A new collision avoidance strategy and its integration with collision detection for five-axis NC machining

Abstract

This paper presents a new collision avoidance strategy and its integration with the collision detection for five-axis NC machining that improves upon the earlier collision detection algorithm developed by the present authors [T.D. Tang, Erik L.J. Bohez, Pisut Koomsap (2007) The sweep plane algorithm for global collision detection with workpiece geometry update for five-axis NC machining. Computer-Aided Design 39(11):1012–1024]. The proposed algorithm automatically detects and corrects the collision based on the biggest collision boxes. The collision detection algorithm is based on the bounding volume and the sweep plane approach. The collision is firstly detected by using the bounding sphere algorithm, and the colliding spheres are then further checked with the sweep plane algorithm. The change of the workpiece geometry is included in the detection process. After the collision detection, the collision data are stored. Only the biggest collision boxes (the boxes with the biggest edge in the X, Y, or Z direction) for each type of collisions are stored. The collision avoidance algorithm corrects the biggest collision based on heuristic strategy. With this strategy, when the biggest collision is corrected, most of other collisions will disappear automatically. Therefore, the time complexity of the collision avoidance strategy is considerably reduced. The algorithm has been implemented in Visual C++ and OpenGL, demonstrated for five-axis machine with two rotary axes on the table (Deckel MAHO 600e) and can be customized to apply for any five-axis CNC machines.