y perceptual. En este articulo se presenta una estrategia para la rehabilitacion de la funcion motora de la extremidad superior que utiliza un brazo robotico para modelar la interaccion con el usuario, anadiendo en la tarea factores para la estimulacion cognitiva. El robot es usado para promover la movilizacion del brazo, mientras se propone al usuario manipular y reconocer objetos. Una webcam colocada en el extremo del robot es utilizada para reconocer los objetos. Ademas, los movimientos del usuario son monitorizados por una camara RGB-D para analizar su calidad. En la fase actual de desarrollo se ha completado la integracion de los componentes para la estrategia. La estrategia propuesta esta orientada a implementar tratamientos personalizables, que incluyan factores cognitivos en el mismo paradigma, y en un entorno mas cercano al de la vida diaria. EnglishAt present, the development of robot-aided systems in rehabilitative procedures has been high. However, there are aspects that limit extended use. A possible cause is that motor gains usually are not transferred to the performance of daily living activities. This fact likely is due to current strategies are enough not promoting the training of other relevant aspects involved in motion generation, such as cognition and perception. This paper presents a strategy for the rehabilitation of upper limb motor function using a robotic arm for modelling the human-robot interaction and adding a cognitive load. The robot is used for promoting the user's arm mobilisation and to propose identification and manipulation of objects. A webcam fixed on the robot gripper is employed to identify the objects. User's movements are registered via an RGB-D camera. In the current development stage, the integration of the strategy components has been successfully performed. The proposed strategy aims to implement customisable treatments that include cognitive aspects in the same paradigm, and into environments similar to the quotidian ones.
今天，机器人系统在康复过程中的辅助发展已经很高。然而，有一些方面限制了更广泛的使用。一个可能的原因可能是运动功能的改善没有直接传递到日常生活活动的发展中。这可能是由于所使用的策略没有促进运动产生的其他基本因素的训练，如认知和知觉。在这篇文章中，我们提出了一种策略，以恢复上肢的运动功能，使用机械手臂模拟与用户的互动，在任务中添加认知刺激因素。机器人被用来促进手臂的移动，同时建议用户操作和识别物体。放置在机器人末端的网络摄像头用来识别物体。此外，用户的动作由RGB-D摄像机监控，以分析其质量。在目前的发展阶段，战略各组成部分的整合已经完成。提出的策略旨在实施可定制的治疗，包括认知因素在相同的范式，并在一个更接近日常生活的环境。目前，机器人辅助系统在康复过程中的发展一直很高。然而，也有限制延长使用的方面。一个可能的原因是，运动的增加通常不会转移到日常生活活动的表现上。这可能是因为目前的战略没有充分促进运动产生所涉及的其他相关方面的培训，例如认知和知觉。This paper代表战略for the康复of上limb motor function using a robotic情景arm for the human-robot interaction and adding a cognitive load。该机器人用于促进用户的手臂动员，并提供对象的识别和操作。= =地理= =根据美国人口普查，这个县的面积为。用户的移动通过RGB-D摄像头记录。在目前的发展阶段，战略组成部分的整合已经成功完成。拟议的战略旨在实施可定制的治疗方法，将认知方面纳入相同的模式和类似的日常环境中。
{"title":"Hacia una estrategia asistida por robot para la recuperación de función motora de extremidad superior con aspectos cognitivos","authors":"E. D. Oña, Bartek Łukawski, A. Jardón, Carlos Balaguer","doi":"10.17979/SPUDC.9788497497169.756","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.756","url":null,"abstract":"espanolA dia de hoy, el desarrollo de sistemas roboticos de asistencia en procedimientos de rehabilitacion ha sido elevado. Sin embargo, existen ciertas aspectos que limitan un uso mas extendido. Una posible causa puede ser que las mejoras en funcion motora no son traspasadas directamente al desarrollo de actividades de la vida diaria. Esto puede provenir de que las estrategias utilizadas no promueven el entrenamiento de otros factores esenciales en la generacion del movimiento, como el cognitivo y perceptual. En este articulo se presenta una estrategia para la rehabilitacion de la funcion motora de la extremidad superior que utiliza un brazo robotico para modelar la interaccion con el usuario, anadiendo en la tarea factores para la estimulacion cognitiva. El robot es usado para promover la movilizacion del brazo, mientras se propone al usuario manipular y reconocer objetos. Una webcam colocada en el extremo del robot es