# Automated Unsupervised 3D Tool-Path Generation Using Stacked 2D Image Processing Technique

^{1}

^{2}

^{*}

## Abstract

**:**

^{®}pseudocode with a G-code interpreter and a simulator. The results showed that the proposed technique produced an automated unsupervised reliable tool-path-generator algorithm and reduced tool wear and costs, by allowing the selection of the tool depth-of-cut as an input.

## 1. Introduction

^{®}) or tool-path generation tools (e.g., Mastercam

^{®}); where, unfortunately, time is wasted while transcoding file formats. Moreover, a STEP based tool-generation approach was implemented in [10], to minimize the file exchange time.

## 2. State of the Art

## 3. Materials and Methods

#### 3.1. Geometric Modeling and Parameter Identification

^{®}software by using STL file reader. The imported model was re-sampled and discretized into (n, m, k) dimension (see Figure 4).

#### 3.2. Model Segmentation and Reconstruction

_{gi}and x

_{gf}are the initial and final grid points that are shifted from the center of cube cells [23].

Algorithm 1: Pseudocode for voxelization as a binary logic |

grid_data = zeros(rx,ry,rz); |

P0 = Facet position |

Nf = Array for normal facets |

for nz = 1 : rz |

for ny = 1 : ry |

for nx = 1 : rx |

% Get the point |

p = [ xa(nx) ; ya(ny) ; za(nz) ]; |

% Find the closest Facet |

vertices_distance = ∑(([ P0(1,:)-p(1) ; P0(2,:)-p(2) ; P0(3,:)-p(3) ])^{2}); |

[v,ind] = min(vertices_distance); |

% Add Point if it is enclosed inside an object |

data = dot(N_f(:,ind),p-P0(:,ind)); |

grid_data(nx,ny,nz) = (data>=0);% logical array size of NxMxK |

end |

end |

end |

#### 3.3. Point Cloud Generation Using Image Processing Techniques

#### 3.4. Tool Path Motion Parameters

^{2}. Actually, it is essential to select the billet dimension with a minimum amount of material removal because it minimizes the machining time, power consumption, and waste of material. In this paper, the billet size is automatically computed by adding an offset value to the absolute difference of maximum and minimum value in both radial and longitudinal motions.

_{f}is the final diameter, and D

_{o}is the initial diameter.

#### 3.5. Tool-Path Generation and Parsing

Algorithm 2: Pseudocode for image processed path generator | |

[z,x] = pixels data from image along(z,x) axis | |

Pcor = []; | % Initializing the dynamic coordinate array |

k = 0; | % Counter |

t = 0; | % Counter |

for i = From Zo to Zf | |

t = t + 1; | % Able to count the number of tool passes |

for j = From Xf to Xo | % Holds true for materials to be removed |

if image(i,j)>0; | |

Pnew = [j,i]; | |

k = k + 1; | % Counts the true pixels to be removed |

else | |

Pnew = []; | % Final product pixels |

end | |

P = Pnew; | |

Pcor = [Pcor;P]; | % Creates a vector of tool path |

end | |

end |

#### 3.6. G-Code Generation

_{i})—the sampling size (N), it is possible to have the Equations (10) and (11).

- G00 X45 Y20 Z00: Rapid movement to coordinates of (45, 20, 0)
- G01 X45 Y20 Z00 F3.5: Linear movement to coordinates of (45, 20, 0)
- G02/G03 X45 Y20 R1.0: Circular motion to coordinates of (45, 20) with a radius of 1.0

## 4. Result and Discussion

^{®}and we adopted the source code of [16] for turning and milling operations.

