{"title":"Enhancing supermarket robot interaction: an equitable multi-level LLM conversational interface for handling diverse customer intents.","authors":"Chandran Nandkumar, Luka Peternel","doi":"10.3389/frobt.2025.1576348","DOIUrl":"10.3389/frobt.2025.1576348","url":null,"abstract":"<p><p>This paper presents the design and evaluation of a comprehensive system to develop voice-based interfaces to support users in supermarkets. These interfaces enable shoppers to convey their needs through both generic and specific queries. Although customisable state-of-the-art systems like GPTs from OpenAI are easily accessible and adaptable, featuring low-code deployment with options for functional integration, they still face challenges such as increased response times and limitations in strategic control for tailored use cases and cost optimization. Motivated by the goal of crafting equitable and efficient conversational agents with a touch of personalisation, this study advances on two fronts: 1) a comparative analysis of four popular off-the-shelf speech recognition technologies to identify the most accurate model for different genders (male/female) and languages (English/Dutch) and 2) the development and evaluation of a novel multi-LLM supermarket chatbot framework, comparing its performance with a specialized GPT model powered by the GPT-4 Turbo, using the Artificial Social Agent Questionnaire (ASAQ) and qualitative participant feedback. Our findings reveal that OpenAI's Whisper leads in speech recognition accuracy between genders and languages and that our proposed multi-LLM chatbot architecture, which outperformed the benchmarked GPT model in performance, user satisfaction, user-agent partnership, and self-image enhancement, achieved statistical significance in these four key areas out of the 13 evaluated aspects that all showed improvements. The paper concludes with a simple method for supermarket robot navigation by mapping the final chatbot response to the correct shelf numbers to which the robot can plan sequential visits. Later, this enables the effective use of low-level perception, motion planning, and control capabilities for product retrieval and collection. We hope that this work encourages more efforts to use multiple specialized smaller models instead of always relying on a single powerful model.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1576348"},"PeriodicalIF":2.9,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12069059/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144046163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Declan Symes, Michael Rose, Emanuel David Nunez Sardinha, Aghil Jafari, Javeed Hussain, Appolinaire Etoundi
{"title":"The Anthro-Thumb: a biomimetic hybrid soft robotic carpometacarpal saddle joint for the thumb.","authors":"Declan Symes, Michael Rose, Emanuel David Nunez Sardinha, Aghil Jafari, Javeed Hussain, Appolinaire Etoundi","doi":"10.3389/frobt.2025.1496073","DOIUrl":"10.3389/frobt.2025.1496073","url":null,"abstract":"<p><p>Robotic hand design is multifaceted, with the design of robotic thumbs often oversimplified to facilitate ease of manufacture, control, and reliability. Despite the extensive development of robotic hands, the need for a more dexterous and anthropomorphic thumb design remains significant, particularly for applications in prosthetics and rehabilitation robotics, where naturalistic movement and adaptability are essential. This paper addresses this gap by exploring the conception, evolution, and evaluation of a unique biomimetic soft thumb. The thumb plays a vital role in hand function, and its unique range of motion is enabled by the carpometacarpal (CMC) saddle joint. By harnessing the biologically accurate mechanisms of the CMC joint, this research aims to enhance the functionality of tendon-driven robotic hands, offering improved dexterity and adaptability for tasks such as grasping and manipulation. The introduced <i>Anthro-Thumb</i> is a biomimetic soft robotic thumb that provides a comprehensive range of motion at the thumb's base while ensuring cost efficiency and reduced mechanical complexity. A comparative analysis with existing robotic thumb designs highlights the advancements of the <i>Anthro-Thumb</i>, particularly in terms of range of motion and cost-effectiveness. Additionally, we discuss the long-term durability and maintenance requirements of the soft robotic materials and components used. When subjected to the Kapandji physiotherapy test, the design secured a commendable score of 9 of 10, with 10 representing the highest level of dexterity achievable by a human thumb. The findings affirm that employing biomimetic soft-structured robotic CMC saddle joints is a promising strategy to address the challenges associated with robotic thumb development in robotic hands.