See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257371739

Augmented Reality in built environment: Classification and implications for

future research

Article in Automation in Construction · July 2013

DOI: 10.1016/j.autcon.2012.11.021

CITATIONS READS

318 4,177

4 authors, including:

Xiangyu Wang Mi jeong Kim

Curtin University Hanyang University
329 PUBLICATIONS 13,648 CITATIONS 208 PUBLICATIONS 3,729 CITATIONS

SEE PROFILE SEE PROFILE

Shih-Chung Jessy Kang

National Taiwan University
281 PUBLICATIONS 4,095 CITATIONS

SEE PROFILE

All content following this page was uploaded by Xiangyu Wang on 17 August 2021.

The user has requested enhancement of the downloaded file.

 Automation in Construction 32 (2013) 1–13

Contents lists available at SciVerse ScienceDirect

Automation in Construction
journal homepage: www.elsevier.com/locate/autcon

Review

Augmented Reality in built environment: Classification and implications

for future research
Xiangyu Wang a, b, d, Mi Jeong Kim b,⁎, Peter E.D. Love a, Shih-Chung Kang c
a
School of Built Environment, Curtin University, Australia
b
Department of Housing and Interior Design, Kyung Hee University, Republic of Korea
c
Department of Civil Engineering, National Taiwan University, Taiwan
d
Australasian Joint Research Centre for Building Information Modelling, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Augmented Reality (AR) has the potential to change how people interact and experience their surrounding
Accepted 8 November 2012 environment. During the last decade a considerable amount of research has been undertaken within the
Available online 28 February 2013
built environment. With this in mind, this paper aims to provide a state-of-the-art review of mainstream
studies undertaken between 2005 and 2011 within the normative literature. We found that a total of 120 ar-
Keywords:
ticles were published in the normative built environment literature within this period. Articles were classi-
Augmented Reality
Built environment
fied according to their concept and theory, implementation, evaluation (effectiveness and usability) and
Review industrial adoption. The classification of the literature has enabled gaps in the AR literature to be identified
Architecture/engineering/construction and future research directions to be proposed.
© 2012 Elsevier B.V. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Enabling technologies of Augmented Reality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.1. Media representation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2. Input mechanism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.3. Output mechanism continuum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.4. Trackers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3. Research approach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3.1. Selection criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.2. Data sources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
3.3. Procedures to extract articles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4. AR classification framework . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
4.1. Classification of articles by publication year . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.2. Classification of articles by journal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
4.3. Layer 1: Concept and Theory Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.4. Layer 2: Implementation Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
4.5. Layer 3: Evaluation Layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
4.6. Layer 4: industrial adoption layer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
5. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.1. Vision of future mobile AR . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.2. Toward context-aware computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
5.3. Evaluating the performance of AR systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
6. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

⁎ Corresponding author.
E-mail address: mijeongkim@khu.ac.kr (M.J. Kim).

0926-5805/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.autcon.2012.11.021
2 X. Wang et al. / Automation in Construction 32 (2013) 1–13

1. Introduction The study in the paper has done a comprehensive review of all the
Augmented Reality tools that are available for research and develop-
ment worldwide, based on the above classification, as shown in
Table 1. This list of tools is a very rich resource that can benefit both re-
“Imagine an environment where most physical objects know where
searchers and developers. The main technological components of
they are, what they are, and can (in principle) network with any
Augmented Reality are briefly explained below as a sense of context
other object. With this infrastructure, reality becomes its own data-
of the later discussion [108].
base. Multiple consensual virtual environments are possible, each
oriented to the needs of its constituency” (Verner Vinge).
2.1. Media representation
Augmented Reality (AR) creates an environment where digital in-
AR systems use more varied media representations than VR. In an
formation is inserted in a predominantly real world view. It originated
AR scene, digital content augments and is registered onto a real back-
from marker-based tracking toolkits (e.g., ARToolkit, ARTag) which
ground (Table 5). The options of media can be text/symbol/indicator,
are used to determine tracking and registration (where to display the
2D image/video, 3D wireframe, 3D data, 3D model, and animation.
digital contents) and the media contents (what digital content to dis-
Each option represents a class of information with common charac-
play). The technological advancements of AR during the past decade
teristics. Also, two or more types of media representation presented
have been significant. We have seen AR migrate from marker-based to
together on a single display, a hybrid representation, may be very
markerless (e.g., D′Fusion in T-Immersion Corp.) and more recently to
useful in some situations wherein the AR system application supports
mobile context-aware methods (e.g., Layar, Junaio, and Wikitube) that
different functions. Observation of AR systems in various domains re-
can bring AR into mobile and field contexts.
veals that many of the applications adopted the hybrid representation
AR is impacting the mobile communications industry by providing a
for multiple-functions.
radical shift in human–computer interaction. AR has been foremost ap-
plied in the area of entertainment, retail, travel, advertising, and social
2.2. Input mechanism
communication. Tons of commercials are starting to come out with AR
or “AR like” elements such as Microsoft XNA and FLARToolkit. Massive
An object in 3D space has six degrees of freedom. To completely
amounts of revenue are predicted to be generated, revolutionizing the
control such an object requires an equal number of degrees of free-
way we live and work. For example, 2014 forecast graph by ABI research
dom in the input signal. There are a number of ways that six dimen-
indicated AR mobile market revenues of $350 M. Juniper Research is
sional (three translational and three rotational) controlling signals
predicting that the market for Augmented Reality services will reach
can be generated. In AR systems, virtual information is presented to
$732 million by year 2014. This is consolidated by the act that several
the user to add to the real world scene and thus the information con-
reputed corporations (IBM, HP, Sony, Nokia, Intel) are going to increase
tent is what matters. For this reason most of the tasks in using an AR
their current efforts in AR. Universities are going to increase their re-
system are information accessing, rather than virtual content (object)
search programs as well toward AR.
manipulation and editing as in a typical VR or 3D CAD environment.
In the last decade AR has received a considerable amount of atten-
Thus the input mechanism discussed here is primarily considering
tion by researchers in the architecture, construction and engineering
object manipulation tasks as a major feature in AR systems. For
(AEC) community. AR can be applied to address a plethora of prob-
more detailed comparison of these input paradigms, readers are re-
lems throughout a construction project's lifecycle, for example, plan-
ferred to [122].
ning, design, safety, and training of crane operators. Yet, before AR
can be applied to construction, a deep and meaningful understanding
2.3. Output mechanism continuum
of the problem at hand is required and knowledge of how the capabil-
ities of new technology may change behavior, processes and tasks
The term output mechanism refers to the devices, or compo-
[29,30,81,108].
nents, used to support presentation of content and MR system re-
This paper provides a state-of-the-art review of mainstream studies
sponses to the user. Like all VR technology, displays in AR need not
undertaken between 2005 and 2011 within the normative literature
be only visual — there is research on creating audio displays for AR
(120 articles). Articles were classified according to the four-layer
[123]. The display can be generally classified as acoustic display, visual
taxonomy: concept and theory, implementation, evaluation (effective-
display, tactile display, etc. Historically, hardware associated with VE
ness and usability) and industrial adoption. The classification of the liter-
presentation has been specialized to render a single facet of human
ature provides a classification framework for the integration and
senses [124]. Almost all work in AR has focused on the visual sense: vir-
classification of the state-of-the-art work in AR research in built envi-
tual graphic objects and overlays. But the augmentation might apply to
ronment. It reveals the research concentration and paucity, and forecast
all other senses as well. For auditory displays, rather than isolating the
future research and development trends.
participant from all sounds in the immediate environment, by means
of a helmet and/or headset, computer-generated signals can instead
2. Enabling technologies of Augmented Reality
be mixed with natural sounds from the immediate real environment.
Haptic displays (that is, information pertaining to sensations such as
[121] classified AR into the following four levels:
touch and pressure) are typically presented by means of some type of
• Level 0 — physical world hyper linking: a simple link from the phys- hand held master manipulator (e.g. Brooks, et al. [125]) or more distrib-
ical world to the virtual world without involving any real time ren- uted glove type devices (Shimoga [126]).
dering and graphics. Examples include bar code and 2D image
recognition. 2.4. Trackers
• Level 1 — marker-based AR: 2D marker AR based on PC and web
cam. Level 1 AR is very challenging for mobile use due to the lack Accurate registration and positioning of virtual objects in the real
of robustness of markers and computing power. environment require accuracy in tracking the user's head as well as
• Level 2 — markerless AR: there is no requirement for markers for sensing the locations of real objects in the environment. The biggest
registration and tracking methods. The tracking is more robust, en- single obstacle to building effective AR systems is the requirement
abling the realization of mobile AR. of accurate, long-range sensors and trackers. In this research, the us-
• Level 3 — augmented vision which is currently being developed. ability issues are addressed while selecting tracking technology. The
 X. Wang et al. / Automation in Construction 32 (2013) 1–13 3

Table 1
Toolkit for Augmented Reality prototyping.

