A GPU based Algorithm for Determining the Optimal Cutting Direction in Deep Mold Machining

Large molds with very deep shape are well used in producing bumpers and inner panels of automobiles. In order to realize the precise and stable machining of such deep molds, 3-axis milling with inclined cutters are often applied. In this paper, we propose a new algorithm for determining the optimal cutting direction in such inclined machining. We introduce a concept of accessibility cone as a measure for evaluating the stability and safety in the inclined machining, and we define the optimal cutting direction as the direction whose corresponding accessibility cone has the maximum peak angle. An accessibility cone for a specific cutting direction can be derived by rendering a silhouette picture of the offset shape of the mold. This computation can be accelerated by using a graphics processing unit (GPU) which is now equipped in most PCs. Proposed algorithm is implemented and an experimental process planning assistance program using this technology is demonstrated.