AI-driven swarm USV operations: A comprehensive bibliometric and analytical review

The rapid advancement of unmanned surface vehicles (USVs) across various domains has triggered increasing research interest, particularly in swarm configurations empowered by artificial intelligence (AI). Despite its rapid development comprehensive bibliometric and analytical reviews focusing specifically on AI-driven swarm USV operations remain limited. To address this gap, this study provides an in-depth bibliometric and analytical review of AI-based approaches for swarm USV operations. Drawing from 254 selected publications indexed in Scopus and Web of Science between 2000 and 2025, the analysis reveals evolving trends, research hotspots, and collaborative patterns with China leading research output (357 total contributions) and formation control emerging as the dominant research focus (77 occurrences). The study classifies swarm USV tasks, highlights dominant AI methodologies (including classical control, intelligent control, nature-based algorithms, machine learning, deep learning, and reinforcement learning) and evaluates their implementation readiness. The paper identifies critical gaps including a significant simulation-reality divide (85.4% simulation-only studies), limited sensor integration frameworks, underdeveloped communication resilience protocols, and insufficient real-world validation standards. The paper concludes with a summary of future directions, highlighting emerging trends, shifting research priorities, and their impact on swarm USV operations. The results provide a roadmap for bridging the simulation-reality gap, developing standardized testing protocols, and establishing regulatory frameworks essential for practical swarm USV deployment in maritime applications.