Frontiers in Robotics and AI最新文献

{"title":"Agentic LLM-based robotic systems for real-world applications: a review on their agenticness and ethics.","authors":"Emmanuel K Raptis, Athanasios Ch Kapoutsis, Elias B Kosmatopoulos","doi":"10.3389/frobt.2025.1605405","DOIUrl":"10.3389/frobt.2025.1605405","url":null,"abstract":"<p><p>Agentic AI refers to autonomous systems that can perceive their environment, make decisions, and take actions to achieve goals with minimal or no human intervention. Recent advances in Large Language Models (LLMs) have opened new pathways to imbue robots with such \"agentic\" behaviors by leveraging the LLMs' vast knowledge and reasoning capabilities for planning and control. This survey provides the first comprehensive exploration of LLM-based robotic systems integration into agentic behaviors that have been validated in real-world applications. We systematically categorized these systems across navigation, manipulation, multi-agent, and general-purpose multi-task robots, reflecting the range of applications explored. We introduce a novel, first-of-its-kind agenticness classification that evaluates existing LLM-driven robotic works based on their degree of autonomy, goal-directed behavior, adaptability, and decision-making. Additionally, central to our contribution is an evaluation framework explicitly addressing ethical, safety, and transparency principles-including bias mitigation, fairness, robustness, safety guardrails, human oversight, explainability, auditability, and regulatory compliance. By jointly mapping the landscape of agentic capabilities and ethical safeguards, we uncover key gaps, tensions, and design trade-offs in current approaches. We believe that this work serves as both a diagnostic and a call to action: as LLM-empowered robots grow more capable, ensuring they remain comprehensible, controllable, and aligned with societal norms is not optional-it is essential.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1605405"},"PeriodicalIF":3.0,"publicationDate":"2025-08-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12402697/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144993944","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":"Design, development, and validation of a non-backdrivable active ankle-foot orthosis for the TWIN lower-limb exoskeleton.","authors":"Raffaele Giannattasio, Nicolò Boccardo, Riccardo Vaccaro, Heeral Bhatt, Stefano Maludrottu, Elena De Momi, Matteo Laffranchi","doi":"10.3389/frobt.2025.1647989","DOIUrl":"10.3389/frobt.2025.1647989","url":null,"abstract":"<p><p>This study's primary objective was to develop an Active Ankle-Foot Orthosis (AAFO) specifically designed for integration into lower-limb exoskeletons. An analysis of human ankle motion is conducted to inform the development process, guiding the creation of an AAFO that aligns with specifics extrapolated by real data. The AAFO incorporates an electric motor with a non-backdrivable transmission system, engineered to reduce distal mass, minimize power consumption, and enable high-precision position control. Capable of generating up to 50 Nm of peak torque, the AAFO is designed to provide support throughout the walking cycle, targeting pathological conditions such as foot drop and toe drag. Performance was first validated through benchtop experiments under unloaded conditions. The AAFO was then integrated into the TWIN lower-limb exoskeleton, employing an optimal trajectory planning method to generate compatible reference trajectories. These trajectories are designed to help the user maintain ground contact during the support phase while ensuring safe toe clearance and minimizing jerk during the swing phase. Finally, the AAFO's performance was assessed in real-world application conditions, with four healthy participants walking with the TWIN lower limb exoskeleton. The results suggest that the proposed AAFO efficiently reduces toe clearance, ensures stable control, and maintains low power consumption, highlighting its suitability for clinical applications.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1647989"},"PeriodicalIF":3.0,"publicationDate":"2025-08-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12399533/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144993972","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":"Train your robot in AR: insights and challenges for humans and robots in continual teaching and learning.","authors":"Anna Belardinelli, Chao Wang, Daniel Tanneberg, Stephan Hasler, Michael Gienger","doi":"10.3389/frobt.2025.1605652","DOIUrl":"10.3389/frobt.2025.1605652","url":null,"abstract":"<p><p>Supportive robots that can be deployed in our homes will need to be understandable, operable, and teachable by non-expert users. This calls for an intuitive Human-Robot Interaction approach that is also safe and sustainable in the long term. Still, few studies have looked at interactive task learning in repeated, unscripted interactions within loosely supervised settings. In such cases the robot should incrementally learn from the user and consequentially expand its knowledge and abilities, a feature which presents the challenge of designing robots that interact and learn in real time. Here, we present a robotic system capable of continual learning from interaction, generalizing learned skills, and planning task execution based on the received training. We were interested in how interacting with such a system would impact the user experience and understanding. In an exploratory study, we assessed such dynamics with participants free to teach the robot simple tasks in Augmented Reality without supervision. Participants could access AR glasses spontaneously in a shared space and demonstrate physical skills in a virtual kitchen scene. A holographic robot gave feedback on its understanding and, after the demonstration, could ask questions to generalize the acquired task knowledge. The robot learned the semantic effects of the demonstrated actions and, upon request, could reproduce those on observed or novel objects through generalization. The results show that the users found the system engaging, understandable, and trustworthy, but with larger variance on the last two constructs. Participants who explored the scene more were able to expand the robot's knowledge more effectively, and those who felt they understood the robot better were also more trusting toward it. No significant variation in the user experience or their teaching behavior was found across two interactions, yet the low return rate and free-form comments hint at critical lessons for interactive learning systems.