Abstract: With the demand-driven and policy-promoted development, the research and application of coal mine robots have become unprecedentedly active. Currently, these robots can autonomously operate in a few scenarios, replacing coal mine operators in performing certain tasks. However, the openness and complexity of the environment and tasks in auxiliary operations necessitate the continued deep involvement of coal mine operators. In the context of human-robot integration and collaboration, it is essential to explore how to fully leverage human intelligence and machine intelligence to optimize the decision-making process in coal mine auxiliary operations. First, relevant concepts and works are reviewed, and decision-making in coal mine auxiliary operations is defined as a human-robot hybrid decision-making problem. Then, based on the human-machine hybrid augmented intelligence paradigm of human in the loop, an XR driven human-robot hybrid decision-making architecture for coal mine auxiliary operations is proposed, and the components and collaborative operation relationship of the architecture are introduced. After that, key technologies and corresponding solutions required to implement this architecture are discussed. Finally, a simulation scenario is constructed using the case of shutdown maintenance for a shearer drum, with application testing conducted under the proposed architecture to verify feasibility and effectiveness. Results indicate that the proposed architecture extends the human-robot hybrid decision-making process into virtual space, enabling a reasonable integration of decision-making between coal mine operators and auxiliary operation robots through the clever transformation and transmission of heterogeneous information flows, thus forming a decision-making model characterized by virtual-real fusion and human-robot co-intelligence, which holds potential for application in coal mine auxiliary operations and similar scenarios in other fields.

Key words: auxiliary operations in coal mines, human-machine hybrid augmented intelligence, human-robot hybrid decision-making, virtual reality, augmented reality