Regular inspection of tunnel ancillary facilities has become a cornerstone of sustainable tunnel operation and maintenance. Conventional manual inspections, however, are often plagued by low efficiency, elevated safety risks, and inherent subjectivity. In contrast, AI-driven diagnostic frameworks offer a transformative approach to enhancing detection precision and operational efficiency, providing a robust foundation for granular tunnel management. Given that the systematic understanding of intelligent detection equipment and algorithms is still evolving, this study provides a comprehensive review of current research progress across three dimensions: detection equipment, object detection algorithms, and defect detection algorithms. Regarding hardware, the architecture of mobile inspection platforms is systematically delineated. An in-depth analysis is conducted on the strengths and constraints of high-resolution visible-light imaging and laser scanning. Furthermore, the integration of multi-sensor fusion and sophisticated in-tunnel positioning technologies is explored. In the realm of object detection, specialized image processing and data augmentation strategies tailored for unique tunnel environments are examined. A comparative performance analysis is performed between two-stage and one-stage detection architectures in the context of ancillary facility recognition. For defect diagnosis, contemporary methodologies are reviewed, with emphasis placed on the application of semantic and instance segmentation for the granular assessment of facility deterioration. Finally, the study identifies core challenges, including the demands of high-speed dynamic inspection, the scarcity of standardized datasets, and insufficient model generalization. To address these, three future research trajectories are proposed: the development of integrated intelligent sensing hardware for high-speed scenarios, the exploitation of generative AI for scalable data augmentation, and the enhancement of algorithmic resilience and robustness within complex tunnel environments.