The Conference at Intelligent Robotics FAIR 2025 will feature multiple tracks showcasing cutting-edge research in robotics, field robotics, ethology-inspired systems, and intelligent technologies. Renowned researchers and industry experts will present breakthrough findings in keynote speeches and presentations, discussing the challenges and opportunities shaping the future.

Naoyuki Kubota

Tokyo Metropolitan University

Talk: Multi-scopic Topological Intelligence for Robotics
Recently, various types of robots have been developed such as mobile robots, robot partners, and mobility support robots. With the rapid progress of sensing technology, network technology, and machine learning technology, we can realize real-time measurement of people and moving objects in human-robot coexisting environments, but we need cyber-physical system integration and analysis. In order to perform multiscale and multiphysics simulations of a real-world phenomenon, we should extract and use topological features and structures from given or measured big data. Therefore, we proposed the concept of topological twin. The goal of topological twin is to (1) extract topological structures implicitly hidden in the real world, (2) explicitly reproduce them in the cyber world, and (3) simulate and analyze real-world phenomena in the cyber world. While we have to deal with the physical dynamics at the microscopic level, we have to deal with spatiotemporal qualitative relationships between objects, people, culture, and knowledge at the macroscopic level. We also need a mesoscopic integration method connecting microscopic and macroscopic topological features. Thus, the topological twin plays an important role in extracting and connecting hidden structures in real world from the multiscopic point of view. Furthermore, we need a multiscopic approach to deal with inference, learning, search, and prediction based on topological and graphical data as a methodology of topological intelligence. In this talk, we first introduce the concept of multiscopic topological twin. Next, we explain various types of topological clustering methods and graph-based methods related to topological intelligence. One of them is Growing Neural Gas (GNG), and we have proposed multi-layer GNG (ML-GNG) to reduce the computational cost of the original GNG and multi-scale batch-learning GNG to improve the convergence property and dynamic adaptation property. Next, I show experimental results of multiscopic topological intelligence for trailer living laboratory, robot partners, multi-legged robots, and mobility support robots. Finally, I discuss the applicability and future directions of multiscopic topological intelligence.

Bio: Naoyuki Kubota is currently a professor in the Department of Mechanical Systems Engineering, the Graduate School of Systems Design, and director of the Community-centric Systems Research Center, Tokyo Metropolitan University, Japan. He is the representative director of the Tokyo Biomarker Innovation Research Association, Japan. He received his Ph.D. degree from Nagoya University, Japan, in 1997. He was a Visiting Professor at University of Portsmouth, UK, Seoul National University, and others. His current interests are in the areas of topological intelligence, coevolutionary computation, spiking neural networks, robot partners, and informationally structured space. He has published more than 600 peer-reviewed journal and conference papers in the above research areas. He received the Best Paper Award of IEEE IECON 1996, IEEE CIRA 1997, and so on. He was an associate editor of the IEEE Transactions on Fuzzy Systems from 1999 to 2010, Editorial Board Member of Advanced Robotics from 2004 to 2007, Editorial Board Member of Journal of Advanced Computational Intelligence and Intelligent Informatics since 2004, the IEEE CIS Intelligent Systems Applications Technical Committee, Robotics Task Force Chair from 2007 to 2014, Editor of ROBOMECH Journal since 2012, IEEE Systems, Man, and Cybernetics Society, Japan Chapter Chair from 2018 to 2021, IEEE Transactions on Affective Computing Steering Committee Member since 2019, and others.

Péter Korondi

University of Debrecen

Talk: Organic Human Robot Interaction (O-HRI)
We often hear that a robot cannot have human feelings and a human soul. Of course, this is true, it cannot be human. Even if we could program human emotions, we should not clone humans either biologically or technically. Nevertheless, we expect to be able to converse with robots in a human-like manner, especially since the advent of ChatGPT and large language models. The contradiction can only be resolved if we define robot feelings and a robot-type soul. For verbal conversation between humans and robots, we should develop robot-type nonverbal metacommunication channels. These can be audio-visual signals, small elementary movements that project the intention, the planned movement. Previously, industrial robots worked in a physically or laser-enclosed room, where only robot specialists could enter. Nowadays, with the advent of cobots, industrial robots have also been released from their cages, but with the rise of service and especially social robots, robots have stepped out of the gates of industrial plants. Robots are increasingly encountering people who are not robotics experts, who have no experience in how to communicate with a robot. The presentation focuses on how we can make communication with robots natural (organic) for non-robotics experts. The presentation introduces the basic concept of organic human-robot interaction and its implemented elements. An HRI can be considered an O-HRI if the robot is capable of verbal and non-verbal communication and has the appropriate psychological characteristics for human-human interaction.
The main elements of O-HRI are: Mistakes and Miscalculations in Behaviour; Imperfection in Behaviour; Showing intentions;Non-verbal meta communication and body language; Flow experience; Perceived competencies; Acceptance; Attachment

Bio: Péter Korondi is a professor at the Department of Electrical Engineering and Mechatronics at the University of Debrecen. He is the head of the cognitive and social robotics group. He received his MSc. in Electrical Engineering in 1984. Between 1993 and 1995, he became acquainted with mechatronics and robotics at the Intelligent Mechatronics Laboratory of Professor Hideki Hashimoto at the University of Tokyo. In the early 2000s, he began working with ethologists on eth-robotics. Their main achievement was MOGI Robi, a robot attached to its owner. After the release of ChatGPT, his interest turned to Organic Human-Robot Interaction.
He is an active volunteer of IEEE and IFAC, and was the chairman of the technical committee related to robotics in both organizations for two terms. He organized several IEEE and IFAC conferences as the general chairman, and was the chairman of the robotics track in several cases.
He has given keynote presentations at several IEEE and IFAC conferences on intelligent space and cognitive and ethological robotics.

