AI-Enabled Underwater Robotics: Current Status, Challenges, and Future Outlook with an Embodied Intelligence Orientation
Zichen Guo, Yanjun Liu, Shuxue Yan, Jian Wang, Qingqing Jiang, Yuhang Zhang, Xiaohui Zhang
Shenyang Institute of Automation Chinese Academy of Sciences
阅读操作
确认中在文库中上传 PDF 后可生成中文音频讲解。
摘要与影响
Driven by growing demands in maritime surveillance, resource exploration, and extreme-environment operations, underwater robots are evolving from conventional automated platforms toward highly autonomous intelligent systems. Unlike terrestrial robots, underwater vehicles operate in complex marine environments characterized by strong hydrodynamic coupling, perceptual degradation, limited communication, and severe resource constraints. Under such conditions, intelligent behavior cannot be fully understood as the outcome of isolated algorithmic modules, but should instead be viewed as the result of continuous closed-loop interactions among the robot body, the environment, and control mechanisms. In this sense, embodied intelligence provides a suitable theoretical framework for understanding the autonomous evolution of underwater robotics. This paper presents a systematic review of AI technologies for underwater robots from an embodied intelligence perspective. It first clarifies the integration logic between underwater robotics and embodied intelligence, and then reviews recent advances across four core functional domains: perception and environment modeling, state monitoring and fault diagnosis, control and guidance, and task planning and decision-making. The paper further discusses extended application scenarios, including multi-robot coordination and intelligent support for mission execution. Finally, it analyzes the main challenges in realizing reliable underwater embodied intelligence, including sensory degradation, data scarcity, embedded resource limitations, and communication bottlenecks, and outlines several future directions such as physics-informed learning, low-power onboard intelligence, cross-modal fusion, and standardized open ecosystems. This review aims to provide a structured reference for the development of underwater robots toward system-level embodied intelligence.
逐年被引趋势
暂无年度引用数据
关键指标
同类平均 = 1
同领域 · 同年份 · 同类型
Google Scholar 与 OpenAlex 的被引统计范围不同,数值存在差异属正常。
AI 辅助阅读
依据:摘要
可就本文提问;依据不足时会说明。
学术脉络
学科主题
工程Underwater Vehicles and Communication Systems
Maritime Navigation and Safety · Water Quality Monitoring Technologies