Algorithms for selecting cutters in multi-part milling problems

Optimal tool selection for 2.5D milling, part 1: a solid-modeling approach for construction of the Voronoi mountain

“…Applications of Voronoi diagrams and medial axes include tool-path generation (Held et al 1994, Veeramani andGau 2000), tool selection (D'Souza et al 2001, Lim et al 2001, Yao et al 2003) and feasibility checking (D'Souza et al 2002), robot motion planning (Takahashi and Schilling 1989, Choset and Burdick 1995a, b, Garber and Lin 2002, shape representation and conversion (Mayya andRajan 1996, Naf et al 1997), mesh generation (Bern and Eppstein 1995, Mishev 1998, Labelle and Shewchuk, 2003, pattern recognition and computer vision (Kise et al 1998, Burge andBurger 2000), medical and biological analysis (Bertin et al 1993, Johnson and Parker 1994, Noever et al 1995, Naef et al 1996, Indermitte et al 2001, offsetting for tolerancing (Feng and Hopp 1991), nearest Optimal tool selection for 2.5D milling, part 1 295 neighbour clustering (Inaba et al 1994, Vaisman et al 1998, crystallography (Christensen and Thomas 1999), and facility location and planning (Robert andToussaint 1990, Ben-Moshe et al 1999). …”

Optimal tool selection for 2.5D milling, part 2: a Voronoi mountain approach for generalized pocket geometries

“…inability to handle open edges and circular edges (Yao et al 2003); . pocket geometry restricted to prismatic pockets with rounded corners, and inter sub-pocket boundaries not addressed (D'Souza et al 2001); .…”

“…The selection problem is formulated as a graph search problem and covers multiple parts and cutters. This work was extended to incorporate an exact area computation using an algorithm for computing the open offset (Yao et al 2003). They also addressed the possibility of analyzing parts with multiple 2.5D features for the target and obstruction profiles and generating an optimal set of tools for multiple such parts.…”

“…One important limitation of many of these approaches is the assumption that machining time depends solely on the machinable area and the diameter of the cutter (Lim et al 2001, Yao et al 2001a, 2003. This assumption can become quite unacceptable in practice.…”

“…Several of these approaches also rely on the offsetoperator approach for finding the accessible regions for a cutter, and are likely to suffer from the common problems arising from the use of conventional offset operations (Yao et al 2003). Other approaches suffer from the limitation of their inability to handle open edges (see figure 2).…”