Gabriella Lakatos

University of hetfordshire

Talk: Credible social robots: From ethologically inspired companion robots to socially acceptable educational robots
The development of credible social robots capable of meaningful and acceptable interaction with humans presents a significant ongoing challenge. This keynote will explore two key approaches. Firstly, I will introduce the approach of ethorobotics - a field leveraging animal behaviour to design socially acceptable robots - examining how principles from animal behaviour, particularly human-dog interaction, can inform the design of companion robots. I will consider principles for designing social robots by aligning robot appearance, behaviours, and functionality with user expectations. I will also introduce a new typology for biologically inspired non-verbal communicative signals for robots to enhance robot transparency and believability. I will highlight how these biologically-grounded approaches contribute to robots being perceived as more natural and trustworthy companions. Secondly, the presentation will transition to the topic of socially credible educational and therapeutic robots, with a particular focus on the humanoid robot Kaspar. I will discuss Kaspar's application in real-world settings to support children, including those with autism spectrum disorder and learning disabilities, in developing crucial skills such as visual perspective-taking, and speech, language, and communication skills. While Kaspar’s design is following a very different approach, its success in fostering engagement and achieving therapeutic goals further demonstrates the importance of social acceptability. By examining these distinct approaches, this keynote will highlight the complex journey towards creating credible social robots for diverse human needs.

Bio: Dr Gabriella Lakatos is a Reader (Associate Professor) in Social- and Ethorobotics at the School of Physics, Engineering and Computer Science at the University of Hertfordshire, United Kingdom, and a member of the Robotics Research Group at the same university. She completed her PhD in the field of Ethology at the Eotvos Lorand University, Hungary, after which she transferred her skills to the field of Etho-robotics and Human-Robot Interactions. She has extensive experience in the fields of human-robot interactions and robot-assisted therapy. Her interest lies in biologically inspired acceptable and credible robot behaviour design, and in trust in social and assistive robotics. She gained her experience in multiple research projects funded by Horizon Europe, EPSRC, UKRI: TAS & UK-RAS, the Baily Thomas Charitable Fund, and the Dinwoodie Charitable Company. Gabriella is an editorial board member of two international journals. She has been co-organiser of several iterations of the Trust, Acceptance and Social Cues in Human-Robot Interaction (SCRITA) workshop at the RO-MAN conference and a guest editor of special issues on the same topic in two international journals.

Eszter Salamon

Parents International

Talk: Do we still need live teachers in the age of AI and robots?
The world was facing a global learning crisis (World Bank 2018.) even before the school closures of 2020-22 that has a number of surprising, but shocking characteristics. It is not only about children with no access to school anymore, but about those who do attend formal education, even receive some kind of school leaving certification, but do not acquire basic literacy and numeracy skills, not to mention other skills necessary for the 21st century. School has little to do with real life which is a multifaceted phenomenon. First of all, school curricula are often overcrowded with skills and academic content that is outdated and without consensus on why they are necessary to teach and learn (see the saber-toothed tiger parable). School is also often sheltered from the outside world meaning that it provides little support and skills development in the field of everyday life situations – present and future – especially for those whose parents are less able to provide such necessary education at home struggling with aspects of everyday life themselves. This has been made worse by more and more school systems trying to ban the most widely accessible digital tool, the mobile phone. At the same time, a concerningly high percentage of families are fully opting out of the formal education system. There is a need for change. In Parents International, an independent research organisation, we have explored and mapped beliefs about education in the process of determining what needs to change in a system. Mapping beliefs and making families contributions to education systems visible is one of the  threads throughout the research we have done in multiple projects in the past 3-4 years. We have extensively researched the situation in Hungary and the Netherlands. This keynote will introduce our results exploring how school systems should become more adjusted to the needs of children of today and the world of tomorrow. We have also explored how the contribution of companies can be evaluated against SDG4 and how collaboration between companies and schools can become more organic to ensure that modern technology is present in schools alongside teachers.

Bio: Eszter Salamon is the Director of Parents International and an inclusion and research ethics consultant for the European School Heads Association. Currently she is also a visiting professor at Vytautas Magnus University in Lithuania. Originally trained as a teacher, she started her career as a teacher at a primary cum secondary school. Later she became an economist and has been teaching in higher and vocational education. She started dealing with rights of the child topics in 1989 and has specialized in students’ rights and parents’ rights in the past 25+ years, being active on European level since 2009 and international level since 2015. She has earned her Doctorate in education leadership focusing on parental empowerment and engagement, and child participation. Her activities and research interests are related to 21st century education, school leadership, parental engagement, quality and inclusion, education-related rights and accountability, and collaboration. Her recent publications include research on education adjusting to the realities of the digital age, teacher and school leader roles and careers, open schooling, innovative and inclusive STE(A)M education, family support systems, and parents’ perceptions of and engagement in education. Her team has developed globally acknowledged practical initiatives on parental engagement, child participation, open schooling, education inclusion. She has been invited to participate at expert groups at the Council of Europe and OECD. In the field of parental engagement, her main area of activities is related to professionals’ perceptions and capacity-building needs for better collaboration. Her work is framed by her experience as a teacher, as a parents, as a school board president as well as a researcher and trainer.

 

• Roundtable talk details comming soon!