Tool Description Source

1. 2D Marker AR — PC and web cam based

ARtoolkit The most common marker-based toolkit for AR tracking http://www.hitl.washington.edu/artoolkit/
ARtag Use arrays of the square ARTag markers added to the environment. http://www.artag.net/
FLARToolKit Web-based AR applications. Flash Actionscript (v3) version of ARToolKit which can be used http://www.artoolworks.com/
to quickly develop web-based AR experiences
Metaio Based on Adobe® Shockwave™ and Flash™, Unifeye Viewer allows developers to produce high- http://www.metaio.com/
quality AR scenarios with low efforts.
Beyond reality Use techniques like image recognition, AR or 3D face tracking, they create stunning experiences http://www.beyondreality.nl/
for the web or standalone applications.
IN2AR Flash Augmented Reality engine can create an Augmented Reality game that can be controlled http://www.beyondreality.nl/
by using the package as a wheel.
Worldviz VR authoring tool with Augmented Reality output and module http://www.worldviz.com/
Junaio The world's most advanced AR browser for mobile AR contents http://www.junaio.com/
MRToolkit An open source MRT C++ Library for marker-based AR applications http://www.cs.ucl.ac.uk/staff/r.freeman/
maxReality™ Is a virtual and augmented reality authoring system that brings Autodesk® 3ds Max or 3ds Max http://www.vuzix.com/ar/maxreality.html
Design® content to life
ALVAR Designed to be as flexible as possible. It offers high-level tools and methods for creating AR http://virtual.vtt.fi/virtual/proj2/
applications with just a few lines of code. The library also includes interfaces for all of the low- multimedia/
level tools and methods,
which makes it possible for the user to develop their own solutions using alternative approaches
or completely new algorithms.
The Eye of Judgment A new style of gameplay where collectable trading cards, embedded with a Cyber Code, are http://us.playstation.com/games-and-
brought to life in the 3D game media/games/the-eye-of-judgment-
ps3.html

2. 2D Marker AR — mobile
Metaio Mobile SDK The most comprehensive solution to create Augmented Reality applications for iPhone, Android, http://www.metaio.com/software/
Symbian and WinMobile devices mobile-sdk/
Int13 Uses the camera of the iPhone to blend the game world and the real world for a truly surprising, http://www.int13.net/en/
immersive and fun experience

3. 3D object recognition — mobile

iVisit Allows you to talk face-to-face with friends from anywhere http://www.ivisit.com/

4. Markerless tools
D′Fusion by T-Immersion Allows for robust tracking of 2D and 3D objects and eliminates the need to create from scratch http://www.t-immersion.com/

5. GPS — compass based AR

Wikitude The very first AR navigation system for smartphones http://www.wikitude.com/
Layar Derived from location based services and works on mobile phones that include a camera, GPS http://www.sprxmobile.com/?s = compass
and a compass.
Sixth Sense A wearable gestural interface that augments the physical world around us with digital http://www.pranavmistry.com/projects/
information and natural gestural interaction sixthsense/
Ovjet A mobile AR application with the recognition of object-based features http://www.ovjet.com/
NRU/Last Minute An AR application that finds nearby restaurants, bars, and shops and displays Zagat ratings in a http://www.zagat.com/About/Index.aspx?
360 degree first-person view menu = PR150
Enkin Displays location-based content in a unique way that bridges the gap between reality and classic http://enkinblog.blogspot.com/
map-like representations.

tracking technologies are organized into the seven types of implemen- showing that approximately 27% of the most cited AR papers from
tation as outlined by Holloway and Lastra [127]. For details of tracking 1998 to 2007 came from the ISMAR conferences. On a more recent
technologies, one can refer to the surveys in Ferrin et al. [128] and work on reviewing AR systems, Van Krevelen and Poelman [132]
Meyer et al. [129]. Also Applewhite (1991) and Holloway and Lastra conducted a very recent survey on AR technologies, applications, and
[127] present an excellent discussion of tracker evaluation by means human factors limitations. As the initial efforts of summarizing AR ad-
of a framework for suitability. vances in built environment, Wang [100] wrote a review summary of
several major research efforts in using AR for architecture, design and
3. Research approach construction. Unfortunately, although there were noted significant
amounts of work in AR in built environment in recent years, there
Azuma [130] published the first survey that defined the AR field, de- was no rigorous paper that reviews the state-of-the-art work of AR in
scribed many problems, and summarized the developments up to that built environment. This paper is timely and tends to fill this gap, as a
point. Zhou et al. [131] conducted a follow-up review to report the milestone summary of the achievements in AR built environment. It
most recent advances in this area. Zhou et al. [132] conducted a survey should be noted that the focus of this review is not on the technical is-
of 276 papers from the past 10 years of ISMAR conferences. Their sur- sues such as implementation errors (such as mis-calibration), techno-
vey was particularly guided by the following three questions. (1) logical problems (such as vertical mismatch in image frames of a
Which areas have been explored in AR? (2) What are the developments stereo display), and fundamental limitations in the design of current
and key problems in these areas? and (3) What are important future HMDs (the accommodation–vergence conflict).
trends for AR research? They also measured their relative impact by cal- Recently, there is a large quantity of work in AR for built environ-
culating the citation rate of the papers. For example, there were 56 AR ment, although the idea of using AR for built environment can be
papers from 1998 to 2007 with an average citation rate of more than dated back to 1996. The nature of the research work has been migrated
10 citations per year. Of these, 15 were from ISMAR conference series, from the early visionary and conceptual framework to lab-based
4 X. Wang et al. / Automation in Construction 32 (2013) 1–13