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1496073"},"PeriodicalIF":2.9,"publicationDate":"2025-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12069996/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143991819","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Monitoring pilots' mental workload in real flight conditions using multinomial logistic regression with a ridge estimator.","authors":"Muhammad Haseeb, Rashid Nadeem, Nazia Sultana, Noman Naseer, Hammad Nazeer, Frédéric Dehais","doi":"10.3389/frobt.2025.1441801","DOIUrl":"https://doi.org/10.3389/frobt.2025.1441801","url":null,"abstract":"<p><p>Piloting an aircraft is a cognitive task that requires continuous verbal, visual, and auditory attentions (e.g., Air Traffic Control Communication). An increase or decrease in mental workload from a specific level can alter auditory and visual attention, resulting in pilot errors. The objective of this research is to monitor pilots' mental workload using advanced machine learning techniques to achieve improved accuracy compared to previous studies. Electroencephalogram (EEG) data were recorded from 22 pilots operating under visual flight rules (VFR) conditions using a six dry-electrode Enobio Neuroelectrics system, and the Riemannian artifact subspace reconstruction (rASR) filter was used for data cleaning. An information gain (IG) attribute evaluator was used to select 25 optimal features out of 72 power spectral and statistical extracted features. In this study, 15 classifiers were used for classification. Multinomial logistic regression with a ridge estimator was selected, achieving a significant mean accuracy of 84.6% on the dataset from 17 subjects. Data were initially collected from 22 subjects, but 5 were excluded due to data synchronization issues. This work has several limitations, such as the author did not counter balance the order of scenario, could not control all the variables such as wind conditions, and workload was not stationary in each leg of the flight pattern. This study demonstrates that multinomial logistic regression with a ridge estimator shows significant classification accuracy (p < 0.05) and effectively detects pilot mental workload in real flight scenarios.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1441801"},"PeriodicalIF":2.9,"publicationDate":"2025-04-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12058475/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144064973","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Precision farming in aquaculture: non-invasive monitoring of Atlantic salmon (<i>Salmo salar</i>) behaviour in response to environmental conditions in commercial sea cages for health and welfare assessment.","authors":"Meredith Burke, Dragana Nikolic, Pieter Fabry, Hemang Rishi, Trevor Telfer, Sonia Rey Planellas","doi":"10.3389/frobt.2025.1574161","DOIUrl":"https://doi.org/10.3389/frobt.2025.1574161","url":null,"abstract":"<p><p>Studies show that Atlantic salmon in captivity adjust their distribution in sea cages based on environmental gradients like temperature, waves, and photoperiod. This study used a computer vision algorithm at three marine farms to analyse fish group swimming behaviour termed \"activity\" (measured in percent), which includes fish abundance, speed, and shoal cohesion. The activity metric inferred the depth distribution of the main fish group and was analysed with respect to environmental conditions to explore potential behavioural drivers and used to assess changes in fish behaviour in response to a stressor, a storm event. During winter conditions, Farms A and B showed distinct thermal stratification, with fish activity demonstrating preference for the warmer lower water column (39.6 ± 15.3% and 27.5 ± 10.2%) over the upper water column (16.3 ± 5.7% and 18 ± 3.3%; p < 0.001). At Farm C, with thermally homogenous water, fish activity was similarly distributed between the upper (18.2 ± 6.9%) and lower (17.7 ± 7.6%) water column. Severe weather increased wave heights, influencing fish horizontal distribution differently at Farms B and C. At Farm B, a deeper site, fish remained in the warmer lower water column and avoided surface waves, while at Farm C, with shallower cages, they moved toward the side of the cage nearest the centre of the farm, presumably less exposed due to nearby cages. Understanding fish behavioural responses to environmental conditions can inform management practices, while using cameras with associated algorithms offers a powerful, non-invasive tool for continuously monitoring and safeguarding fish health and welfare.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1574161"},"PeriodicalIF":2.9,"publicationDate":"2025-04-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12056675/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144041778","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Motonobu Aoki, Francesco Rea, Giuseppe Di Cesare, Giulio Sandini, Takura Yanagi, Atsushi Takamatsu, Tomohiro Yamamura
{"title":"Effects of gentle and rude vitality forms in social robots on humans during cognitive multitasking.","authors":"Motonobu Aoki, Francesco Rea, Giuseppe Di Cesare, Giulio Sandini, Takura Yanagi, Atsushi Takamatsu, Tomohiro Yamamura","doi":"10.3389/frobt.2025.1305685","DOIUrl":"https://doi.org/10.3389/frobt.2025.1305685","url":null,"abstract":"<p><p>Designing a social humanoid robot to enhance human cognitive multitasking in a mixed human-robot team is anything but straightforward. In fact, the robot's presence and behavior can either improve or impair human performance. In this study, we examined how different vitality forms-expressed through the robot's actions, speech and facial expressions-affect cognitive multitasking. Analysis of human facial expressions and skin conductance data revealed that a robot exhibiting a gentle vitality form fostered a more positive and relaxed state than one displaying a rude vitality form. Moreover, the gentle vitality form improved human performance on tasks involving short-term memory and continuous target tracking. To our knowledge, this is the first study to explore the long-term impact of vitality forms. Overall, our findings suggest that properly designing a social humanoid robot's vitality forms can significantly enhance cognitive multitasking performance in human-robot teams.