utilizada para reconocer los objetos. Ademas, los movimientos del usuario son monitorizados por una camara RGB-D para analizar su calidad. En la fase actual de desarrollo se ha completado la integracion de los componentes para la estrategia. La estrategia propuesta esta orientada a implementar tratamientos personalizables, que incluyan factores cognitivos en el mismo paradigma, y en un entorno mas cercano al de la vida diaria. EnglishAt present, the development of robot-aided systems in rehabilitative procedures has been high. However, there are aspects that limit extended use. A possible cause is that motor gains usually are not transferred to the performance of daily living activities. This fact likely is due to current strategies are enough not promoting the training of other relevant aspects involved in motion generation, such as cognition and perception. This paper presents a strategy for the rehabilitation of upper limb motor function using a robotic arm for modelling the human-robot interaction and adding a cognitive load. The robot is used for promoting the user's arm mobilisation and to propose identification and manipulation of objects. A webcam fixed on the robot gripper is employed to identify the objects. User's movements are registered via an RGB-D camera. In the current development stage, the integration of the strategy components has been successfully performed. The proposed strategy aims to implement customisable treatments that include cognitive aspects in the same paradigm, and into environments similar to the quotidian ones.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122784984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.655
Ernesto Aranda Escolástico, María Guinaldo Losada, Sebastián Dormido Cantó, L. Colombo
espanolEn este articulo se estudia el problema del control de formaciones para sistemas multi-agente gobernados por una dinamica lagrangiana. Cada agente observa un conjunto convexo, el cual debe alcanzar, siendo el objetivo del grupo alcanzar la formacion deseada en estos conjuntos cuando la comunicacion entre los agentes esta sujeta a retardos. Se propone una nueva ley de control para alcanzar la formacion en los conjuntos objetivo, mientras que la velocidad de los agentes converge a cero. Finalmente, se han validado los resultados teoricos mediante un ejemplo de simulacion. EnglishWe study the problem of targeted formation control for multi-agent systems which evolves under the dynamics provided by a Lagrangian function. Each agent observes a convex set as a target and the goal of the swarm is to achieve the desired formation within these sets while the measurements from other agents are received with delay. A novel control law is proposed to achieve the formation in all targeted sets while the velocities of the agents are driven to zero. Applications and simulation results are given to validate the theoretical results.
本文研究了拉格朗日动力学控制下多智能体系统的构造控制问题。每一个凸,注意到该制度应该实现,目标集团联合编队飞行中这些当这个受重度行为者之间的沟通。提出了一种新的控制律来实现目标集的形成，同时agent的速度收敛到零。最后，通过仿真实例验证了理论结果。EnglishWe study the problem of货运地层之下multi-agent systems which evolves动态控制提供by a Lagrangian function。各agent convex注意到规定的and the goal of the swarm is to the的地层起这些sets而measurements内收到的其他用品are with delay。提出了一项新的控制法，以实现对所有目标组别的训练，同时将特工的速度降至零。应用和模拟结果是给予validate the理论成果。
{"title":"Control de formaciones sobre regiones objetivo de sistemas multi-agente lagrangianos con retardo","authors":"Ernesto Aranda Escolástico, María Guinaldo Losada, Sebastián Dormido Cantó, L. Colombo","doi":"10.17979/SPUDC.9788497497169.655","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.655","url":null,"abstract":"espanolEn este articulo se estudia el problema del control de formaciones para sistemas multi-agente gobernados por una dinamica lagrangiana. Cada agente observa un conjunto convexo, el cual debe alcanzar, siendo el objetivo del grupo alcanzar la formacion deseada en estos conjuntos cuando la comunicacion entre los