DISA - LR - Comunicaciones en Congresos y otros eventos

Permanent URI for this collection

http://dspace.uc3m.es/handle/10016/11852

Browse

Now showing 1 - 20 of 40

Guest Editorial: "Humans and Humanoids Face to Face"
(World Scientific Publishing, 2015-09) Balaguer Bernaldo de Quirós, Carlos; Asfour, Tamim; Metta, Giorgio; Yokoi, Kazuhito; Monje Micharet, Concepción Alicia
Dynamic semantic ontology generation: a proposal for social robots
(Universitat Jaume I, 2021) Sevilla Salcedo, Javier; Quispe-Flores, Miguel Ángel; Carrasco-Martínez, Sara; Gomez-Jimenez, Jaime; Castillo Montoya, José Carlos; Castro González, Álvaro; Malfaz Vázquez, María Ángeles; Salichs Sánchez-Caballero, Miguel; Ministerio de Ciencia, Innovación y Universidades (España)
During a human-robot interaction by dialogue/voice, the robot cannot extract semantic meaning from the words used, limiting the intervention itself. Semantic knowledge could be a solution by structuring information according to its meaning and its semantic associations. Applied to social robotics, it could lead to a natural and fluid human-robot interaction. Ontologies are useful representations of semantic knowledge, as they capture the relationships between objects and entities. This paper presents new ideas for ontology generation using already generated ontologies as feedback in an iterative way to do it dynamically. This paper also collects and describes the concepts applied in the proposed methodology and discusses the challenges to be overcome.
Modelos de lenguaje natural para robots sociales
(Servizo de Publicacións Universidade da Coruña, 2022) Sevilla Salcedo, Javier; Carrasco Martinez, Sara; Castillo Montoya, José Carlos; Castro González, Álvaro; Salichs Sánchez-Caballero, Miguel; Comunidad de Madrid; Ministerio de Ciencia e Innovación (España)
Hoy en día, los robots se están abriendo paso en nuestras vidas en muchos campos. Uno de ellos es el social, en el que encontramos robots capaces de interactuar con las personas y realizar diferentes actividades. El habla de un robot social es un elemento central en la interacción humano-robot; para parecer natural y amigable, debe evitar los textos predefinidos repetitivos. Con esta premisa, junto con el importante crecimiento en el campo de los modelos de generación de lenguaje natural, este artículo explora las capacidades de los modelos de lenguaje natural para conducir a una interacción humano-robot más fluida y abrir un abanico de nuevas oportunidades y aplicaciones. Tras la implementación de las diferentes aplicaciones, se ha observado el potencial de la integración de la generación de lenguaje natural en el campo de la robótica. [Abstract] Nowadays, robots are making their way into our lives in many fields. One of them is social, where we find robots capable of interacting with people and performing different activities. The speech of a social robot is a central element in human-robot interaction; to sound natural and friendly, it must avoid repetitive predefined texts. With this premise, together with the significant growth in the field of natural language generation models, this paper explores the capabilities of natural language models to lead to a more fluent human-robot interaction and open up a range of new opportunities and applications. Following the implementation of different applications, the potential of integrating natural language generation in the field of robotics has been observed.
