proceedings of the workshop on user interaction techniques...
TRANSCRIPT
Proceedings of the workshop on User Interaction Techniques for Future Lighting Systems
September 5, 2011: Lisbon, Portugal
in conjunction with Interact
Organizers: Dzmitry Aliakseyeu1, Jon Mason
1, Bernt Meerbeek
1, Harm
van Essen2, Serge Offermans
2, Andrés Lucero
3
1 Philips Research Europe, 5656 Eindhoven, The Netherlands
2 TU Eindhoven, Industrial Design department, Eindhoven, The Netherlands
3 Nokia Research Center, Visiokatu 1, 33720 Tampere, Finland
{dzmitry.aliskseeyeu, jon.mason, bernt.meerbeek}@philips.com
{h.a.v.essen, s.a.m.offermans}@tue.nl, [email protected]
URL: http://interactingwithlight.id.tue.nl/
2
Contents
User Interaction Techniques for Future Lighting Systems .................................................. 3
Dzmitry Aliakseyeu, Jon Mason, Bernt Meerbeek, Harm van Essen, Serge Offermans,
Andrés Lucero
Semantic Light ...................................................................................................................................... 6
Zary Segall, Chad Eby and Pietro Lungaro
Interacting with Light Apps and Platforms .................................................................................. 9
Serge Offermans, Harm van Essen, Berry Eggen
The future of interaction with light and lighting dynamics ..................................................... 14
Jettie Hoonhout, Lillian Jumpertz, Jon Mason
User Interface for Task Lighting in Open Office ....................................................................... 23
Koen van Boerdonk, Jon Mason, Dzmitry Aliakseyeu
MeShirt: concepts for provocation and promotion ................................................................... 28
Alessandrini Andrea, Erik Grönvall, Paola Manuli, Valentina Senesi, Simona Melaragni and
Maria Teresa Oliviero
Towards Efficient Illumination Control for Underground Parking ..................................... 32
Paulo Carreira and Renato Nunes
3
User Interaction Techniques for Future Lighting
Systems
Dzmitry Aliakseyeu1, Jon Mason
1, Bernt Meerbeek
1, Harm van Essen
2, Serge
Offermans2, Andrés Lucero
3
1 Philips Research Europe, 5656 Eindhoven, The Netherlands 2 TU Eindhoven, Industrial Design department, Eindhoven, The Netherlands
3 Nokia Research Center, Visiokatu 1, 33720 Tampere, Finland
{dzmitry.aliskseeyeu, jon.mason, bernt.meerbeek}@philips.com
{h.a.v.essen, s.a.m.offermans}@tue.nl, [email protected]
Abstract. LED-based lighting systems have introduced radically new
possibilities in the area of artificial lighting. Being physically small the LED
can be positioned or embedded into luminaires, materials and even the very
fabric of a building or environment. Together with new functionality and
flexibility comes complexity; the simple light switch is not anymore sufficient
to control our light. This workshop explores new ways of interacting with light.
The goal is to define directions for new forms of user interaction that will be
able to support the emerging LED-based lighting systems.
Keywords: Lighting; User Interaction; LED; Smart lighting
1 Introduction
The Light Emitting Diode (LED) has caused a profound change within the lighting
industry. This is due in part to the LED‟s key properties of being physically small,
highly efficient, digitally controlled and soon, very cheap to manufacture. Being
physically small the LED can be positioned or embedded into luminaires, materials
and even the very fabric of a building or environment [1]. The price to pay for all this
functionality and flexibility is complexity. In the past, the single light bulb was
controlled using a single switch; on and off. LED-based lighting systems can easily
consist of hundreds separate light sources, with each source having many individually
controllable parameters including colour, intensity, and saturation. With this high
complexity, end-users cannot be expected to fully control all aspects of the lighting
system. One direction that is being explored is to enrich lighting systems with sensor
networks that will enable automatic lighting control that is based on contextual
information [2]. However in many situations, such as setting up an atmospheric light,
an explicit user interaction will still be required. Moreover, as functionality and
complexity of light systems grow, the mapping between the sensors data and the
desired light outcome will become fuzzy and will require an explicit user interaction
for fine tuning the outcome or for adjusting the mapping between sensor input and
light output. Thirdly, explicit interaction can be desired to allow users to feel in
control while interacting with intelligent lighting systems. The light switch therefore
in many situations will need to be replaced by novel forms of interactions that offer
4
richer interaction possibilities such as tangible, multi-touch, or gesture-based user
interfaces. As proliferation of LED light continues, it becomes more important to go
beyond scattered design efforts [2, 3] and systematically study user interaction with
emerging lighting systems. The goal of this workshop is to take the first steps in this
direction.
2 Goals of the workshop
The focus of this workshop is on formulating key research challenges for user
interaction with future lighting systems, creating initial design guidelines, and
proposing novel interaction techniques for these systems. The goals of the workshop:
1. Make a first step toward expanding the design space of interactive technologies to
include new forms of decorative, ambient, and task lighting.
2. Identify key challenges of UI for controlling new forms of lighting systems.
3. Establish a link with existing interaction paradigms that can be (re-)used for
control of future lighting systems.
During the workshop, we would like to address and discuss the following questions:
What design opportunities for interactive technology exist in the context of the
new forms of lighting?
What forms and types of (existing) interaction are suited for emerging lighting
systems (in particular tangible, gesture and multi-display interaction techniques)?
What forms of interaction are best suited for a global control (e.g. atmosphere)
and what for a point control (e.g. task lighting)?
How to balance between explicit user control and internal system control?
How to use a lighting infrastructure as ambient displays and how to combine it
with its primary function i.e. illumination?
What is the impact of the proposed interaction techniques for complex lighting
systems in other domains? What is the generalizability of these techniques?
To address the workshop questions we are inviting researchers to submit position
papers that discuss or present new forms of interaction techniques for lighting. This
topic deals with the research and design of new forms of user interaction or adaptation
of existing ones to emerging lighting systems. The topic should attract researchers and
designers working on new forms of interaction, LED lighting, smart lighting systems
and who are interested in exploring UI for new emerging types and forms of
luminaires and lighting systems.
3 Organizers
The workshop organizers are all active researchers in the area of user interaction, light
control and light perception specifically focusing on new forms of interaction and
collectively have considerable experience in organizing workshops.
Dzmitry Aliakseyeu is a senior scientist in the Human Interaction and Experiences
group of Philips Research. Prior to this he has held position of Assistant Professor at
5
the Industrial Design department of the Eindhoven University of Technology. His
research interests are in new forms of interaction and user interaction in the areas of
lighting and sleep.
Bernt Meerbeek is a senior scientist at Philips Research in Eindhoven. He holds a
Professional Doctorate in Engineering on User-System Interaction from Eindhoven
University of Technology. His research interests are in user interaction solutions for
intelligent systems, ranging from smart consumer appliances to smart environments
and lighting systems. Bernt is actively involved in user experience research at the
ExperienceLab research facility.
Harm van Essen is an assistant professor in the Department of Industrial Design at
Eindhoven University of Technology. His research interests are in interaction design,
design methodology, and the integration of technology and insights from social
sciences in product design. Design with and for lighting is an important research
topic, especially related to the development of new interaction styles to present and
communicate relevant information in an unobtrusive way, including end-user
programming, design opportunities for decentralized systems and ambient displays.
Serge Offermans is a PhD Student at the Eindhoven University of Technology,
department of Industrial Design. His PhD project is situated in the Intelligent Lighting
Institute and focuses on the interaction with novel lighting platforms and applications,
especially in multi-user environments.
Andrés Lucero is a senior researcher at Nokia Research Center in Tampere, Finland.
He has a background in Visual Communication Design (MA), User-System
Interaction (PDEng), and Human-Computer Interaction (PhD). His interests lie in the
areas of user-driven innovation, mobile interactions, and design research.
Jon Mason is a senior scientist at Philips Research in Eindhoven. His work at Philips
has included the design of new user interaction means for lighting in the retail, office
and hospitality contexts. His interests include UI design, design methodology, and
the inclusion of art in design.
References
1. Price, C. Light Fantastic, Digital Home Magazine, November, (2003).
2. Bhardwaj, S.; Ozcelebi, T.; Lukkien, J.; Smart lighting using LED luminaries. In proc. of
PERCOM Workshops, IEEE, 654 – 659, (2010)
3. Lucero, A., Lashina, T., and Terken, J. Reducing Complexity of Interaction with Advanced
Bathroom Lighting at Home. I-COM 5(1), Oldenbourg, 34-40 (2006).
