Edge AI and robotics are increasingly integrated in real-world systems that demand low-latency inference, efficient computation, and robust autonomy. However, deploying AI at the edge brings unique challenges, including limited computational resources, evolving regulatory constraints, and the need for trustworthy, safe, and explainable models. 

As Europe advances toward sovereign, human-centric AI systems, new approaches are required to ensure that intelligent agents operating at the edge remain reliable, sustainable, and aligned with ethical and legal frameworks such as the AI Act. 

This workshop brings together researchers from Edge AI, embedded learning, and robotics to explore the intersection of technical design and responsible deployment. It focuses on architectures and methods that enable privacy-preserving, bias-aware, and energy-efficient AI at the edge, while ensuring adaptability and safety in robotic applications. 

The event is jointly organized by four EU-funded Networks of Excellence (NoEs) - dAIEDGE, euROBIN, ELIAS, and ENFIELD - and aims to foster collaboration across these initiatives and the broader HiPEAC community

Submission Guidelines

Papers will be reviewed by the workshop's technical program committee according to criteria regarding a submission's quality, relevance to the workshop's topics, and, foremost, its potential to spark discussions about directions, insights, and solutions on the topics mentioned above. Research papers, case studies, position papers and work-in-progress are all welcome. 

Papers should be in double column IEEE format of between 4 and 8 pages including references. Papers should be uploaded as PDF and not anonymized. At least one of the authors must register at the conference and attend the workshop to present the paper if it is accepted. 

List of Topics

Architectures and Frameworks for Edge AI 

• Distributed and hierarchical AI architectures for robotic systems 
• Edge–cloud continuum design and orchestration 
• Lightweight neural networks and model compression for embedded devices 
• Hardware-aware AI acceleration (FPGA, RISC-V, neuromorphic, etc.) 
• Real-time inference and scheduling for autonomous robots 

Trustworthy and Explainable Edge Intelligence 

• Explainable and interpretable AI for robotic decision-making 
• Uncertainty estimation and safety assurance for edge perception and control 
• Robustness against adversarial conditions and sensor noise 
• Verification, validation, and certification of AI components at the edge 
• Compliance with EU regulatory frameworks (e.g., AI Act, Ethics Guidelines) 

Sustainable and Energy-Efficient AI 

• Low-power AI computation and green optimization techniques 
• Energy-aware learning and adaptive workloads for embedded systems 
• Sustainable lifecycle management of AI models and hardware 
• Metrics and benchmarks for energy efficiency in edge robotics 

Learning and Adaptation at the Edge 

• Continual, federated, and transfer learning for robotics 
• On-device and online learning strategies 
• Adaptive multi-modal perception for dynamic environments 
• Collaborative intelligence among distributed robotic agents 

Applications and Use Cases 

• Edge-enabled robotics for manufacturing, logistics, healthcare, and agriculture 
• Human–robot interaction and shared autonomy 
• Swarm and multi-robot coordination using distributed intelligence 
• Trustworthy AI for inspection, maintenance, and environmental monitoring 

Cross-Cutting Themes 

• Open datasets, benchmarks, and reproducible research for edge robotics 
• Security, privacy, and data governance in edge–robotic ecosystems 
• Integration of large or foundation models at the edge 
• Socio-technical, ethical, and legal dimensions of autonomous systems 

Organizing Committee

• Cecilia Zanazzo, Università di Trento 
• Georgios Kavallieratos, Norwegian University of Science and Technology (NTNU) 
• Lydia Belkadi, KU Leuven 
• Mohamed Selim, German Research Center for Artificial Intelligence (DFKI) 
• Rebecca Schedl-Warpup, German Aerospace Center (DLR)

Venue

The workshop will be held in HiPEAC 2026 Conference in Kraków, Poland.

Contact

All questions about submissions should be emailed to mohamed.selimdfki.de