agentes esta sujeta a retardos. Se propone una nueva ley de control para alcanzar la formacion en los conjuntos objetivo, mientras que la velocidad de los agentes converge a cero. Finalmente, se han validado los resultados teoricos mediante un ejemplo de simulacion. EnglishWe study the problem of targeted formation control for multi-agent systems which evolves under the dynamics provided by a Lagrangian function. Each agent observes a convex set as a target and the goal of the swarm is to achieve the desired formation within these sets while the measurements from other agents are received with delay. A novel control law is proposed to achieve the formation in all targeted sets while the velocities of the agents are driven to zero. Applications and simulation results are given to validate the theoretical results.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"80 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114643485","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.741
Á. Madridano, A. Al-Kaff, David Martín, A. Escalera, J. M. Armingol
espanolLos avances en robotica han provocado la aparicion del uso cooperativo y autonomo de distintos tipos de vehiculos,lo que permite completar de manera mas eficiente diferentes tipos de trabajos. Uno de los aspectos fundamentales relacionados con esta colaboracion es la planifcacion de rutas seguras, de tal manera, que se pueda garantizar el movimiento libre de colisiones de cada uno de los vehiculos por el area de trabajo. Por este motivo, se propone una planificacion de rutas multiples para un enjambre de Vehiculos Aereos No tripulados basado en la creacion de Hojas de Rutas Probabilisticas. El objetivo del trabajo es la implementacion de un metodo para la construccion de trayectorias seguras para cada uno de los agentes del enjambre en diversas situaciones, tales como: alcanzar un unico objetivo de manera conjunta, o dividirse para alcanzar diferentes ubicaciones con los de diversos vehiculos que conforman el enjambre (caso etiquetado y no etiquetado). Para ello, se emplearia Hojas de Rutas Probabilisticas, demostrando que aportan una solucion eficiente al problema de planificacion de multiples rutas. Ademas, se propone la integracion de esta solucion en una arquitectura en ROS que permita, en primera instancia obtener resultados dentro del entorno de simulacion Gazebo y, en un futuro, una sencilla implementacion de los algoritmos presentados en una ota de vehiculos real. Los resultados obtenidos muestran un rendimiento optimo en cuanto a tiempo de computo para las distintas situaciones propuestas y, la posible integracion en un framework versatil como ROS EnglishAdvances in robotics have resulted in the appearance of cooperative and autonomous use of dfferent types of vehicles, in order to complete different types of work more effciently. One of the fundamental aspects that allows to achieve this collaboration is the planning of safe routes, in such a way that it is possible to guarantee the free movement of collisions of each one of the vehicles by the work area. For this reason, in this work, it is proposed to plan multiple routes for a swarm of Unmanned Aerial Vehicles based on the creation of Probabilistic Road Maps. The objective is the implementation of a method that allows the construction of safe trajectories for each one of the agents of the swarm when a joint objective is desired, a labeled objective or when there are several objectives to be completed and several vehicles are available to do so ( unlabeled ), demonstrating that probabilistic roadmaps constitute a fast and effective method for the establishment of multiple routes. In addition, it is proposed to integrate this solution in an ROS architecture that allows, in the first instance, to obtain results within the Gazebo simulation environment and, in the future, a simple implementation of the algorithms presented in a real eet of vehicles. The results obtained show optimum performance in terms of computing time for the different situations proposed and possible integrati
