Collision Detection of Virtual Powered Support Groups Under Complex Floors

Abnormal relative position and attitude offset may cause powered supports to collide during movement, affecting the stability of a longwall operation. This study proposed a collision detection method for virtual powered support groups in the undulating strata condition. First, a support workspace was developed by defining key points, and the Denavit–Hartenberg (DH) matrix coordinate system of powered supports was established to solve the key points. Then, based on the intersection rule of plane line segments, collision detection between powered supports was realized by calculating the coordinates of the key points. According to actual coal face data, a virtual simulation platform is established, and the sequential movement of support groups on the virtual floor was realized. Finally, contrast experiments of different virtual floor models were conducted, and the Mesh floor, which is closer to the actual coal seam, was selected for subsequent virtual simulation. The collision detection experiments of powered supports were conducted in both virtual and physical environments, and the results validated the feasibility of the collision detection method. This method improves the virtual simulation degree of longwall coal mining, and provides simulation reference for the attitude adjustment of actual powered support groups before moving.