Abstract: Intelligent roadway excavation is a key component in achieving the construction goals of smart coal mines. This paper elucidates the connotation of intelligent roadway excavation and, through an analysis of the characteristics and current development of rapid excavation systems—including tunnel boring machine, bolter-miners, continuous miners, and integrated excavation-bolting machines—provides an in-depth examination of the critical technologies required to accelerate its development. These technologies encompass precise equipment localisation, directional excavation, autonomous cross-section cutting and shaping, intelligent ground support, collaborative control of equipment fleets, fault and safety early-warning for working-face equipment, geological condition perception, and intelligent working-face management and control.To address the requirements for equipment localisation, environmental perception, intelligent control, and decision-making in complex underground conditions, this work proposes cooperative-target-based visual localisation, “vision+” multi-source fusion integrated localisation, and targetless visual localisation techniques oriented towards roadway environmental features. In addition, visual+deep-learning-based methods for roadway cross-section fracture detection and roadway geometric reconstruction are developed, thereby enhancing the autonomous navigation and environmental perception capabilities of underground excavation equipment. Technologies for directional excavation and deviation correction, trajectory planning and cutting control, virtual-teaching-based memory cutting, bolt-hole recognition, multi-boom coordinated support control, and equipment-fleet collision early warning are proposed. These advances address the key control challenges in the excavation process, including difficulties in accurately regulating equipment pose, dynamically planning cutting trajectories, ensuring coordination during support operations, and avoiding spatial interference among multiple machines, thereby enabling adaptive control of both excavation and support. By integrating digital twin technology into cutting control, collision detection and early warning, and collaborative control decision-making, a digital twin-driven intelligent management and control system for tunneling faces has been constructed, achieving real-time mapping between physical entities and virtual models, collaborative decision-making among equipment groups, fault diagnosis of key equipment parts, visual management of the operation process, and risk prediction. A development framework and research directions for intelligent roadway excavation technology are proposed, including equipment positioning based on multi-modal information fusion, intelligent control technologies for complex operating conditions, AI-driven virtual-physical collaborative intelligence for the working face, and digital-twin-driven virtual commissioning techniques. These advancements aim to achieve highly automated, unmanned, and intelligent rapid excavation of roadways.