6
Semantic Light
Zary Segall1, Chad Eby
1,2, and Pietro Lungaro
1
1Kungliga Tekniska högskolan, Stockholm, Sweden
{segall, chad, pietro}@kth.se 2Florida State Univeristy, Tallahassee, Florida, U.S.A.
Abstract. This position paper introduces the concept of Semantic Light, the
Semantic Light research program at KTH in Stockholm, and the background
intellectual property related to the project.
Keywords: semantic lighting, human-aware lighting, context-aware lighting,
multimedia, M2M, DLP, sensors.
1 Introduction
Light is fundamentally relevant to the human experience. The natural cycle of
diurnal light and dark moderates our patterns of sleeping and waking, and artificial
light sources help us work, play and learn in times and places when the sun is not
available. Across cultures, light and illumination has been used as a metaphor for
knowledge, understanding and reason and, practically, as a channel for
communication.
With the advent of high-intensity RGB LEDs, DLP micro projectors and
inexpensive cameras and sensors, light fixtures have had the potential to evolve into
“smarter” devices better suited to improve people‟s productivity and quality of living.
But, despite technological advances, light sources today tend to be static or responsive
to only relatively coarse control. More significantly, existing lighting is utterly
agnostic in terms of both the surfaces and objects illuminated, and the characteristics
of individuals‟ eyes. In addition, the information-carrying potential of light and
lighting is generally not considered; with digital light processing, a light source may
convey information as basic as a signal lantern or as complex as text and video.
2 IP and Research
Our work relates to a light delivery system and, more particularly, to a semantic
lighting system that delivers appropriate light (modulating qualities of spectrum,
intensity, color, contrast, temperature, angle, focus, text, image, animation, video and
other data) to a two or three dimensional subject by analyzing the properties of the
subject to be lit (nature, dimensions, shape, textual and image content, texture,
contrast, reflectance, refraction, specularity, etc.), the existing illumination, the eye
characteristics of the human user and the relative position and orientation of the
7
subject with respect to both the source of light and the human user. The semantic light
communicates meaning to the user by overlaying information by means of color, text,
image, animation, video and other data. In addition, the semantic light delivers
dynamic light that is changing in time and in synchronization with the semantic of the
task requiring illumination. Additionally, the semantic light stores and transmits
information relating to the subject and the human user to other semantic lights over
wireless networks, coordinating lighting in both time and space. More details are
available in U.S. patent “Semantic Light,” patent number 7,564,368, filed July 2009.
Semantic Light and Info Media is a way to make lighting smarter across a number
of awareness aspects. The first is human-aware lighting, which seeks to
algorithmically understand the physiological lighting needs of individuals. This may
include tailoring illumination to eye function in terms of intensity and frequency
response, or generating personalized phototherapy based on light spectrum exposure
patterns and also includes managing subjective individual preferences. The second
application area is that of context-aware task lighting. This takes the form of heuristic
analysis for optimal contrast lighting, dynamic radiometric compression for shadow-
free lighting, and other such adjustments that emphasize specific qualities and
quantities of light that enhance visual perception relating to particular tasks. The third
application area is semantic information-aware lighting which can add meaning in the
form of an overlay, everything from low-density ambient information to highly
focused, high-density media in an augmented reality fashion is possible and may be
modulated by task semantics and physical context.
3 Researchers
Zary Segall is a KTH Chair Professor in Scalable Mobile Services and the director
of the Mobile Media and Services Lab. Prior of joining the University of Maryland as
a Distinguished Professor, he was at University of Oregon as Professor and
Department Head, and a Professor at Carnegie Mellon University. As part of his
research activity, he had developed theoretical methods and practical systems for
parallel processing, highly dependable systems, networking and wearable information
systems. This work led to software licensing to IBM, AT&T, GE and NASA and to
applications to parallel processing, NASA missions, Air Traffic Control and
telecommunication services. His current research work is in Human Aware Wearable
Computing, Context Awareness and Mobile Systems and Services. Dr. Segall is a
fellow of the IEEE Computer Society, and a Fulbright Distinguished IT Chair.
Petro Lungaro holds a M.Sc. in Telecommunication Engineering from Politecnico
di Milano and a M.Sc. in Electrical Engineering from KTH. He has recently defended
his PhD thesis, which focused on exploring the area of context-aware and
opportunistic content provision in cellular network. His interests include radio
resource management, novel content provision paradigms and content-awareness both
at network and terminal sides.
Chad Eby is an assistant professor of Art and Design and co-director of the
Facility for Arts Research at Florida State University, and guest researcher at KTH,
Stockholm. His work explores the intersections of art and technology, particularly
8
issues surrounding mapping, translation and representation and the subtle interface
between humans and machines.
9
Interacting with Light Apps and Platforms
Serge Offermans, Harm van Essen, Berry Eggen
Eindhoven University of Technology, Department of Industrial Design
Postbus 513, 5600MB, Eindhoven, The Netherlands
{s.a.m.offermans, h.a.v.essen, j.h.eggen}@tue.nl
Abstract. In the near future, highly dynamic light sources will be embedded in
the areas in which we live and work, as well as in the objects within these areas.
Furthermore, all these light elements will be connected, and digitally controlled.
We believe that this development will turn our environments into lighting
platforms that will not only allow us to see our surroundings and perform our
tasks, but that will support many other functions and activities. This vision
requires us to rethink the way we interact with light.
Keywords: Light, Apps, Platforms, Interaction
1 Introduction
Modern LED light sources are small, energy efficient and durable. They are highly
dynamic and properties such as brightness, color (temperature), direction, and focus
can be can be easily controlled and adapted to our desires. The nature of this new type
of lighting will allow light sources to be embedded in the environments in which we
live and work. These light sources will be connected to each other, and can be
considered as a single platform.
Although the developments in artificial lighting are very rapid, the way we interact
with these light sources has hardly changed since the invention of electric light. We
still use switches to turn on a single lamp or a pre-defined group of lamps. In some
cases we are able to gradually dim our lights, or even to choose a preferred color.
Control is usually provided via switches or turning knobs, and in some cases using
novel sensors such as capacitive touch. Dynamically controlling settings like intensity
and color (temperature) of more complex sets of light sources is almost exclusively
done via complicated systems that either work autonomously (e.g. using timers), use
predefined presets, or are supposed to be used by trained professionals (e.g. for use in
theatres and clubs).
We envision a future in which we will no longer control individual light sources,
but rather interact with our environments as a whole. These environments can be seen
as lighting platforms that can provide various services. We will explore three areas
within this vision: the opportunities for lighting, the development of services and
platforms, and the development of new interaction styles for user-system interaction.
10
Opportunities for lighting. Light platforms will not be restricted to provide just task
lighting. The practical function of providing visibility to be able to perform a task is
just one of the roles of light. Studies on the effects of light can provide useful starting
points to find innovative concepts. This relates to physiological (or biological) effects
of lights or to psychological effects (perception and cognition) [1]. Besides the effects
on people, the use of light as an information medium is a promising direction. We
already get a lot of information from the light in our environment. Outside light subtly
tells us something about the time of the day. The lights in the neighbors‟ house lets
you know they are home. The term „Information decoration‟ [2] describes a class of
ambient displays [3] in which aesthetically presented information that unobtrusively
informs the user in the periphery of its attention. Light could very well function as a
medium in this concept. But if light is used as an information medium, then how will
this information be understood? And how is informative light distinguished from
illuminative light?
Apps and Platforms. Modern electronic products often serve multiple purposes.
They are systems that provide various services to the user. This trend is most obvious
in smart-phones where the platform (device) provides as many services as the user
desires. An added value for these platforms is especially created by the „user
generated content‟ for these platforms; meaning that new services can emerge and
build on other services. Lighting solutions could benefit from a similar structure: the
applications will determine the actual function and value of the system at a given
moment. Lighting solutions will shift from single-function luminaires towards light
„apps‟ and „platforms‟. These platforms can provide us with suitable atmospheres,
information about our environment, support our social connections, increase our
health and well-being and support our activities in many other ways. Applications can
be provided by experts, but perhaps also by the end-user themselves. Exploring the
potential of end-user programming of light apps is therefore an area of interest.