西班牙机器人技术的进步导致了不同类型车辆的合作和自主使用的出现，这使得更有效地完成不同类型的工作成为可能。与这种合作相关的一个基本方面是规划安全路线，以确保每辆车在工作区域的自由碰撞移动。因此，提出了一种基于概率航路图创建的无人机群多航路规划方法。工作的目标是实施一方法为肯尼亚的安全途径为执法人员每人挤满各种情况,包括:实现一个单一的共同目标,或者分享实现不同位置与各种机动车记录构成苹果蜂(情况下标签和标签)。为了实现这一目标，我们将使用概率路径表，证明它们为多路径规划问题提供了一个有效的解决方案。此外，我们提出将该解决方案集成到ROS体系结构中，首先允许在露台模拟环境中获得结果，并在未来简单地实现在真实车辆ota中提出的算法。结果显示性能optimo至于时间的电脑所提议的不同情况,是否为integracion framework versatil作为凯皮EnglishAdvances in机器人向民兵in the appearance of cooperative and成行use of dfferent types of车辆,in order to complete types of work more effciently好处。实现这一合作的一个关键方面是规划安全道路，即确保每辆车在工作区域内的碰撞自由移动。为此，在本工作中，建议根据概率道路图的创建，为一群无人飞行器规划多条道路。民间is The implementation of a method that allows The construction of safe trajectories for每one of The制剂of The swarm when a joint目标是理想的,labeled目标or when there are若干目标to be completed和若干车辆are to do so (unlabeled),表明这概率roadmaps的a fast and effective method for The establishment of multiple route。此外，还建议将该解决方案集成到ROS架构中，以便在第一次实例中在凉亭模拟环境中获得结果，并在未来简单地实现在车辆的实际效果中提出的算法。结果表明，在不同情况下的计算时间方面具有最佳性能，并可能集成到一个通用框架(如ROS)中。
{"title":"Método de planificación de trayectorias múltiples para enjambre de UAVs","authors":"Á. Madridano, A. Al-Kaff, David Martín, A. Escalera, J. M. Armingol","doi":"10.17979/SPUDC.9788497497169.741","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.741","url":null,"abstract":"espanolLos avances en robotica han provocado la aparicion del uso cooperativo y autonomo de distintos tipos de vehiculos,lo que permite completar de manera mas eficiente diferentes tipos de trabajos. Uno de los aspectos fundamentales relacionados con esta colaboracion es la planifcacion de rutas seguras, de tal manera, que se pueda garantizar el movimiento libre de colisiones de cada uno de los vehiculos por el area de trabajo. Por este motivo, se propone una planificacion de rutas multiples para un enjambre de Vehiculos Aereos No tripulados basado en la creacion de Hojas de Rutas Probabilisticas. El objetivo del trabajo es la implementacion de un metodo para la construccion de trayectorias seguras para cada uno de los agentes del enjambre en diversas situaciones, tales como: alcanzar un unico objetivo de manera conjunta, o dividirse para alcanzar diferentes ubicaciones con los de diversos vehiculos que conforman el enjambre (caso etiquetado y no etiquetado). Para ello, se emplearia Hojas de Rutas Probabilisticas, demostrando que aportan una solucion eficiente al problema de planificacion de multiples rutas. Ademas, se propone la integracion de esta solucion en una arquitectura en ROS que permita, en primera instancia obtener resultados dentro del entorno de simulacion Gazebo y, en un futuro, una sencilla implementacion de los algoritmos presentados en una ota de vehiculos real. Los resultados obtenidos muestran un rendimiento optimo en cuanto a tiempo de computo para las distintas situaciones propuestas y, la posible integracion en un framework versatil como ROS EnglishAdvances in robotics have resulted in the appearance of cooperative and autonomous use of dfferent types of vehicles, in order to complete different types of work more effciently. One of the fundamental aspects that allows to achieve this collaboration is the planning of safe routes, in such a way that it is possible to guarantee the free movement of collisions of each one of the vehicles by the work area. For this reason, in this work, it is proposed to plan multiple routes for a swarm of Unmanned Aerial Vehicles based on the creation of Probabilistic Road Maps. The objective is the implementation of a method that allows the construction of safe trajectories for each one of the agents of the swarm when a joint objective is desired, a labeled objective or when there are several objectives to be completed and several vehicles are available to do so ( unlabeled ), demonstrating that probabilistic roadmaps constitute a fast and effective method for the establishment of multiple routes. In addition, it is proposed to integrate this solution in an ROS architecture that allows, in the first instance, to obtain results within the Gazebo simulation environment and, in the future, a simple implementation of the algorithms presented in a real eet of vehicles. The results obtained show optimum performance in terms of computing time for the different situations proposed and possible integrati","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131301194","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.663
