Intelligent robotic die polishing system using a neural network

A robotic die polishing system was designed and demonstrated conceptually in the Computer Integrated Manufacturing Laboratory of the Pennsylvania State University. Die polishing has traditionally been a labor intensive field requiring skilled machinists. Yet some of the time consuming aspects of die polishing can be transferred to a neural network. The robotic die polishing system uses a neural network specially developed to accommodate orientation changes and achieve shift invariance (Multi Net Invariant Network-MNIN). The MNIN discerns patterns and create strategies for polishing rough-machined dies that initially have a range of unpredictable surface finishes due to machining variations including differences from tool changes and spindle vibrations. Unique aspects of this project include use of a vision sensor to acquire surface texture patterns similar to those anticipated in die polishing and a neural network to make decisions on robot path planning to polish each individual die with respect to the particular pattern of roughness observed.