User-System Interaction. In contrast to the relation between a light switch and a
light bulb going on and off, the relation between a person and a light platform with
many light sources that each have various properties is not straight forward.
Controlling dynamic lighting that comes from many different sources creates the need
for new forms of interaction. Furthermore, the many different functions that such a
lighting system will perform in a dynamic context, creates additional challenges for
the interaction. A comprehensive interaction paradigm is required that allows the
various users to control, configure and switch between applications, and interact with
services that go well beyond controlling the lighting conditions.
The approach taken in this project is that of design-research [4]. We will develop
and explore new interaction styles with various light platforms and applications.
Evaluations of these systems will allow us to identify the common or valuable
elements of these interactions. Following iterations of the designs will allow us to
work towards a new paradigm for the interaction with light apps. In the remainder of
this paper we will introduce a case in which we will explore our research question in a
11
research through design methodology and we conclude the paper with formulating the
initial directions of research.
2 Case: the Modern Office
Offices are becoming more open and dynamic. People work on flexible desks in open
spaces, and the office provides new types of spaces for activities focused on for
instance social interaction, (informal) meetings and relaxation. The needs of the „new‟
office workers could be addressed and supported by dynamic lighting solutions. As
the spaces are so dynamic and serve many different purposes, addressing these needs
requires different applications that could all run on the same platform. The modern
office is therefore a suitable initial context to explore the opportunities of light apps
and the interaction with the platform.
An interesting type of space in the modern office is the „breakout area‟. This is an
area where people can have informal meetings, sit down to read, have a brainstorm or
just have a coffee. This space is particularly interesting as it serves so many different
functions, and can therefore benefit from a high variety of lighting solutions.
Furthermore it is an area in which can be experimented with light without disrupting
any regular desk work or regulations. Lighting solutions could support the different
activities in the breakout area by providing a suitable atmosphere or stimulate for
instance concentration, creativity, or relaxation. Light could also be used to create
separate zones in the breakout area to support the use of the area for multiple
activities that go on simultaneously. Finally, light could be used as an information
source, providing information about the use of the area, the people in it, or about other
things that are relevant to the different activities. An example of such a system was
developed by Occhialini et al. [5]. Their system supported timekeeping in meetings
using an unobtrusive lighting pattern on the wall that constantly informs the people
about the progress of their meeting.
Fig. 1. Prototype by Occhialini et al.; Light indicating the progress in a meeting
We have taken this concept and implemented it in an initial light app on a new
light platform element. This same platform element was also used for another app that
12
provides atmospheric lighting, and could again be used for numerous other
applications.
Fig. 2. Initial element of a lighting platform running the „meeting timer‟ app (left) and the
„atmospheric lighting‟ app (right)
3 Directions for Research
Considering the areas described in our vision on light platforms and the case of the
modern office, we can formulate directions for research within these areas; Light
Concepts, Service Infrastructure and User-System Interaction.
Directions for Light Concepts. As with the Smartphone, a large portion of the added
value is in the applications. Therefore, an important direction for research and
development is the exploration of interesting concepts for applications with light. We
will work according to user centered methods, exploring the (latent) needs of
Breakout users. For instance applying the knowledge about biological effects of lights
on our circadian rhythms can lead to more productive or comfortable environments,
while knowledge about psychological effects may lead to lighting applications that
support the inter-personal relations during a meeting. The use of light as an
information carrier is another starting point. Maybe lighting can tell you something
about the activity of your colleagues or the upcoming weather. Lighting could also be
used to display presence and availability information, or to communicate non-verbal
cues during meetings or activities.
Directions for Service Infrastructure. This is the area that will actually allow the
concept of light apps to become a reality. This is both a technical issue in terms of
connectivity, repositories and so on, but also a business and user issue. How will
people develop apps? Who will be app developers? What critical mass of apps has to
be developed for a community to take off? Can development take place on different
levels (experienced users, and novel users)? What standards and protocols need to be
established, and what will they look like? What business model will underlie this
concept? How will we integrate sensor networks and third party lighting equipment?
Where will people download or buy their apps?
13
Directions for User-system Interaction. The current focus of the design-research
efforts in the project is directed towards user-system interaction. It is tempting to say
that people will use their Smartphone to control the light. However, there are much
more attractive or suitable ways of controlling these light apps, depending on various
parameters such as the context and frequency of use. Especially the field of Tangible
Interaction [6] offers advantages such as direct control, and potentially meaningful
interaction. There are several challenges to be addressed in terms of interaction design
[7]. How to develop several comprehensive interaction styles that allow people to use
and configure very different applications in a meaningful way? What is the balance
between autonomous system behavior and user control?
Conclusion
In the future, our environments will contain numerous embedded light sources that
will together form light platforms on which various applications will run depending
on the current usage of an area. Besides interesting light concepts and a service
infrastructure, a new paradigm for interaction with these environments is required in
order to benefit from its full potential. Our current research frames these questions.
References
1. Knoop, M. (Martine): Dynamic lighting for well-being in work places: Addressing the
visual, emotional and biological aspects of lighting design., http://repository.tue.nl/666147,
(2006).
2. Eggen, B., Mensvoort, K.: Making Sense of What Is Going on “Around”: Designing
Environmental Awareness Information Displays. In: Markopoulos, P., De Ruyter, B., and
Mackay, W. (eds.) Awareness Systems. pp. 99-124. Springer London, London (2009).
3. Pousman, Z., Stasko, J.: A taxonomy of ambient information systems. Proceedings of the
working conference on Advanced visual interfaces - AVI ‟06. p. 67. , Venezia, Italy
(2006).
4. Edelson, D.C.: Design Research: What We Learn When We Engage in Design. Journal of
the Learning Sciences. 11, 105 (2002).
5. Occhialini, V., Essen, H., Eggen, B.: Design and evaluation of an ambient display to
support time management during meetings. Presented at the 13th IFIP Conference on
Human-Computer Interaction, INTERACT (to be published) (2011).
6. Ullmer, B., Ishii, H.: Emerging frameworks for tangible user interfaces. IBM Syst. J. 39,
915-931 (2000).
7. Bellotti, V., Back, M., Edwards, W.K., Grinter, R.E., Henderson, A., Lopes, C.: Making
sense of sensing systems: five questions for designers and researchers. Proceedings of the
SIGCHI conference on Human factors in computing systems: Changing our world, changing
ourselves. pp. 415-422. ACM, Minneapolis, Minnesota, USA (2002).
14
The future of interaction with light and lighting
dynamics.
Jettie Hoonhout*, Lillian Jumpertz**, Jon Mason*
* Philips Research Europe, 5656 Eindhoven, The Netherlands
{jettie.hoonhout, jon.mason}@philips.com
** Utrecht University, Department of Experimental Psychology
Abstract. Traditionally luminaire design required careful consideration of
the materials and forms that provided a surround or shade for an incandescent
lamp. The control and user interaction (UI) for such a luminaire was often a
switch or an analogue dial to vary the light output, and it has been this way for
almost a century. Within the next decade this is likely to change dramatically
with the introduction of the Light Emitting Diode (LED) into most modern
professional and domestic luminaires. The main reason for this change is that
the LED‟s attributes provide far greater design freedom than was available
before and the result is an opportunity for new lighting, products and services.
In contrast and on a darker note, an increase in design freedom can also bring
with it complexity and confusion for designers and end users alike. In this
paper, the key game changing attributes of the LED that contribute towards this
new design freedom, but also towards the potential complexity of future
luminaire design, are summarized. To facilitate with reducing at least one
aspect of this aforementioned complexity a research study into dynamic lighting
was undertaken; this provides an example for the type of exploratory research
work that may be required in the future to understand further the LED and its
potential. The paper concludes with a discussion into how these factors may
affect the UI for both the designer and end user.
Keywords. Lighting UI, LEDs, 3D luminaire design, perception
1 Introduction
The lighting domain is undergoing a major change as new LED technology is
superseding the much older incandescent, compact fluorescent and the halogen lamp
types. This is due to in part to the LEDs being more efficient, digitally controllable
and much smaller than these earlier lamp types. Nevertheless, it is almost inevitable
that with new design freedom this also induces new complexities that may need to be
overcome, and this is the focus of this paper.
The paper has been divided into three parts. The first part is a description of how
the LED‟s attributes over the traditional lamp types may affect the future of luminaire
design. The second part is an overview of a research study into the perception of
15
dynamic lighting which is just one example of the complexities that designers in the
future may face. The concluding part is a discussion on how these aforementioned
aspects may affect the future of UI for both the designer and the user.
