In the paper, a new volume-based feature recognition method has been developed to recognize both isolated and interacting undercut features in a uniform way. Firstly, a set of potential undercut are generated by performing the regularized difference between the part and its convex hull. The optimal parting direction is then determined by minimizing the number of undercuts. At last, undercut features are identified from the set of potential features with respect to the optimal parting dirction. However, instead of by the first step, the potential features of an interacting feature are generated by two-stages; volume decomposition, in which the volume to be recognized is decomposed into convex cells by intersecting it with half spaces of its faces having concave edges; and reconstruction of features, in which potentiaol featrues are reconstructed through systematically connecting the small cells using the non-directional blocking graph (NDBG) of the cells and the part. Moreover, multiple interpretations of an interacting feature can be easily generated by simply subtracting potential external undercut features from each other in different orders.
展开▼
