JEC COMPOSITES MAGAZINE - Issue #142 - September/October 2021 - 59

lmproved manual composite manufacturing with an augmented-reality system
Fig. 3: Operator's view when using the smart glasses with the Augmented Lay-Up
software
Fig. 2: User interface with display of ply position contour and additional information
feature, which includes a toolbox of quality assurance functions tailored to
the customer's specifications. In addition, this approach keeps the software
applications lean due to the reduction of functions, contents, processes and
data in a single module and thus enables an optimal implementation the
software in terms of stability, usability, performance and accuracy (Figure 2).
The Augmented Lay-Up assistance system is generally platform-independent;
however, an appropriate hardware has to fulfil requirements concerning
performance, field of view, resolution, connectivity, control options and usable
sensors. For the initial development at the Fraunhofer IPT, a first-generation
Microsoft HoloLens was selected due to its good compliance with
the mentioned criteria. Comparable solutions that were considered include
the Meta 2 from Schenker Tech, the R-7 from ODG and the BT-300 from
Epson. In the development, Vuforia was used for object recognition and the
visualization of support contours was realized with Microsoft's Unity and
Mixed Reality Toolkit.
Validation of the Augmented Lay-Up technology
After conceptualising and implementing the software, the assistance system
was validated by the manufacturing of a demonstrator with different manual
lay-up operations at the Fraunhofer IPT. In general, a good support for
both wet lay-up and prepreg lay-up processes was observed. When starting
the software, the user is in a higher-level menu, in which both implemented
software features can be accessed. After starting the Cutting feature, the environment
in the worker's field of view is scanned for plane surfaces. When
the user stands directly in front of the work surface and has it in view, the
selected two-dimensional ply geometry appears. In a menu displayed on
the side of the work surface, it is possible to switch between geometries to
be cut, e.g. to make optimum use of the remaining preform area and consequently
minimize waste and material costs. The same effect can be achieved
by moving and rotating the cut lines, which are displayed in green, by hand
gestures to maximise the used preform area. Cutting along the projected
lines can be performed intuitively and after a short practice period, geometric
deviations from the target geometry can be eliminated.
When the Lay-Up feature is started, the field of view of the smart glasses is
scanned for suitable three-dimensional surfaces based on the stored mould
geometry. After selecting the current layer, the deposition contour is displayed
on the mould and additional information such as the layer number,
the fibre orientation, the distances of the current ply from the tool edges
and the material designation are provided at the side of the working area.
A movement of the ply visualises the deposition direction. This approach
saves time and gives the worker enhanced control over the performed task
(Figure 3).
During the validation phase, optimisation potentials for the software were
identified based on IPT staff's experience and through a user study. In genFig.
4: Validation of the Augmented Lay-Up system by manufacturing of demonstrator
parts
eral, the received feedback highlighted the intuitive application, the accessibility
of the mould from different directions and the short period of getting
used to the new method. The presentation of the cutting and lay-up lines,
the stability of the application - especially concerning object recognition -
and the position of information display could all be optimized based on the
findings of this phase. The prototype Augmented Lay-Up system allows
for decreased manufacturing time with reduced cutting and lay-up errors
while minimizing material waste and can be used for different components,
processes and materials. The current limitations are mainly traceable to the
smart glasses used for the initial implementation. The small field of view, the
processor power and the battery life of the used glasses limited the applicability,
especially for more complex parts. However, due to the platform-independent
programming approach, current and future technological development
of smart devices promises to remedy these limitations (Figure 4).
Demonstration and feasibility study
In conclusion, the Augmented Lay-Up assistance system is well suited for
the support of workers in manual composite lay-up procedures. By visualising
cutting geometries, deposition lines and additional information, time
and costs can be saved compared to existing solutions for plybook display.
Current and future progress in smart glass technology promise to provide
optimisation potential and extend the software's capabilities. For interested
companies that would like to get to know the software, the Fraunhofer
IPT offers a feasibility study in which the assistance system is demonstrated
using the current prototype setup. Ideally, potential users can test the assistance
systems themselves during a demonstration with IPT experts. At the
same time, the customer's use case will be discussed and analysed to design
concepts for the optimum development of a tailored Augmented Lay-Up
system for the application and products concerned.
Acknowledgements
This project was supported by the Fraunhofer Internal Programmes
under Grant No. MEF 603 023.
More information:
https://www.ipt.fraunhofer.de/en/Competencies/Productionmachines/
Projects/smartlayup.html
N°142 September - October 2021 / jec composites magazine
59
https://www.ipt.fraunhofer.de/en/Competencies/Productionmachines/Projects/smartlayup.html
JEC COMPOSITES MAGAZINE - Issue #142 - September/October 2021

Table of Contents for the Digital Edition of JEC COMPOSITES MAGAZINE - Issue #142 - September/October 2021