2 Future LED Luminaires
For over a century the traditional methods for producing artificial light have been
incandescent or gas discharge, but this about to change with the advent of the LED.
The LED produces its light via the movement of electrons in a semiconductor
material. Digitally controllable, small size, variety in light output and efficiency are
some of key benefits, but also challenges that the LED introduces to the lighting
industry. These are summarized below.
Lighting in the future will become digital [4] since LEDs are semiconductors and
require digital input signals rather than analogue signals that many of the previous
lamp types used. This single attribute enables the LED to be programmed and
controlled via software. Furthermore, the connection of LEDs to other software
programs and devices, such as sensors, media output, networked inputs/outputs for
example is highly plausible.
While the digitization of light will result in an increase in design freedom it will
also increase the complexity of lighting and luminaire design. Designers will have to
consider more than just materials and optics but also the software and programming
that will control the LEDs within the luminaire. The light output can now become
dynamic to suit individual user‟s needs; however, the designer will have to understand
how to achieve this and the effects of different dynamic lighting on humans.
Dynamic lighting is when there is a variation in one or more lighting parameters such
as brightness, hue, saturation, source, etc. A flashing warning light or dappled
sunlight through a tree‟s branches are examples of a simple and a more complex form
of dynamic lighting.
Another key attribute of the LED is its size, which typically ranges from 2mm to
20mm. This is not the complete picture however, since the LED must also be
accompanied with a heat sink, optics and additional electrical components to drive
and protect the LEDs. Despite this, the physical size of LED luminaires is reducing
rapidly they will have a size advantage over the traditional lamp types. This enables
the LED to be hidden or embedded into materials and for the first time designers can
produce luminaires without necessarily requiring the normal lamp housings or
fittings.
LEDs are available in a variety of lumen outputs, colour temperatures and colours.
This variety, in conjunction with the LEDs smaller size, enables designers to position
different LED types within a single luminaire making it easier to produce multiple
light effects from a single luminaire.
Not only are LEDs more energy efficient than the traditional lamps they are more
energy efficient for a longer time. With life spans of up to 50,000 hours LEDs and
the LED luminaires are becoming practically maintenance free since there may be no
requirement to replace the LEDs for the entire useful life of the product.
16
The LED is bringing new freedom to luminaire designers but also, as described
briefly in this section, it is introducing more complexity. This phenomenon has
occurred before in the design domain with the introduction of additive manufacture
where designers must unlearn much of their previous knowledge on manufacturing
limitations in order to embrace this new technology [5]. In a similar fashion, the
advantages, flexibility and variability of the LED are simultaneously a blessing and a
curse.
An example of a LED luminaire is shown in figure 1. This was designed and built
at Philips Research as a exploration piece into new luminaire design. Aesthetically
the form reflects the past with the bulb shape of the glass and the illuminated coil in
the centre being analogous to the incandescent lamp coils. This luminaire was
designed to utilize the LED as much as possible. The LEDs (n=34) are hidden within
the luminaire‟s structure and the light is transmitted via optical light output guides.
The light output could be static, dimmable, and also highly dynamic. The lumianrie
was also designed to be connected to external sensors that could provide input for the
light output, e.g. when to illuminate and how. Programming the dynamics was a time
consuming task since the dynamic output was based on the designer‟s own intuition
and understanding of what may or may not „work‟; but this approach is not only time
consuming, it is also a rather unsystematic and unscientific approach. Insights into
how people may perceive certain dynamic lighting effects may have facilitated in
reducing the time required to design this luminaire since it would have contributed
towards reducing some of the complexity.
Fig. 1. Example of a LED luminaire
3 Research to Reduce Complexity
The aforementioned luminaire designed at Philips Research is an example of applying
the new design freedom that LEDs provides as well as highlighting areas of
complexity such as designing the dynamic lighting. Up until now many dynamic
17
lighting installations have relied on intuition and the personal aesthetic judgment of
the designer to determine whether a dynamic sequence of LEDs is sufficient for the
end luminaire. For example, refer to the sparkling LEDs used as Christmas tree
decorations. So, although dynamic lighting is already deliberately and intuitively used
for different environments – other examples include e.g. disco lighting, used to add to
the exciting atmosphere in a club; and twinkling lighting effects incorporated into the
ceiling of saunas to add to the effect of relaxing people – no research investigating
these effects of dynamic lighting has been found. The only studies on the application
of dynamic lighting that focus on the effect of different lighting variables, focus on
the perception of danger in warning signals (e.g. [2]), not on supporting different
atmospheres or providing a mood/feeling. The researchers involved in this study
were unable to retrieve any studies on this topic despite extensive literature searches;
this study was therefore largely exploratory in character. The aim was to investigate if
designers could be provided with guidelines for dynamic LED lighting to express
particular moods to be recognized as such by people when viewing such a luminaire.
NB at this point in time we do not want to claim that these moods might be evoked in
people – that might be a next step, for now we would be quite content with expressing
certain moods in a lighting sequence, for people to recognize and appreciate.
Designers may then be able, with more confidence, to create a luminaire that portrays
the desired dynamics that have the essence of an emotion and should help to reduce
some of the complexity of using LEDs and dynamic lighting in luminaire design.
The study described in this paper was conceived to explore the perception of
dynamic light patterns created by an array of individually controlled LEDs within a
single luminaire. The aim was to determine if certain sequences of dynamics would
indeed be perceived by a population as being identical to the mood intended by the
designer.
3.1 Design of the Study
For the purpose of the study, a luminaire was built which enabled various dynamic
light sequences to be presented to participants. The first requirement was to develop
dynamic light sequences with the intention that they would represent different moods
e.g. ranging from calm to more exciting. These were then presented to participants to
investigate two aspects. The first was could the participants differentiate between the
dynamic light sequences presented. The second was to determine if they perceived
the different sequences as the designer had originally intended and whether they were
appealing.
The luminaire that was built consisted of: a black square board (600 x 600 mm); 17
flexible rods suspended underneath; and 24 pairs of LEDs mounted onto these rods
(see Figure 2). For the test, 8 different dynamic light sequences were created, using
the circumplex model of affect [1] as the starting point (see Figure 3). Four distinct
moods of the circumplex model were chosen: serene (highly positive, low arousal),
calm (slightly positive, very low arousal), happy (highly positive, high arousal), and
excited (slightly positive, very high arousal). Since the purpose was to create
appealing, pleasant light sequences, only moods on the positive side of the
unpleasant-pleasant scale were used as starting points, although it could have been
18
interesting as well to include one or more sequences based on the negative side of the
scale, for comparison sake; however, in order to keep the number of sequences
presented to the participants within a reasonable range, the decision was made not to
include „unpleasant‟ sequences.
Fig. 2. The luminaire used in the study
Fig. 3. The circumplex model of affect (Russell, 1980). The horizontal axis represents unpleasant-pleasant and the vertical axis represents arousing-sleepy.
19
Participants (n=23 11M + 12F) saw the 8 sequences in a random order. For each
sequence, they were asked to first say out loud the things that immediately came to
mind while looking at the sequence. Next, they were asked to fill out the Self-
Assessment Manikin (SAM), which consists of 3 subscales – Happy-Unhappy,
Excited-Calm, Controlled-In Control, and is based on the circumplex model of affect
[3]. Lastly, they answered some questions about each sequence, such as what they
liked or disliked about it, and if and how they would want to improve it. At the end of
the test session, participants answered general questions about the appeal of such a
luminaire, and for what situations and purposes they thought it would be fitting.
4 Findings and Discussion
The first aspect investigated was to determine whether or not the participants could
differentiate between the dynamic light sequences presented to them. The results of
the (SAM) were analysed for each of the three scales separately in order to find out
whether the 8 sequences differed significantly on these scales. For the Happy-
Unhappy scale and the Excited-Calm scale, there was a significant difference between
sequences (Friedman‟s test: χ2(7) = 28.67, p < .001 and χ
2(7) = 50.52, p < .001,
respectively).
Therefore, these results indicate that the participants could indeed see differences
between the dynamic light sequences presented to them. For example they could
determine that one light sequence was more or less „excited‟ than another.
The next aspect was to determine whether the participants would perceive the
dynamic light sequences as the designer had intended: were the sequences
representing serene, calm, happy and excited described as being so?