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1305685"},"PeriodicalIF":2.9,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12053143/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143991818","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Investigation of the conditions for continuous information conveyance by two autonomous conversational agents.","authors":"Takamichi Isowa, Kohei Ogawa, Satoshi Sato, Tomonori Kubota, Hiroshi Ishiguro","doi":"10.3389/frobt.2025.1417488","DOIUrl":"https://doi.org/10.3389/frobt.2025.1417488","url":null,"abstract":"<p><strong>Introduction: </strong>In recent years, information conveyance through conversation using agents such as robots and avatars has gained attention. Among them, conversation by two agents has been shown to encourage effective information conveyance. Previous studies have also demonstrated that incorporating subjective information, such as emotions, into conversations enhances this effect. Therefore, a medium for information conveyance involving two autonomous agents and including subjective information is expected to be effective.</p><p><strong>Methods: </strong>In this study, such a medium was implemented, and the conditions necessary for it to convey information continuously were investigated. Objective information was defined as the content of existing news, and subjective information was defined as the preference toward the news. A frame structure was used for organizing objective information, and a network structure was used for subjective information. A method was developed to autonomously obtain both types of information. This knowledge was then distributed to two agents, who exchanged it and attempted to understand each other through conversation.</p><p><strong>Results: </strong>Experiments were conducted to determine whether the subjective information obtained autonomously by the agents was as natural and consistent as that of humans. Further experiments examined the conditions for enabling continuous information conveyance using the medium.</p><p><strong>Discussion: </strong>The results indicated that conveying important information first and using robots rather than text were effective strategies for maintaining continuous information conveyance.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1417488"},"PeriodicalIF":2.9,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12044221/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144020081","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Factors influencing subjective opinion attribution to conversational robots.","authors":"Yuki Sakamoto, Takahisa Uchida, Midori Ban, Hiroshi Ishiguro","doi":"10.3389/frobt.2025.1521169","DOIUrl":"https://doi.org/10.3389/frobt.2025.1521169","url":null,"abstract":"<p><p>The usefulness of conversational robots has been demonstrated in various fields. It is suggested that expressing subjective opinions is essential for conversational robots to stimulate users' willingness to engage in conversation. However, a challenge remains in that users often find it difficult to attribute subjective opinions to robots. Therefore, this study aimed to examine the factors influencing the attribution of subjective opinions to robots. We investigated robot and human factors that may affect subjective opinion attribution to robots. Furthermore, these factors were investigated in four different cases, adopting a combination of the robots' types and control methods, considering actual scenarios of robot usage. The survey was conducted online, and the participants completed a questionnaire after receiving instructions regarding the robot. The results revealed that the perception of the robots' sensory capabilities significantly influenced the attribution of subjective opinions in all the surveyed cases. Additionally, in the case of an autonomous small robot, there is a possibility that participants' self-perception of their judgment abilities might also impact their subjective opinion attribution to the robot. The findings highlight the importance of aligning subjective opinion utterances in conversational robots with user perceptions of the robot's sensory capabilities. They also emphasized the significance of exploring how users' self-perceptions influence their perceptions of robots. These insights provide valuable guidance for designing conversational strategies and speech generation in robots that engage in the exchange of subjective opinions with humans.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1521169"},"PeriodicalIF":2.9,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12040941/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144044867","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"An open-source reproducible chess robot for human-robot interaction research.","authors":"Renchi Zhang, Joost de Winter, Dimitra Dodou, Harleigh Seyffert, Yke Bauke Eisma","doi":"10.3389/frobt.2025.1436674","DOIUrl":"https://doi.org/10.3389/frobt.2025.1436674","url":null,"abstract":"<p><p>Recent advancements in AI have accelerated the evolution of versatile robot designs. Chess provides a standardized environment for evaluating the impact of robot behavior on human behavior. This article presents an open-source chess robot for human-robot interaction research, specifically focusing on verbal and non-verbal interactions. The OpenChessRobot recognizes chess pieces using computer vision, executes moves, and interacts with the human player through voice and robotic gestures. We detail the software design, provide quantitative evaluations of the efficacy of the robot, and offer a guide for its reproducibility. An online survey examining people's views of the robot in three possible scenarios was conducted with 597 participants. The robot received the highest ratings in the robotics education and the chess coach scenarios, while the home entertainment scenario received the lowest scores. The code is accessible on GitHub: https://github.com/renchizhhhh/OpenChessRobot.