Pasar al contenido principal

ROB-IN: Robots para la asistencia continua y personalizada capaces de explicar-se a si mismos

PLEC2021-007859

•
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 en Líneas Estratégicas
Año convocatoria 2021
Plan Estatal 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Centro beneficiario AGENCIA ESTATAL CONSEJO SUPERIOR DE INVESTIGACIONES CIENTIFICAS (CSIC)

Resultados relacionados

Resultados totales (Incluyendo duplicados): 35
Encontrada(s) 1 página(s)

DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Implications of robot backchannelling in cognitive therapy

  • Andriella, Antonio
  • Torras, Carme
  • Alenyà, Guillem
Trabajo presentado en la 14th International Conference on Social Robotics , celebrada en Florencia (Italia), del 13 al 16 de diciembre de 2022, The social ability of humans to provide active feedback during conversations is known as backchannelling. Recent work has recognised the importance of endowing robots with such social behaviour to make interactions more natural. Nonetheless, very little is known about how backchannelling should be designed in order to be detected and whether it can have an impact on users' behaviour and performance in cooperative tasks. In this article, we aim at evaluating the legibility of robot's backchannelling behaviour on Persons with Dementia (PwDs) and its effect on their performance when playing cognitive training exercises. Aiming to do so, a TIAGo robot was endowed with backchannelling behaviour generated by combining verbal and non-verbal cues. To evaluate our system, two user studies were carried out, in which the social signal was provided first by a human therapist and later on by a robot. Results indicate that patients were capable of identifying such kind of feedback. Nonetheless, our findings pointed out a significant difference in terms of performance between the two studies. They reveal how patients in the study with the robot overused the feedback to obtain the correct answer, putting in place a cheating mechanism that has led them to significantly worsen their performance. We conclude our work by discussing the implications of our findings when deploying robots in sensitive roles and possible solutions to address such unexpected behaviours., This project has received funding from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 801342 (Tecniospring INDUSTRY). This project has also been supported by the European Union’s Horizon 2020 under ERC Advanced Grant CLOTHILDE (no. 741930); by MCIN/AEI/10.13039/501100011033 by the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859); by the Research Council of Norway under the project SECUROPS (INT-NO/0875); and by the “European Union NextGenerationEU/PRTR” through CSIC’s Thematic Platforms (PTI+ Neuro-Aging).
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Generating predicate suggestions based on the space of plans: an example of planning with preferences

  • Canal, Gerard
  • Torras, Carme
  • Alenyà, Guillem
Task planning in human–robot environments tends to be particularly complex as it involves additional uncertainty introduced by the human user. Several plans, entailing few or various differences, can be obtained to solve the same given task. To choose among them, the usual least-cost plan criteria is not necessarily the best option, because here, human constraints and preferences come into play. Knowing these user preferences is very valuable to select an appropriate plan, but the preference values are usually hard to obtain. In this context, we propose the Space-of-Plans-based Suggestions (SoPS) algorithms that can provide suggestions for some planning predicates, which are used to define the state of the environment in a task planning problem where actions modify the predicates. We denote these predicates as suggestible predicates, of which user preferences are a particular case. The first algorithm is able to analyze the potential effect of the unknown predicates and provide suggestions to values for these unknown predicates that may produce better plans. The second algorithm is able to suggest changes to already known values that potentially improve the obtained reward. The proposed approach utilizes a Space of Plans Tree structure to represent a subset of the space of plans. The tree is traversed to find the predicates and the values that would most increase the reward, and output them as a suggestion to the user. Our evaluation in three preference-based assistive robotics domains shows how the proposed algorithms can improve task performance by suggesting the most effective predicate values first., This work has been partially supported by the ERC project Clothilde (ERC-2016-ADG-741930); by MCIN/AEI/10.13039/501100011033 under the project CHLOE-GRAPH (PID2020-118649RB-l00); by the European Union NextGenerationEU/PRTR under the project ROB-IN (PLEC2021-007859); and by the CHIST-ERA project COHERENT (EPSRC EP/V062506/1). Gerard Canal has also been supported by the Spanish Ministry of Education, Culture and Sport by the FPU15/00504 doctoral grant and by the Royal Academy of Engineering and the Office of the Chief Science Adviser for National Security under the UK Intelligence Community Postdoctoral Research Fellowship programme., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Social robot-delivered customer-facing services: an assessment of the experience, 由社交机器人提供的面向客户的服务:对体验的分析

  • Forgas Coll, Santiago
  • Huertas García, Rubén
  • Andriella, Antonio
  • Alenyà, Guillem
The ability to install social intelligence protocols in robots in order for them to exhibit conversational skills has made them ideal tools for delivering services with a high cognitive and low emotional load. Little is known about how this capability influences the customer experience and the intention to continue receiving these services. Experiences were assessed in a study simulating customer-facing service delivery, and the constructs of the technology readiness index and stated gender were analysed as possible moderators in a quasi-experiment. Hedonic quality was the most relevant factor explaining attitude, and attitude explained intention to use as well as social influence. As for the constructs of technological readiness and gender, optimism and innovativeness seem to be the most likely candidates for moderating the other variables. The most optimistic and the most innovative route would be for the main actors to continue adapting to social robot technology in the future., This work was partially funded by the European Union’s Horizon 2020 under ERC Advanced Grant CLOTHILDE (no. 741930); by MCIN/AEI/10.13039/501100011033 and by the ‘European Union NextGenerationEU/PRTR’ under the project ROB-IN (PLEC2021-007859); by the Research Council of Norway under the project SECUROPS (INT-NO/0875); and by the ‘European Union NextGenerationEU/PRTR’ through CSIC’s Thematic Platforms (PTI + Neuro-Aging); A. Andriella was supported by the European Union’s Horizon 2020 research and innovation programme PRO-CARED (no.801342) under Marie Skłodowska-Curie grant agreement (Tecniospring INDUSTRY) and the Government of Catalonia’s Agency for Business Competitiveness (ACCIÓ)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Condition-based Design of Variable Impedance Controllers from User Demonstrations