prototype proof and evaluation. It is necessary to have a comprehensive 3.2. Data sources
and thorough review of the recent significant AR work in its application
in built environment. The main objectives of this review are to: A list of target leading international journals in AEC/FM was iden-
tified and searched, as listed in Table 2 with the number of papers
• provide a classification framework for the integration and classifica- identified. In order to not miss the relevant articles in other domains,
tion of the state-of-the-art work in AR research in built environment for example pure AR technical journals and conferences, which might
• reveal the research concentration and paucity have small number of relevant papers dealing with AR for built envi-
• summarize the issues and challenges, and suggest how they can be ronment, we have searched the following six online databases. The
pursued number of articles by each online database is as follows: Science Di-
• forecast future research and development trends. rect, IEEE Xplore, Springer Link Online Libraries, ACM Digital Library,
Wiley InterScience, Ingenta Journals, and EBSCO (Electronic Journal
Service). Journals that are more computer science focused, narrow
in focus to a specific niche, were excluded from the list. The relevant
3.1. Selection criteria
journals identified from this search include Presence, and the relevant
conferences identified from this search include ISMAR and IV.
Each paper was reviewed by the authors to determine its eligibility
and level of relevance based on the following criteria: Only articles pub-
3.3. Procedures to extract articles
lished from 2005 to 2011 are considered to assure the recent work. The
selected paper must be AR work in areas of built environment. Those ar-
Research reviews of a specific paradigm are an established form of
ticles that concentrated solely on AR rather than their applications in
enquiry within the information system discipline [77]. For example
built environment were not included. For example, AR papers for prod-
Hong et al. [40] used the approach to conduct a research review in the
uct design or product assembly are out of the scope. Considering that AR
area of context-aware computing. The identification of articles within
for built environment is an emerging area and there are not significant
the journal and conference list involved keyword searches and an ex-
body of journal articles, we also looked into the leading built environ-
haustive search on the title of the journals and conferences. The search
ment research conferences. Both the leading international journals
was initialized using the keywords “augmented”, “augmented reality”,
and conferences (Table 2 listed the searched journals and conferences)
“mixed” and “mixed reality”. Articles that appeared to fit into the cate-
are searched for quality assurance purpose.
gory were verified by firstly reading the abstract and then the entire ar-
ticle to extract the main findings and emphasis of the article. Many
articles that were initially identified via the title or keywords were
discarded in this process if either the abstract or the article itself did
not primarily focus on AR. For example, although having the keyword
Table 2 “augmented”, “augmented cognition” papers in cognitive psychology
The list of searched Journals and conferences and the number of articles.
area are not eligible because they have no AR contents. This process
Number of identified initially over 154 papers and after the screening process,
papers only 120 articles were determined for the content analysis.
Journal List Two independent researchers proposed their classification frame-
Advanced Engineering Informatics 1 works. Each of the two researchers conducted their own content analy-
American Society of Civil Engineers Journal (ASCE) 10 sis based on their classification frameworks. Then through a matching
Artificial Intelligence for Engineering Design, Analysis and 1
Manufacturing (AIEDAM)
process, the resolution of inconsistencies forms one classification
Automation in Construction 8 framework. Once each category of the classification framework was fi-
CoDesign 1 nalized, one of the original researchers and a third researcher, who
Design Studies 1 was not involved in developing the classification, allocated all of the ar-
International Journal of Architectural Computing 7
ticles to the framework. The Cohen's Kappa statistic (Krippendorf,
Journal of Information Technology in Construction 8
Presence: Teleoperator and Virtual Environments 1 [133]) was used to analyze the correspondence between the re-
searchers' allocation of articles to the categories in the classification
Conference List framework. The result shows that there was a high degree of reliability
ACADIA (Association for Computer Aided Design in Architecture) 4 between the researchers (0.9). This was well above the recommended
ASCE (American Society of Civil Engineers) Conferences 8
CAAD Futures (Computer Aided Architectural Design Futures) 5
Cohen's Kappa level of 0.80 (Krippendorf, [133]).
CAADRIA (International Conference on the Association for 14
Computer-Aided Architectural Design Research in Asia)
CDVE (Cooperative Design, Visualization, and Engineering) 2
4. AR classification framework
CIB W78 (International Council for Research and Innovation in 3
Building and Construction) The classification framework of AR system as shown consists of
CONVR (International Conference on Construction Applications 17 the following three parts:
of Virtual Reality)
DDSS (The International Conference in Design & Decision Support 1 1. Application area of built environment: Architecture and Design,
System in Architecture and Urban Planning) Landscape and Urban Planning, Engineering-related, Construction,
eCAADe (Conference on Education in Computer Aided 9
Facility Management, Life Cycle Integration, and Training and
Architectural Design in Europe)
IASDR(Conference on international Association of Societies of 1 Education.
Design Research) 2. The architecture layer framework of AR systems in built environment
ICCCBE (The International Conference on Computing in Civil and 4 as shown in Fig. 1 is developed based on the standard architecture
Building Engineering
layer criteria for emerging information technology concepts or tools
ISARC (International Association for Automation and Robotics in 12
Construction)
[40], for example the context-aware computing, mobile computing
ISMAR (International Symposium on Mixed and Augmented 1 and pervasive computing. The architecture consists of 4 layers:
Reality) • Concept and Theory Layer: consists of Algorithm, Development
IV (Information Visualization) 1 Specifications, Conceptual Framework, Evaluation Framework,
Total 120
and Technology Transfer.
 X. Wang et al. / Automation in Construction 32 (2013) 1–13 5

Fig. 1. The architecture layer framework of AR systems.

• Implementation Layer: include hardware and software layers; Construction published the most articles on AR systems (8 articles,
Hardware Layer: means display, input, computing and trackers; 6.7%) respectively. ASCE journals published 7 articles (5.8%), and In-
Software Layer: involves Agent-based, Intelligence, Knowledge- ternational Journal of Architectural Computing published 10 articles
based, Information design, and Interaction design. Each of them (8.3%). In terms of conferences, the largest outlet for AR is CONVR (In-
can be further divided into sub-categories. We use the taxonomy ternational Conference on Construction Applications of Virtual Reality)
[108] to classify the system features into sub-categories. The tax- (17 articles, 14.2%), followed by CAADRIA (International Conference
onomy has been integrated with the classification framework as on the Association for Computer-Aided Architectural Computing) (14
shown in Table 3. articles, 11.7%). This distribution shows us that most of AR papers
• Evaluation Layer: divided into effectiveness and usability catego- were published in digital or IT-related journals in AEC. The other ar-
ries. ticles are spread across the remaining journals highlighting that
• Industry Adoption Layer: has the prototype been tested in industry? there are many facets of AR in built environment research that are
3. Other technical criteria: relevant to a wide cross section of journals.
• Theory of human factors: does the work involve theory of Fig. 3 depicted the number of AR articles in each built environ-
human factors? ment area from year 2005 to 2011, which provides a longitudinal per-
• Built Environment domain knowledge: does the work consider- spective for identifying patterns, trends, and themes in relation to
ably rely the AR work on built environment domain knowledge? research areas in the quantity of publication. This time frame does
• Individual AR systems v.s. collaborative AR system not cover the early stages of AR work in built environment, but focus-
• Stationary AR v.s. mobile AR es on the most recent advances from 2005 to 2011. The period is ex-
• Types of display used tensive enough to identify the emergence of literature in each area.
• Self developed v.s. commercial software Architecture/design and construction are the leading adopters. It is
• Self evaluation v.s. comparative evaluation. not surprising that architecture and design takes the leading position
(45 articles, 37.5%). Undoubtedly architect was the first in built envi-
4.1. Classification of articles by publication year ronment to embrace AR, partially because they are visual-oriented
and appreciate the needs of visual aids. The door was then opened
The number of articles by publication year is depicted in Fig. 2. Nu- for a new set of inquiring minds to join architects who were already
merous AR articles have grown considerably in year 2009. The number taking a look at the possibilities of virtual visualization. The exciting
in 2009 has become 2 times more than the average number of articles in visualization opportunities presented by Augmented Reality technol-
previous years. The number of articles before and after 2009 is similar in ogies have captured the attention of a growing number of researchers
quantity. The number of articles published remained under 18 per year from the broad architecture, engineering, construction and facilities
(except year 2007) until 2009 when 26 were published. Surprisingly, management (AEC/FM) domain.
there was a significant drop in numbers in year 2010 and climbed up Construction takes the second place in its popularity of AR applica-
rapidly in 2011. It is very clear that the efforts and interests in AR re- tions (55 articles, 45.8%). It is a little surprising that AR for construction
search in built environment were increased and this trend will is catching up the pace of the architecture and design. Construction is
continue. interested in AR, at least more than other earlier visualization technolo-
gies such as Virtual Reality. This is partially because construction de-
4.2. Classification of articles by journal mands more on onsite work, which requires the visualization to be
mobile as well as grounded or context-aware into the real construction
The journals and conferences we searched are categorized in site, which justifies the adoption and promise of AR as the tool.
Table 2. Most of them were related to the Architecture, Design, Engi- Training and Education has 5 articles (4.2%). Urban Planning and
neering, Construction, and Facility Management. As Table 2 shows, Landscape has 6 articles (5.5%) and Engineering-related have 3 arti-
Automation in Construction and Journal of Information Technology in cles (2.5%). There are very few papers in FM (5, 4.2%) and life cycle
6 X. Wang et al. / Automation in Construction 32 (2013) 1–13

Table 3
Number of articles by each subject of the classification framework from 2005 to 2011.

Classification criteria Number of articles Relative percentage (%) Global percentage (%)

Application 120 100 100.0

Architecture and Design 45 37.5 37.5
Urban Planning and Landscape 7 5.8 5.8
Engineering 3 2.5 2.5
Construction 55 45.8 45.8
Facility Management 5 4.2 4.2
Life Cycle Integration 0 0.0 0.0
Training and Education 5 4.2 4.2
Layer 1. Concept and Theory 63 100 52.5
1.1 Algorithm and Modeling 22 34.9 18.3
1.2 Conceptual Framework 32 50.8 26.7
1.3 Evaluation Framework 4 6.3 3.3
1.4 Technology Adoption 9 14.3 7.5
Layer 2. Implementation 73 100 60.8
2.1 Software 63 86.3 52.5
2.1.1 Agent-based AR 1 1.4 0.8
2.1.2 Knowledge-based AR 0 0.0 0.0
2.1.3 Contents design 48 65.8 40.0
2.1.4 Interaction design 16 21.9 13.3
2.2 Hardware 50 68.5 41.7
2.3 Case study 12 16.4 10.0
Layer 3. Evaluation 19 100 15.8
3.1 Effectiveness 10 52.6 8.3
3.2 Usability 2 10.5 1.7
3.3 Effectiveness + usability 7 36.8 5.8
Layer 4. Industry Adoption 0 0.0 0.0
AR system
Single-user system 61 50.8 50.8
Collaborative system 6 5.0 5.0
AR toolkit
Prototype 32 26.7 26.7
Commercial 33 27.5 27.5
Evaluation method
Self-evaluation 9 7.5 7.5
Comparative evaluation 12 10.0 10.0
Stationary 38 31.7 31.7
Mobile 27 22.5 22.5
Human factors 27 22.5 22.5
Doman knowledge 32 26.7 26.7

integration (0, 0.0%). Although FM has not attracted as much research example, a paper might have significant “contents design” contribution
attention and efforts as the other areas, AR is a very promising tool. and have a significant evaluation on both effectiveness and usability,
AR in life cycle integration is still in the process of early exploration the sum of the “number of articles” column in Table 3 is not 120. The
stage and does not attract much attention. numbers of articles in Table 3 tally the number of articles which falls
Fig. 4 shows the number of articles at each architecture layer from to each category. The relative percentage within subject indicates the
year 2005 to 2011. The number of articles, relative percentage within percentage of papers within one category. The global percentage
subject, and global percentage of all subjects by each category of the means that the number of papers in total number. Tables 4, 6 and 7
classification framework are listed in Table 3. Considering the fact
that a certain paper might fall into several categories meanwhile, for