Neural Style Transfer with Twin-Delayed DDPG for Shared Control of Robotic Manipulators
(IEEE, 2022-05-23) Fernández Fernández, Raúl; Aggravi, Marco; Robuffo Giordano, Paolo; González Víctores, Juan Carlos; Pacchierotti, Claudio; Comunidad de Madrid; Agencia Estatal de Investigación (España); Universidad Carlos III de Madrid
Neural Style Transfer (NST) refers to a class of algorithms able to manipulate an element, most often images, to adopt the appearance or style of another one. Each element is defined as a combination of Content and Style: the Content can be conceptually defined as the “what” and the Style as the “how” of said element. In this context, we propose a custom NST framework for transferring a set of styles to the motion of a robotic manipulator, e.g., the same robotic task can be carried out in an “angry”, “happy”, “calm”, or “sad” way. An autoencoder architecture extracts and defines the Content and the Style of the target robot motions. A Twin Delayed Deep Deterministic Policy Gradient (TD3) network generates the robot control policy using the loss defined by the autoencoder. The proposed Neural Policy Style Transfer TD3 (NPST33) alters the robot motion by introducing the trained style. Such an approach can be implemented either offline, for carrying out autonomous robot motions in dynamic environments, or online, for adapting at runtime the style of a teleoperated robot. The considered styles can be learned online from human demonstrations. We carried out an evaluation with human subjects enrolling 73 volunteers, asking them to recognize the style behind some representative robotic motions. Results show a good recognition rate, proving that it is possible to convey different styles to a robot using this approach
A pilot study on the design considerations and user impressions of an assistive affordable device
(Consejo Superior de Investigaciones Científicas, 2016-05-26) Oña Simbaña, Edwin Daniel; Jardón Huete, Alberto; Balaguer Bernaldo de Quirós, Carlos; Martínez Martínez, Ariadna; Sánchez Herrera, Patricia; Miangolarra Page, Juan Carlos; Comunidad de Madrid
PRESSMATIC is a portable electromechanical device which was designed to assist people whose manual dexterity has been impaired for any reason (ageing, illness, paralysis, etc.). Its goal is restoring the loss of functionali ty of the thumb and forefinger in certain tasks that require pinching movements (e.g. grabbing scissors, tweezers, etc.). In this paper, a study on design considerations and user experience of PRESSMATIC is present ed. For that purpose, pilot trials were conducted at healthcare facilities. Target users evaluated several prototypes of PRESSMATIC and their opinions were registered. Based on this, the original design specifications were reviewed. First, the methodology used in the study is presented. Par ticipants, devices and tasks are described too. Then, trial results are shown and the system requirements will be discussed taking into account user ex perience. Finally, the study conclusions are shown.
New trends and challenges in the automatic generation of new tasks for humanoid robots
(Consejo Superior de Investigaciones Científicas (CSIC), 2016) Fernández Fernández, Raúl; González Víctores, Juan Carlos; Balaguer Bernaldo de Quirós, Carlos; Comunidad de Madrid
In this paper, the study and implementation of a task generation system that uses the information obtained from the user and already known cases is presented. One of the main objectives of the system is to introduce a new approach in robotics that takes into account the physical limitation of teaching and learning time, and thus the amount of knowledge that a robot can obtain of a given environment (tasks, objects, user preferences...), as a critical bottleneck of any robotic system. For this, the study of the Case Based Reasoning (CBR) problem is presented. Additionally, Base Trajec-tory Combination (BATC), a novel trajectory combination method based on a simplified CBR structure, using trajectories instead of high-level tasks, is proposed and explained. Finally, this system is tested with Move-it! as the simulation environment, using the humanoid robot TEO from Universidad Carlos III de Madrid as the robotic platform. The results of these experiments are also presented with the corresponding conclusions and future research lines.
Quick, Stat!: a statistical analysis of the Quick, Draw! Dataset
(ARGESIM, 2019) Fernández Fernández, Raúl; González Víctores, Juan Carlos; Estévez Fernández, David; Balaguer Bernaldo de Quirós, Carlos; Comunidad de Madrid
The Quick, Draw! Dataset is a Google dataset with a collection of 50 million drawings, divided in 345 categories, collected from the users of the game Quick, Draw!1. In contrast with most of the existing image datasets, in the Quick, Draw! Dataset, drawings are stored as time series of pencil positions instead of a bitmap matrix composed by pixels. This aspect makes this dataset the largest doodle dataset available at the time. The Quick, Draw! Dataset is presented as a great opportunity to researchers for developing and studying machine learning techniques. Due to the size of this dataset and the nature of its source, there is a scarce of information about the quality of the drawings contained. In this paper a statistical analysis of three of the classes contained in the Quick, Draw! Dataset is depicted: mountain, book and whale. The goal is to give to the reader a first impression of the data collected in this dataset. For the analysis of the quality of the drawings a ClassificationNeuralNetworkwas trained to obtain a classification score. Using this classification score and the parameters provided by the dataset, a statistical analysis of the quality and nature of the drawings contained in this dataset is provided.
Towards clothes hanging via cloth simulation and deep convolutional networks
(Argesim, 2021-03) Estévez Fernández, David; González Víctores, Juan Carlos; Fernández Fernández, Raúl; Balaguer Bernaldo de Quirós, Carlos; Comunidad de Madrid
People spend several hours a week doing laundry, with hanging clothes being one of the laundry tasks to be performed. Nevertheless, deformable object manipulation still proves to be a challenge for most robotic systems, due to the extremely large number of internal degrees of freedom of a piece of clothing and its chaotic nature. This work presents a step towards automated robot clothes hanging by modeling the dynamics of the hanging task via deep convolutional models. Two models are developed to address two different problems: determining if the garment will hang or not (classification), and estimating the future garment location in space (regression). Both models have been trained with a synthetic dataset formed by 15k examples generated though a dynamic simulation of a deformable object. Experiments show that the deep convolutional models presented perform better than a human expert, and that future predictions are largely influenced by time, with uncertainty influencing directly the accuracy of the predictions.
