Enhancing Autonomous System Security and Resilience With Generative AI: A Comprehensive Survey

This is the authors' abstract. We don't add key points for this paper.

This survey explores the transformative role of Generative Artificial Intelligence (GenAI) in enhancing the trustworthiness, reliability, and security of autonomous systems such as Unmanned Aerial Vehicles (UAVs), self-driving cars, and robotic arms. As edge robots become increasingly integrated into daily life and critical infrastructure, the complexity and connectivity of these systems introduce formidable challenges in ensuring security, resilience, and safety. GenAI advances from mere data interpretation to autonomously generating new data, proving critical in complex, context-aware environments like edge robotics. Our survey delves into the impact of GenAI technologies—including Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), Transformer-based models, and Large Language Models (LLMs)—on cybersecurity, decision-making, and the development of resilient architectures. We categorize existing research to highlight how these technologies address operational challenges and innovate predictive maintenance, anomaly detection, and adaptive threat response. Our comprehensive analysis distinguishes this work from existing reviews by mapping out the applications, challenges, and technological advancements of GenAI and their impact on creating secure frameworks for autonomous systems. We discuss significant challenges and future directions for integrating these technologies within security frameworks to address the evolving landscape of cyber-physical threats, underscoring the potential of GenAI to make autonomous systems more adaptive, secure, and efficient.