Pasar al contenido principal

SISTEMAS DE INTELIGENCIA ARTIFICIAL EXPLICABLES PARA LA SALUD Y EL BIENESTAR

PID2019-104829RA-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 de I+D+i (Generación de Conocimiento y Retos Investigación)
Año convocatoria 2019
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario UNIVERSIDAD DE LAS ISLAS BALEARES

Resultados relacionados

Resultados totales (Incluyendo duplicados): 4
Encontrada(s) 1 página(s)

Repositori Institucional UIB
Publicación Artículo científico (JournalArticle)

Sobre el reconocimiento de emociones y la precisión de los clasificadores

  • Roig-Maimó, Maria Francesca
  • Mascaró Oliver, Miquel
  • Amengual Alcover, Esperança
  • Mas-Sansó, Ramon
[spa] El reconocimiento de expresiones faciales es un tópico muy popular desde el surgimiento de la Inteligencia Artificial. Desde el punto de vista de la Interacción Persona-Ordenador también existe un gran interés, ya que las emociones transmitidas a través de las expresiones aportan una gran cantidad de información a la hora de tratar la comunicación no verbal. Hoy en día, las redes neuronales son uno de los sistemas de aprendizaje computacional más utilizados para reconocer y analizar emociones y, en general, los principales esfuerzos se dirigen a entrenar modelos que alcancen el máximo rendimiento posible en términos de precisión. Pero, en realidad, los humanos no son perfectos en el reconocimiento de emociones. En este trabajo, planteamos la cuestión de si el objetivo a perseguir al entrenar tales modelos debe basarse únicamente en alcanzar altos valores de precisión de reconocimiento o si también debemos centrarnos en tratar de emular el comportamiento humano. Intentamos llegar a una respuesta a través de comparar los resultados de clasificación de emociones en dos experimentos: uno con participantes humanos y otro con una red neuronal convolucional., [eng] The recognition of facial expressions is a very popular topic since the emergence of Artificial Intelligence. From the point of view of Human-Computer Interaction, there is also great interest since the detected facial expression provides a large amount of information to recognize its associated emotion. Nowadays, neural networks are one of the most widely used computational learning systems for recognizing and analyzing emotions from facial expressions. Usually, the main efforts are directed towards obtaining the maximum performance of the system in terms of high accuracy. But in reality, humans are not that good at distinguishing between emotions from static facial expressions. In this paper, we raise the question of whether the goal to pursue when training models should be based solely on achieving high values of accuracy or whether we should also focus on trying to emulate human behavior. We try to provide an answer through two experiments: one with human participants and the other one with a convolutional neural network.
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Cognitive interaction analysis in human–robot collaboration using an assembly task

  • Chacón, Alejandro
  • Ponsa Asensio, Pere
  • Angulo, Cecilio
In human–robot collaborative assembly tasks, it is necessary to properly balance skills to maximize productivity. Human operators can contribute with their abilities in dexterous manipu-lation, reasoning and problem solving, but a bounded workload (cognitive, physical, and timing) should be assigned for the task. Collaborative robots can provide accurate, quick and precise physical work skills, but they have constrained cognitive interaction capacity and low dexterous ability. In this work, an experimental setup is introduced in the form of a laboratory case study in which the task performance of the human–robot team and the mental workload of the humans are analyzed for an assembly task. We demonstrate that an operator working on a main high-demanding cognitive task can also comply with a secondary task (assembly) mainly developed for a robot asking for some cognitive and dexterous human capacities producing a very low impact on the primary task. In this form, skills are well balanced, and the operator is satisfied with the working conditions., This work has been co-financed by the European Regional Development Fund of the European Union in the framework of the ERDF Operational Program of Catalonia 2014–2020, grant number 001-P-001643. Professor Cecilio Angulo has been partly supported by the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 825619 (AI4EU). Professor Pere Ponsa has been partly supported by the Project EXPLainable Artificial INtelligence systems for health and well-beING (EXPLAINING) (PID2019-104829RA-I00/AEI/10.13039/501100011033).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Usability study through a human-robot collaborative workspace experience

  • Chacón, Alejandro
  • Ponsa Asensio, Pere
  • Angulo, Cecilio
The use of collaborative robots (cobots) in industrial and academic settings facilitates physical and cognitive interaction with operators. This framework is a challenge to determine how measures on concepts, such as usability, can be adapted to these new environments. Usability is a quality attribute prevalent in the field of human-computer interaction concerning the context of use and the measure of effectiveness, efficiency, and satisfaction of products and systems. In this work, the importance of the role of benchmarking usability with collaborative robots is discussed. The introduced approach is part of a general methodology for studying people and robots’ performance in collaboration. It is being designed and developed on a concrete experience into a human-robot collaborative workspace. Outcomes from the study include a list of steps, resources, recommen-dations, and some customized questionnaires to obtain cobot-oriented usability analysis and case study results., This work has been co-financed by the European Regional Development Fund of the European Union in the framework of the ERDF Operational Program of Catalonia 2014–2020, grant number 001-P-001643. Prof. Cecilio Angulo has been partly supported by the European Union’s Horizon 2020 research and innovation programme under grant agreement No. 825619 (AI4EU). Prof. Pere Ponsa has been partly supported by the Project EXPLainable Artificial INtelligence systems for health and well-beING (EXPLAINING) (PID2019-104829RA-I00/AEI/10.13039/501100011033).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Towards a cognitive assistant supporting human operators in the Artificial Intelligence of Things

  • Angulo, Cecilio
  • Chacón, Alejandro
  • Ponsa Asensio, Pere
Internet of Things (IoT) systems are becoming increasingly complex due to heterogeneity of devices and requirements for real-time processing and decision making. In this context, Artificial Intelligence (AI) technologies provide powerful capabilities for endowing IoT devices with intelligent services, leading to the so-called Artificial Intelligence of Things (AIoT). The operator is in the middle of this complexity, trying to understand the situation and make effective real-time decisions. Hence, human factors, especially cognitive ones, are a major issue to be addressed. The human cognitive part must be framed together with intelligent artefacts, requiring a systematic approach in the domain of joint cognitive systems. New software development methods in the form of assistants and wizards are necessary to help operators to be context-aware and reduce their technical workload regarding coding or computer-oriented skills, focusing on the task or service at hand. Building on previous research on the role of the human worker in an AIoT environment, this article analyses the described situation in terms of human cyber–physical systems, with the aim of proposing a conceptual framework for these assistance systems at the cognitive level. Two illustrative examples are described to validate the effectiveness of the proposed framework in collaborative tasks., This work has been partly funded by the European Regional Development Fund of the European Union within the framework of the ERDF Operational Program of Catalonia 2014–2020, grant number 001-P-001643. Prof. Pere Ponsa has been partly supported by the Project EXPLainable Artificial INtelligence systems for health and well-beING (EXPLAINING), Spain (PID2019-104829RA-I00/AEI/10.13039/501100011033)., Peer reviewed