Abstract: Because the coal mining is still in the stage of partially unmanned operation, human factor still play an important role in a fully mechanized mining system. Based on the current intelligent equipment and technologies, the intelligent fully mechanized coal face should rely on the deep human-machine cooperation between human(including patrol workers, centralized control workers and maintenance workers),equipment and environment to achieve the goal of safe and efficient operation. Therefore, based on the authors' long-term practice in the ‘XR +' field, the intelligent operation mode of human-machine collaboration in a fully mechanized mining face from the perspective of ‘human-centered manufacturing' is explored. Firstly, the working environment of three types of underground jobs and ground jobs are compared and analyzed, and the path of how to upgrade them to intelligent operators in the future is obtained. The framework of multi-type cooperative operation mode of a fully mechanized mining face based on a multi-function parallel system is put forward, and the global virtual scene with a deep fusion of ‘human, machine and environment' is established. The intelligent deduction is carried out, and the cooperative interaction between three types of work and the multi-function virtual parallel system is designed. Then some key technologies are studied, such as parametric virtual miner model, real-time linkage between real miners and virtual miners, inspection gesture set design and underground collaborative maintenance system and so on. Through the multi-functional parallel system, the relevant tests had been carried out for multi-type training, virtual operation of inspection workers, optimization of centralized control workers, coordinated operation of maintenance workers and multi-type collaborative operation. The real-time interaction between three types of work and the multi-functional parallel system is realized. Finally, a collaborative operation mode of ‘global details and control, push results by intelligent machine and multi-type collaborative decision-making' is formed to jointly conduct high-quality remote monitoring, intervention, inspection and maintenance operations for improving the efficiency and reliability of a fully mechanized mining face.