  • San Miguel, Alberto
  • Puig, Vicenç
  • Alenyà, Guillem
This paper presents an approach to ensure conditions on Variable Impedance Controllers through the off-line tuning of the parameters involved in its description. In particular, we prove its application to term modulations defined by a Learning from Demonstration technique. This is performed through the assessment of conditions regarding safety and performance, which encompass heuristics and constraints in the form of Linear Matrix Inequalities. Latter ones allow to define a convex optimisation problem to analyse their fulfilment, and require a polytopic description of the VIC, in this case, obtained from its formulation as a discrete-time Linear Parameter Varying system. With respect to the current state-of-art, this approach only limits the term definition obtained by the Learning from Demonstration technique to be continuous and function of exogenous signals, i.e. external variables to the robot. Therefore, using a solution-search method, the most suitable set of parameters according to assessment criteria can be obtained. Using a 7-DoF KINOVAGEN3 manipulator, validation and comparison against solutions with relaxed conditions are performed. The method is applied to generate Variable Impedance Controllers for a pulley belt looping task, inspired by the Assembly Challenge for Industrial Robotics in World Robot Summit 2018, to reduce the exerted force with respect to a standard (constant) Impedance Controller. These controllers fulfil a set of safety constraints, namely stability, bounds on task variables and maximum response overshooting; and their performance is determined by the User Preference heuristic, which allows to intuitively define the desired compliant behaviour along the task. In the context of the task, this is used to generate new controllers for one-off modifications of the nominal belt looping task setup without new demonstrations., This work is partially supported by MCIN/ AEI /10.13039/501100011033 and by the ”European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859); and by MCIN/ AEI /10.13039/ 501100011033, Spain, under the project CHLOE-GRAPH (PID2020- 119244GB-I00). Authors also want to thank Adriá Colomé and Edoardo Caldarelli for their comments and help throughout this work., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Gendered Human–Robot Interactions in Services

  • Forgas Coll, Santiago
  • Huertas García, Rubén
  • Andriella, Antonio
  • Alenyà, Guillem
The outbreak of Covid-19 precipitated the use of service robots in customer-facing services as a replacement for employees to avoid human-to-human contact. However, this development has not resolved the debate as to whether robots should be characterized with gender attributes or simply be genderless. This study explores whether endowing a robot with gender attributes makes it more acceptable as a service provider among stated men and women. To this end, an experiment was conducted at a public fair in which a gendered robot simulated the provision of a service to customers, which consisted of offering them advice, hints, and messages of encouragement to help complete a eudaemonic puzzle. A parsimonious version of the Almere model was used to estimate acceptance of the technology. The findings reveal that for both stated men and women, the main drivers for accepting the female-coded robot are perceived usefulness and social influence, although women attach greater importance to social influence. For the male-coded robot, perceived usefulness and social influence are the main arguments for women, while for men they are enjoyment, perceived usefulness and, negatively, ease of use. In addition, different indirect effects between stated sexes are also identified. In summary, men and women consider different factors when accepting robots of each gender., This work was partially funded by the European Union’s Horizon 2020 programme under ERC Advanced Grant CLOTHILDE (no. 741930); and by the European Union’s Horizon 2020 research and innovation programme under the Marie Sk?odowska-Curie grant agreement No 801342 (Tecniospring INDUSTRY); by MCIN/ AEI /https://doi.org/10.13039/501100011033 and by the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859); by the Research Council of Norway under the project SECUROPS (INT-NO/0875); and by the “European Union NextGenerationEU/PRTR” through CSIC’s Thematic Platforms (PTI + Neuro-Aging). Open Access funding provided thanks to the CRUE-CSIC agreement with Springer Nature., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Correction to: Gendered Human–Robot Interactions in Services (International Journal of Social Robotics, (2023), 15, 11, (1791-1807), 10.1007/s12369-023-01035-8)

  • Forgas Coll, Santiago
  • Huertas García, Rubén
  • Andriella, Antonio
  • Alenyà, Guillem
This paper was inadvertently published with incorrect funding information. Please find the incorrect and correct versions below. Springer wishes to apologize for any inconvenience caused. Incorrect funding information: This work was partially funded by the European Union’s Horizon 2020 programme under ERC Advanced Grant CLOTHILDE (no. 741,930) and under the Marie Skłodowska-Curie grant agreement (no. 712,949) (TECNIOspring PLUS); by MCIN/ AEI /https://doi.org/10.13039/501100011033 and by the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859); by the Research Council of Norway under the project SECUROPS (INT-NO/0875); and by the “European Union NextGenerationEU/PRTR” through CSIC’s Thematic Platforms (PTI + Neuro-Aging). Correct funding information: This work was partially funded by the European Union’s Horizon 2020 programme under ERC Advanced Grant CLOTHILDE (no. 741,930); and by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 801,342 (Tecniospring INDUSTRY); by MCIN/ AEI /https://doi.org/10.13039/501100011033 and by the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859); by the Research Council of Norway under the project SECUROPS (INT-NO/0875); and by the “European Union NextGenerationEU/PRTR” through CSIC’s Thematic Platforms (PTI + Neuro-Aging)., This work was partially funded by the European Union’s Horizon 2020 programme under ERC Advanced Grant CLOTHILDE (no. 741,930); and by the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement No 801,342 (Tecniospring INDUSTRY); by MCIN/ AEI /https://doi.org/10.13039/501100011033 and by the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859); by the Research Council of Norway under the project SECUROPS (INT-NO/0875); and by the “European Union NextGenerationEU/PRTR” through CSIC’s Thematic Platforms (PTI + Neuro-Aging)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Adaptive Human-Robot Collaboration: Evolutionary Learning of Action Costs Using an Action Outcome Simulator

  • Izquierdo-Badiola, Silvia
  • Alenyà, Guillem
  • Rizzo, Carlos
Trabajo presentado en el 32nd IEEE International Conference on Robot and Human Interactive Communication (RO-MAN), celebrado en Busan (República de Corea), del 28 al 31 de agosto de 2023, One of the main challenges for successful human-robot collaborative applications lies in adapting the plan to the human agent's changing state and preferences. A promising solution is to bridge the gap between agent modelling and AI task planning, which can be done by integrating the agent state as action costs in the task planning domain. This allows for the plan to be adapted to different partners, by influencing the action allocation. The difficulty then lies in setting appropriate action costs. This paper presents a novel framework to learn a set of planning action costs considering the preferred actions for an agent based on their state. An evolutionary optimisation algorithm is used for this purpose, and an action outcome simulator is developed to act as the black-box function, based on both an agent model and an action type model. This addresses the challenge of collecting data in HRC real-world scenarios, accelerating the learning for posterior fine-tuning in real applications. The coherence of the models and the simulator is proven through a conducted survey, and the learning algorithm is shown to learn appropriate action costs, producing plans that satisfy both the agents' preferences and the prioritised plan requisites. The resulting system is a generic learning framework integrating components that can be easily extended to a wide range of applications, models and planning formalisms., This work has been partially supported by MCIN/ AEI /10.13039/501100011033 under the project CHLOE-GRAPH (PID2020- 119244GB-I00); by MCIN/ AEI /10.13039/501100011033 and by the ”European Union (EU) NextGenerationEU/PRTR under the project ROB-IN (PLEC2021-007859), Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Exploring the Potential of a Robot-Assisted Frailty Assessment System for Elderly Care