Samuel Felipe Baltanas Molero, J. R. Sarmiento, J. Gonzalez-Jimenez
espanolLas expectativas de implantacion de vehiculos autonomos en nuestra sociedad (coches, UAVs, robots, etc.) esta impulsando la demanda de profesionales en el campo de la robotica movil. Dada la diversidad y complejidad de las materias que abarca la robotica, e.g. probabilidad, algebra lineal, calculo diferencial, mecanica, programacion, inteligencia artificial, etc., se hace necesario contar con material didactico innovador que facilite el aprendizaje y posibilite el desarrollo de problemas practicos soportados por sus correspondientes fundamentos teoricos. Este trabajo presenta una coleccion de documentos interactivos para el aprendizaje en robotica, basados en la novedosa tecno- log a Jupyter Notebook, la cual permite al docente incluir en el mismo documento texto, ecuaciones matematicas, recursos visuales como imagenes o videos, enlaces externos, e implementacion de codigo. De este modo, los documentos interactivos introducen temas fundamentales de la robotica movil situando al estudiante en un mejor contexto, incorporando los conceptos necesarios de distintas materias, y permitiendole interactuar y analizar los resultados obtenidos. Esta herramienta esta siendo integrada en diversos cursos de la Universidad de M alaga, entre ellos la asignatura obligatoria de Robotica del grado de Ingenieria Informatica (Computacion). EnglishThe recent popularity of mobile robotics has resulted in an increasing demand of practitioners with a solid theoretical and practical background in the eld. Nevertheless, the diversity of related topics e.g. probability, algebra, computing, etc. complicates the development of learning materials to reinforce these fundamentals. This paper presents a colection of interactive notebooks targeted at robotics courses, making use of the Jupyter Notebook framework, which allows the use of rich text components whithin a document, e.g. mathematical equations, multimedia content such as images or videos, external links and interactive code. Therefore, these interactive documents are able to introduce the mobile robotics fundamentals, in a proper context, interweaving the core concepts of the different elds, and allowing the student to analyze and interact with the results. The developed notebooks are meant to become a useful tool in academia, particularly in our courses at the University of Malaga, where they are being integrated.
在我们的社会中实施自动驾驶汽车(汽车、无人机、机器人等)的期望正在推动对移动机器人领域专业人员的需求。考虑到机器人所涵盖的学科的多样性和复杂性，如概率、线性代数、微分计算、力学、编程、人工智能等，有必要有创新的教学材料，以促进学习，并使实际问题的发展得到相应的理论基础的支持。这项工作提出了一个机器人学习的交互式文档集合，基于新的技术日志到木星笔记本，它允许教师包括文本，数学方程，视觉资源，如图像或视频，外部链接，和代码实现在同一个文档。通过这种方式，交互式文档介绍了移动机器人的基本主题，将学生置于一个更好的环境中，结合不同学科的必要概念，并允许他们互动和分析获得的结果。该工具被整合到M alaga大学的各种课程中，包括计算机工程学位(计算)机器人的必修课。最近EnglishThe popularity of mobile机器人in an需求不断增加了练习者with a solid理论实践背景in the eld。然而，相关科目的多样性，如概率、代数、计算等，使学习材料的开发更加复杂，以加强这些基础知识。本文介绍了一套针对机器人课程的交互式笔记本，使用Jupyter笔记本框架，允许在文档中使用富文本组件，例如数学方程、图像或视频等多媒体内容、外部链接和交互式代码。因此，这些交互式文档能够在适当的背景下介绍移动机器人的基本原理，将不同的elds的核心概念交织在一起，使学生能够分析并与结果进行互动。开发出来的笔记本将成为学术界的有用工具，特别是在马拉加大学我们的课程中，它们被整合在那里。