Fig. 2. Number of AR publications from 2005 to 2011. Fig. 3. Number of articles in each built environment area from 2005 to 2011.
 X. Wang et al. / Automation in Construction 32 (2013) 1–13 7

Artificial Intelligence methodology that provides the foundation for AR

to become intelligent systems. Conceptual framework gives users a gen-
eral idea of what the systems will look like and how it will implement
systems, for example, the general capabilities of the systems, a user in-
terface system functions, system (data) flow, and system management.
Future implementation work can be built upon such conceptual
roadmap.
In the Layer 1 “AR Concept and Theory” as shown in Tables 3 and 4,
the category can be divided into four subjects. Most of the articles that
fall into this category related to “Conceptual Framework” (32 articles)
and “Modeling and Algorithm” (22 articles). 26.7% (32 articles) of the
papers focused on the conceptual framework of a proposed AR system
with sophistication. 18.3% (22 articles) of papers have significant
work on the development of a new algorithm or modeling efforts
from engineering principles. Interestingly, 3.3% (4 articles) concerned
the development of an AR evaluation framework or guidelines that
can form the basis of heuristic guidelines of usability evaluation for
the tasks in built environment. Certainly, these evaluation guidelines
can be adapted and customized for their adoption into the context of
Fig. 4. Number of articles in each classification category from 2005 to 2011. built environment. 7.5% (9 articles) of the papers studied the issues
and potentials of technology transfer and adoption in industry. This
number is reasonable and not surprising, considering the nature of
represent all references of the AR articles. As for the Layer 1, 52.5% this study. Strategic alternatives or business models for gaining the rev-
(63 articles) of reviewed articles have significant concept and theory in- enue by using AR systems are very few.
volved in their studies. Layer 2 Implementation has the highest percent- Most of the implemented AR systems are single user-based (61 arti-
age of AR articles (73 articles, 60.8%). This implies that the AR cles, 50.8%) which is much more than collaborative AR systems (6 arti-
technology has mature enough to enable implementation either from cles, 5.0%). Therefore, more attention should be spent on collaborative
off-the-shelf commercial packages or self-development from lower systems, considering that any IT tool eventually goes to collaborative
software infrastructure. 15.8% (19 articles) of the papers have signifi- circumstance as groupware. Multidisciplinary collaboration is a trend
cant evaluation (Layer 3) of the effectiveness and/or usability studies and will become a future standard. 26.7% (32 articles) of implemented
through scientific and formal approach. Unfortunately, none of the AR systems were self-developed systems. Self-development means
work has significant industrial adoption or piloting. These four percent- that the AR is prototyped predominantly based on their own efforts or
ages have already depicted a full picture of what happened and the significant further development efforts were spent, if a commercial
trend: it becomes a mature practice to implement AR for built environ- toolkit was adopted.
ment; their applications have attracted formal evaluation to quantita- This partially relates to the modeling efforts in the first layer “algo-
tively demonstrate their benefits and usefulness; and wide industrial rithm” of the classification architecture. 27.5% (33 articles) were based
adoption of the AR research outcomes has not been realized yet, but on the customization of off-the-shelf commercial or open source AR
started to attract attention. The AR application in built environment software/toolkit. This is unsurprising, as the readiness and sophistica-
has far left the technology trigger phase, and now into the phase of tion of commercial software is not sufficient yet to fulfill the system
peak of inflated expectations. AR in built environment is behind after and use requirements of their applications in built environment. Fur-
their applications in general public and entertainment, as the latter thermore, technologies/tools related to AR are not standardized yet,
has been in the phase of trough of disillusionment (SPRXMobile). considering the high percentage of self-developed AR software as the
The number of publications within each of the above categories over basis for prototyping. 31.7% (38 systems) are stationary system, basical-
the seven-year period is shown in Fig. 4. The most striking trend is the ly marker-based desktop or indoor AR. 22.5% (27 systems) are mobile
increase in the number of system implementation related articles systems, which basically are markerless AR systems relying on GPS or
with studies over the period. Concept and Theory has dropped from other hybrid tracking systems. This number shows that a relative bal-
the year 2007 for a couple of years and is now picking up in the year ance has been made in the efforts made between stationary AR and mo-
2011. bile AR. Actually more attention and efforts were attracted to mobile AR
recently.
4.3. Layer 1: Concept and Theory Layer
4.4. Layer 2: Implementation Layer
This represents the references according to the detailed categories in
Concept and Theory Layer in Table 3. “Concept and Theory” involves a In terms of the Layer 2 Implementation, there are much more arti-
new concept of how AR should and can be adopted to solve a problem cles dealing with software side (73 articles, 60.8%) of AR as compared
in one particular area of built environment, new theories and founda- to the hardware side (50 articles, 41.7%). There are many articles related
tion to construct and apply AR for built environment. Algorithm is an es- to the contents design of AR systems (48 articles, 40.0%). The articles
sential part in developing the AR systems and involves research such as significantly addressing interaction design have only 16 articles

Table 4
Articles on Concept and Theory Layer.

Classification criteria References

Concept and Theory Algorithm and Modeling [63,39,82,119,7,54,2,64,21,93,94,33,22,26,85,6,74,84,102,1,3,87,25,93,94]

Conceptual Framework [83,108,106,71,32,23,69,98,99,75,14,70,62,60,57,76,12,13,18,51,52,42,43,50,111,72,36,86,38,103–105,45,115,88,51,52,112]
Evaluation Framework [80,29,108,30]
Technology Adoption [80,29,30,95,109,27,100,55,28]
8 X. Wang et al. / Automation in Construction 32 (2013) 1–13

Table 5 expert evaluations of AR systems. Emerging AR devices and services in-

Number of articles classified by the categories of AR implementations (enabling troduce a new level of complexity when it comes to usability evaluation.
technologies).
Some researchers investigated general guidelines for evaluating AR sys-
Categories of AR implementations Number of papers Relative percentage tems, while others have evaluated novel AR interfaces and ways of
Media representation interacting. It was suggested that a hybrid evaluation method com-
a. Text, symbol and indicator 3 4.3 prised cognitive walkthrough and heuristic evaluation for AR interface.
b. 2D image/video 6 8.7 Only 19 articles (15.8%) have rigorous evaluation work, among
c. 3D wireframe 2 2.9
which 10 articles (8.3%) are related to effectiveness evaluation of per-
d. 3D data 2 2.9
e. 3D model 51 73.9 formance of using AR for built environment tasks and only 2 papers
f. Animation 5 7.3 (1.7%) of the evaluation work focused on usability evaluation only,
Total 69 100.0 and 7 papers (5.8%) have conducted both effectiveness and usability
Computing unit evaluations. There should be more usability evaluation as additional
a. Desktop, 29 47.5
to the effectiveness evaluation. Further a subcategory of evaluation
b. Tablet/laptop 21 34.4
c. Hand-held 9 14.8 is to look at if it is self evaluation or comparative evaluation. Self eval-
d. Wearable computers 2 3.3 uation means to evaluate the effectiveness and system performance
Total 61 100.0 itself. Comparative evaluation means to compare the AR tool with a
Interaction devices
well-established benchmark (e.g., a typical work method/tool or
a. 2D-based input 53.7 47.5
b. 2D imitated controller input 1 1.7 CAD software). Comparative evaluation should be done more because
c. Hand-held free space input 6 9.8 it is fair and objective to have comparison. The number of papers
d. Gestural input 4 6.6 which conducted self-evaluation is 9 (7.5%) and the number of papers
e. Tangible input 21 34.4 which conducted comparative evaluation is 12 (10.0%).
f. Embodied input 0 0.0
Surprisingly, as shown in Table 3, only 22.5% (27 articles) papers
Total 61 100.0
Registration method considerably involve human factor model theories, practices. Like-
a. Video see through, 46 93 wise, 26.7% (32 articles) of papers have serious domain knowledge
b. Optical see through 3 7 studies, extracted and applied when implementing their AR systems
Total 49 100.0
for built environment. This percentage is indeed low and expected
Display
a. Monitor-based display 22 38.6
to be higher. Otherwise, the efforts will simply be proof-of-concept
b. Monitor-based stereo glasses 0 0.0 rather than proof-of-benefit. Unlike computer science and computer
c. Head-mounted display with 25 43.9 engineering researchers who have a technology development per-
monitor-based output spective, the AEC/FM community has a user perspective that must
d. Video see-through HMD output 2 3.5
be developed as part of our contribution to shaping the development
e. Optical see-through HMD output 8 14.0
Total 57 100.0 of these technologies and ultimately realizing their adoption into
Trackers practice. Opportunities exist for improving practice through new effi-
a. Magnetic 4 6.9 ciencies and through devising new ways of executing work tasks. Our
b. Mechanical 0 0.0
attention to human resource capabilities and the attendant human
c. Acoustic 0 0.0
d. Inertial 1 1.7
factors can yield successful technology development decisions and in-
e. Optical 39 67.3 tegration. Table 7 lists all the references of paper that belong to this
f. Global tracking system: GPS 5 8.6 category.
g. Hybrid system 9 15.5
Total 58 100.0

