Ensuring Safety and Reliability in Intelligent Autonomous Systems Through Verification and Certification

Artificial Intelligence (AI) is reshaping the field of autonomous systems by enabling intelligent agents that can perceive, reason, and act in dynamic and uncertain environments. These systems are central to domains such as healthcare, robotics, transportation, and smart infrastructure, where they interact with and control complex physical processes through integrated cyber-physical architectures. Supported by cloud-based technologies, autonomous systems can scale effectively, coordinate actions across distributed environments, and process massive volumes of data in real time. However, as adoption accelerates, new challenges arise in ensuring that such systems behave in ways that are safe, reliable, and aligned with human and societal intentions. The unpredictability of data-driven models, the opacity of deep learning systems, and the increasing interconnectedness of AI-driven applications demand new strategies to build confidence and trust in their operation.Recent advances have introduced verification and validation techniques that combine formal methods with AI-driven analytics to enhance assurance. Model checking, runtime monitoring, and simulation-based testing now form critical foundations for detecting errors and guaranteeing compliance with safety and ethical standards. Complementary innovations such as federated learning, blockchain-based trust mechanisms, and cloud-enabled digital twins further enrich system capabilities but also introduce new security, privacy, and scalability concerns. Despite progress, major gaps persist in the formal verification of learning-enabled systems, cross-domain certification, and the standardization of safety frameworks across distributed AI environments. Therefore, there is an urgent need for unified, scalable, and practical approaches that can ensure intelligent agents operate robustly under variable and adversarial conditions.This Research Topic aims to promote the development of trustworthy, certifiable, and reliable AI-enabled autonomous systems that can perform safely in large-scale, cloud-integrated environments. The primary objectives are to bridge the gap between adaptive learning capabilities and formal assurance requirements by advancing methods for verification, certification, and runtime validation. Key goals include improving frameworks for system-level safety, developing methods for analyzing uncertainty in AI behaviors, and integrating control, modeling, and formal verification perspectives. This collaborative effort will unify expertise from AI research, software engineering, automation, and systems verification to establish shared tools, principles, and metrics for building resilient intelligent systems that interact seamlessly with both digital and physical worlds.The scope of this Research Topic encompasses theoretical models, applied methodologies, and cross-domain practices for verifying and certifying intelligent autonomous systems. It focuses on frameworks that ensure correctness, security, and reliability in cloud-enabled, distributed, and cyber-physical contexts.To gather further insights into these pressing challenges and opportunities, we welcome articles addressing, but not limited to, the following themes:• Verification and certification of Agentic AI and autonomous systems• Formal methods and hybrid system modeling for cyber-physical and control systems• Neural network verification, robustness analysis, and safety guarantees• Reachability analysis and formal safety assurance of learning-enabled systems• Simulation, validation, and runtime monitoring of system behaviors• Integration of verification techniques with AI and control-driven decision-making• Verification of distributed and cloud-based AI systems, including federated learning and digital twins• Privacy, security, and trust mechanisms, including blockchain-enabled systems and access control• Anomaly detection, reliability analysis, and secure operation in intelligent systems• Verification challenges in communication systems, including 6G and semantic communication• AI-driven optimization and resource management with correctness guarantees• AI for quantum computing and hybrid classical–quantum systems• Standardization, benchmarking, and tool development for verification and certification Topic editors: Ratan Lal, Dr. Rabindra K. Barik, Amrendra Singh Yadav Source: https://www.frontiersin.org/research-topics/80817/ensuring-safety-and-reliability-in-intelligent-autonomous-systems-through-verification-and-certification