Figure 4 shows the box plots of the SAM for the happy-unhappy rating scale. The
expectation was that all the 8 light sequences would be perceived as being more
pleasant than un-pleasant with regard to the circumplex model. It was also expected
that the serene (1 and 5) and the happy (3 and 7) would be rated as being more happy
since those sequences were more energetic than the calm sequences but less so than
the excited.
The results showed that the serene sequence (2) was considered to be the most
happy along with excited (4). The happy sequences (3 and 7) were considered to be
relatively less happy than expected.
20
Fig. 4. Boxplots of the scores on the Happy-Unhappy subscale of the SAM. High scores represent unhappy values. Sequences: 1=serene; 2=calm; 3=happy; 4=excited;
5=serene; 6=calm; 7=happy; 8=excited. Sequences 1-4 used a “random” programming of the dynamics, sequences 5-8 used a repeated patterns program.
Figure 5 shows the box plots of the SAM for the excited-calm rating scale. The
expectation here was that the excited (4 and 8) and the happy (3 and 7) sequences
would be rated as relatively excited. In contrast, the calm (2 and 6) and the serene (1
and 5) sequences would be rated as being more calm. Indeed, the results did show
this to be case; however, the differences between serene and calm, and happy and
excited respectively were less distinct.
These results show that it is possible to create dynamic light sequences which
express various moods. More specifically, people appear to be able to discriminate
between sequences which are perceived as low in arousal and those which are
perceived as high in arousal.
Fig. 5. Boxplots of the scores on the Excited-Calm subscale of the SAM. High scores represent calm values. Sequences: 1=serene; 2=calm; 3=happy; 4=excited;
5=serene; 6=calm; 7=happy; 8=excited. Sequences 1-4 used a “random” programming of the dynamics, sequences 5-8 used a repeated patterns program.
0
1
2
3
4
5
6
7
8
9
10
1 2 3 4 5 6 7 8
Rating H
appy-
Unha
pp
y
Settings
0
1
2
3
4
5
6
7
8
9
10
1 2 3 4 5 6 7 8
Rating E
xcited -
Calm
Settings
21
This study was also an opportunity to question the participants, who had just
experienced dynamic lighting, on how much they liked the dynamics and to explore
the idea of owning or using a luminaire with this kind of functionality. Overall, the
participants did like the sequences (mean 7.4), enjoyed them (mean 7.6), and they
expressed an interest in owning such a luminaire (mean 7.3). Only one participant did
not like the sequences at all; the rest of the participants either liked them, or saw the
potential in them which is impressive considering they were viewing a technical
prototype.
Several reasons were given by the participants why they liked the dynamic light
sequences: because it can be used to create different atmospheres and to affect
emotions and behaviour; because it can be used in different sequences and for
different uses; because it is different from other luminaries; because of its variety; and
because it is decorative, pretty, dynamic, interesting, surprising, unpredictable, and
nice to look at. Participants who did not like or enjoy the sequences mostly mentioned
the same factors as were described above when reviewing positive and negative
features of the sequences: high frequency and abruptness. Furthermore they
mentioned restlessness and distraction.
Finally, environments or situations that were seen as appropriate by the participants
for a luminaire like this were in the home, hotels, shops, outdoor places and waiting
rooms.
With findings such as these there appears to be potential for applying more precise
dynamic light sequences/guidelines to the design of luminaires or lighting
installations in order to help reduce the complexity. This work is also an example for
how research can facilitate design in this domain and can perhaps be applied to UI
and other aspects of LED luminaire design.
5 Future of UI for LED
Thus far in this paper, a number of attributes of LED technology has been described
which presents the designer (and user) with more design freedom but also increased
complexity. An example of how research can be applied to reduce aspects of this
complexity has also been described but, how could this facilitate UI?
How this question is answered depends on who the user is. In this dynamic
lighting case, the user could easily be the designer who needs to interact with software
that composes the dynamics for the LEDs. The software could follow the guidelines
found in the previously described study and provide the designer with inputs and
suggestions for the type of mood wish to create.
Alternatively, the guidelines from the study could provide designers with more
confidence that a certain dynamic sequence of LEDs is perceived in a particular way
by end users and thus the designer can design the luminaire‟s UI to match that
perception. A happy dynamic should not perhaps be activated using a very slow and
relaxing UI style.
As end users become more familiar with software and media connectivity, they too
may wish to design their own lighting and lighting dynamics. Luminaire designers
may design the basic lumianire for which the end users then program the light output
22
they wish to have. The software the end users experience should also have guidelines
to ensure that they too can produce high quality results. This may appear farfetched,
but when considering that an LED luminiare may last for 22 years* the luminaire may
reasonably outlive the UI. The UI may need to be designed to be updated or replaced
with a more suitable and state of the art technology. In a similar way that the
incandescent lamp provided companies with regular turnover when the lamps needed
replacing, the UI means may take on this role. Users could choose to change their UI
to suit their changing needs and enable them to design or purchase new behaviors for
their luminaires.
Research is needed to help provide the necessary guidelines to support the future of
lighting and the UI for lighting so that designers and even end users can operate and
change their lighting in a meaningful and understandable way. The balance between
those who create the luminaires and those who merely operate could easily change in
the future due to the LED and the digitization of light.
* (50,000hrs/6hrs usage perday)/365 = 22years approximate lifetime.
References
1. Russell, J.A. (1980). A circumplex model of affect. Journal of Personality and Social
Psychology, 39(6), 1161-1178.
2. Chan, A.H.S, & Ng, A.W.Y. (2009). Perceptions of implied hazard for visual and auditory
signals. Safety Science, 47, 346-352.
3. Bradley, M.M. & Lang, P.J. (1994). Measuring emotion: The self-assessment manikin and
the semantic differential. Journal of Behavior Therapy and Experimental Psychiatry,
25(1), 49-59
4. Aarts, E. 2011. Liberation of Light. TEDx Talks Presentation, 13.05.11. URL:
http://www.youtube.com/watch?v=FZ93BxvIzGY [Accessed 22.07.11].
5. Campbell, I.R. 2011. Additive Manufacture. [Presentation] (Invited visit to Philips
Research, 14 July 2011).
23
User Interface for Task Lighting in Open Office
Koen van Boerdonk, Jon Mason, Dzmitry Aliakseyeu
Philips Research Europe, 5656 Eindhoven, The Netherlands
{koen.van.boerdonk, jon.mason, dzmitry.aliakseyeu}@philips.com
Abstract. In this paper we explore the use of LED-based lighting in the office
environment. We have identified that the ideal office lighting should be capable
of illuminating the entire desk as well as parts of the desk (varying beam sizes
and locations), and enable control of the illuminance level and the colour
temperature. We have implemented two user interfaces for controlling these
parameters: one with individual control of light parameters and another –
preset-based where all parameters are set simultaneously based on the selected
light scene. The evaluation of these interfaces have shown that majority of
participants prefer the interface with individual control.
Keywords: Lighting; User Interaction; LED; Open office; Personal Desk
Luminaire
1 Introduction
Nowadays many offices have static lighting systems that maintain a constant light
level, which is defined by national and international standards. These standards ensure
proper lighting conditions in the office for the average office worker, but neglect
interpersonal differences with regard to office lighting. Included in these standards are
for example minimum illuminance (lux) levels in relation to specific tasks; however,
multiple office workers may be working on different tasks at the same time in the
same space and they may require different illuminance levels. This is confirmed by
preceding research that have shown that when given a choice many office workers
select illuminance levels that are different from those prescribed in the standards [1-
7].
With the introduction of LED-based lighting systems it is now easier to include a
wider spread of controllable lighting parameters within a single luminaire than just
the illuminance level; however it is unknown what parameters, in addition to
illuminance, office workers would want to control. Such parameters include the
colour temperature, hue, saturation, as well as the location and adjustability of the
light beam (size, shape etc.).
24
Fig. 1. Luminaries used in the study
This position paper is organized as follows: we first describe a study that was
carried out to identify what parameters of task lighting office workers would want to
control; we then present the design and implementation of two user interfaces for
lighting control; and finally we discuss the preliminary results of the user evaluation.
2 User Study
To identify the parameters of task lighting that should be controllable, a qualitative
user test with 20 participants (flex office workers) was set up. In this study a projector
was used to project (imitate) different light parameters and light settings (combination
of parameters) onto the desk to help people see precisely what the light output would
be like. The study revealed that the ideal office lighting should be capable of
illuminating the entire desk as well as parts of the desk (varying beam sizes and
locations), and enable control of the illuminance level and the colour temperature.