4.6. Layer 4: industrial adoption layer

(13.3%). There is only 1 paper that incorporates agent into AR. There is One critical element in AR research has been the constant need to
no knowledge-based AR. There are 12 papers (10.0%) which had stud- prototype, demonstrate, and evaluate the variety of platforms, tools,
ied their AR system in case studies. and experiences that are documented in research and tutorials. It
can all sound great on paper and be exquisitely and deceptively cap-
4.5. Layer 3: Evaluation Layer tured in photography and video. However, any representation other
than hands-on practical experience doesn't tell the whole story. The
Evaluation is classified as two types: effectiveness evaluation and results/benefits gained from using the implemented AR tools in prac-
usability evaluation. Effectiveness evaluation concerns the productivity tical projects can educate or convince industrial to accept and even
improvement, performance time, number of errors, and other quantita- widely adopt the tool into their daily practice routines. Therefore
tive indicators of how effectively a particular AR system can facilitate a the industrial adoption practice is critical in assessing the effective-
certain task or activity. Usability category involves investigating the ness of the tool. Therefore, this criterion is treated as one category
user needs based on user interviews, field evaluations with users, and in the classification framework.

Table 6
Articles on the Implementation Layer.

Classification criteria References

Implementation Software Agent-based AR [98,99]

Knowledge-based AR [81,39,82,119,1,3,48,7,2,91,61,66,64,20,67,89,58,4,78,46,21,35,96,65,79,15–17,110,33,19,22,41,9,118,85,6,5,120,114,24,
contents design 103–105,56,37,47,31,34,97,117,15–17]
Interaction design [63,103–105,107,108,53,101,113,68,92,90,11,8,44,59,87,73]
Hardware [63,81,103–105,39,82,119,107,1,3,48,7,2,116,53,101,68,61,66,64,20,89,58,4,92,90,78,75,11,8,21,35,96,65,98,99,79,15–17,
110,33,19,22,41,9,118,6,5,120,114,44,24,103–105,56,37,47,31,97,117,59,87,10,49,42,43,15–17]
Case [39,1,3,95,2,68,66,67,11,24,38,115,34]
study
 X. Wang et al. / Automation in Construction 32 (2013) 1–13 9

Table 7 almost instantly (Rice [136]). Inspired from The Decade of Ubiquity,
Articles at the Evaluation Layer. the vision of future construction site can be: The physical objects in
Classification criteria References the construction site and the site itself around you become your display
Evaluation Effectiveness [48,116,53,101,90,22,118,120,31,15–17]
and interface to interact with information. Every physical construction
Usability [107,97] resources or constructed components such as equipment, tool, labor,
Effectiveness +usability [81,103–105,78,98,99,79,19,42,43] personnel, material, and work package in progress will be tagged, la-
beled, interpreted, monitored, tracked, and connected and will interact
with each other, in a real-time manner, to create a Cyberspace. The vi-
Unfortunately, no articles have significantly involved or engaged in- sion is that it will influence the current industries, professions, and mil-
dustrial practitioner and test their system in practical project. This lions of jobs, including AEC/FM. To make it happen, it apparently
shows a need to establish stronger tied connections between industry requires a convergence of a variety of technologies and domain knowl-
and academia. The dissemination of real-world applications directly edge. Ubiquitous computing in future construction also has the same
within the AEC/FM community is void, which actually can help redefine implications and trends as in their applications in other pioneering in-
industry practices and obtain feedback to help define future research dustries. The following are some of the key features that future AR
agenda. It should be noted that the case studies that only collect the trend will address:
practitioners' opinions of a particular AR system, ask practitioners to
participate a study, or using data/information from a practical industrial • Standardization of mobile communications protocols and data for-
project as the basis for AR prototyping and implementing, did not fall mats exchanged between different mobile AR systems.
into this category. In order to be qualified of being in this category, the • Service-oriented architecture (SOA) to complement the current
study must test AR systems in practical industrial circumstances/project server–client communication architecture in AR.
throughout. • Mobile multi-user AR system, synchronous and/or asynchronous
• Image recognition, image mining, and the use of intelligent agents
5. Discussion for more intelligent mobile AR applications for construction con-
texts
Aside from some games like Sony's Eye of Judgment, most of the AR • Natural human-AR interaction, for example, from simple example
systems involved more or less technical issues such as registration er- of gestural interfaces, to more sophisticated embodied interaction,
rors, computing resources, and latency. There are some noted efforts considering the fact that construction workers have to carry equip-
as pioneering piloting projects to preliminarily explore and validate ment, tools, drawings, etc.
the potentials of using markers. Level 1 AR (marker-based) is usually • User-friendly authoring tools for developing AR applications for
limited to either desktop or some stationary static objects with certain built environment use.
range and basically determine two things: where to display and what
to display. We need to move away from the desktop AR and start paying 5.2. Toward context-aware computing
more attention to another two things: where you are and what is
around you. This unavoidably requires context-aware ubiquitous and AR is a hot topic in the mobile space, with platforms and services
pervasive computing and make AR as ubiquitous AR. As for markerless on iPhone and Android platforms, and it represents the next genera-
examples, probably the most well-known examples of Level 2 AR is tion as location-aware applications move toward the plateau (Fenn
Mobilizy's Wikitude AR for the Android platform. The mobile device be- and Raskino [137]). To provide most relevant service/information
comes a lens that gives us the sensation of looking through and seeing to the users, applications and services should be aware of their con-
the world around us layered with information, data, and visualizations, texts and automatically adapt to their changing contexts—known as
which we usually termed as “digital layers” (Mistry and Maes [134]). context-awareness (Bolchini, Schreiber, and Tanca, [138]; Dey,
For example, MIT's “Sixth Sense” tried to project images onto objects [139]; Zhu, Mutka, & Ni, [140]). An entity is a person, place, or object
(project-based AR) instead of augmenting what you see (Mistry and that is considered relevant to the interaction between a user and an
Maes [135]). application, including location, time, activities, and the preferences
Meanwhile, from the technological availability side, the current em- of each entity (Dey, 2001). In order to fulfill the real potentials of
phasis on mobile communications demonstrated the power of enabling AR, it has to be integrated with context aware concept to acquire
technology such as handheld computers (for example, smart phones and utilize information on the context of a device in order to provide
and iPhone) to impact AR innovation. The transformation of mobile services that are appropriate to the particular people, place, time,
phones into mobile internet devices (MIDs) with powerful processors, event, etc. This integration will make dumb AR system which is
3D graphics, and GPS functionality has already changed the way we where we are now (delivers a static piece of information linked to
think, communicate, and interact with media. The handheld device's an object or place), to an intelligent AR augmentation means (uses
small GUI screen is no longer a miniature desktop but an experiential context to deliver dynamic links).
portal into the world, transforming invisible intelligence into visible Context aware includes two items: one is the status information of
augmentation (ISMAR 2009). AR can combine mainstream consumer people, place, time and event. The other thing is the cognitive status
products such as cameras (as trackers), cell phones (as computer of users such as attentions and comprehension. Most efforts of AR are
power), and GPS tag (tracking technology) together as small as in focused on the first item such as accuracy tracking. The second item
everybody's pocket to employ ubiquitous AR to influence how we has been less investigated but, recently, some of the literatures which
work and live. Other technological elements such as 4 G standard, sen- focus on cognitive context have been introduced. However, these efforts
sor networks and context delivery architecture are evolving and will are still insufficient to establish context-aware AR systems that reflect
play a key role in expanding the impact of AR in the physical world. cognitive context. To provide suitable and personalized service for the
user, more studies need to be conducted in this area. One of the largest
5.1. Vision of future mobile AR hurdles of bringing Building Information Modeling (BIM) into onsite is
the problem with large quantity of data. How to deal with concurrently
It is known that 2010 to 2020 will become The Decade of Ubiquity enormous digital data, information and knowledge that have different
which is defined as the next ten years where every aspect of our lives formats to serve suitable AR service to users? Essentially AR is the visu-
will be permeated by digital, mobile, media, data, information, aug- alization end. Any data fed into this end has to be pre-processed in a
mented, virtual, and so forth. It will be everywhere and accessible manner to make the data make sense to the user.
10 X. Wang et al. / Automation in Construction 32 (2013) 1–13