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1436674"},"PeriodicalIF":2.9,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12040702/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144016403","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"How the presence of others shapes the user experience of service robots.","authors":"Stefan Tretter, Pia von Terzi, Sarah Diefenbach","doi":"10.3389/frobt.2025.1538711","DOIUrl":"https://doi.org/10.3389/frobt.2025.1538711","url":null,"abstract":"<p><p>In the age of mobile and self-service technologies, human-computer interaction (HCI) often takes place in public settings. Such interactions can be considered a performance in front of others, when the relationship with potential observers may affect user preferences for different interaction styles. From a psychological perspective, public interactions may feel embarrassing or disturbing, but they also provide the opportunity for favorable self-presentation or connection with others. The present study investigated how the presence of different observers (i.e., acquaintance, stranger) emphasizes different psychological needs and, in turn, affects preferences for more or less expressive interactions with a service robot. Results show that users' need for relatedness was higher when imagining a robot interaction with close observers, while popularity was more important with unknown observers. Relatedness was directly linked to a preference for more expressive interactions, regardless of the expected outcome. In contrast, popularity led to stronger expressivity preferences only when users anticipated a successful interaction for which they could take credit. Our research provides valuable insights into the impact of user-observer-relationship on public HCI, and can inspire designers to take into account how present others and users' expectation of successful outcomes may call for different degrees of expressivity in interaction design.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1538711"},"PeriodicalIF":2.9,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12040957/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144003381","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Bilal Msallem, Lara Veronesi, Florian Samuel Halbeisen, Michel Beyer, Adrian Dragu, Florian Markus Thieringer
{"title":"Evaluation of angulation and distance deviation for robot-guided laser osteotomy - a follow-up study on digital high-tech procedures.","authors":"Bilal Msallem, Lara Veronesi, Florian Samuel Halbeisen, Michel Beyer, Adrian Dragu, Florian Markus Thieringer","doi":"10.3389/frobt.2025.1559483","DOIUrl":"https://doi.org/10.3389/frobt.2025.1559483","url":null,"abstract":"<p><strong>Background and objective: </strong>Conventional osteotomy tools, including drills and saws, have been associated with several limitations, such as restricted cutting geometry and the risk of heat-induced necrosis, which affects bone healing. Laser-based osteotomy systems have emerged as a promising solution for these constraints. This study aims to evaluate the accuracy of robot-guided laser osteotomy compared to conventional cutting-guided osteotomy based on surface scanning.</p><p><strong>Materials and methods: </strong>Ten 3D printed mandibular models were used to perform segmentectomy. Five models were treated with conventional osteotomies employing a cutting-guided saw technique, while the remaining five were subjected to laser osteotomy. Initially conducted using root mean square (RMS) values, the analysis has been expanded to reevaluate the angulation and distance deviation outcomes.</p><p><strong>Results: </strong>Precision analysis of the upper cutting plane revealed a statistically significant difference in distance deviation between the laser osteotomy group (0.48 mm) and the conventional osteotomy group (0.78 mm). In terms of angulation deviation, the laser osteotomy group exhibited, both in the upper and lower cutting planes, statistically significant results (2.19° and 2.86°) compared to the osteotomy group (5.15° and 8.12°).</p><p><strong>Conclusion: </strong>Based on the observed angulation and distance deviations, it can be concluded that robot-guided laser systems achieve significantly higher accuracy in osteotomies than conventional cutting-guided systems currently available. Consistent with the findings of a prior study, these results confirm that robot-guided laser osteotomy provides substantial advantages, facilitating the seamless integration of precise virtual preoperative planning with exact execution in the human body.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1559483"},"PeriodicalIF":2.9,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12040628/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144034508","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}