RobOntics 2025: International Workshop on Ontologies for Autonomous Robots Royal Continental Hotel, Naples, Italy, December 10, 2025 |
| Conference website | https://robontics.github.io/ |
| Submission link | https://easychair.org/conferences/?conf=robontics2025 |
| Submission deadline | October 15, 2025 |
International Workshop on Ontologies for Autonomous Robotics (ROBONTICS) @ AIxRobotics (https://www.aixrobotics.org/), Naples, Italy (hybrid event -- online participation also possible)
We encourage researchers interested in the fields of robotics and knowledge engineering to submit research and position papers by *October 15th*.Researchers with accepted papers will be invited to present at the RobOntics 2025 workshop.
**WORKSHOP MOTIVATION**
RobOntics focuses on autonomous agents informed by knowledge-driven approaches, with particularemphasis on formal systems like ontologies, knowledge graphs, and their integration with neuralarchitectures. The workshop aims to foster interaction across robotics, ontology, knowledgerepresentation and reasoning, and neural computing to investigate promising cognitively-inspiredknowledge based architectures, neuro-symbolic approaches and review progress in knowledge-drivenrobotics.
Today, the convergence of symbolic knowledge representation and neural learning is reshapingrobotics and standardization efforts for intelligent robotic systems. Many open problems involveautonomous agents that must seamlessly integrate symbolic reasoning with neural perception andlearning while operating in natural, artificial, or socio-technical environments. Research projects inhealthcare assistance, logistics, autonomous driving, and human-robot collaboration increasinglyrequire robots that can both learn from experience and reason with explicit knowledge in realistichuman environments.
One of the key challenges is developing cognitive-inspired architectures that combine the flexibility ofneural learning with the interpretability and structure of symbolic knowledge. Robotic agents needreactive world models that can rapidly adapt to dynamic environments while maintaining coherentknowledge representations. Further, knowledge-driven and commonsense knowledge-driveninteraction capabilities must enable robots to understand, predict, and respond to human intentions andbehaviors in natural, contextually-appropriate ways. Such knowledge should be reusable acrossdifferent agents and scenarios while remaining accessible and modifiable by human operators.
To garner trust, ensure dependability, and enable effective human-robot collaboration, hybridintelligence systems must provide transparent explanations of their reasoning processes and offerintuitive interfaces for knowledge inspection and modification.
**Special Topic of RobOntics 2025**
This edition of RobOntics is particularly interested in three interconnected themes:
*Neuro-symbolic architectures for autonomous agents*: Exploring architectures that integrate neuralnetworks with symbolic reasoning systems, enabling robots to learn from data while maintaininginterpretable knowledge structures and logical reasoning capabilities.
*Cognitive-inspired world models for knowledge-driven reactive agents*: Investigating howcognitive principles can inform the development of dynamic world models that support rapid,knowledge-informed reactions to environmental changes while maintaining consistency with learnedand encoded knowledge.
*Knowledge-driven human-robot interaction*: Examining how ontological knowledge representationcan enhance robots' understanding of human behavior, intentions, and social contexts to enable morenatural, effective, and trustworthy human-robot collaboration.
**IMPORTANT DATES**
- Submission deadline: October 15th, 2025- Notification: November 7th, 2025- Camera ready version: November 28th, 2025- Workshop: TBA (December 8th-10th, 2025)
**LIST OF TOPICS (partial)**
Participants are invited to submit original papers (5-8 pages + references) on the topics of particular interest described above, but contributions are also welcome on topics such as:
- Foundational issues:
- Are there some ontological approaches better suited than others for autonomous robotics? why?
- How should we ontologically model notions like capability, action, interaction, context etc. in robotics?
- How can ontology be used to model culture, cultural knowledge and cultural behavior?
- Robustness:
- How can ontologies be used to help robots cope with the variety and relatively fluid structure of human environments?
- Ontologies in the perception-action loop:
- What roles can ontology play in autonomous manipulation?
- How can ontology be used to support machine learning for object classification?
- Interactivity:
- How can knowledge about other agents present in the environment be modelled?
- How can ontology be used to model the flow of an interaction, e.g., in the case of shared tasks?
- Normed behavior:
- How can we ontologically represent norms and cultural expectations?
- How can expectations be acquired? would they be the same for robots and for humans?
- Explainability:
- Decision chains are very complex; how can these be organized and presented at various levels of detail for the benefit of a human user?what is an explanation? what is a good explanation? how it be generated from a collection of knowledge items?
**WORKSHOP CO-CHAIRS (alphabetical order)**
- Daniel Beßler, Institute for Artificial Intelligence, University of Bremen, Germany
- Stefano De Giorgis, Department of Artificial Intelligence, Vrije Universiteit Amsterdam, Netherlands
- Mihai Pomarlan, Digital Media Lab, University of Bremen, Germany
- Nikos Tsiogkas, KU Leuven, Belgium
**SUBMISSION INFORMATION**
Beside regular papers, position and survey papers are also welcome. All the contributions to the workshop must be submitted according to the CEUR format (laTeX template: http://ceur-ws.org/Vol-XXX/CEURART.zip overleaf template: https://www.overleaf.com/latex/templates/template-for-submissions-to-ceur-workshop-proceedings-ceur-ws-dot-org/wqyfdgftmcfw). More information about CEUR style submissions is available at https://ceur-ws.org/HOWTOSUBMIT.html.
Papers will be refereed and accepted on the basis of their merit, originality, and relevance. Each paper will be reviewed by at least two Program Committee members. Papers must be submitted electronically in PDF, using this link:
- https://easychair.org/conferences/?conf=robontics2025
**PUBLICATION**
Accepted works will be published in an open access CEUR volume as part of the new IAOA series (see http://ceur-ws.org/iaoa.html).