5.3. Evaluating the performance of AR systems • There is very few work that studied the issues and potentials of tech-
nology transfer and adoption in industry.
It is essential to improve AR systems based on performance evalua- • Relatively lower amount of work addressed interaction design in AR.
tion. Various models such as Balanced Score Card (BSC), IS Success • Most of AR systems (73.9%) used “3D models” as the media represen-
Model, Control Objectives for Information and Related Technology tation. The percentage of using other media such as “text, symbol and
(COBIT), Information Economics, and so on, have been applied to sug- indicator”,“2D image”,“3D wireframe”,“3D data”, and “animations” are
gest measures that can evaluate the performance of information sys- evenly distributed.
tems by considering diverse perspectives. As mentioned above, there • Most of AR systems (47.5%) used “Desktop” AR and 3.3% used “wear-
are very few efforts in developing the evaluation framework and meth- able computers”. Most of AR systems (47.5%) used “2D-based input”
od for AR systems in built environment. Due to the unique characteris- and then followed by “Tangible input” (34.4%).
tics of AR systems such as portability, mobility, nomadicity, information • Most of AR systems (93%) used “video see-through” as the registration
load, congruence, cognitive load and so on, the current technology for method, while “optical see-through” has 7%.
evaluating information systems (IS) does not provide techniques to for- • Most of AR systems (67.3%) used “optical” as the trackers and then
mally define, verify, implement, and analyze them. Therefore, research followed by “Hybrid system” (15.5%).
on the model and performance measure reflecting the characteristics • Most of the work did not considerably involve human factor model
of AR systems should be investigated. theories, practices and did not involve serious domain knowledge
This partially relates to the modeling efforts in the first layer “algo- from built environment. Our attention to human resource capabilities
rithm” of the classification architecture. 30% (33 articles) were based and the attendant human factors can yield successful technology de-
on the customization of off-the-shelf commercial or open source AR velopment decisions and integration.
software/toolkit. This is unsurprising, as the readiness and sophistica- • The dissemination of real-world applications directly within the AEC/
tion of commercial software is not sufficient yet to fulfill the system FM community is void. This shows a need to establish stronger tied
and use requirements of their applications in built environment. Fur- connections between industry and academia.
thermore, technologies/tools related to AR are not standardized yet, • Most of the implemented AR systems are single user-based AR systems,
considering the high percentage of self-developed AR software as much more than collaborative AR systems.
the basis for prototyping. 32.7% (36 systems) are stationary system, • The readiness and sophistication of commercial software is not suffi-
basically marker-based desktop or indoor AR. 21.8% (24 systems) cient yet to fulfill the system and use requirements of their applications
are mobile systems, which basically are markerless AR systems rely- in built environment. Furthermore, technologies/tools related to AR are
ing on GPS or other hybrid tracking systems. This number shows not standardized yet, considering the high percentage of self-developed
that a relatively balance has been made in the efforts made between AR software as the basis for prototyping.
stationary AR and mobile AR. Actually more attention and efforts • There is a strong trend of transiting from Level 2 to Level 3 AR. The fu-
were attracted to mobile AR recently. ture AR must be context aware and becomes ubiquitous AR.
The major findings from the significant review are: • For construction use, BIM integration with AR is the trend. AR will be the
site-extension of BIM concept and approach. AR will maximize the po-
• As compared to other areas, architecture/design and construction tentials of BIM in construction site, and eventually becomes a digital
have received most of the attention of AR research. The field of AR toolkit for construction workers.
for FM and life cycle integration is in its infancy, but has great poten- • More research on performance measuring method reflecting the char-
tial. acteristics (portability, nomadicity, congruence, cognitive load, etc.) of
• Massive amounts of revenue are predicted to be generated in AR, rev- AR systems should be conducted.
olutionizing the way we live and work. Both industry and universities
are trying to increase their research programs toward AR.
• Level 2 AR (marker-based AR) is becoming popular and dominant, as
6. Conclusions
moving from the technology trigger phase to just before the actual
peak of inflated expectations. The next phase marker-based AR will
This paper reviewed the 120 articles on AR in built environment
get into soon is the Trough of Disillusionment where interest wanes
between 2005 and 2011, with a literature classification framework.
as experiments and implementations fail to deliver.
This review details the wide variety of built environment applications
• 120 articles were identified as the AR work in built environment be-
for which AR systems are now being developed and tested. The pub-
tween 2005 and 2011. Most of articles for AR in built environment
lished studies, for the most part, still consist of lab-based controlled
were published in digital or IT-related journals and conferences in
studies without industrial practitioner-based control groups, geared
AEC/FM.
toward feasibility or proof of concept testing. However, this type of de-
• Most of the papers reviewed (60.9%) have system implementation
sign is appropriate for testing a new technology in the early stages of
component, which implies that the AR technology has mature
its application. An AR system, no matter how sophisticated it is itself,
enough to enable implementation either from off-the-shelf commer-
might fail if it is not appropriate adopted for a practice by considering
cial packages or self-development from lower software infrastruc-
the user requirements behind, for example. However, how well the
ture.
new AR systems work in this regard will depend very much on how fa-
• There are more efforts on proof-of-concept rather than proof-of-
miliar the developers and users are with the scientific rationale behind
benefit. Only 15.5% of the papers have significant evaluation of the ef-
their applications in the built environment area. This understanding
fectiveness and/or usability studies through scientific and formal ap-
will be the key to designing appropriate system features and success-
proach. None of the work has significant industrial adoption or piloting.
ful AR support.
• AR applications in built environment are behind after their applications
in general public and entertainment.
• There is very few work investigating integrating Artificial Intelligence
methodology (agent and knowledge-based system) into AR to make 7. Acknowledgements
it intelligent.
• There is very few work focused on the development of an AR evalua- This research was supported by Basic Science Research Program
tion framework or guidelines that can form the basis of heuristic through the National Research Foundation of Korea(NRF) funded by
guidelines of usability evaluation for the tasks in built environment. the Ministry of Education, Science and Technology (2011-0005888).
 X. Wang et al. / Automation in Construction 32 (2013) 1–13 11