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1605652"},"PeriodicalIF":3.0,"publicationDate":"2025-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12381527/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974414","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":"AUSPEX: An integrated open-source decision-making framework for UAVs in rescue missions.","authors":"Björn Döschl, Kai Sommer, Jane Jean Kiam","doi":"10.3389/frobt.2025.1583479","DOIUrl":"10.3389/frobt.2025.1583479","url":null,"abstract":"<p><p>Unmanned aerial vehicles (UAVs) have become paramount for search and rescue (SAR) missions due to their ability to access hazardous and challenging environments and to rapidly provide cost-effective aerial situational awareness. Nevertheless, current UAV systems are designed for specific tasks, often focusing on benchmarking use cases. Therefore, they offer limited adaptability for the diverse decision-making demands of SAR missions. Furthermore, commercially available integrated UAV systems are non-open-source, preventing further extension with state-of-the-art decision-making algorithms. In this paper, we introduce Automated Unmanned Aerial Swarm System for Planning and EXecution (AUSPEX), which is a holistic, modular, and open-source framework tailored specifically for enhancing the decision-making capabilities of UAV systems. AUSPEX integrates diverse capabilities for knowledge representation, perception, planning, and execution with state-of-the-art decision-making algorithms. Additionally, AUSPEX considers the heterogeneity of available UAV platforms and offers the possibility of including off-the-shelf and generic UAVs, with an open architecture into the AUSPEX ecosystem. The framework relies only on open-source components to ensure transparency, as well as system scalability and extensibility. We demonstrate AUSPEX's integration with the Unreal Engine-based simulation framework REAP for software-in-the-loop validation and a platform-independent graphical user interface (AUGUR). We demonstrate how AUSPEX can be used for generic scenarios in SAR missions while highlighting its potential for future extensibility.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1583479"},"PeriodicalIF":3.0,"publicationDate":"2025-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12378040/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974374","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":"The robot that stayed: understanding how children and families engage with a retired social robot.","authors":"Zhao Zhao, Rhonda McEwen","doi":"10.3389/frobt.2025.1628089","DOIUrl":"10.3389/frobt.2025.1628089","url":null,"abstract":"<p><strong>Introduction: </strong>Social robots are increasingly introduced into homes as short-term educational or entertainment tools for children. However, their physical presence and social roles may persist long after their intended use has ended. This study explores how families continue to engage with a child-focused educational robot years after its original deployment.</p><p><strong>Methods: </strong>We conducted a retrospective follow-up study with 19 families who participated in a 2021 in-home deployment of a reading companion robot for preschool-aged children. In 2025, we revisited these families through in-depth interviews to investigate how the robot had been integrated, re-purposed, or preserved over time.</p><p><strong>Results: </strong>Despite the children outgrowing the robot's instructional content, 18 families had retained the robot. Families described transitions in its role-from an educational device to a symbolic household member-characterized by emotional attachment, care-taking behaviors, and affection. The robot was re-framed as a memory object, integrated into new routines, or passed on ceremonially, akin to a \"retirement.\"</p><p><strong>Discussion: </strong>Our findings reveal three key themes explaining the robot's enduring presence: (1) emotional attachment and personification, (2) symbolic and nostalgic value, and (3) practical re-purposing within household routines. This study contributes to long-term human-robot interaction literature by extending domestication theory and emphasizing the importance of designing for the full life cycle of social robots-including end-of-life transitions. It underscores how social robots can become meaningful companions and enduring artifacts of family identity, long after their functional use has ended.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1628089"},"PeriodicalIF":3.