The invention discloses an area tree-based underground roadway point cloud data quick query method. The method comprises the steps of (1) dividing a roadway space region; (2) according to the dividedroadway space region, constructing a topology structure of an area tree; (3) extracting a central axis of a roadway point cloud; (4) according to the area tree, establishing corresponding relationships between distances between the roadway point cloud and the central axis; (5) performing area segmentation on the roadway point cloud according to different distances between the roadway point cloud and the central axis along the central axis; and (6) determining a segmented region where a to-be-queried point is located, thereby quickly searching for the to-be-queried point. The point cloud data organization and management efficiency can be effectively improved; the to-be-queried point is quickly searched for; and the surrounding rock deformation monitoring efficiency and precision are improved.