References [29] P.S. Dunston, X. Wang, A hierarchical taxonomy of AEC operations for mixed reality
applications, Journal of Information Technology in Construction 16 (2011) 433–444.
[30] P.S. Dunston, X. Wang, An iterative methodology for mapping mixed reality
[1] A.H. Behzadan, V.R. Kamat, Reusable modular software interfaces for outdoor technologies to AEC operations, Journal of Information Technology in Construc-
augmented reality applications in engineering, Computing in Civil Engineering tion 16 (2011) 509–528.
2007, American Society of Civil Engineers, ASCE, 2007. [31] S. EL-Tawil, V.R. Kamat, Rapid reconnaissance of post-disaster building damage
[2] A.H. Behzadan, V.R. Kamat, Automated generation of operations level construction using augmented situational visualization, in: 17th Analysis and Computation
animations in outdoor augmented reality, in: Journal of Computing in Civil Engi- Specialty Conference, American Society of Civil Engineers, ASCE, 2006.
neering, 23(4), American Society of Civil Engineers, ASCE, 2009, pp. 405–417. [32] M. Germen, S. Artut, et al., The representation and navigation of complex data,
[3] A.H. Behzadan, V.R. Kamat, Georeferenced registration of construction graphics in: The 24th Conference on Education in Computer Aided Architectural Design
in mobile outdoor augmented reality, in: Journal of Computing in Civil Engineer- in Europe: Communicating Space(s) Volos, Greece, 2006.
ing, 21(4), American Society of Civil Engineers, ASCE, 2007, pp. 247–258. [33] M. Golparvar-Fard, Y. Fukuchi, et al., D4AR-4 dimensional augmented reality-
[4] P. Anders, W. Lonsing, AmbiViewer: A Tool for Creating Architectural Mixed Re- models for automation and interactive visualization of construction progress
ality, Association for Computer Aided Design in Architecture, 2005. monitoring of civil infrastructure projects, in: International Conference on
[5] A.H. Behzadan, V.R. kamat, Animation of construction activities in outdoor aug- Construction Applications of Virtual Reality, 2010.
mented reality, in: The International Conference on Computing in Civil and [34] M. Golparvar-Fard, S. Savarese, et al., Interactive visual construction progress
Building Engineering, 2006. monitoring with D4AR-4D augmented reality-models, Construction Research
[6] A.H. Behzadan, V.R. Kamat, Integrated information modeling and visual simula- Congress 2009: Broadening Perspectives, American Society of Civil Engineers,
tion of engineering operations using dynamic augmented reality scene graphs, ASCE, 2009.
Journal of Information Technology in Construction (2011) 259–278. [35] L. Gruber, S. Gauglitz, et al., The city of sights: design, construction, and mea-
[7] A.H. Behzadan, B.W. Timm, et al., General-purpose modular hardware and soft- surement of an augmented reality stage set, in: International Symposium in
ware framework for mobile outdoor augmented reality applications in engi- Mixed and Augmented Reality (2010), Seoul, Korea, 2010.
neering, Advanced Engineering Informatics 22 (1) (2008) 90–105. [36] N. Gu, X. Wang, Applying augmented reality for data interaction and collabora-
[8] D. Belcher, B. Johnson, MxR: A Physical Model-Based Mixed Reality Interface for tion in BIM, in: International Conference on the Association for Computer-Aided
Design Collaboration, Simulation, Visualization and Form Generation, Associa- Architectural Design Research in Asia, 2010.
tion for Computer Aided Design in Architecture, 2008. [37] M. Hakkarainen, C. Woodward, et al., Software architecture for mobile mixed re-
[9] L. v Berlo, D.A. Helmholt, et al., C2B: augmented reality on the construction site, ality and 4D BIM interaction, in: International Council for Research and Innova-
in: Interational Conference on Construction Applications of Virtual Reality, 2009. tion in Building and Construction, 2009.
[10] J. Chang, H. Hsu, et al., Development of a road monitoring and reporting system based [38] A. Hammad, E. Mozaffari, et al., Framework for mixed reality applications in civil
on location-based services and augmented reality technologies, in: Journal of Perfor- engineering, in: The International Conference on Computing in Civil and Build-
mance of Constructed Facilities, 25, American Society of Civil Engineers, ASCE, 2011. ing Engineering, 2006.
[11] C.-T. Chen, T.-W. Chang, 1:1 spatially augmented reality design environment, in: [39] A. Hammad, H. Wang, et al., Distributed augmented reality for visualising collab-
The International Conferences on Design & Decision Support Systems in Archi- orative construction tasks, in: Journal of Computing in Civil Engineering, 23(6),
tecture and Urban Planning, 2006. American Society of Civil Engineers, ASCE, 2009, pp. 418–427.
[12] I.R. Chen, M.A. Schnabel, Retrieving lost space with tangible augmented reality, [40] J.-y Hong, E.-h. Suh, et al., Context-aware systems: a literature review and clas-
in: International Conference on the Association for Computer-Aided Architec- sification, Expert Systems with Applications 36 (2009) 8509–8522.
tural Design Research in Asia, 2009. [41] L. Hou, X. Wang, Application of augmented reality technology in improving as-
[13] R. Chen, x. Wang, Tangible augmented reality for design learning: an implemen- sembly task proficiency, in: Interational Conference on Construction Applica-
tation framework, in: International Conference on the Association for Computer- tions of Virtual Reality, 2010.
Aided Architectural Design Research in Asia, 2008. [42] L. Hou, X. Wang, Experimental framework for evaluating cognitive workload of
[14] R. Chen, X. Wang, Tangible augmented reality: a new design instructional tech- using AR system in general task, in: International Symposium on Automation
nology, Computer Aided Architectural Design Futures, 2009. and Robotics in Construction, 2011.
[15] Y.-C. Chen, H.-L. Chi, et al., Use of tangible and augmented reality models in engi- [43] L. Hou, X. Wang, Experimental Framework for Evaluating Cognitive Workload of
neering graphics courses, in: Journal of Professional Issues in Engineering Educa- Using AR System in General Assembly Task, in: International Association for Au-
tion & Practice, 137, American Society of Civil Engineers, ASCE, 2011, pp. 267–276. tomation and Robotics in Construction (ISARC 2011), 2011, pp. 625–630.
[16] Y.-C. Chen, H.-L. Chi, et al., A Smart Crane Operations Assistance System Using [44] Y. Huang, K.-S. Ho, An integrated environment of representing digital antiques, in:
Augmented Reality Technology, in: International Association for Automation International Conference on Construction Applications of Virtual Reality, 2008.
and Robotics in Construction (ISARC 2011), 2011, pp. 643–649. [45] J.L. Izkara, J. Pérez, et al., Mobile augmented reality, an advanced tool for the con-
[17] Y.C. Chen, H.L. Chi, et al., A smart crane operations assistance system using aug- struction sector, in: International Council for Research and Innovation in Build-
mented reality technology, in: International Symposium on Automation and Ro- ing and Construction, 2007.
botics in Construction, 2011. [46] G.k. e K_nayo_lu, Using audio-augmented reality to assess the role of sound-
[18] Y.L. Chen, P.C. Hung, Intuitive augmented reality navigation system design — scape in environmental perception: an experimental case study at UC Berkeley
implementation by Next-Gene20 project, in: International Conference on the campus, in: The 27th Conference on Education in Computer Aided Architectural
Association for Computer-Aided Architectural Design Research in Asia, 2009. Design in Europe: Communicating Space(s), 2009.
[19] H.L. Chi, Y.C. Chen, et al., Development of tele-operated robotics crane for im- [47] V.R. Kamat, S. EI-Tawil, Rapid post-disaster evaluation of building damage using
proving erection operation in construction, in: International Conference on Con- augmented situational visualization, Construction Research Congress 2005:
struction Applications of Virtual Reality, 2010. Broadening Perspectives, American Society of Civil Engineers, ASCE, 2005.
[20] S.Y. Choo, K.S. Heo, et al., Augmented reality-effective assistance for interior [48] V.R. Kamat, M.A. herif El-Tawil, Evaluation of augmented reality for rapid assess-
design: focus on tangible AR study, in: The 27th Conference on Education in ment of earthquake-induced building damage, in: Journal of Computing in Civil
Computer Aided Architectural Design in Europe: Communicating Space(s) Engineering, 21(5), American Society of Civil Engineers, ASCE, 2007,
Istanbul, Turkey, 2009. pp. 303–310.
[21] D.H.J. Chung, S.Z. Zhiying, et al., Outdoor mobile augmented reality for past and [49] M.A. Khan, D. Andy, Using geo-located augmented reality for community evalu-
future on-site architectural visualizations, Computer Aided Architectural Design ation, Computer Aided Architectural Design Futures (CAAD Futures 2011), 2011.
Futures, 2009. [50] C. Kim, H. Lim, et al., Mobile computing platform for construction site manage-
[22] F. Dai, S. Dong, et al., Photogrammetry assisted rapid measurement of earthquake- ment, in: International Symposium on Automation and Robotics in Construc-
induced building damage, in: International Conference on Construction Applica- tion, 2011.
tions of Virtual Reality, 2010. [51] M.J. Kim, J.H. Lee, et al., Design Collaboration for Intelligent Construction Man-
[23] B. Dave, J. Moloney, Augmenting time and space in design inquiries, in: International agement in Mobile Augmented Reality, International Association for Automa-
Conference on the Association for Computer-Aided Architectural Design Research in tion and Robotics in Construction (ISARC 2012), 2011.
Asia, 2009. [52] M.J. Kim, J.H. Lee, et al., Design collaboration for intelligent construction man-
[24] J.M.S. Dias, A.C. Lopes, et al., Seamless indoor and outdoor location, guidance and agement in mobile augmented reality, in: International Symposium on Automa-
visualization in mobile computing, in: Interational Conference on Construction tion and Robotics in Construction, 2011.
Applications of Virtual Reality, 2005. [53] M.J. Kim, M.L. Maher, The impact of tangible user interfaces on spatial cognition
[25] S. Dong, C. Feng, et al., Occulusion handling method for ubiquitous augmented during collaborative design, Design Studies 29 (3) (2008) 222–253.
reality using reality capture technology and GLSL, in: Journal of Computing in [54] R. Lakaemper, A.M. Malkawi, Integrating robot mapping and augmented build-
Civil EngineeringAmerican Society of Civil Engineers, ASCE, 2011, pp. 494–503. ing simulation, in: Journal of Computing in Civil Engineering, 23(8), American
[26] S. Dong, V.R. Kamat, Resolving incorrect visual occlusion in outdoor augmented Society of Civil Engineers, ASCE, 2009, pp. 384–390.
reality using TOF camera and OpenGL frame buffer, in: Interational Conference [55] S. Lee, D.H. Shin, et al., Functional requirement of an AR system for supervision,
on Construction Applications of Virtual Reality, 2010. commenting, and strategizing in construction, in: International Conference on
[27] P.S. Dunston, X. Wang, Mixed reality-based visualization interfaces for architec- Construction Applications of Virtual Reality, 2010.
ture, engineering, and construction industry, in: Journal of Construction Engi- [56] S. Lee, M.T. Turkaslan, et al., The development of an augmented reality-based data
neering and Management, 131(12), American Society of Civil Engineers, ASCE, visualization system for commissioning of air handling units, in: International Con-
2005, pp. 1301–1309. ference on Construction Applications of Virtual Reality, 2007.
[28] P.S. Dunston, X. Wang, An iterative methodology for mapping mixed reality tech- [57] J. Lertlakkhanakul, I. Lee, et al., Using the mobile augmented reality techniques
nologies to AEC operations, Construction Research Congress 2005: Broadening Per- for construction management, in: International Conference on the Association
spectives, American Society of Civil Engineers, ASCE, 2006. for Computer-Aided Architectural Design Research in Asia, 2005.
12 X. Wang et al. / Automation in Construction 32 (2013) 1–13

