Abstract: Human–machine shared navigation control for maritime autonomous surface ships (MASSs) integrate advanced technologies such as perception enhancement, intelligent decision-making, and autonomous control, forming a novel navigation control paradigm characterized by efficient cooperation between ship operators and autonomous navigation systems, dynamic authority allocation, and intuitive human-machine interaction. It has become a key technological direction for supporting the development of the next-generation intelligent maritime transportation system and enhancing maritime operational safety. Motivated by the practical application demands of MASSs, this paper focuses on the intrinsic characteristics of deep coupling among the core components of human–machine shared navigation control. From a full-process perspective covering modeling, perception, decision-making, navigation control, and fault-tolerant assurance, it systematically reviews the major research progress in this field and summarizes the representative architectures of human-machine shared navigation control. In particular, the state of the art is analyzed in key areas including ship operator behavior modeling and risk quantification, enhanced navigation situation awareness, intelligent navigation decision-making, cooperative navigation control, and fault diagnosis and fault-tolerant control, as well as the cross-stage coupling and coordination mechanisms among these modules. Finally, future trends in human-machine shared navigation control for MASSs are discussed, with particular emphasis on potential interdisciplinary research directions involving large language model-driven human–machine interaction, digital twin modeling, self-learning intelligent decision-making, and dynamic game-based allocation of control authority.