  • Civit, Aniol
  • Andriella, Antonio
  • Antonio, Maite
  • Javierre, Casimiro
  • Boqué, Concepción
  • Alenyà, Guillem
Frailty assessment plays a pivotal role in providing older adults care. However, the current process is time-consuming and only measures patients' completion time for each test. This paper introduces a set of algorithms to be used in robots to autonomously perform frailty assessments. In doing so we aim at reducing therapists' burden and provide additional frailty-related metrics that can enhance the effectiveness of diagnosis. We conducted a pilot study with 22 elderly participants and compared our system's performance with that of medical professionals to assess its precision. The results demonstrate that our approach achieved performances close to that of its human counterpart. This research represents an important step forward in the integration of social robotics in healthcare, offering potential benefits for patient care and clinical decision-making., This work was supported by the project ROBIN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the ”European Union NextGenerationEU/PRTR”; by the EU-founded project grant agreement No 101070930 (VALAWAI); and the 23S06141-001 FRAILWATCH project funded by Barcelona Ciencia 2020-2023 Plan. A. Civit has been supported by AGAUR-FI ajuts (2023 FI-3 00065) Joan Or´o of the Secretariat of Universities and Research of the Department of Research and Universities of the Generalitat of Catalonia and the European Social Plus Fund., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

How do people intend to disclose personal information to a social robot in public spaces?

  • Aryania, Azra
  • Huertas García, Rubén
  • Forgas-Coll, Santiago
  • Angulo, Cecilio
  • Alenyà, Guillem
Social robots interacting with people in public spaces may access and collect their personal information, which raises privacy concerns regarding the disclosure of personal information. This paper aims to investigate factors impacting individuals' intention to disclose personal information to a social robot in public spaces and evaluate the actual disclosure during the interaction with the robot. For this purpose, a model is proposed to predict people's intentions to disclose information to a social robot. We conducted our experiment at a public festival with more than 100 participants using the social robot ARI. The findings reveal the substantial impact of factors including risk beliefs, trusting beliefs, perceived enjoyment, and social influence on the intention to disclose personal information. Moreover, they reveal that although only a small percentage (6.20%) of people had the intention to disclose information to the social robot, most participants (98.00%) finally disclosed their personal information., This work was partially funded by the Research Council of Norway under the project SECUROPS (INT-NO/0875), by MCIN/ AEI /10.13039/501100011033 and by the European Union NextGenerationEU/PRTR under the project ROB-IN (PLEC2021-007859), and by Generalitat de Catalunya 2021 SGR 00514., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Introducing Social Robots to Assess Frailty in Older Adults

  • Civit, Aniol
  • Andriella, Antonio
  • Barrue, Cristian
  • Antonio, Maite
  • Boqué, Concepción
  • Alenyà, Guillem
Frailty is a crucial indicator in determining the well-being of older adults in terms of their health. With the growing number of elderly people, the demand for geriatricians is increasing, which means that they have less time to spend with each patient. The current methods for frailty assessment use simple tests that are time-consuming and do not require specific medical expertise. To address this issue, this paper proposes the use of social robots to assess frailty autonomously. It presents a practical proposal that defines the robot's behavior and explains the design and implementation concepts. Finally, it discusses some of the challenges that may arise from introducing social robots as frailty evaluators., This work was supported by the project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR"; project CHLOE-GRAPH PID2020-118649RB-I00 funded by MCIN/AEI /10.13039/501100011033; and the 23S06141-001 FRAILWATCH project funded by Barcelona Ciencia 2020-2023 Plan., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

PlanCollabNL: Leveraging Large Language Models for Adaptive Plan Generation in Human-Robot Collaboration

  • Izquierdo-Badiola, Silvia
  • Canal, Gerard
  • Rizzo, Carlos
  • Alenyà, Guillem
"Hey, robot. Let's tidy up the kitchen. By the way, I have back pain today". How can a robotic system devise a shared plan with an appropriate task allocation from this abstract goal and agent condition? Classical AI task planning has been explored for this purpose, but it involves a tedious definition of an inflexible planning problem. Large Language Models (LLMs) have shown promising generalisation capabilities in robotics decision-making through knowledge extraction from Natural Language (NL). However, the translation of NL information into constrained robotics domains remains a challenge. In this paper, we use LLMs as translators between NL information and a structured AI task planning problem, targeting human-robot collaborative plans. The LLM generates information that is encoded in the planning problem, including specific subgoals derived from an NL abstract goal, as well as recommendations for subgoal allocation based on NL agent conditions. The framework, PlanCollabNL, is evaluated for a number of goals and agent conditions, and the results show that correct and executable plans are found in most cases. With this framework, we intend to add flexibility and generalisation to HRC plan generation, eliminating the need for a manual and laborious definition of restricted planning problems and agent models., This work was partially financed by MCIN/ AEI /10.13039/501100011033 and by the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859); and by the project COHERENT funded by MCIN/ AEI /10.13039/501100011033, Spain, the “European Union NextGenerationEU/ PRTR”, Spain, (PCI2020-120718-2), and by UKRI (EP/V062506/1)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Navigating autonomy, privacy, and ageism in robot home care with aged users: A preliminary analysis of ROB-IN

  • Liedo Fernández, Belén
In this article, I propose an ethical analysis of assistive domestic robots for older users. In doing so, I illustrate my inquiry with the example of ROB‐IN assistive robot. ROB‐IN is a Spanish project which is devoted to developing a robot that will perform in the private home of nondependent, aged users. It is aimed to help people in their daily activities and contribute to appropriate health monitoring. One of their potentially most useful features is related to data gathering and sharing. For the inquiry on the ethical underpinnings of this case, I develop a framework for domestic assistive robots for competent older adults drawn on the ethics of care. I assess that this type of robots could be ethically appraised attending to their impact on the well‐being and autonomy of users. I approach autonomy from a relational perspective, and I delve into the relationship between autonomy and well‐being through the concept of paternalism. I argue that this type of assistive robots should never act paternalistically. Given ROB‐IN great implications regarding privacy, I subsequently explore the ways in which the privacy of users should be respected in their interaction with assistive robots, focusing on the relation with autonomy and well‐being. Lastly, I highlight the need for avoiding ageism. This investigation focuses on aged users, but it is suggested that the situation of caregivers should be also the object of further investigations., This investigation has been developed under Grant FPU19/06027 funded by the Spanish Ministry of Universities. It has been partially carried out during a research stay at the Institute of Industrial Robotics and Informatics of the Spanish National Research Council (Barcelona, Spain), funded by the AI-HUB research network, and under the project ROB-IN (PLEC2021-007859) funded by MCIN/AEI/10.13039/501100011033 and by the European Union (EU) NextGenerationEU/PRTR. The publication fee has been supported by the CSIC, Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Unfolding the Literature: A Review of Robotic Cloth Manipulation