Assessment of Manual Dexterity in VR: Towards a Fully Automated Version of the Box and Blocks Test
(IOS Press, 2019-08-12) Oña Simbaña, Edwin Daniel; García A., Jaime; Raffe, William; Jardón Huete, Alberto; Balaguer Bernaldo de Quirós, Carlos
In recent years, the possibility of using serious gaming technology for the automation of clinical procedures for assessment of motor function have captured the interest of the research community. In this paper, a virtual version of the Box and Blocks Test (BBT) for manual dexterity assessment is presented. This game-like system combines the classical BBT mechanics with a play-centric approach to accomplish a fully automated test for assessing hand motor function, making it more accessible and easier to administer. Additionally, some variants of the traditional mechanics are proposed in order to fully exploit the advantages of the chosen technology. This ongoing research aims to provide the clinical practitioners with a customisable, intuitive, and reliable tool for the assessment and rehabilitation of hand motor function.
RoboCom++: Rethinking Robotics for the Robot Companion of the Future
(CEA-IFAC, 2017-06-08) Monje Micharet, Concepción Alicia; Balaguer Bernaldo de Quirós, Carlos; Comunidad de Madrid; Ministerio de Economía y Competitividad (España)
The main objective of the RoboCom++ proposal is to lay the foundation for a future global interdisciplinary research programme (e.g., a FET-Flagship project) on a new science-based transformative Robotics, to be launched by the end of the H2020 Programme. RoboCom++ will gather the community and organise the knowledge necessary to rethink the design principles and fabrication technologies of future robots. RoboCom++ will aim at developing the cooperative robots (or Companion Robots) of the year 2030, by fostering a deeply multidisciplinary, transnational and federated effort. The mechatronic paradigm adopted today, although successful, may prevent a wider use of robotic systems. For example, system complexity increases with functions, leading to more than linearly increasing costs and power usage and decreasing robustness. RoboCom++ will pursue a radically new design paradigm, grounded in the scientific studies of intelligence in nature. This approach will allow achieving complex functionalities in a new bodyware with limited use of computing resources, mass and energy, with the aim of exploiting compliance instead of fighting it. Simplification mechanisms will be based on the concepts of embodied intelligence, morphological computation, simplexity, and evolutionary and developmental approaches.
HUMASoft: Diseño y Control de Eslabones Blandos para Robots Humanoides
(CEA-IFAC, 2017-06-08) Monje Micharet, Concepción Alicia; Balaguer Bernaldo de Quirós, Carlos; Ministerio de Economía y Competitividad (España)
El principal objetivo del proyecto HUMASoft es el desarrollo de un nuevo tipo de eslabones para crear robots humanoides más blandos que reúnan las características de simplicidad, accesibilidad y seguridad. Estos eslabones blandos podrán emplearse indistintamente en varias extremidades de los robots humanoides, como brazos, cuello y espina dorsal, bajo los condicionantes de escalabilidad, controlabilidad de su rigidez e integrabilidad. Para conseguir este objetivo, el proyecto propone los siguientes sub-objetivos: 1) diseño y desarrollo de un prototipo de eslabón blando, con definición del material y del sistema de actuación. Como resultado se obtendrá el prototipo electromecánico con la premisa de fácil integración en la estructura rígida de un robot humanoide; 2) sistema de control reconfigurable embebido del eslabón blando, usando técnicas de control fracccionario y robusto. Como resultado se obtendrá un controlador fácilmente integrable en i_TEO; 3) sustitución (integración) de varios eslabones del robot humanoide de tamaño natural TEO por eslabones blandos escalados para brazos, cuello y espina dorsal. Como resultado se tendrá un nuevo humanoide blando; y 4) evaluación final del sistema, desarrollando nuevas métricas para el análisis del comportamiento del robot blando, sobre todo en la interacción humano-robot.