## 5. Conclusions and Future Work

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- Altintas, Y. Manufacturing Automation by Yusuf Altintas. Camb. Core
**2012**. [Google Scholar] [CrossRef] - Zhou, M.; Zheng, G.; Chen, Z.C. An automated CNC programming approach to machining pocket with complex islands and boundaries by using multiple cutters in hybrid tool path patterns. Int. J. Adv. Manuf. Technol.
**2016**, 83, 407–420. [Google Scholar] [CrossRef] - Wang, X.; Fu, X.; Li, C.; Kang, M. Tool path generation for slow tool servo turning of complex optical surfaces. Int. J. Adv. Manuf. Technol.
**2015**, 79, 437–448. [Google Scholar] [CrossRef] - Filice, L.; Ambrogio, G.; Gaudioso, M. Optimised tool-path design to reduce thinning in incremental sheet forming process. Int. J. Mater. Form.
**2013**, 6, 173–178. [Google Scholar] [CrossRef] - Lechevalier, D.; Shin, S.-J.; Rachuri, S.; Foufou, S.; Lee, Y.T.; Bouras, A. Simulating a virtual machining model in an agent-based model for advanced analytics. J. Intell. Manuf.
**2019**, 30, 1937–1955. [Google Scholar] [CrossRef] [PubMed] - Jouandeau, N. Rapidly-exploring Sorted Random Tree: A Self Adaptive Random Motion Planning Algorithm. In Informatics in Control, Automation and Robotics: Selected Papers from the International Conference on Informatics in Control, Automation and Robotics 2007; Filipe, J., Cetto, J.A., Ferrier, J.-L., Eds.; Springer: Berlin/Heidelberg, Germany, 2009; pp. 63–73. [Google Scholar] [Green Version]
- Liu, Y.; Guo, X.; Li, W.; Yamazaki, K.; Kashihara, K.; Fujishima, M. An intelligent NC program processor for CNC system of machine tool. Robot. Comput. Integr. Manuf.
**2007**, 23, 160–169. [Google Scholar] [CrossRef] [Green Version] - Ridwan, F.; Xu, X.; Ho, F.C.L. Adaptive execution of an NC program with feed rate optimization. Int. J. Adv. Manuf. Technol.
**2012**, 63, 1117–1130. [Google Scholar] [CrossRef] - Tönshoff, H.K.; Janocha, H.; Seidel, M. Image Processing in a Production Environment. CIRP Ann.
**1988**, 37, 579–590. [Google Scholar] [CrossRef] - Reduction of Production Cycle Time by Optimising Production and Non-Production Components of Time in the Metalworking Industry: A Case Study. Available online: http://www.scielo.org.za/scielo.php?script=sci_arttext&pid=S2224-78902016000100015 (accessed on 10 August 2019).
- Ullah, A.S.; Harib, K.H. Tutorials for Integrating CAD/CAM in Engineering Curricula. Educ. Sci.
**2018**, 8, 151. [Google Scholar] [CrossRef] - Mikołajczyk, T.; Kłodowski, A.; Mrozinski, A. Camera-based Automatic System for Tool Measurements and Recognition. Procedia Technol.
**2016**, 22, 1035–1042. [Google Scholar] [CrossRef] [Green Version] - Posada, J.R.D.; Schneider, U.; Sridhar, A.; Verl, A. Automatic Motion Generation for Robotic Milling Optimizing Stiffness with Sample-Based Planning. Machines
**2017**, 5, 3. [Google Scholar] [CrossRef] - Dupont, P.E.; Derby, S. An Algorithm for CAD-Based Generation of Collision-Free Robot Paths. In CAD Based Programming for Sensory Robots; Springer: Berlin/Heidelberg, Germany, 1988; pp. 433–465. [Google Scholar]
- Xu, K.; Li, Y.; Xiang, B. Image processing-based contour parallel tool path optimization for arbitrary pocket shape. Int. J. Adv. Manuf. Technol.
**2019**, 102, 1091–1105. [Google Scholar] [CrossRef] - Chitsaart, C.; Rianmora, S.; Rattana-Areeyagon, M.; Namjaiprasert, W. Automatic Generating CNC-Code for Milling Machine. World Acad. Sci. Eng. Technol. J. Ind. Manuf. Eng.
**2013**, 7, 2607–2613. [Google Scholar] - Konobrytskyi, D. Automated CNC Tool Path Planning and Machining Simulation on Highly Parallel Computing Architectures. Ph.D. Thesis, Clemson University, Clemson, SC, USA, 2013. [Google Scholar]
- Zhao, Q.; Huang, X.; Wang, C.; Yao, Y.; Hu, Q. NC tool path generation approach based on STL. In Proceedings of the 2011 IEEE International Symposium on IT in Medicine and Education, Guangzhou, China, 9–11 December 2011; pp. 228–231. [Google Scholar]
- Qu, X. Raster milling tool-path generation from STL files. Rapid Prototyp. J.
**2006**, 12, 4–11. [Google Scholar] [CrossRef] - Hassaballah, M.; Abdelmgeid, A.A.; Alshazly, H.A. Image Features Detection, Description and Matching. In Image Feature Detectors and Descriptors: Foundations and Applications; Awad, A.I., Hassaballah, M., Eds.; Springer International Publishing: Cham, Switzerland, 2016; pp. 11–45. [Google Scholar]
- Vatani, M.; Rahimi, A.R.; Brazandeh, F. An Enhanced Slicing Algorithm Using Nearest Distance Analysis for Layer Manufacturing. Proc. World Acad. Sci. Eng. Technol.
**2009**, 3, 6. [Google Scholar] - Eragubi, M. Slicing 3D CAD Model in STL Format and Laser Path Generation. Int. J. Innov. Manag. Technol.
**2013**, 4, 410–413. [Google Scholar] [CrossRef] - Szucki, M.; Suchy, J. A voxelization based mesh generation algorithm for numerical models used in foundry engineering. Metall. Foundry Eng.
**2012**, 38, 43. [Google Scholar] [CrossRef]

