AI at the network edge promises real-time perception and decision-making in safety-critical domains such as aerial robotics, autonomous vehicles, and 5G-enabled infrastructures. Yet, operating under resource constraints, dynamic, and adversarial conditions exposes edge AI systems to fragility, inefficiency, and security risks that threaten their safe operation. GuardAI, a Horizon Europe project, introduces a framework for resilient and trustworthy edge AI that unites three pillars: adversarial robustness, context-enhanced inference, and security-by-design. Initial project results include a diffusion-based adversarial purification framework optimized for real-time operation, lightweight deep unrolling architectures for LiDAR super-resolution with built-in outlier removal, and robust uncertainty quantification modules to improve confidence calibration. It further develops a context-enhanced inference engine that integrates visual, spatial, and operational context across multi-agent systems, and a risk-aware defense recommender that autonomously selects mitigation strategies based on evolving threat landscapes. Through representative Use Cases, covering monitoring with Unmanned Aerial Vehicle, decentralized 5G network analytics, and secure perception in connected autonomous vehicles, GuardAI demonstrates how robust and adaptive AI can be achieved within stringent edge constraints. Together, these technologies lay the groundwork for a new generation of secure, context-aware, and certifiable AI systems that can be trusted to operate autonomously in the physical world.

Unmanned Aerial Vehicles (UAVs) have become essential in search and rescue operations, especially in disaster management scenarios. Their effective navigation and the integration of a plethora of sensors assist in efficient person detection, making them an essential technological tool to first responders. Multi-UAV systems extend these benefits by using coordinated strategies to cover large areas efficiently, reducing overall mission response time and enhancing its success. Despite these advantages, challenges remain in ensuring the safety, security, and dependability of (mutli-)UAV missions. Issues such as navigation risks, potential cyber threats, and hardware-/software-related reliability issues can impact the mission results. Additionally, UAVs are highly constrained devices with limited battery capacity, requiring the use of lightweight technologies. In this paper, we present part of the results of the SESAME project, an EU multi-partner project that aims to develop safe and secure multi-robot Systems. In particular, we present some of the developed SESAME Executable Digital Dependability Identities (EDDI) technologies based on Markov models, statistical distance measures, and other advanced approaches for enhancing safety, security and dependability of the UAV platform and underlying models. These EDDI technologies are seamlessly integrated using the ConSerts framework in a multi-UAV platform and tested using search and rescue scenarios. The results demonstrate significant improvements in multi-UAV safety, with an availability rate of 91% and a search and rescue algorithmic accuracy of 99.8%. Additionally, the system achieves precise detection of spoofing attacks, using collaborative localization as a mitigation technique to guide the UAV to a safe landing, even in the absence of GPS signals,