  • Longhini, Alberta
  • Wang, Yufei
  • Garcia-Camacho, Irene
  • Blanco-Mulero, David
  • Moletta, Marco
  • Welle, Michael
  • Alenyà, Guillem
  • Yin, Hang
  • Erickson, Zackory
  • Held, David
  • Borràs, Julia
  • Kragic, Danica
The realm of textiles spans clothing, households, healthcare, sports, and industrial applications. The deformable nature of these objects poses unique challenges that prior work on rigid objects cannot fully address. The increasing interest within the community in textile perception and manipulation has led to new methods that aim to address challenges in modeling, perception, and control, resulting in significant progress. However, this progress is often tailored to one specific textile or a subcategory of these textiles. To understand what restricts these methods and hinders current approaches from generalizing to a broader range of real-world textiles, this review provides an overview of the field, focusing specifically on how and to what extent textile variations are addressed in modeling, perception, benchmarking, and manipulation of textiles. We conclude by identifying key open problems and outlining grand challenges that will drive future advancements in the field., This work was supported by the European Research Council (ERC-BIRD); the European Union’s Horizon Europe program through projects SoftEnable (HORIZON-CL4-2021-DIGITALEMERGING-01-101070600) and IRE (HORIZON-CL4-2023-DIGITAL-EMERGING-01- 101135082); the Swedish Research Council, the Knut and Alice Wallenberg Foundation, and the National Science Foundation under NSF CAREER grant IIS-2046491; and the project ROBIN PLEC2021-007859 funded by MCIN/AEI/10.13039/501100011033 and by European Union NextGenerationEU/PRTR., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Artículo científico (JournalArticle)

Effects of Privacy Warning on the Intention to Disclose Personal Information During Interaction with a Robot in Public Spaces

  • Aryania, Azra
  • Huertas García, Rubén
  • Forgas-Coll, Santiago
  • Angulo, Cecilio
  • Alenyà, Guillem
Social robots in public spaces can potentially gather personal data. This leads to privacy concerns and prompts users to consider whether to disclose their information to the robot during interaction with the robot. This paper aims to explore the effects of privacy warnings on people’s intentions to disclose their personal information and their actual disclosure when interacting with a robot in public spaces. We propose a model to estimate individuals’ disclosure intentions toward a robot and evaluate the impact of privacy warnings during interaction. We conducted an experiment with more than 100 participants interacting with a robot to assess the proposed model. Our results indicate that factors such as risk beliefs, trusting beliefs, perceived enjoyment, and social influence significantly influence individuals’ intentions to disclose information. In general, a robot equipped with privacy warnings receives greater acceptance than a robot without privacy warnings, and participants who receive warnings disclose more low-privacy information. However, there is no significant difference in disclosure between participants who interacted with a robot with privacy warnings and those who interacted with a robot without privacy warnings. Overall, disclosure behavior does not significantly differ between the two groups, suggesting that privacy warnings do not effectively reduce disclosure information., This work was partially funded by the Research Council of Norway under the project SECUROPS (INT-NO/0875); by MCIN/ AEI /10.13039/501100011033, by the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859) and by the European Union under the project ARISE (HORIZON-CL4- 2023-DIGITAL-EMERGING-01101135959)., Peer reviewed
DIGITAL.CSIC. Repositorio Institucional del CSIC
Publicación Otros (Other)

Inside-out states of garments

  • Jiménez Schlegl, Pablo
IRI-TR-24-01, Cloth items such as garments can be potentially in in nite di erent con gurations, due to their deformability. For the practical purposes of manipulation, however, it is possible to discretize the space of deformations into a set of equivalent states. This is pursued in the present work, where the complex problem of recovering the canonical con guration of partially (or completely) reversed garments is addressed. Without claiming to solve the hard perceptual and manipulative challenges of this type of task {which are also thoroughly described{, this work should rather be viewed as a pioneering e ort to formalize the high-level strategies that aim at solving this canonical con guration-recovering task., This work has been partially funded by the European Union Horizon 2020 Programme under grant agreement no. 741930 (CLOTHILDE), Project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/ PRTR" and Project CHLOE-Graph PID2020-118649RB-I00 funded by MCIN/ AEI /10.13039/501100011033, Peer reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Unfolding the literature: a review of robotic cloth manipulation

  • Longhini, Alberta
  • Wang, Yufei
  • Garcia Camacho, Irene|||0000-0002-1210-7345
  • Blanco Mulero, David|||0000-0003-1866-2144
  • Moletta, Marco
  • Welle, Michael C.
  • Alenyà Ribas, Guillem|||0000-0002-6018-154X
  • Yin, Hang
  • Erickson, Zackory
  • Held, David
  • Borràs Sol, Júlia|||0000-0002-1662-2037
  • Kragic, Danica
The realm of textiles spans clothing, households, healthcare, sports, and industrial applications. The deformable nature of these objects poses unique challenges that prior work on rigid objects cannot fully address. The increasing interest within the community in textile perception and manipulation has led to new methods that aim to address challenges in modeling, perception, and control, resulting in significant progress. However, this progress is often tailored to one specific textile or a subcategory of these textiles. To understand what restricts these methods and hinders current approaches from generalizing to a broader range of real-world textiles, this review provides an overview of the field, focusing specifically on how and to what extent textile variations are addressed in modeling, perception, benchmarking, and manipulation of textiles. We conclude by identifying key open problems and outlining grand challenges that will drive future advancements in the field., This work was supported by the European Research Council (ERC-BIRD); the European Union’s Horizon Europe program through projects SoftEnable (HORIZON-CL4-2021-DIGITALEMERGING-01-101070600) and IRE (HORIZON-CL4-2023-DIGITAL-EMERGING-01- 101135082); the Swedish Research Council, the Knut and Alice Wallenberg Foundation, and the National Science Foundation under NSF CAREER grant IIS-2046491; and the project ROBIN PLEC2021-007859 funded by MCIN/AEI/10.13039/501100011033 and by European Union NextGenerationEU/PRTR., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Introducing social robots to assess frailty in older adults

