The presentation did at the XXXVIII Iberian Latin-American Congress on Computational Methods in Engineering regarding a technological approach to enhance machine setup efficiency through computer vision. For further information, please refer to this page: http://bit.ly/cilamce2017
Computational visual system to reduce setup time in CNC vertical machining centers
1. Smart Setup
Computational visual system to reduce setup time in CNC
vertical machining centers
Paulo Araujo, Romulo Lins, Alexandre Acácio, Cedéia Araújo, Rovilson Mafalda
November 6, 2017
Federal University of ABC
Graduate Program in Engineering and Innovation Management
3. Problem statement
Setup activities are time-consuming!
All steps needed to leave a
machine fully operational
They are costly and do not add
value to the final product
4. Problem statement
Setup activities are time-consuming!
All steps needed to leave a
machine fully operational
They are costly and do not add
value to the final product
The faster, the better
Less setup time Companies & Customers
5. CNC machine setup
CNC Computer Numerical Control
Machine axis movement controlled
by computer
6. CNC machine setup
CNC Computer Numerical Control
Machine axis movement controlled
by computer
The number of axis may vary
depending on the type of machine
7. CNC machine setup
CNC Computer Numerical Control
Machine axis movement controlled
by computer
z
x
y
The number of axis may vary
depending on the type of machine
We are focused on 3-D vertical
machines
10. CNC machine setup
Machine
initialization
Machine home
definition
Tools
verification
Machine
reference
definition
Part zero
definition
CNC program
loading
Program
execution
Machine
shutdown
Main objective of the project
12. CNC machine setup
Once the part zero point is
defined the machine controller
calculate all axis movements
relative to it
13. CNC machine setup
Once the part zero point is
defined the machine controller
calculate all axis movements
relative to it
There are different
methods to define the
part zero point
14. CNC machine setup
Once the part zero point is
defined the machine controller
calculate all axis movements
relative to it
There are different
methods to define the
part zero point
Methods may vary in
cost, accuracy, ease of
use, etc
16. Proposed system
Template to physically locate the
part zero point
Reference features to facilitate 3-D
location recovery
Stereo system to recover 3-D data
from the scene and locate the part
zero point in a single run
17. Proposed system
Before further investments it was conduct some laboratory experiments
3-D printed template 2 webcams+
Test: Measure the distance between 2 reference features and compare
it with the theoretical value
Algorithms and cameras’ calibration were implemented using Matlab
22. Results
Theoretical measure = 55.000 mm
Deviation = 0.145 mm ≡ 2.64%
Factors that may have contributed to this deviation:
§ Poor cameras quality
§ 3-D printer precision
§ Template finishing
Experimental measure = 54.855 mm
23. Conclusions
Once all calibration is done the proposed system showed to be a quick
and affordable method to define the position of the part zero.
24. Conclusions
Once all calibration is done the proposed system showed to be a quick
and affordable method to define the position of the part zero.
Cheaper materials during the prototype phase resulted in a
measurement with small deviation, stimulating further studies on this
kind of application.
25. Conclusions
Once all calibration is done the proposed system showed to be a quick
and affordable method to define the position of the part zero.
Cheaper materials during the prototype phase resulted in a
measurement with small deviation, stimulating further studies on this
kind of application.
Depending on the final accuracy, this system would be elected for a
certain group of manufacturing processes.
26. Conclusions
Once all calibration is done the proposed system showed to be a quick
and affordable method to define the position of the part zero.
Cheaper materials during the prototype phase resulted in a
measurement with small deviation, stimulating further studies on this
kind of application.
Depending on the final accuracy, this system would be elected for a
certain group of manufacturing processes.
Further developments would include the integration of this system
with the machine’s controller in order to accelerate even more the
machine setup.