**Figure 1.**Diagram of research design and workflow. In this workflow, a geometric model was imported into a simulation environment and processed into numerical values.

**Figure 3.**A cube with vertices and facets indices. A triangulation of vertices is represented using edges, and these combinations can be given in matrices form.

**Figure 4.**(

**a**) CAD model in STL file. (

**b**) 3D model in MATLAB environment. (

**c**) The discretization approach.

**Figure 5.**A model contained in a grid (

**a**) 2D; (

**b**) 3D; (

**c**) 3D discretized and voxelized; (

**d**) 2D projection of (

**c**).

**Figure 6.**Block diagram of CAD to BMP processing. The STL reader imports CAD model into the MATLAB working space and extracts vertice, edge, and facet data. The image processing algorithm function further processes CAD data and creates a BMP output.

**Figure 8.**Converted CAD model from STL to BMP image file. (

**a**) 2D projection of a 3D model on 2D grid container, (

**b**) Intensity image for contained models, (

**c**) Binary scale of image (b).

**Figure 9.**(

**a**) CAD model imported as STL file format and (

**b**) Converted CAD model from STL to BMP image file.

**Figure 10.**Schematic diagram of turning parameters and demonstration. (

**a**) Geometric description and parameter definitions for turning operation, (

**b**) Axis assignment convention and demonstration of motion types.

**Figure 12.**Image processing technique based milling operation simulation. (

**a**) Hypersurface visualization and process definition for milling operation of circular path. (

**b**) Tool-path planning and plotting.

**Figure 13.**Image processing technique based turning operation simulation. (

**a**) 3D model (front view), (

**b**) 2D binary image, (

**c**) path generated model).

**Figure 14.**How accuracy of models can be affected by segmentation size. (

**a**) 16 divisions, (

**b**) 32 divisions, (

**c**) 64 divisions.

**Figure 15.**Simulation for the turning process using MATLAB script; (

**a**) Path visualization for tool pass; (

**b**) Simulation processes for turning operations.

Parameter | STEP | STL | IGES |
---|---|---|---|

Memory (Bytes) | 16.7 K | 684 | 21.3 K |

3D | Yes | Yes | Yes |

Vertices | 18 | 36 | 18 |

Faces | 6 | 12 | 6 |

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Tuli, T.B.; Cesarini, A.
Automated Unsupervised 3D Tool-Path Generation Using Stacked 2D Image Processing Technique. *J. Manuf. Mater. Process.* **2019**, *3*, 84.
https://doi.org/10.3390/jmmp3040084

**AMA Style**

Tuli TB, Cesarini A.
Automated Unsupervised 3D Tool-Path Generation Using Stacked 2D Image Processing Technique. *Journal of Manufacturing and Materials Processing*. 2019; 3(4):84.
https://doi.org/10.3390/jmmp3040084

**Chicago/Turabian Style**

Tuli, Tadele Belay, and Andrea Cesarini.
2019. "Automated Unsupervised 3D Tool-Path Generation Using Stacked 2D Image Processing Technique" *Journal of Manufacturing and Materials Processing* 3, no. 4: 84.
https://doi.org/10.3390/jmmp3040084