• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.1
• /
• pp.46-53
• /
• 2007

NC (Numerical Control) code for the tool path needs to be generated efficiently for machining of free form pockets with arbitrary wall geometry on a three axis CNC machine. The unified rough and finish cut algorithm and the tool motion is graphically simulated in Part 1. In this paper, a grid based 3D navigation algorithm simulated in Part 1 for generating NC tool path data for both linear interpolation and a combination of linear and circular interpolation for three-axis CNC milling of general pockets with sculptured bottom surfaces is experimentally performed and verified.

• Journal of Korea Society of Industrial Information Systems
• /
• v.9 no.1
• /
• pp.7-16
• /
• 2004

The tool path needs to be determined in an efficient manner to generate the final NC (numerical control) code for efficient machining. This is particularly important in machining free form pockets with an arbitrary wall geometry on a three-axis CNC machine. Many CAD/CAM systems use linear interpolation to generate NC tool paths for curved surfaces. However, this needs to be modified to improve the smoothness of the machined bottom surface, reduce machining time and CL (cutter location) file size. Curved machining can be a solution to reduce these problems. The unified rough and finish cut algerian and the tool motion is graphically simulated. In this paper, a grid based 3D navigation algorithm for generating NC tool path data for both linear interpolation and a combination of linear and circular interpolation for three-axis CNC milling of general pockets with sculptured bottom surfaces is developed.

• Journal of Advanced Navigation Technology
• /
• v.12 no.5
• /
• pp.494-501
• /
• 2008

In this paper, to merge whole 3D information of an occluded body from view point, the new image merging algorithm is introduced after obtaining images of body on the turn table from 4 directions. The two images represented by polygon meshes are merged using weight bipartite matching method with different weights according to coordinates and axes based on minimum distance since two images merged don't present abrupt variation of 3D coordinates and scan direction is one direction. To obtain entire 3D information of body, these steps are repeated 3 times since the obtained images are 4. This proposed method has advantage 200 - 300% searching time reduction rather than conventional branch and bound, dynamic programming, and hungarian method though the matching accuracy rate is a little bit less than these methods.