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12370748/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974360","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":"Editorial: Advancing soft, tactile, and haptic technologies: recent developments for healthcare applications.","authors":"Antonia Tzemanaki, Benjamin Ward-Cherrier, Matteo Cianchetti, Jelizaveta Konstantinova","doi":"10.3389/frobt.2025.1658613","DOIUrl":"10.3389/frobt.2025.1658613","url":null,"abstract":"","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1658613"},"PeriodicalIF":3.0,"publicationDate":"2025-08-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12371425/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974371","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":"Editorial: Human-robot interaction in industrial settings: new challenges and opportunities.","authors":"Sara Bouraine, Mehdi Ammi, Kurt Geihs, Abdelfetah Hentout, Abderraouf Maoudj, Fouad Yacef","doi":"10.3389/frobt.2025.1652426","DOIUrl":"10.3389/frobt.2025.1652426","url":null,"abstract":"","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1652426"},"PeriodicalIF":3.0,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12360944/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144974397","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":"Towards fluid human-agent collaboration: From dynamic collaboration patterns to models of theory of mind reasoning.","authors":"Florian Schröder, Fabian Heinrich, Stefan Kopp","doi":"10.3389/frobt.2025.1532693","DOIUrl":"10.3389/frobt.2025.1532693","url":null,"abstract":"<p><p>Collaborating in real-life situations rarely follows predefined roles or plans, but is established on the fly and flexibly coordinated by the interacting agents. We introduce the notion of fluid collaboration (FC), marked by frequent changes of the tasks partners assume or the resources they consume in response to varying requirements or affordances of the environment, tasks, or other agents. FC thus necessitates dynamic, action-oriented Theory of Mind reasoning to enable agents to continuously infer and adapt to others' intentions and beliefs in real-time. In this paper, we discuss how FC can be enabled in human-agent collaboration. We introduce Cooperative Cuisine, an interactive environment inspired by the game <i>Overcooked!</i> that facilitates human-human and human-agent collaboration in dynamic settings. We report results of an empirical study on human-human collaboration in CoCu, showing how FC can be measured empirically and that humans naturally engage in dynamically established collaboration patterns with minimal explicit communication and relying on efficient mentalizing. We then present an approach to develop artificial agents that can effectively participate in FC. Specifically, we argue for a model of dynamic mentalizing under computational constraints and integrated with action planning. We present first steps in this direction by addressing resource-rational and action-driven ToM reasoning.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1532693"},"PeriodicalIF":3.0,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12353729/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144875991","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":"Intelligent maneuver decision-making for UAVs using the TD3-LSTM reinforcement learning algorithm under uncertain information.","authors":"Tongle Zhou, Ziyi Liu, Wenxiao Jin, Zengliang Han","doi":"10.3389/frobt.2025.1645927","DOIUrl":"10.3389/frobt.2025.1645927","url":null,"abstract":"<p><p>Aiming to address the complexity and uncertainty of unmanned aerial vehicle (UAV) aerial confrontation, a twin delayed deep deterministic policy gradient (TD3)-long short-term memory (LSTM) reinforcement learning-based intelligent maneuver decision-making method is developed in this paper. A victory/defeat adjudication model is established, considering the operational capability of UAVs based on an aerial confrontation scenario and the 3-degree-of-freedom (3-DOF) UAV model. For the purpose of assisting UAVs in making maneuvering decisions in continuous action space, a model-driven state transition update mechanism is designed. The uncertainty is represented using the Wasserstein distance and memory nominal distribution methods to estimate the detection noise of the target. On the basis of TD3, an LSTM network is utilized to extract features from high-dimensional aerial confrontation situations with uncertainty. The effectiveness of the proposed method is verified by conducting four different aerial confrontation simulation experiments.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1645927"},"PeriodicalIF":3.0,"publicationDate":"2025-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12355034/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144875989","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":"Burrowing and unburrowing in submerged granular media through fluidization and shape-change.","authors":"Aniruddha Nayak, Hoseung Seo, Nick Gravish, Michael T Tolley","doi":"10.3389/frobt.2025.1546407","DOIUrl":"10.3389/frobt.2025.1546407","url":null,"abstract":"<p><p>Subterranean exploration in submerged granular media (GM) presents significant challenges for robotic systems due to high drag forces and the complex physics of GM. This paper introduces a robotic system that combines water-jet-based fluidization for self-burrowing in submerged environments and an untethered, volume-change mechanism for burrowing out. The water-based fluidization approach significantly reduces drag on the robot, allowing it to burrow into GM with minimal force. To burrow out, the robot uses a soft, inflatable bladder that undergoes periodic radial expansion, inspired by natural systems such as razor clams. Experimental results demonstrate that increased water flow rates accelerate the burrowing process, while the unburrowing mechanism is effective at varying depths. Comparisons between pneumatic and hydraulic untethered systems highlight trade-offs in terms of operational time and unburrowing speed. This work advances the capabilities of robots in underwater environments, with potential applications in environmental monitoring and underwater archaeology.</p>","PeriodicalId":47597,"journal":{"name":"Frontiers in Robotics and AI","volume":"12 ","pages":"1546407"},"PeriodicalIF":3.0,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12351326/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144875988","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}