scieee AI-readable full text Open interactive document viewer

Italian Workshop on Artificial Intelligence and Robotics (AIRO2025) Focus Theme: Planning and Learning for Autonomous Robotics

Finzi, Alberto; Beraldo, Gloria; Castellini, Alberto; Patrizi, Fabio

Abstract

The AIRO workshop, organized by the Working Group on Artificial Intelligence and Robotics of the Italian Association for Artificial Intelligence (AIxIA), provides a forum to discuss the integration of AI techniques in autonomous robotics. The 2025 I-RIM edition of the workshop will focus on the synergistic combination of planning and learning, which is increasingly recognized as a key enabler for advanced robot autonomy. These techniques are crucial across several domains of robotics and are pervasively exploited at multiple levels of robotic architectures — from high-level decision making and task planning to low-level control and adaptive behaviors. As robots increasingly move beyond structured lab settings and into open-world applications, the ability to combine high-level planning mechanisms with effective methods for learning from experience and demonstration becomes essential. The combination of planning and learning is instrumental in shaping robot intelligence, enabling robots to plan complex action sequences, learn new tasks, and continuously adapt to dynamic environments. The workshop aims to bring together researchers working at this intersection, with contributions in (but not limited to) the following areas: - Decision-theoretic and symbolic planning for robotics - Learning-based control and adaptive behaviors - Integration of learning and task planning in robot architectures - Task and motion planning with learning components - Task Teaching and Learning from Demonstrations - Reinforcement learning, imitation learning, and online adaptation - Applications in real-world scenarios: HRI, service robotics, autonomous exploration In continuity with previous editions, AIRO2025 will also showcase contributions from the broader AI & Robotics community in Italy, particularly those involved in the PNRR MUR FAIR project (Future AI Research), to encourage national collaborations and cross-contamination between research domains.

Full text

1. Workshop Title: Italian Workshop on Artificial Intelligence and Robotics (AIRO2025) Focus Theme: Planning and Learning for Autonomous Robotics 2. Organizers Alberto Finzi, Corresponding Organizer, University of Naples “Federico II”, Italy, [email protected] Gloria Beraldo, National Research Council, CNR, Rome, Italy, [email protected] Alberto Castellini, University of Verona, Italy, [email protected] Fabio Patrizi, Sapienza University, Rome, Italy, [email protected] 3. Workshop Format: Both invitation and open for submission 4. Workshop Objectives and Main Topics: The AIRO workshop, organized by the Working Group on Artificial Intelligence and Robotics of the Italian Association for Artificial Intelligence (AIxIA), provides a forum to discuss the integration of AI techniques in autonomous robotics. The 2025 I-RIM edition of the workshop will focus on the synergistic combination of planning and learning, which is increasingly recognized as a key enabler for advanced robot autonomy. These techniques are crucial across several domains of robotics and are pervasively exploited at multiple levels of robotic architectures — from high-level decision making and task planning to low-level control and adaptive behaviors. As robots increasingly move beyond structured lab settings and into open-world applications, the ability to combine high-level planning mechanisms with effective methods for learning from experience and demonstration becomes essential. The combination of planning and learning is instrumental in shaping robot intelligence, enabling robots to plan complex action sequences, learn new tasks, and continuously adapt to dynamic environments. The workshop aims to bring together researchers working at this intersection, with contributions in (but not limited to) the following areas: ● Decision-theoretic and symbolic planning for robotics ● Learning-based control and adaptive behaviors ● Integration of learning and task planning in robot architectures ● Task and motion planning with learning components ● Task Teaching and Learning from Demonstrations ● Reinforcement learning, imitation learning, and online adaptation ● Applications in real-world scenarios: HRI, service robotics, autonomous exploration In continuity with previous editions, AIRO2025 will also showcase contributions from the broader AI & Robotics community in Italy, particularly those involved in the PNRR MUR FAIR project (Future AI Research), to encourage national collaborations and cross-contamination between research domains. 5. Keywords 2025 I-RIM Conference October 17-19, Rome, Italy ISBN: 9788894580570 10.5281/zenodo.17629630 37 Task and Motion Planning in Robotics, Reinforcement Learning for Robot Autonomy, Learning from Demonstration and Imitation, Collaborative Task Teaching and Execution, Neuro-Symbolic Integration, Generative AI for Planning and Learning, Cognitive and Developmental Robotics, Robot Software Architectures 6. Preferred Date: 17 October 7. List of Speakers The list of speakers will be determined after the distribution of the call for contributions. We aim to encourage participation from individuals involved in current National PNRR MUR projects related to AI & Robotics, such as FAIR. The invited speakers will be also selected from the authors that have submitted their work to the ongoing special issue “Planning and Learning for Autonomous Robotics" in the Journal of Robotics and Autonomous System, such as Francesco Amigoni, Politecnico Milano, [email protected] (to be confirmed) Alessandro Farinelli, University of Verona, [email protected], Luca Iocchi, Sapienza University, [email protected]iroma1.it (to be confirmed). 8. Expected Number of Attendees 30 based on the previous editions 9. Target Audience ( E.g., sector experts, general public.) Sector experts from academia and industry 10. Workshop Outcomes (E.g., publications, videos, special issues, future events.) The workshop will be recorded, and the video will be published on the AIRO website (https://www.airo-aixia.it/home). The accepted submissions will be included in the IRIM proceedings. Future events include the next AIRO workshop series at the AIxIA conference. 11. Plan for Special Issue or Session The AIRO community is already curating the special issue "Planning and Learning for Autonomous Robotics" (Robotics and Autonomous Systems). Based on the outcomes of the workshop, we plan to evaluate a follow-up initiative to continue disseminating research results, either as an extension of the current issue or as a new publication effort. 12. Call for Extended Abstracts Once the workshop is confirmed, the call for contributions will be shared with potential speakers and distributed through various mailing lists related to AI & Robotics and current PNRR projects. Submission instructions will be posted on our website (https://www.airoaixia.it/home). 13. Workshop Webpage (Provide the link or planned location. This will be linked in the I-RIM program.) https://www.airo-aixia.it/home 14. Notes on Registration The two registration will be used or by the organizers or by external invited speaker 38