  • Civit Bertran, Aniol|||0009-0000-6255-3362
  • Andriella, Antonio|||0000-0002-6641-6450
  • Barrué Subirana, Cristian|||0000-0002-3342-1085
  • Antonio, Maite
  • Boqué, Concepción
  • Alenyà Ribas, Guillem|||0000-0002-6018-154X
Frailty is a crucial indicator in determining the well-being of older adults in terms of their health. With the growing number of elderly people, the demand for geriatricians is increasing, which means that they have less time to spend with each patient. The current methods for frailty assessment use simple tests that are time-consuming and do not require specific medical expertise. To address this issue, this paper proposes the use of social robots to assess frailty autonomously. It presents a practical proposal that defines the robot’s behavior and explains the design and implementation concepts. Finally, it discusses some of the challenges that may arise from introducing social robots as frailty evaluators., This work was supported by the project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR"; project CHLOE-GRAPH PID2020-118649RB-I00 funded by MCIN/AEI /10.13039/501100011033; and the 23S06141-001 FRAILWATCH project funded by Barcelona Ciencia 2020-2023 Plan., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Estudio, informe, memoria (Report)

Inside-out states of garments, Technical report IRI-TR-24-01

  • Jiménez Schlegl, Pablo|||0000-0003-3627-4938
Cloth items such as garments can be potentially in infinite different configurations, due to their deformability. For the practical purposes of manipulation, however, it is possible to discretize the space of deformations into a set of equivalent states. This is pursued in the present work, where the complex problem of recovering the canonical configuration of partially (or completely) reversed garments is addressed. Without claiming to solve the hard perceptual and manipulative challenges of this type of task –which are also thoroughly described–, this work should rather be viewed as a pioneering effort to formalize the high-level strategies that aim at solving this canonical configuration-recovering task., Cloth items such as garments can be potentially in infinite different configurations, due to their deformability. For the practical purposes of manipulation, however, it is possible to discretize the space of deformations into a set of equivalent states. This is pursued in the present work, where the complex problem of recovering the canonical configuration of partially (or completely) reversed garments is addressed. Without claiming to solve the hard perceptual and manipulative challenges of this type of task –which are also thoroughly described–, this work should rather be viewed as a pioneering effort to formalize the high-level strategies that aim at solving this canonical configuration-recovering task., This work has been partially funded by the European Union Horizon 2020 Programme under grant agreement no. 741930 (CLOTHILDE), Project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the ”European Union NextGenerationEU/PRTR” and Project CHLOE-Graph PID2020-118649RB-I00 funded by MCIN/AEI /10.13039/501100011033
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

ReSeTOX: Re-learning attention weights for toxicity mitigation in machine translation

  • García Gilabert, Javier
  • Escolano Peinado, Carlos|||0000-0001-6657-673X
  • Ruiz Costa-jussà, Marta
Our proposed method, RESETOX (REdo SEarch if TOXic), addresses the issue of Neural Machine Translation (NMT) generating translation outputs that contain toxic words not present in the input. The objective is to mitigate the introduction of toxic language without the need for retraining. In the case of identified added toxicity during the inference process, RESETOX dynamically adjusts the key-value self-attention weights and re-evaluates the beam search hypotheses. Experimental results demonstrate that RESETOX achieves a remarkable 57% reduction in added toxicity while maintaining an average translation quality of 99.5% across 164 languages. Our code is available at: https: //github.com/mt-upc/ReSeTOX, The work by Javier García Gilabert and Carlos Escolano has been funded by the Spanish Ministerio de Ciencia e Innovación and the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Exploring the potential of a robot-assisted frailty assessment system for elderly care

  • Civit Bertran, Aniol|||0009-0000-6255-3362
  • Andriella, Antonio|||0000-0002-6641-6450
  • Antonio, Maite
  • Javierre Garcés, Casimiro
  • Boqué, Concepción
  • Alenyà Ribas, Guillem|||0000-0002-6018-154X
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Frailty assessment plays a pivotal role in providing older adults care. However, the current process is time- consuming and only measures patients’ completion time for each test. This paper introduces a set of algorithms to be used in robots to autonomously perform frailty assessments. In doing so we aim at reducing therapists’ burden and provide additional frailty-related metrics that can enhance the effectiveness of diagnosis. We conducted a pilot study with 22 elderly participants and compared our system’s performance with that of medical professionals to assess its precision. The results demonstrate that our approach achieved performances close to that of its human counterpart. This research represents an important step forward in the integration of social robotics in healthcare, offering potential benefits for patient care and clinical decision-making., This work was supported by the project ROBIN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR"; by the EU-founded project grant agreement No 101070930 (VALAWAI); and the 23S06141-001 FRAILWATCH project funded by Barcelona Ciencia 2020-2023 Plan., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Artículo científico (JournalArticle)

Impact of design transparency on trust and data sharing during human-robot interactions in public places

  • Aryania, Azra
  • Alenyà Ribas, Guillem|||0000-0002-6018-154X
  • Chockalingam, Sabarathinam
  • Rødsethol, Hanne Kristine
The prevalence of social robots is increasing, with examples such as customer service robots in malls and airports. This trend highlights the importance of transparency, particularly in data-sharing interactions with social robots operating in public spaces, where users may be asked to provide personal information to receive personalized experiences. This article investigates how design transparency influences user trust and data-sharing behavior in human-robot interactions. We conducted an experiment with 143 participants who interacted with the social robot ARI under two transparency conditions: low and high transparency. In the low-transparency condition, participants were informed about the data being collected and could choose to save or delete it. In the high-transparency condition, the robot additionally indicated the sensitivity level of each data item: low (e.g., scenario preference), medium (e.g., name and e-mail), and high (e.g., religious beliefs), allowing participants to make more informed decisions. Participants were presented with two scenarios: exploring city events and discovering local attractions. They received personalized recommendations based on their preferences, with the option to provide personal data (name, phone number, e-mail) for possible future communication. After the interaction, participants decided whether to save or delete the data they had shared. The results indicated that while transparency did not significantly affect trust in the robot, it influenced data-sharing behavior. In particular, participants in the high-transparency condition demonstrated more cautious behavior, opting to save less data and delete more. Furthermore, the results showed that both sensitivity levels and transparency influenced the participants’ data-sharing choices. Low-level sensitivity data led to the highest rates of saving and the lowest rates of deleting, while medium-level sensitivity data showed the opposite pattern. These findings highlight the need to align data categorization with user perceptions to address data sharing concerns more effectively., This work is partially funded by the Research Council of Norway (RCN) through the project “User-centred Security Framework for Social Robots in Public Space (SecuRoPS),” with the project number 321324 (INT-NO/0875), byMCIN/AEI/10.13039/501100011033, and by the European Union NextGenerationEU/PRTR under the project ROB-IN(PLEC2021-007859)., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

