Generating Milling Tool Paths for Prismatic Parts Using Genetic Programming

Abstract

The automatic generation of milling tool paths traditionally relies on applying complex tool path generation algorithms to a geometric model of the desired part. For parts with unusual geometries or intricate intersections between sculpted surfaces, manual intervention is often required when normal tool path generation methods fail to produce efficient tool paths. In this paper, a simplified model of the machining process is used to create a domain-specific language that enables tool paths to be generated and optimised through an evolutionary process -formulated, in this case, as a genetic programming system. The driving force behind the optimisation is a fitness function that promotes tool paths whose result matches the desired part geometry and favours those that reach their goal in fewer steps. Consequently, the system is not reliant on tool path generation algorithms, but instead requires a description of the desired characteristics of a good solution, which can then be used to measure and evaluate the relative performance of the candidate solutions that are generated. The performance of the system is less sensitive to different geometries of the desired part and doesn't require any additional rules to deal with changes to the initial stock (e.g. when rest roughing). The method is initially demonstrated on a number of simple test components and the genetic programming process is shown to positively influence the outcome. Further tests and extensions to the work are presented.

Automatic Tool Path Generation in Contour Milling Using Genetic Algorithm

“…The shape and dimensions of the initail workpiece (marked blu in Figure 1) is predetermined and as such, represents one of the input data in the model. In order to determine the path of the tool, it is necessary to find a measure of the discretization of the surface being processed (Barclay, Dhokia, and Nassehi 2015) i.e. to determine the size of the set of points and the position of the points that represent the positioning of the center of the tool, having in mind that the tool must be positioned at each point of that set, but also that its movement leads to processing the contour to the final size, that is, to enable the final processing.…”

“…Discretization can be modelled aspixelizationmapping of the surface being processed with a large number of squares (Kumar and Khatak, 2018) basically cover the surface being processed very well, but then their number, that is, the resolution of the grid of squares goes up to 0.01 mm is the usual resolution of today's CNC machines. Such a data set is very large for processing by metaheuristic methods, especially GA, which can lead to the fact that it is not possible to find a solution that converges to the optimal one in an acceptable time (Barclay, Dhokia and Nassehi, 2015). Other approach is surface discretization by placing up an equidistant grid of points on the surface to be processed, but this method gives good results only in cases of rough contour milling (Nassehi, Essink, and Barclay, 2015).…”

“…The generated tool path follows the shape of the contour, that is, it enables the processing of any shape of the contour, but the model does not foresee path optimization. The model that integrates toolpath generation and optimization is provided in the research ofBarclay, Dhokia, Nassehi (2015) The model is based on the discretization or pixelization of the surface, which is processed by levels, that is, the tool path is generated and optimized for the value of the axial depth of cut. The authors state that the problem of path generation in milling can be viewed as a TSP problem (Traveling Salesman Problem) that can be efficiently solved by applying GA. Like Kumar and Khatak (2018) the surface to be processed is modelled with a mesh with a finite number of squares.…”

Tool Path Optimization for Complex Cavity Milling Based on Reinforcement Learning Approach

“…The main focus of this paper is on 2D-TPO in the presence of complex cavity features. Barclay et al [8], for example, used genetic programming techniques to develop a generic program which automatically generates the optimal tool path based on the contour shape of a given 2D milling area. Zhou et al [5] developed a genetic algorithm (GA) based tool path generation scheme specifically for complex cavity milling tasks.…”