The participants were not interested in multiple light beams, coloured light or
adjusting the light beam‟s shape (rectangular or round spots). Participants also
indicated that controlling the light should be undertaken in a fast and simple way.
3 Design and implementation
Based on the results of the study two distinct types of UI were explored:
UIs with individual control of illuminance level, color temperature and light
location. This UI was realized as a form of multi-touch desktop interaction.
UIs based on presets where all parameters are set simultaneously based on
the selected light scene. This was realized using a handle that was tilted
from one side to another.
25
3.1 Touch UI
With the touch interface a user can set their task lighting by controlling three
independent light parameters. The interface uses the desk‟s surface as a „touchpad‟ in
which the selectable options are projected onto the desk‟s surface by the luminaire.
Capacitive sensors attached under the table are used to capture users‟ input.
The light is set in three steps, at every step the system projects a variation of one
light parameter across the desk (see figure 2). The user has to touch the desk to set a
light parameter he or she wants, after which the next parameter is projected. Since
light options are projected by the luminaire itself, the users can see directly what light
they will get.
To switch on the luminaire and to change the parameters when luminaire is already
on the user would touch the desk where the two virtual buttons are projected (see
Figure 3).
Fig. 2. The three steps of the touch interface: setting illuminance, color temperature and light
location (two hands are used to set this parameter)
Fig. 3. Virtual buttons for on/off and settings control
3.1 Preset UI
Using the preset based interface the user controls the light by activating light scenes
using a handle that can be tilted over 180 degrees (see figure 4). The angle in which
the handle is tilted determines the light scene provided. Four main light scenes (calm,
neutral, energetic and productive) are divided across the tilting range, but all the
angles between any two scenes will result in a mixture of those scenes. This means
26
users can have more freedom when selecting the light they would like to have.
Fig. 4. Preset based UI: (left) physical handle; (right) scenes and angles
4 Evaluation
These two UIs were evaluated for one week by four participants from Philips IT;
Philips IT practice flex space working in their offices and thus these participants were
familiar with the need for flexible lighting. The goal of the study was to compare
these two interfaces to determine which type of control means the participants would
prefer for the longer term: precise interaction versus more abstract, preset-based
interaction. Two desks with the two UIs were setup in a standard office (see figure 1)
where two participants undertook their real work for one week. This was repeated for
the second pair of participants.
Each study started with an introduction meeting in which the participants were
instructed about the study and the user interfaces they would be using during the
week. In the middle of the week the participants were asked to swap desks, so that
each participant could experience both UIs. For the whole week the participants
worked with the luminaries while performing their regular working tasks. At the end
of the week the participants were asked their opinion of the two interfaces in an
individual interview sessions.
The results showed that three of the four participants preferred to control individual
parameters (precise) over the control of predefined light scenes (abstract). The fourth
person did not have a preference regarding the UI but stressed that there should
always be enough light to work by. The three participants with a preference for
controlling individual parameters have commented that they would also prefer to be
able to save their personal settings for easy recall.
During the interviews four main factors that influenced the users‟ preferences were
identified: time to set up the light; ease of use; expected duration of work at that desk;
and the quality of initial light conditions for a specific task. Although all the
participants considered the same four factors the judgement was subjective and varied
from person to person.
27
References
1. Kate E. Charles; Alison J. Danforth; Jennifer A. Veitch; Christina Zwierzchowski; Byron
Johnson; Karen Pero, (2004), Workstation Design for Organizational Productivity
2. Veitch, J.A., (2001). Lighting quality contributions from bio-psychological processes.
Journal of the Illuminating Engineering Society, 30(1), 3-16
3. Gornicka, G.B. (2008) Lighting at work. Environmental study of direct effects of lighting
level and spectrum on psychophysiological variables. Dissertation, Eindhoven University
of Technology, Eindhoven;
4. Boyce, P.R. (2004). Lighting research for interiors: the beginning of the end or the end of
the beginning. Lighting Research and Technology, 36 (4), 283-294
5. de Kort, YAW; Smolders, KCHJ, (2010). Effects of dynamic lighting on office workers:
First results of a field study with monthly alternating settings
6. Veitch, J. A.; Newsham, G. R. (2000). Preferred luminous conditions in open-plan offices:
Research and practice recommendations. Lighting Research and Technology, 32, 199-212.
7. van Bommel, WJM; van den Beld, GJ, (2004). Lighting for work: a review of visual and
biological effects, Lighting Res. Technol. 36,4 (2004) pp. 255–269
MeShirt: concepts for provocation and promotion
Alessandrini Andrea1, Erik Grönvall2, Paola Manuli1, Valentina Sanesi1, Simona Melaragni1, Maria Teresa Oliviero1,
1 University of Siena, Communication Science Department, via Roma 56,
53100 Siena, Italy 2 Aarhus University, Department of Computer Science, Aabogade 34
DK-8200 Aarhus N, Denmark
[email protected], [email protected], [email protected], [email protected], [email protected], [email protected]
Abstract. People have a will and need to express themselves. During the last years, technologies such as Web 2.0 and blogs have made it easier for people to create and make publically available their thoughts, reflections and ideas. However, in contradiction to the Web 2.0 paradigm, this content and its creation is strongly connected to a server infrastructure and to some degree hard to use by people with limited computer skills. MeShirt is a design concept of an interactive t-shirt that allows text and simple graphics to be visualized, using the fabric as display. A number of interaction modalities have been investigated that can make MeShirt a distributed, but personal, information system. The MeShirt design allows users in real-time to express their opinions and comment on the world around them, in a citizen journalism spirit. MeShirt provoke, inspire and create a stage for people to share, discuss and to find solutions for local problems and situations. The project initiated with a benchmark activity of enabling technologies, followed by scenario creations, attribute listing and other creative methods. A series of mock-up sessions, both internally and with potential users has been carried out to make the concept stronger.
Keywords: Personal expressions, t-shirt, wearable lighting, networked, distributed, contagious, society
1 Introduction
MeShirt is a design concept that allows us to explore how humans express themselves, communicate and how ideas spread in an ‘ad-hoc’ human driven network. The core of the MeShirt concept is an interactive t-shirt that allows text and simple graphics to be visualized, for example connected to Twitter feeds, using the fabric as display. The concept also allows a number of interactions through touch and permits users in real time to project their opinions and comments to the world around them, in a citizen journalism spirit. We intend citizen journalism as “playing an active role in the process of collecting, reporting, analyzing and disseminating news and information" [1] MeShirt provides a stage for people to share ideas or to discuss and
28
to find solutions about local problems and situations. In this way MeShirt became a distributed, but personal, information system.
Spuri describes the communicative potentiality through verbal messages, graphic symbols, photos, etc. that can be embodied in a T-shirt [2]. Indeed, a T-shirt can communicate dreams, passions, believes and the way people (i.e. the T-shirt wearers) are. MeShirt support the user’s will to express herself, but still allow a continuous adaptation and personalization of the T-shirt; so people will be able to assemble an outfit that connect the traditional usage of clothing with its potential new usage: ad-hoc communication, information or entertainment [3].
Figure. 1: : MeShirt system picture and mockup
The MeShirt (see Figure 1) concept supports the users’ will to express themselves through two different interaction modalities, where one modality does not exclude the other. 1) “single user modality”; users receive new text to be displayed on MeShirt from different information beacons like a library, bookstore, special events such as a festival or by input from a smartphone, for example utilizing Twitter or Facebook updates. 2) “multi user modality”; where two or more users exchange text through body contact like a handshake. ‘Contamination’, (exchanging or spreading a message from one T-shirt to another) is possible using any body surface, such as the hands, finger, arms, lips, face, legs or torso, thanks to MeShirt’s bluetooth technology and Smart phone creating a Personal Area Network with the Me-Shirt. MeShirt offers a range of usage scenarios, such as in festivals, games and education but in this paper the focus is on a t-shirt to express oneself - telling a story incessantly evolving through interaction with other inhabitants of the city. MeShirt can be continuously personalised so users can always wear their latest ideas and thoughts. The MeShirt concept took inspiration from Dan Sperber [4]; he says that ideas govern our behaviour and they can be transmitted from one person to another. This property of ideas permits them to be spread and changed, i.e. ideas are contagious like diseases. MeShirt aims to take advantage of this mechanism to help people to express themselves, to quickly spread their thoughts and ideas in a simple, distributed fashion.