{"title":"Colección de Jupyter Notebooks para cursos de robótica móvil","authors":"Samuel Felipe Baltanas Molero, J. R. Sarmiento, J. Gonzalez-Jimenez","doi":"10.17979/SPUDC.9788497497169.663","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.663","url":null,"abstract":"espanolLas expectativas de implantacion de vehiculos autonomos en nuestra sociedad (coches, UAVs, robots, etc.) esta impulsando la demanda de profesionales en el campo de la robotica movil. Dada la diversidad y complejidad de las materias que abarca la robotica, e.g. probabilidad, algebra lineal, calculo diferencial, mecanica, programacion, inteligencia artificial, etc., se hace necesario contar con material didactico innovador que facilite el aprendizaje y posibilite el desarrollo de problemas practicos soportados por sus correspondientes fundamentos teoricos. Este trabajo presenta una coleccion de documentos interactivos para el aprendizaje en robotica, basados en la novedosa tecno- log a Jupyter Notebook, la cual permite al docente incluir en el mismo documento texto, ecuaciones matematicas, recursos visuales como imagenes o videos, enlaces externos, e implementacion de codigo. De este modo, los documentos interactivos introducen temas fundamentales de la robotica movil situando al estudiante en un mejor contexto, incorporando los conceptos necesarios de distintas materias, y permitiendole interactuar y analizar los resultados obtenidos. Esta herramienta esta siendo integrada en diversos cursos de la Universidad de M alaga, entre ellos la asignatura obligatoria de Robotica del grado de Ingenieria Informatica (Computacion). EnglishThe recent popularity of mobile robotics has resulted in an increasing demand of practitioners with a solid theoretical and practical background in the eld. Nevertheless, the diversity of related topics e.g. probability, algebra, computing, etc. complicates the development of learning materials to reinforce these fundamentals. This paper presents a colection of interactive notebooks targeted at robotics courses, making use of the Jupyter Notebook framework, which allows the use of rich text components whithin a document, e.g. mathematical equations, multimedia content such as images or videos, external links and interactive code. Therefore, these interactive documents are able to introduce the mobile robotics fundamentals, in a proper context, interweaving the core concepts of the different elds, and allowing the student to analyze and interact with the results. The developed notebooks are meant to become a useful tool in academia, particularly in our courses at the University of Malaga, where they are being integrated.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"18 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128900328","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.491
Manuel G. Satué, Luis Fernando Mendoza Castaño, M. Linares, F. Rubio
espanolEl presente trabajo trata sobre el modelado y simulacion de una instalacion de seguimiento solar fotovoltaico para produccion de energia. Se parte de una instalacion real con unas especificaciones de funcionamiento muy estrictas en cuanto al seguimiento del sol, debido a que alberga paneles solares fotovoltaicos de alta concentracion, donde pequenas desviaciones en el apuntamiento suponen grandes diferencias en la generacion de energia. El objetivo es la obtencion de un modelo matematico que sirva como simulador para ser utilizado como primera herramienta en el diseno y ajuste de los controladores o estrategias de seguimiento que se implementaran en la instalacion fotovoltaica. EnglishThe present work deals with the modelling and simulation of a photovoltaic solar tracking installation for energy production. It starts from a real installation with very strict performance specifications in terms of tracking the sun, because it houses high concentration photovoltai solar panels, where small deviations in pointing represent large diffeerences in power generation. The objective is to obtain a mathematical model that serves as a simulator and is used as the first tool in the design and adjustment of the controllers or tracking strategies that will be implemented in the photovoltaic installation.