Design and performance validation of a cable-driven soft robotic neck
(Universidad de Valladolid, 2018-06-01) Nagua Cuenca, Luis Fernando; Monje Micharet, Concepción Alicia; Muñoz Yáñez-Barnuevo, Jorge; Balaguer Bernaldo de Quirós, Carlos; Ministerio de Economía y Competitividad (España)
The purpose of this paper is to design a soft robotic neck prototype with two Degrees of Freedom (DOF). It is mainly aimed to investigate, study and design a mechanism that allows to simulate the movements of a human neck, concretely the movements of flexion, extension and lateral bending. To archieve these movements, the design is made based on a cable-driven mechanism, validating the design of spring, through which it will be possible to obtain the sketch of the components that make up the soft neck and then its manufacture in a 3D printer. Another important aspect for the development of the project is the load weight that the soft neck can support, in order to size the motors that are needed for the operation of the parallel mechanism. In addition, the analysis of its mathematical model for the control system that will be implemented in future work is carried out.
Robust motion control of a soft robotic system using fractional order control
(Springer, 2017-07-27) Deutschmann, Bastian; Ott, Christian; Monje Micharet, Concepción Alicia; Balaguer Bernaldo de Quirós, Carlos
This work presents a novel control approach for a tendon-driven soft robotic system. The soft robotic system composed of a silicon continuum, tendons and antagonistic actuation yields a highly complex mechanical model. As the high complexity is not feasible here, a linear time invariant system is approximated instead for the controller design. A fractional order PDalfa controller is applied to meet performance and the high robustness requirements due to the neglected nonlinear dynamics. Simulation and experimental data confirm a superior performance of the FO controller while exhibiting a higher robustness to model mismatches and better disturbance rejection properties.
Evaluación multi-ZMP para tareas de transporte de objetos en robots humanoides
(Universidade Da Coruña, 2019-09-01) García Haro, Juan Miguel; Martínez de la Casa Díaz, Santiago; Hernández Vicen, Juan; Balaguer Bernaldo de Quirós, Carlos; Comunidad de Madrid; Ministerio de Economía y Competitividad (España)
Este artículo presenta un enfoque de control postural de un robot humanoide camarero. En investigaciones anteriores, se propusieron dos métodos para tratar esta complejidad. La primera fue una mejora para el control del equilibrio del cuerpo (locomoción) y la segunda fue un método para aplicar los conceptos clásicos del equilibrio del cuerpo para transportar objetos en una bandeja (manipulación). Este enfoque se basa en el concepto de un sistema de evaluación multi-ZMP para controlar la estabilidad del objeto y el cuerpo. Ambos métodos se desarrollaron de forma independiente, evitando perturbaciones entre ellos. La integración en una arquitectura de control postural de todo el cuerpo es un reto para el rendimiento de ambos méetodos, debido a la gran influencia entre ellos. En este artículo, se presentan ambos métodos para tratar la complejidad de la tarea humanoide. Es decir, cómo estos métodos se han integrado en un controlador postural humanoide y los resultados de la interacción.
A First Approach to a Proposal of a Soft Robotic Link Acting as a Neck
(Universidade da Coruña, 2018-09-05) Nagua Cuenca, Luis Fernando; Muñoz Yáñez-Barnuevo, Jorge; Monje Micharet, Concepción Alicia; Balaguer Bernaldo de Quirós, Carlos; Ministerio de Economía y Competitividad (España)
The purpose of this paper is to design a soft robotic neck prototype with two Degrees of Freedom (DOF) and propose a control system based on a fractional order PD controller (FPD). The neck will be able to perform movements of flexion, extension and lateral bending. To achieve these movements, the design is made based on a cable-driven mechanism, with components easy to manufacture in a 3D printer. Simulations are performed to validate the feasibility of the developed parallel robot prototype and the robustness of the proposed control scheme to mass changes at the tip.
Balance computation of objects transported on a tray by ahumanoid robot based on 3D dynamic slopes
(IEEE, 2019-01-01) García Haro, Juan Miguel; Martínez de la Casa Díaz, Santiago; Balaguer Bernaldo de Quirós, Carlos; Comunidad de Madrid; Ministerio de Economía y Competitividad (España)
Humanoid robots are designed to perform tasks in the same way than humans do. One of these tasks is to act as a waiter serving drinks, food, etc. Transporting all these items can be considered a manipulation task. In this application, the objects are transported over a tray, without grasping them. The consequence is that the objects are not firmly attached to the robot, which is the case in grasping. Then, the complexity of robotics grasping is avoided but stability issues arise. The problem of keeping balance of the object transported by a robot over a tray is discussed in this paper. The approach presented is based on the computation of the Zero Moment Point (ZMP) of the object, which is modelled as a three dimensional Linear Inverted Pendulum Model (3D-LIPM). The use of force-torque sensors located at the wrist enables ZMP computation, but the main problem to be solved is how the robot should react when the object losses balance. One strategy is to rotate the tray to counteract the rotation of the object. This rotation has to be proportional to the ZMP variation and the object's rotation angle. This issue is solved by applying the concept of three dimensional dynamic slopes. It helps to avoid kinematic problems and make balance computation independent from the angle of the tray.