TALP-UPC at ProbSum 2023: fine-tuning and data augmentation strategies for NER

  • Torrero Salas, Neil
  • Sant Muniesa, Gerard
  • Escolano Peinado, Carlos|||0000-0001-6657-673X
This paper describes the submission of the TALP-UPC team to the Problem List Summarization task from the BioNLP 2023 workshop. This task consists of automatically extracting a list of health issues from the e-health medical record of a given patient. Our submission combines additional steps of data annotationwith finetuning of BERT pre-trained language models. Our experiments focus on the impact of finetuning on different datasets as well as the addition of data augmentation techniques to delay overfitting., This work was funded by Spanish State Research Agency (AEI) project PID2019-107579RBI00 (AEI/10.13039/501100011033) and the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

NYAM: the role of configurable engagement strategies in robotic-assisted feeding

  • Barrué Subirana, Cristian|||0000-0002-3342-1085
  • Suárez Hernández, Alejandro
  • Inzitari, Marco
  • Ribera Solé, Aida
  • Alenyà Ribas, Guillem|||0000-0002-6018-154X
© 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM., In some contexts, like geriatric hospitals, the number of patients requiring assistance with feeding is very high and robots may be an effective tool for caregivers to provide better assistance. This article introduces NYAM, a robot designed to aid in the feeding process for individuals. Our robot is equipped with a mechanism to effectively recapture the person's attention whenever necessary. The mechanism is easily adjustable by the caregivers, allowing the straightforward customisation of the feeding service. The approach was evaluated, within a geriatric hospital, with 9 patients who used the robot for 5 consecutive days. We argue that incorporating enhanced social aspects into the robot is imperative to enhance the effectiveness and acceptance of this solution. © 2024 Copyright held by the owner/author(s), This work was supported by the project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR"; project CHLOE-GRAPH PID2020-118649RB-I00 funded by MCIN/ AEI /10.13039/501100011033; and project PIONEER from PSPV., Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Toxicity in multilingual machine translation at scale

  • Ruiz Costa-jussà, Marta
  • Smith, Eric
  • Ropers, Christophe
  • Licht, Daniel
  • Maillard, Jean
  • Ferrando Monsonís, Javier|||0000-0002-2637-0961
  • Escolano Peinado, Carlos|||0000-0001-6657-673X
Machine Translation systems can produce different types of errors, some of which are characterized as critical or catastrophic due to the specific negative impact that they can have on users. In this paper we focus on one type of critical error: added toxicity. We evaluate and analyze added toxicity when translating a large evaluation dataset (HOLISTICBIAS, over 472k sentences, covering 13 demographic axes) from English into 164 languages. An automatic toxicity evaluation shows that added toxicity across languages varies from 0% to 5%. The output languages with the most added toxicity tend to be low-resource ones, and the demographic axes with the most added toxicity include sexual orientation, gender and sex, and ability. We also perform human evaluation on a subset of 8 translation directions, confirming the prevalence of true added toxicity. We use a measurement of the amount of source contribution to the translation, where a low source contribution implies hallucination, to interpret what causes toxicity. Making use of the input attributions allows us to explain toxicity, because the source contributions significantly correlate with toxicity for 84% of languages studied. Given our findings, our recommendations to reduce added toxicity are to curate training data to avoid mistranslations, mitigate hallucination and check unstable translations. WARNING: this paper contains examples of toxicity that may be offensive or upsetting in nature., Javier Ferrando and Carlos Escolano are supported by the Spanish Ministerio de Ciencia e Innovación through the project PID2019-107579RB-I00/AEI /10.13039/501100011033 and the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859), Peer Reviewed
UPCommons. Portal del coneixement obert de la UPC
Publicación Contribución de congreso (ConferenceOutput)

Adaptive human-robot collaboration: evolutionary learning of action costs using an action outcome simulator

  • Izquierdo Badiola, Silvia
  • Alenyà Ribas, Guillem|||0000-0002-6018-154X
  • Rizzo, Carlos
© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, One of the main challenges for successful human- robot collaborative applications lies in adapting the plan to the human agent’s changing state and preferences. A promising solution is to bridge the gap between agent modelling and AI task planning, which can be done by integrating the agent state as action costs in the task planning domain. This allows for the plan to be adapted to different partners, by influencing the action allocation. The difficulty then lies in setting appropriate action costs. This paper presents a novel framework to learn a set of planning action costs considering the preferred actions for an agent based on their state. An evolutionary optimisation algorithm is used for this purpose, and an action outcome simulator is developed to act as the black-box function, based on both an agent model and an action type model. This addresses the challenge of collecting data in HRC real-world scenarios, accelerating the learning for posterior fine-tuning in real applications. The coherence of the models and the simulator is proven through a conducted survey, and the learning algorithm is shown to learn appropriate action costs, producing plans that satisfy both the agents’ preferences and the prioritised plan requisites. The resulting system is a generic learning framework integrating components that can be easily extended to a wide range of applications, models and planning formalisms., This work has been partially supported by MCIN/ AEI /10.13039/501100011033 under the project CHLOE-GRAPH (PID2020- 119244GB-I00); by MCIN/ AEI /10.13039/501100011033 and by the ”European Union (EU) NextGenerationEU/PRTR under the project ROB-IN (PLEC2021-007859), Peer Reviewed
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

TALP-UPC at ProbSum 2023: fine-tuning and data augmentation strategies for NER

  • Torrero Salas, Neil
  • Sant Muniesa, Gerard
  • Escolano Peinado, Carlos
This paper describes the submission of the TALP-UPC team to the Problem List Summarization task from the BioNLP 2023 workshop. This task consists of automatically extracting a list of health issues from the e-health medical record of a given patient. Our submission combines additional steps of data annotationwith finetuning of BERT pre-trained language models. Our experiments focus on the impact of finetuning on different datasets as well as the addition of data augmentation techniques to delay overfitting., This work was funded by Spanish State Research Agency (AEI) project PID2019-107579RBI00 (AEI/10.13039/501100011033) and the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859), Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