{"title":"Modelado de un seguidor solar con paneles de alta concentración","authors":"Manuel G. Satué, Luis Fernando Mendoza Castaño, M. Linares, F. Rubio","doi":"10.17979/SPUDC.9788497497169.491","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.491","url":null,"abstract":"espanolEl presente trabajo trata sobre el modelado y simulacion de una instalacion de seguimiento solar fotovoltaico para produccion de energia. Se parte de una instalacion real con unas especificaciones de funcionamiento muy estrictas en cuanto al seguimiento del sol, debido a que alberga paneles solares fotovoltaicos de alta concentracion, donde pequenas desviaciones en el apuntamiento suponen grandes diferencias en la generacion de energia. El objetivo es la obtencion de un modelo matematico que sirva como simulador para ser utilizado como primera herramienta en el diseno y ajuste de los controladores o estrategias de seguimiento que se implementaran en la instalacion fotovoltaica. EnglishThe present work deals with the modelling and simulation of a photovoltaic solar tracking installation for energy production. It starts from a real installation with very strict performance specifications in terms of tracking the sun, because it houses high concentration photovoltai solar panels, where small deviations in pointing represent large diffeerences in power generation. The objective is to obtain a mathematical model that serves as a simulator and is used as the first tool in the design and adjustment of the controllers or tracking strategies that will be implemented in the photovoltaic installation.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126162408","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.842
V. Román, Sergio Cebollada López, L. Castelló, M. Flores, A. Gil, Óscar Reinoso García
espanolLos robots autonomos deben ser competentes en tareas de localizacion y creacion de mapas. Una de las posibles soluciones para estos problemas es emplear descriptores de apariencia global que obtienen un unico vector que describe globalmente una imagen panoramica previamente adquirida. Comparando los descriptores HOG y Gist, el proposito de este trabajo es el estudio de un nuevo modo de utilizar estos descriptores con la finalidad de sustituir el modo de empleo actual o lo que sera mas interesante combinar ambos modos para obtener un metodo de mayor calidad. Este trabajo se realiza con imagenes reales tomadas en un campo de trabajo heterogeneo en el que simultaneamente conviven personas y robots, por ello en las imagenes hay cambios de iluminacion, de mobiliario y personas que ocluyen parte de la escena. Como conclusiones se obtiene que el descriptor HOG tiene un resultado similar empleando ambos modos mientras que con el descriptor Gist se obtiene un peor resultado al utilizar el nuevo modo de ventanas verticales. EnglishMap building and localization are two im portant tasks that autonomous mobile robots have to deal with. In the last decades it has appeared many studies and techniques to approach these problems. This work brie y describes some possible solutions and it recommends global appearance descriptors to carry our localization tasks. The aim is to obtain a unique vector that describes globally the panoramic image. The mobile robot is able to estimate its position and to create maps using these methods, the result will be independent of the robot orientation. This paper goal is to use a brand new way to use the global appearance descriptors. The nal result could be to use this new way or, a more interesting result, use a combination between both ways taken as a result a third better way. This work will consist on a comparison between both global appearance method ways. The study has been carried out using real images taken in an heterogeneous atmosp here where humans and robots work toget her, for that reason light condition changes and humans can apppear on the scene.
自主机器人必须精通本地化和地图创建任务。解决这些问题的一种可能的方法是使用全局外观描述符，它获得一个单一的向量，全局描述先前获得的全景图像。通过比较HOG和Gist描述符，本文的目的是研究一种使用这些描述符的新方法，以取代当前的使用方法，或者更有趣的是将这两种方法结合起来，以获得更高质量的方法。这个作品是在一个异质的工作场所拍摄的真实图像，在那里人与机器人同时共存，所以在图像中有照明、家具和人的变化，遮住了部分场景。结论是，HOG描述符在使用两种模式时得到相似的结果，而Gist描述符在使用垂直窗口的新模式时得到较差的结果。建筑和本地化是自主移动机器人必须处理的两个重要任务。在过去的几十年里，出现了许多研究和技术来解决这些问题。本工作描述了一些可能的解决方案，并推荐了全球外观描述符来完成我们的本地化任务。= =地理= =根据美国人口普查，这个县的总面积为，其中土地和(0.964平方公里)水。The mobile机器人能够去判断其position and to create地图利用这些方法,因此will be independent of The orientation机器人。本文的目标是使用一种全新的方式来使用全局外观描述符。全国宇空实验室因此可以be to use this new way or more interesting不满,use a过分between both途径而因此第三a better way。本文将对这两种整体外观方法进行比较。该研究使用的是在人类和机器人一起工作的异质大气中拍摄的真实图像，因此光线条件发生了变化，人类可以在现场拍摄。
