Seite - 107 - in Proceedings of the OAGM&ARW Joint Workshop - Vision, Automation and Robotics
Bild der Seite - 107 -
Text der Seite - 107 -
Novel HumanMachine Interaction with Sticky Notes for Industrial
Production
Gernot Stuebl, Thomas Poenitz, Harald Bauer, and Andreas Pichler1
Abstract—In this paper we present a 3D documentation
systemwhichutilizes newhumanmachine interaction concepts
on the example of virtual sticky notes. Using different tracking
techniques the virtual notes can be attached to a physical
object and are displayed on a tablet in an Augmented Reality
way. Themain intention is to strengthen the interplay between
construction and production of industrial machines as the
virtual notes are synchronized with a production lifecycle
management system.
I. PROBLEM DESCRIPTION
An essential part of machine manufacturing is the in-
terplay between construction and production. Often this
connection leaks information in both ways: the construction
team changes details in the last minute, while in production
things are mounted in a different order or way as it was
intended. Since in the end both sides have to synchronize
their knowledge this results mainly in a mass of notes
often stuck on the machine or even worse written on the
machine itself. This industrial spotlight paper presents a
novel development integrating latest technologies to manage
position based notes digitally in an Augmented Reality based
way.
II. STATE-OF-THE-ART
Although in media one can see photos of tablets showing
Augmented Reality (AR) overlays of shop floors, industrial
grade AR documentation systems are rare. To our knowledge
the most similar system available to our proposition is
Docufy [1]. It is an AR interface to a dedicated content
management system and used to display technical content
like manuals in a read-only way. The mapping of the data
to 3D is managed via a manual registration step of interest
points originating from a priori known CAD data. When the
tablet is moving, the interest points are tracked and the 3D
data adapts to the new viewpoint.
III. PROPOSED SYSTEM
In this paper we propose a system which enables a user to
view, edit and add virtual sticky notes to a machine during
assembling by using Augmented Reality techniques. The
notes are position dependent and synchronized with the 3D
database of a production lifecycle management (PLM) sys-
tem.Since theconstruction teammainlyworkswith thePLM
system, they immediately have access to the information and
may modify existing or add new notes directly.
1PROFACTOR GmbH, 4407 Steyr-Gleink, Im Stadtgut A2, Austria
{Forename.Surname}@profactor.at A. Tracking System
For the tracking we pursue a multi-modality strategy,
which utilizes a combination of
• a commercially available infrastructure-based tracking
system,
• fiducial Augmented Reality markers,
• and a visual real-time tracking system.
The infrastructure-based tracking system is the main source
of 6 degree-of-freedom (DOF) data for the system. See
Figure 1 for a tablet enhanced with a tracking system
receiver. The two major extensions to the state-of-the-art
Fig. 1. Tablet with attached tracking system receiver on the left side. This
enables a 6 degree-of-freedom positioning in space.
concern the tracking part as well as the way a user can
interact with the system.
The initial registration is done with an AR marker system
based on the work of Garrido-Jurado et al. [2]. Fiducial
markers are preferred to feature based approaches as the
presence of stable features on industrial objects cannot be
guaranteed. After that the tracking data is transformed to the
coordinate system of the machine’s CAD model which is
provided by the PLM software. This allows the attachment
of notes to positions in the CAD data.
A special approach is required for movable parts of the
machine, like panels. When mounted to the machine, their
relative positions to the base CAD data can be determined
using attached AR markers.
An additional feature was built in to handle unmounted
sub-assemblies. They can be annotated like any other ma-
chine parts, however they have to be pre-identified in a
manual step. This is the input of a real-time tracker which
is an extension of Akkaladevi et al. [3].
When a user likes to add a new note to the machine he/she
has two possibilities to define the position: either the desired
position is touched with the measuring tip mounted on the
107
Proceedings of the OAGM&ARW Joint Workshop
Vision, Automation and Robotics
- Titel
- Proceedings of the OAGM&ARW Joint Workshop
- Untertitel
- Vision, Automation and Robotics
- Autoren
- Peter M. Roth
- Markus Vincze
- Wilfried Kubinger
- Andreas Müller
- Bernhard Blaschitz
- Svorad Stolc
- Verlag
- Verlag der Technischen Universität Graz
- Ort
- Wien
- Datum
- 2017
- Sprache
- englisch
- Lizenz
- CC BY 4.0
- ISBN
- 978-3-85125-524-9
- Abmessungen
- 21.0 x 29.7 cm
- Seiten
- 188
- Schlagwörter
- Tagungsband
- Kategorien
- International
- Tagungsbände