NYAM: the role of configurable engagement strategies in robotic-assisted feeding

  • Barrué Subirana, Cristian
  • Suárez Hernández, Alejandro
  • Inzitari, Marco
  • Ribera Solé, Aida
  • Alenyà Ribas, Guillem
© 2024 Copyright held by the owner/author(s). Publication rights licensed to ACM., In some contexts, like geriatric hospitals, the number of patients requiring assistance with feeding is very high and robots may be an effective tool for caregivers to provide better assistance. This article introduces NYAM, a robot designed to aid in the feeding process for individuals. Our robot is equipped with a mechanism to effectively recapture the person's attention whenever necessary. The mechanism is easily adjustable by the caregivers, allowing the straightforward customisation of the feeding service. The approach was evaluated, within a geriatric hospital, with 9 patients who used the robot for 5 consecutive days. We argue that incorporating enhanced social aspects into the robot is imperative to enhance the effectiveness and acceptance of this solution. © 2024 Copyright held by the owner/author(s), This work was supported by the project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR"; project CHLOE-GRAPH PID2020-118649RB-I00 funded by MCIN/ AEI /10.13039/501100011033; and project PIONEER from PSPV., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

ReSeTOX: Re-learning attention weights for toxicity mitigation in machine translation

  • García Gilabert, Javier
  • Escolano Peinado, Carlos
  • Ruiz Costa-jussà, Marta
Our proposed method, RESETOX (REdo SEarch if TOXic), addresses the issue of Neural Machine Translation (NMT) generating translation outputs that contain toxic words not present in the input. The objective is to mitigate the introduction of toxic language without the need for retraining. In the case of identified added toxicity during the inference process, RESETOX dynamically adjusts the key-value self-attention weights and re-evaluates the beam search hypotheses. Experimental results demonstrate that RESETOX achieves a remarkable 57% reduction in added toxicity while maintaining an average translation quality of 99.5% across 164 languages. Our code is available at: https: //github.com/mt-upc/ReSeTOX, The work by Javier García Gilabert and Carlos Escolano has been funded by the Spanish Ministerio de Ciencia e Innovación and the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859), Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Impact of design transparency on trust and data sharing during human-robot interactions in public places

  • Aryania, Azra
  • Alenyà Ribas, Guillem
  • Chockalingam, Sabarathinam
  • Rødsethol, Hanne Kristine
The prevalence of social robots is increasing, with examples such as customer service robots in malls and airports. This trend highlights the importance of transparency, particularly in data-sharing interactions with social robots operating in public spaces, where users may be asked to provide personal information to receive personalized experiences. This article investigates how design transparency influences user trust and data-sharing behavior in human-robot interactions. We conducted an experiment with 143 participants who interacted with the social robot ARI under two transparency conditions: low and high transparency. In the low-transparency condition, participants were informed about the data being collected and could choose to save or delete it. In the high-transparency condition, the robot additionally indicated the sensitivity level of each data item: low (e.g., scenario preference), medium (e.g., name and e-mail), and high (e.g., religious beliefs), allowing participants to make more informed decisions. Participants were presented with two scenarios: exploring city events and discovering local attractions. They received personalized recommendations based on their preferences, with the option to provide personal data (name, phone number, e-mail) for possible future communication. After the interaction, participants decided whether to save or delete the data they had shared. The results indicated that while transparency did not significantly affect trust in the robot, it influenced data-sharing behavior. In particular, participants in the high-transparency condition demonstrated more cautious behavior, opting to save less data and delete more. Furthermore, the results showed that both sensitivity levels and transparency influenced the participants’ data-sharing choices. Low-level sensitivity data led to the highest rates of saving and the lowest rates of deleting, while medium-level sensitivity data showed the opposite pattern. These findings highlight the need to align data categorization with user perceptions to address data sharing concerns more effectively., This work is partially funded by the Research Council of Norway (RCN) through the project “User-centred Security Framework for Social Robots in Public Space (SecuRoPS),” with the project number 321324 (INT-NO/0875), byMCIN/AEI/10.13039/501100011033, and by the European Union NextGenerationEU/PRTR under the project ROB-IN(PLEC2021-007859)., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Toxicity in multilingual machine translation at scale

  • Ruiz Costa-jussà, Marta
  • Smith, Eric
  • Ropers, Christophe
  • Licht, Daniel
  • Maillard, Jean
  • Ferrando Monsonís, Javier
  • Escolano Peinado, Carlos
Machine Translation systems can produce different types of errors, some of which are characterized as critical or catastrophic due to the specific negative impact that they can have on users. In this paper we focus on one type of critical error: added toxicity. We evaluate and analyze added toxicity when translating a large evaluation dataset (HOLISTICBIAS, over 472k sentences, covering 13 demographic axes) from English into 164 languages. An automatic toxicity evaluation shows that added toxicity across languages varies from 0% to 5%. The output languages with the most added toxicity tend to be low-resource ones, and the demographic axes with the most added toxicity include sexual orientation, gender and sex, and ability. We also perform human evaluation on a subset of 8 translation directions, confirming the prevalence of true added toxicity. We use a measurement of the amount of source contribution to the translation, where a low source contribution implies hallucination, to interpret what causes toxicity. Making use of the input attributions allows us to explain toxicity, because the source contributions significantly correlate with toxicity for 84% of languages studied. Given our findings, our recommendations to reduce added toxicity are to curate training data to avoid mistranslations, mitigate hallucination and check unstable translations. WARNING: this paper contains examples of toxicity that may be offensive or upsetting in nature., Javier Ferrando and Carlos Escolano are supported by the Spanish Ministerio de Ciencia e Innovación through the project PID2019-107579RB-I00/AEI /10.13039/501100011033 and the “European Union NextGenerationEU/PRTR” under the project ROB-IN (PLEC2021-007859), Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Introducing social robots to assess frailty in older adults

  • Civit Bertran, Aniol
  • Andriella, Antonio
  • Barrué Subirana, Cristian
  • Antonio, Maite
  • Boqué, Concepción
  • Alenyà Ribas, Guillem
Frailty is a crucial indicator in determining the well-being of older adults in terms of their health. With the growing number of elderly people, the demand for geriatricians is increasing, which means that they have less time to spend with each patient. The current methods for frailty assessment use simple tests that are time-consuming and do not require specific medical expertise. To address this issue, this paper proposes the use of social robots to assess frailty autonomously. It presents a practical proposal that defines the robot’s behavior and explains the design and implementation concepts. Finally, it discusses some of the challenges that may arise from introducing social robots as frailty evaluators., This work was supported by the project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR"; project CHLOE-GRAPH PID2020-118649RB-I00 funded by MCIN/AEI /10.13039/501100011033; and the 23S06141-001 FRAILWATCH project funded by Barcelona Ciencia 2020-2023 Plan., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Exploring the potential of a robot-assisted frailty assessment system for elderly care