A robust control method for the elbow of the humanoid robot TEO based on a fractional order controller
(IEEE, 2019-01-07) Muñoz Yáñez-Barnuevo, Jorge; Monje Micharet, Concepción Alicia; Martín Monar, Fernando; Balaguer Bernaldo de Quirós, Carlos; Ministerio de Economía y Competitividad (España)
This paper presents a novel method for the control of the elbow joint of the humanoid robot TEO, based on a fractional order PD controller. Due to the graphical nature of the proposed method, a few basic operations are enough to tune the controller, offering very competitive results compared to classic methods. The experiments show a robust performance of the system to mass changes at the tip of the humanoid right arm.
Diseño de un exoesqueleto por segmentos para evaluación y tratamiento del miembro superior
(2016-05) Alguacil, Isabel; Monge, E.; Copaci, Dorin Sabin; Blanco Rojas, María Dolores; Pérez de Heredia, Marta; Collado, S.
Se trata del primer exoesqueleto para codo actuado con fibras de SMA. La carencia de motores permite reducir drásticamente el peso y conseguir un funcionamiento silencioso, características que permiten aumentar la confortabilidad del sistema. Frente a otros exoesqueletos actualmente implementados en servicios especializados de rehabilitación, de elevado peso y no portátil, es factible el diseño de equipos de bajo coste guiados mediante sistemas informáticos.
Obtención del modelo dinámico simbólico de robots ramificados utilizando grupos de Lie y grafos
(Comite Español De Automatica (CEA-IFAC), 2016) Escalera Piña, Juan Antonio; Abu-Dakka, Fares Jawad Moh D; Alhama Blanco, Pablo José; Abderrahim Fichouche, Mohamed
En este articulo se presenta una formulación matricial simbólica para el modelado dinámico de robots ramificados, los cuales están compuestos por varias cadenas cinemáticas lineales abiertas. El método propuesto utiliza la mecánica geométrica basada en la teoría de Screws y grupos de Lie para derivar la ecuación de movimiento de Newton-Euler geométrica. La formulación es válida para cualquier robot formado por cuerpos rígidos acoplados mediante juntas de un grado de libertad (por tanto, rotacionales y/o prismáticas) sin formar cadenas cinemáticas cerradas. Bajo estas condiciones, estos robots pueden ser representados en forma única como un grafo de tipo árbol dirigido. Finalmente combinando la teoría de grafos con la mecánica geométrica se obtiene el modelo dinámico complete de robots ramificados. Ademas, la formulación propuesta presenta los parámetros intrínsecos del robot explícitamente en términos aislados. De este modo, la ecuación resultante se puede utilizar en algoritmos tales como identificación, simulación y control.
Exoesqueleto actuado por SMA para movilización de la muñeca
(2016-09-07) Copaci, Dorin Sabin; Blanco Rojas, María Dolores; López Guerra, Iago; Collado Vázquez, Susana; Pérez de Heredia, Marta
La mano juega un papel fundamental en la relación del ser humano con su entorno, siendo las tareas de manipulación de enorme importancia en la vida diaria. Pero la mano es una extremidad frágil, un gran número de enfermedades y trastornos musculo-esqueléticos y neurológicos derivan en la pérdida total o parcial de la movilidad de esta extremidad. De aquí la importancia de las terapias de rehabilitación dirigidas a recuperar y/o mantener su movilidad. En este artículo se presenta el diseño preliminar de un dispositivo robótico para la rehabilitación de la muñeca con 2 g.d.l., con actuadores basados en fibras de SMA (Shape Memory Alloy). Gracias a este sistema de actuación el exoesqueleto es extremadamente ligero, de actuación silenciosa y con una estructura muy simple. El número de actuadores y el diseño preliminar se han calculado a partir de la simulación del comportamiento biomecánico del cuerpo humano con un tipo específico de pacientes. Exoesqueleto actuado por SMA para movilización de la muñeca (PDF Download Available).

Browse

Recent Submissions