We considered performances that play with lights and luminated words (e.g. Jenny Holzer [5] and Lozano Hemmer [6]) and also wearable technologies (e.g. Hug Shirt [7] and Lumalive [8]). Based on a technology feasibility study, we identified two example technologies that could allow a future development, changing a t-shirt into an interactive and personalized MeShirt: Lumalive and Red Tacton [9]. Lumalive is a photonic textile, which uses cloth as a lighted graphic display medium. Developed by Philips Research Technologies, it displays graphic, text and animation. Lumalive integrates flexible arrays of multi-coloured LEDS into fabrics. Red Tacton is a human
29
area networking technology that uses the surface of the human body as a data communication path. This could allow communication between two MeShirt wearers through touch.
2 Best Solution Scenario
In order to explore the potentiality of MeShirt, we developed the Best Solution scenario, an envisioned system to find the “best” solution to citizen problems (See Figure 2). Best Solution offers a voice to the local community to bring forward important local problems in a way which provoke and promote solutions. Through this system each person has the possibility to make visible a local problem and spread it wearing MeShirt. This requires a MeShirt owner to define an important problem via a mobile or online system, and then she/he suggests a possible solution to that specific problem. I.e. how she/he would like to solve a specific problem proposing a personal suggestion, provocation, or promotion by making it visible on MeShirt.
Figure. 2: Best Solution scenario
Other citizens can respond to the problem notifications visualizing different solutions on their own MeShirts, or they can adopt a solution proposed by someone else. In this last case, the system counts the number of identical solutions for the same problem. The main idea is that one person can propose and be the carrier of a solution and his mission is to ‘contaminate’ as many people as possible to adopt the same solution to a local problem, and hence became the Best Solution. Through simple touch, like a handshake, a solution spreads over town, from one MeShirt to another. All problems, provocations and solutions are presented online, accessible both from a mobile app and on the official web site of the municipality. From here, the municipality hear and valuate the citizen’s voice through Best Solution, in fact, the most diffused and adopted solution, i.e. the more ‘infected’ solution among citizens will be sent and evaluated by the mayors office who will be made aware of this particular citizen’s problem, and the proposed Best Solution. The MeShirt owners with the solution that in one certain period of time was most displayed on local MeShirts, will be awarded with the text “Best Solution”, which will be send to their MeShirts by the Municiplity. Best Solution stimulate and involve everyone by communicating that your contribution has been essential to solve a specific problem that now, thanks to
30
MeShirt has become an issue involving the whole town. This is a new way to expand the potentiality of citizen journalism that informs municipality and institutions about the most worrying/important problems, according to the citizens’. Citizens play an important role because their active and provocative participation make them reflect on their real and perceived problems. Thanks to MeShirt citizens can contribute to make their city and neighbourhood more pleasant using a playful, simple, and unobtrusive system.
3 Discussion
The MeShirt design concept opens up for a new perspective of citizen journalism. The concept explores scenarios for supporting people’s need to spread ideas and believes. MeShirt allows users in real-time to express their opinions and comment on the world around them, in a citizen journalism spirit. Indeed, the MeShirt concept can be perceived as distributed, real-time citizen journalism tool. The design concept has been evaluated through workshops and role-playing together with potential users. We developed scenarios, one being Best Solution, based on the idea to allow people to create and contaminate ideas through the functionality enabled by MeShirt. MeShirt provides people with a platform to actively and easily contribute to the citizen journalism practice on a local prospective, governing a democratic society. We like to permit everyone to express themselves in real-time, simply walking in the city, through a t-shirt. Through Me-shirt, users can upload a text or a simple graphic that is immediately visible, and through gesture and touch different ways to interact are enabled, allowing information exchange between numbers of MeShirts. To enable users to also change text through a simple handshake, this system can make the expression more easily for common laypeople. The work presented in this paper provides one example of use, but MeShirt is designed as an open-ended system, not trying to limit the users to merely one application field.
References
[1] Citizen journalism - [Online]. Available: http://en.wikipedia.org/wiki/Citizen_journalism. [2] C. Spuri, T-shirt, il tatuaggio di stoffa. Storia e attualità formato maglietta, vol. 7. Tunué,
2006. [3] E. H. L. Aarts e S. Marzano, The new everyday: Views on ambient intelligence. 010
Publishers, 2003. [4] D. Sperber, Explaining culture: A naturalistic approach. Wiley-Blackwell, 1996. [5] Jenny Holzer - Projections. [Online]. Available: www.jennyholzer.com. [6] Rafael Lozano-Hemmer. [Online]. Available: www.lozano-hemmer.com. [7] CuteCircuit � Wearable Technology. [Online]. Available: www.cutecircuit.com. [8] Bring spaces alive - Philips. [Online]. Available:
www.lighting.philips.com/main/application_areas/luminous-textile/index.wpd. [9] RedTacton. [Online]. Available: www.redtacton.com.
31
Towards Efficient Illumination Control forUnderground Parking
Paulo Carreira and Renato Nunes
INESC-ID and Instituto Superior Tecnico, Avenida Prof. Cavaco Silva, Tagus Park,2780-990 Porto Salvo, Portugal
{paulo.carreira, renato.nunes}@ist.utl.pt
Abstract. Increasing the illumination efficiency of underground park-ing is challenging mainly because traditional control strategies have lim-ited applicability to this type of spaces. LED illumination is making itsway into underground parking due to their greater lighting efficiency.However, LEDs bring along new control possibilities that are yet to beexplored and can further increase the energetic efficiency of undergroundparking. In this paper we address this problem and propose new illumi-nation control strategies that leverage the unique features of LEDs andtake into account the specificity of parkings and their usage patterns.
1 Introduction
On a variety of facilities the illumination of underground parking is a greatspender of electricity. For example, the illumination of a commercial mall consistsof thousands of points of light, half of those are installed in underground parking.Thus, improving the illumination efficiency of underground parking is of utmostimportance. Although conceivably simple, improving the illumination efficiencyof underground parking can be quite challenging due to the constraints posed onthe lighting scheme by safety and marketing. Efficiency is frequently sacrificedto make the customers feel welcome. It has been observed that to feel safe,people need to be able to see clearly the surroundings and recognize a face asfriendly or unfriendly at a distance of 60m and, moreover, there should be nodark or shadowed corners. Technically speaking, the illumination of undergroundparking requires (i) a comfortable average horizontal luminance level with (ii) agood light uniformity and (iii) a good chromatic restitution [5]. To meet theserequirements in a cost effective way fluorescent illumination has been prevalentin underground parkings.
LEDs are slowly making their way into underground parking, mostly in lampretrofitting energy efficiency initiatives. Studies show that LED light sourcesat the same brightness can save up to 50% of energy. However, to achieve thesame uniformity as a fluorescent lamp, the placement of LED luminaries must bereconfigured which is very labour intensive. Although this issue will be overcomein the coming years with technological advances in the optic components ofluminaries, we believe that it is possible to take illumination efficiency in thesespaces one step further.
32
Another way to increase energy efficiency of illumination is to take advantageof automated control techniques [1]. Illumination systems performing daylightharvesting or occupancy detection have been shown to achieve reductions ofenergy consumption around 30% [2]. In underground parking there is virtuallyno daylight, which prevents daylight harvesting technique from working, andspaces are usually ample with hundreds of meters in line of sight, that undermineoccupancy detection. Thus, traditional illumination control strategies do notapply to underground parking or apply only to a very limited extent.
We conjecture that some features of LEDs open an array of control pos-sibilities: LEDs start instantaneously, they can be dimmed at a very low costand their color characteristics can be controlled digitally. Moreover they have agood chromatic restitution, which helps in identifying faces and activities at adistance for safety purposes requiring lower luminance. Finally, developments inLED technology also point to the possibility of installing a large number of lowcost units that will lend themselves to computerised individual control.
Our paper is organized as follows. In Section 2 we present the backgroundconcepts related to illumination control and then, in Section 3 we briefly overviewthe traditional illumination control strategies. Section 4 discusses the new tech-niques we propose and finally Section 5 closes with challenges and directions forfurther research.