  • Civit Bertran, Aniol
  • Andriella, Antonio
  • Antonio, Maite
  • Javierre Garcés, Casimiro
  • Boqué, Concepción
  • Alenyà Ribas, Guillem
© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works., Frailty assessment plays a pivotal role in providing older adults care. However, the current process is time- consuming and only measures patients’ completion time for each test. This paper introduces a set of algorithms to be used in robots to autonomously perform frailty assessments. In doing so we aim at reducing therapists’ burden and provide additional frailty-related metrics that can enhance the effectiveness of diagnosis. We conducted a pilot study with 22 elderly participants and compared our system’s performance with that of medical professionals to assess its precision. The results demonstrate that our approach achieved performances close to that of its human counterpart. This research represents an important step forward in the integration of social robotics in healthcare, offering potential benefits for patient care and clinical decision-making., This work was supported by the project ROBIN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the "European Union NextGenerationEU/PRTR"; by the EU-founded project grant agreement No 101070930 (VALAWAI); and the 23S06141-001 FRAILWATCH project funded by Barcelona Ciencia 2020-2023 Plan., Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Artículo científico (JournalArticle)

Unfolding the literature: a review of robotic cloth manipulation

  • Longhini, Alberta
  • Wang, Yufei
  • Garcia Camacho, Irene
  • Blanco Mulero, David
  • Moletta, Marco
  • Welle, Michael C.
  • Alenyà Ribas, Guillem
  • Yin, Hang
  • Erickson, Zackory
  • Held, David
  • Borràs Sol, Júlia
  • Kragic, Danica
The realm of textiles spans clothing, households, healthcare, sports, and industrial applications. The deformable nature of these objects poses unique challenges that prior work on rigid objects cannot fully address. The increasing interest within the community in textile perception and manipulation has led to new methods that aim to address challenges in modeling, perception, and control, resulting in significant progress. However, this progress is often tailored to one specific textile or a subcategory of these textiles. To understand what restricts these methods and hinders current approaches from generalizing to a broader range of real-world textiles, this review provides an overview of the field, focusing specifically on how and to what extent textile variations are addressed in modeling, perception, benchmarking, and manipulation of textiles. We conclude by identifying key open problems and outlining grand challenges that will drive future advancements in the field., This work was supported by the European Research Council (ERC-BIRD); the European Union’s Horizon Europe program through projects SoftEnable (HORIZON-CL4-2021-DIGITALEMERGING-01-101070600) and IRE (HORIZON-CL4-2023-DIGITAL-EMERGING-01- 101135082); the Swedish Research Council, the Knut and Alice Wallenberg Foundation, and the National Science Foundation under NSF CAREER grant IIS-2046491; and the project ROBIN PLEC2021-007859 funded by MCIN/AEI/10.13039/501100011033 and by European Union NextGenerationEU/PRTR., Peer Reviewed, Postprint (published version)
Recercat. Dipósit de la Recerca de Catalunya
Publicación Contribución de congreso (ConferenceOutput)

Adaptive human-robot collaboration: evolutionary learning of action costs using an action outcome simulator

  • Izquierdo Badiola, Silvia
  • Alenyà Ribas, Guillem
  • Rizzo, Carlos
© 2023 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting /republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works, One of the main challenges for successful human- robot collaborative applications lies in adapting the plan to the human agent’s changing state and preferences. A promising solution is to bridge the gap between agent modelling and AI task planning, which can be done by integrating the agent state as action costs in the task planning domain. This allows for the plan to be adapted to different partners, by influencing the action allocation. The difficulty then lies in setting appropriate action costs. This paper presents a novel framework to learn a set of planning action costs considering the preferred actions for an agent based on their state. An evolutionary optimisation algorithm is used for this purpose, and an action outcome simulator is developed to act as the black-box function, based on both an agent model and an action type model. This addresses the challenge of collecting data in HRC real-world scenarios, accelerating the learning for posterior fine-tuning in real applications. The coherence of the models and the simulator is proven through a conducted survey, and the learning algorithm is shown to learn appropriate action costs, producing plans that satisfy both the agents’ preferences and the prioritised plan requisites. The resulting system is a generic learning framework integrating components that can be easily extended to a wide range of applications, models and planning formalisms., This work has been partially supported by MCIN/ AEI /10.13039/501100011033 under the project CHLOE-GRAPH (PID2020- 119244GB-I00); by MCIN/ AEI /10.13039/501100011033 and by the ”European Union (EU) NextGenerationEU/PRTR under the project ROB-IN (PLEC2021-007859), Peer Reviewed, Postprint (author's final draft)
Recercat. Dipósit de la Recerca de Catalunya
Publicación

Inside-out states of garments, Technical report IRI-TR-24-01

  • Jiménez Schlegl, Pablo
Cloth items such as garments can be potentially in infinite different configurations, due to their deformability. For the practical purposes of manipulation, however, it is possible to discretize the space of deformations into a set of equivalent states. This is pursued in the present work, where the complex problem of recovering the canonical configuration of partially (or completely) reversed garments is addressed. Without claiming to solve the hard perceptual and manipulative challenges of this type of task –which are also thoroughly described–, this work should rather be viewed as a pioneering effort to formalize the high-level strategies that aim at solving this canonical configuration-recovering task., Cloth items such as garments can be potentially in infinite different configurations, due to their deformability. For the practical purposes of manipulation, however, it is possible to discretize the space of deformations into a set of equivalent states. This is pursued in the present work, where the complex problem of recovering the canonical configuration of partially (or completely) reversed garments is addressed. Without claiming to solve the hard perceptual and manipulative challenges of this type of task –which are also thoroughly described–, this work should rather be viewed as a pioneering effort to formalize the high-level strategies that aim at solving this canonical configuration-recovering task., This work has been partially funded by the European Union Horizon 2020 Programme under grant agreement no. 741930 (CLOTHILDE), Project ROB-IN PLEC2021-007859 funded by MCIN/ AEI /10.13039/501100011033 and by the ”European Union NextGenerationEU/PRTR” and Project CHLOE-Graph PID2020-118649RB-I00 funded by MCIN/AEI /10.13039/501100011033, Preprint