{"title":"Evaluación de nuevos modos de empleo de los descriptores de apariencia global en tareas de localización","authors":"V. Román, Sergio Cebollada López, L. Castelló, M. Flores, A. Gil, Óscar Reinoso García","doi":"10.17979/SPUDC.9788497497169.842","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.842","url":null,"abstract":"espanolLos robots autonomos deben ser competentes en tareas de localizacion y creacion de mapas. Una de las posibles soluciones para estos problemas es emplear descriptores de apariencia global que obtienen un unico vector que describe globalmente una imagen panoramica previamente adquirida. Comparando los descriptores HOG y Gist, el proposito de este trabajo es el estudio de un nuevo modo de utilizar estos descriptores con la finalidad de sustituir el modo de empleo actual o lo que sera mas interesante combinar ambos modos para obtener un metodo de mayor calidad. Este trabajo se realiza con imagenes reales tomadas en un campo de trabajo heterogeneo en el que simultaneamente conviven personas y robots, por ello en las imagenes hay cambios de iluminacion, de mobiliario y personas que ocluyen parte de la escena. Como conclusiones se obtiene que el descriptor HOG tiene un resultado similar empleando ambos modos mientras que con el descriptor Gist se obtiene un peor resultado al utilizar el nuevo modo de ventanas verticales. EnglishMap building and localization are two im portant tasks that autonomous mobile robots have to deal with. In the last decades it has appeared many studies and techniques to approach these problems. This work brie y describes some possible solutions and it recommends global appearance descriptors to carry our localization tasks. The aim is to obtain a unique vector that describes globally the panoramic image. The mobile robot is able to estimate its position and to create maps using these methods, the result will be independent of the robot orientation. This paper goal is to use a brand new way to use the global appearance descriptors. The nal result could be to use this new way or, a more interesting result, use a combination between both ways taken as a result a third better way. This work will consist on a comparison between both global appearance method ways. The study has been carried out using real images taken in an heterogeneous atmosp here where humans and robots work toget her, for that reason light condition changes and humans can apppear on the scene.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"293 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116423766","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2020-06-11DOI: 10.17979/SPUDC.9788497497169.048
J. Cordoba, Oscar Déniz Suárez, Gloria Bueno García
Breast biopsies are crucial in the process of detec ing a wide range of diseases such as breast cancer. The evaluation of these biopsies is performed by trained pathologists that are often overworked due to the increasing number of pathologies requested. Automatic tumour detection techniques have been developed, achieving very good results. In this work, we propose to classify breast biopsies in all the different types of tissue present in them. The tissue types were identified by hand-labeling them following the indications of an expert pathologist. Afterward, they were trained with diffeerent convolutional neural networks such as GoogleNet, AlexNet, SqueezeNet and DenseNet. Out of these four networks, GoogleNet outperformed all of them achieving 95.4% of accuracy. Finally, we tried to identify why the networks were underperforming while also suggesting how results could be improved.
{"title":"Deep learning for the automatic classification of tissue types in breast biopsies","authors":"J. Cordoba, Oscar Déniz Suárez, Gloria Bueno García","doi":"10.17979/SPUDC.9788497497169.048","DOIUrl":"https://doi.org/10.17979/SPUDC.9788497497169.048","url":null,"abstract":"Breast biopsies are crucial in the process of detec ing a wide range of diseases such as breast cancer. The evaluation of these biopsies is performed by trained pathologists that are often overworked due to the increasing number of pathologies requested. Automatic tumour detection techniques have been developed, achieving very good results. In this work, we propose to classify breast biopsies in all the different types of tissue present in them. The tissue types were identified by hand-labeling them following the indications of an expert pathologist. Afterward, they were trained with diffeerent convolutional neural networks such as GoogleNet, AlexNet, SqueezeNet and DenseNet. Out of these four networks, GoogleNet outperformed all of them achieving 95.4% of accuracy. Finally, we tried to identify why the networks were underperforming while also suggesting how results could be improved.","PeriodicalId":189601,"journal":{"name":"XL Jornadas de Automática: libro de actas (Ferrol, 4-6 de septiembre de 2019)","volume":"49 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2020-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121434897","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：