2 Zoning and Flow control
A visual task is any task that can be assessed according to lighting requirements.It can be parking a vehicle, walking on the parking, or deciding to buy a productat a shop. Different visual tasks require distinct lighting. Illumination controlsystems are capable of creating lighting scenarios with varying degrees of detailthrough two fundamental control strategies1. Controlling groups of luminariesarranged into zones independently from one another, a technique is known aszoning, and they are also capable of creating light scenarios with varying levelsof intensity, a technique known as flow control.
Since lighting of different areas (zones) can be controlled independently, in-stead of illuminating the whole space, only the area where a visual task is takingplace is lit while surrounding areas can save energy. Zoning can be coarser orfiner. The coarser type of zoning control consists of allowing the control of onlylarge sets of luminaries at once, while the finest type of zoning consists of allow-ing the individual control of each luminary. Finner zoning is more flexible andenables higher energy savings.
Flow control consists of changing the amount of light in a given space. Thereare basically two types of flow control: discrete and continuous. Common imple-mentation of discrete flow control consists of simple on/off control. Another formof discrete control is bi-level dimming obtained by actuating on dual-lamp lumi-naries or interleaving luminaries on and off. Continuous flow control is obtained
1 For now, we are leaving color out of our discussion.
33
by dimming lighting intensity using specific electronic ballasts. Unoccupied ar-eas can be dimmed down. Discrete flow control often results in a weak lightuniformity with annoying shadows. Thus to achieve a comfortable light level,the space has to be illuminated in excess. Therefore, continuous flow control ismore energy efficient.
To be cost effective, fluorescent lighting control currently deployed are coarsezoned and more discrete in terms of flow control. LEDs are more akin to finezoning and continuous flow control.
3 Traditional control techniques
Daylight harvesting refers to controlling illumination, i.e., artificial light to takeadvantage of natural light. By definition, underground parking has little or nonatural light available, which makes this technique impracticable.
Occupancy-based control aims at switching off illumination when the spaceis unoccupied and has proven to be an effective way to reduce energy consump-tion [3]. Occupancy is detected through passive infrared sensors which are knowto be unreliable and fail to detect occupancy whenever a human subject staysidle. Therefore a timer is used to uphold the on state for a period of time es-timated to be sufficient for the visual task to be completed. If the period istoo long, illumination ends up staying on longer than needed; if the period istoo short, illumination gets turned off leaving the occupant in darkness. Thisis known as a false off. In underground parking, pedestrians can be idle on awalkway, or simply stay in the car without leaving. Also it is difficult to assurea complete coverage of all the space. So, it is very hard in practice to be surethat the space is unoccupied. Since false offs are unacceptable, occupancy-basedillumination control is somewhat limited.
Scheduled shutdown can be used to completely turn-off illumination at clos-ing hours. The scope of this functionality is limited in our setting since illumi-nation will be turned off only after the emptiness of the parking is verified by ahuman.
4 Innovative LED-based underground lighting control
The new control strategies that we propose take advantage not only of LEDfeatures but also of aspects related to the reaction of the human eye to light.A linear decrease in luminance is perceived logarithmic In fact people cannotreliably perceive reductions of 20% in luminance level [4]. Another observationis that we have a greater sensitivity to luminance level transitions (contrast) bycomparison with a global luminance level (brightness). From the former obser-vation we conjecture that it may be feasible to achieve interesting energy savingswith marginal costs on visual comfort, while the latter points to the possibility ofmaintain lower luminance levels in some areas as long as luminance transitionsare smoothed. We also note that the eye takes time to adapt when transitioningfrom a brighter environment to a darker environment. Hence, luminance levels
34
along access paths should be set in progressive intensity to help the eye adapt.Finally, the perception of luminance decreases with the distance to the sourceof light (differences in intensity levels become indistinguishable at a distance).Conceivably, lower luminance can be applied to zones that are farther from theobserver.
We envision implementing a fine zoning and continuous flow control illumina-tion system consisting of a tight matrix LED luminaries driven by a distributedcontrol system. This control system aggregates occupancy information from amesh of inexpensive occupancy and car parking sensors installed trough out thespace. The underlying idea is to minimize the overall illumination intensity whilemaximizing visual comfort for the users of the space. The system aims at creatingthe most adjusted visual scenario for each focus of activity such as a car arrivingor leaving or pedestrians walking. The control system will take advantage of aset of illumination control strategies that can be summarised as follows:
Adaptive compensation consists of using progressive light levels to assist theeye adapting to luminance variance while commuting between spaces. Tran-sitions between the inside and outside of the parking should be brighterduring the day and dimmed during the night period, accompanying the vari-ations of external light levels. The illumination of access walkways to themall can be managed progressively to help the eyes adapt to the increasedambient light level of the mall. This feature enables the parking area to bemaintained at a lower overall intensity level.
Occupancy prediction consists of predicting when a space is about to beoccupied or unoccupied and create the appropriate illumination scenario,which will be dimmer when the space is unoccupied thus saving energy onvacancy. During certain periods certain zones of the parking alternate fre-quently between occupied and vacant and this technique takes advantage ofshort vacancy intervals frequent in parking that are not explored with flu-orescent illumination to save the lamp lifetime. When someone is about toenter the space it should be instantly illuminated. For example, dependingon the place where a car was parked may hold the illumination on that areauntil the guest crosses the access door. However, for some reason the guestmay not cross the door. In that case the system, after a certain time, maystart to dim slowly the illumination to save energy. Occupancy sensors maybe spread on the parking to revert the dimming process whenever activityis detected.
Activity spotlight refers to illuminating with higher intensity the area whereactivity (a visual task) is taking place. We conjecture that users are morelikely to accept a lower global intensity levels as long as the luminance level ishigher near the place where they are standing. This feature has the benefit ofhighlighting any other activity in the parking making the persons feel safer.This technique saves energy by keeping the luminaries in between the areasof activity at a lower luminance level.
Progressive spacial dimming consists of progressively dimming the luminar-ies farther from the focus of activity. Since parkings are characterized by
35
being open spaces with long lines of sight, in order to ditch shadows, a greatnumber of luminaries have to be switched on whenever activity is detected.Dimming can be used to a greater extent if the structure of the space arehighlighted. Users tend to feel more comfortable whenever they feel they cancorrectly read the space on visual contact. Energy savings will follow fromdimming luminaries progressively with the distance to the source of activity.
5 Conclusions
Although the replacement of current fluorescent illumination by LEDs alreadyoffer efficiency gains, we argued that current illumination control techniques areof limited applicability to underground parking and proposed a new range ofcontrol strategies that can further increase the energy efficiency of LED-basedunderground illumination.
In this paper we championed illumination control systems for undergroundparking which explore the individualized control with dimming (fine zoning andcontinuous flow) features of LED illumination. We conjecture that using thesestrategies it is possible to reduce the overall luminance levels of parkings, withcorresponding energy savings, while maintaining visual comfort levels.
Implementing on such system presents several challenges. Since there is nocheap way to accurately determine whether the space is vacant or occupied, weenvision integrating information coming from parking sensors, passive infra-redsensors and door sensors. Another source of uncertainty is how to determine theminimum comfortable luminance levels and whether a minimum safe luminancemust be kept at all times for security reasons. It is also unclear what will bethe impact of these techniques in the perception of the users. These issues, webelieve, will have to be determined experimentally.
We are currently in the process of formalizing a consortium some of the majornational mall management companies to implement a pilot test of the ideas wehave presented.
References
1. C. DiLouie. Advanced Lighting Controls: Energy Savings, Productivity, Technologyand Applications. The Fairmon Press, Inc., third ed. edition, 2006.
2. A. D. Galasiu, M. R. Atif, and R. A. MacDonald. Impact of window blinds ondaylight-linked dimming and automatic on/off lighting controls. Solar Energy,76(5):523 – 544, 2004.
3. X. G. D. T. G. Henze and C. Waters. The performance of occupancy-based lightingcontrol systems: A review. Lighting Research and Technology, 42(4):415–431, 2010.
4. T. Shikakura, H. Morikaewa, and Y. Nakamura. Research on the perception oflighting fluctuation in luminous officies environment. Journal of the IlluminatingEngineering Institue of Japan, 85(5):346–351, 2001.
5. E. Tetri and W. Pohl. Concepts and techniques for energy efficient lighting solutions.In P. Bertoldi and B. Atanasiu, editors, Proc. of the IEECB Focus 2008 ImprovingEnergy Efficiency in Commercial Buildings Conference, volume 2, pages 613–624,2008.
36