[58] R.-H. Liang, Y.-M. Huang, Visualizing bits as urban semiotics, in: International Research Congress 2010. Innovation for Reshaping Construction Practice Banff
Conference on the Association for Computer-Aided Architectural Design Re- Alberta, Canada, 2010.
search in Asia, 2009. [89] Tomohiro Fukuda, M. Kawaguchi, et al., Development of the environmental de-
[59] F. Liarokapis, I. Greatbatch, et al., Mobile augmented reality techniques for sign tool “tablet MR” on-site by mobile mixed reality technology, in: The 24th
geovisualisation, in: 9th International Conference on Information Visualisation, Conference on Education in Computer Aided Architectural Design in Europe:
London, 2005. Communicating Space(s) Volos, Greece, 2006.
[60] H.C. Lin, Y.T. Shen, et al., IP++: computer-augmented information portal in [90] C. Tonn, D. Donath, et al., Simulating the atmosphere of spaces — the AR-based
place, in: International Conference on the Association for Computer-Aided Ar- support of 1:1 colour sampling in and within existing buildings, in: The 25th
chitectural Design Research in Asia, 2005. Conference on Education in Computer Aided Architectural Design in Europe:
[61] Y. Liu, Y. Wang, et al., Key issues for AR-based digital reconstruction of Communicating Space(s), 2007.
Yuanmingyuan garden, Presence: Teleoperator and Virtual Environments 15 (3) [91] C. Tonn, F. Petzold, et al., Spatial augmented reality for architecture — designing
(2006) 336–340. and planning with and within existing buildings, International Journal of Archi-
[62] W. Lonsing, P. Anders, Three-dimensional computational structures and the real tectural Computing 6 (1) (2009) 41–58.
world, in: International Conference on the Association for Computer-Aided Ar- [92] C. Tonn, F. Petzold, et al., Put on your glasses and press right mouse button
chitectural Design Research in Asia (2011), Hong Kong, 2011. AR-based user interaction using laser pointer tracking, in: The 26th Conference
[63] A.M. Malkawi, R.S. Srinivasan, A new paradigm for human-building interaction: the on Education in Computer Aided Architectural Design in Europe: Communicating
use of CFD and augmented reality, Automation in Construction 14 (1) (2005) 71–84. Space(s), 2008.
[64] M. Mani Golparvar-Fard, S.D. Feniosky Peña-Mora, et al., Application of D4AR — [93] N.T. Trung, D.Q. Truong, et al., A Generation Step for Force Reflecting Control of a
a 4-dimensional augmented reality model for automating construction progress Pneumatic Excavator Based on Augmented Reality Environment, International
monitoring data collection, processing and communication, Journal of Informa- Association for Automation and Robotics in Construction (ISARC 2011), 2011.
tion Technology in Construction 14 (2009) 129–153. [94] N.T. Trung, D.Q. Truong, et al., A generation step for force reflecting control of a
[65] J. Moloney, Screen based augmented reality for architecture, in: International pneumatic excavator based on augmented reality environment, in: International
Conference on the Association for Computer-Aided Architectural Design Re- Symposium on Automation and Robotics in Construction, 2011.
search in Asia, 2007. [95] J. Verlinden, I. Horv\, et al., Analyzing Opportunities for Using Interactive Aug-
[66] J. Moloney, B. Dave, From abstraction to being there: mixed reality at the early stages mented Prototyping in Design Practice, Artificial Intelligence for Engineering
of design, International Journal of Architectural Computing 9 (1) (2011) 1–16s. Design, Analysis and Manufacturing 23 (3) (2009) 289–303.
[67] E.-K. Nada, D.P. Giovanni, et al., Digital reconstruction as a means of understanding [96] T.P. Viet, C.S. Yeon, et al., AR: an application for interior design, in: International
a building's history, in: The 24th Conference on Education in Computer Aided Ar- Conference on the Association for Computer-Aided Architectural Design Re-
chitectural Design in Europe: Communicating Space(s) Volos, Greece, 2006. search in Asia, 2009.
[68] G. Papagiannakis, N. Magnenat-Thalmann, Mobile augmented heritage: enabling [97] R. Wang, X. Wang, Experimental investigation of co-presence factors in a mixed
human life in ancient Pompeii, International Journal of Architectural Computing 5 reality-mediated collaborative design system, in: Proceedings of the 6th Inter-
(2) (2007) 396–415. national Conference on Cooperative Design, Visualization, and Engineering,
[69] P.D. Peter Anders, Designing Mixed Reality: Perception, Projects and Practice, Springer-Verlag, Luxembourg, 2009.
Association for Computer Aided Design in Architecture, 2007. [98] X. Wang, Agent-based augmented reality system for urban design: framework
[70] F. Petzold, O. Bimber, et al., CAVE without CAVE: on-site visualization and design and experimentation, in: International Conference on the Association for Computer-
support in and within existing building, in: The 25th Conference on Education in Aided Architectural Design Research in Asia, 2007.
Computer Aided Architectural Design in Europe: Communicating Space(s), 2007. [99] X. Wang, Mutually augmented virtual environments for architectural design and
[71] V.T. Phan, S.Y. Choo1, Augmented reality-based education and fire protection for collaboration, Computer Aided Architectural Design Futures, 2007.
traditional Korean buildings, International Journal of Architectural Computing [100] X. Wang, Augmented reality in architecture and design: potentials and chal-
8 (1) (2010) 75–91. lenges for application, International Journal of Architectural Computing 7 (2)
[72] C. Po Han, N.T.L. Truc, Computer-aided visual communication for way-finding in (2009) 309–326.
emergence in indoor environment, in: International Symposium on Automation [101] X. Wang, R. Chen, An experimental study on collaborative effectiveness of aug-
and Robotics in Construction, 2008. mented reality potentials in urban design, CoDesign 5 (4) (2009) 229–244.
[73] F. Salim, P. Jaworski, et al., Informing Architecture and Urban Modeling with [102] X. Wang, P.S. Dunston, Real time polygonal data integration of Cad/augmented
Real-world Data on 3D Tangible Interfaces and Augmented Displays, The Associ- reality in architectural design visualization, Computing in Civil Engineering
ation for Computer Aided Design in Architecture (ACADIA 11), 2011. 2005, American Society of Civil Engineers, ASCE, 2005.
[74] P. Santos, T. Gierlinger, et al., Display and rendering technologies for virtual and [103] X. Wang, P.S. Dunston, Compatibility issues in augmented reality systems for
mixed reality design review, in: International Conference on Construction Appli- AEC: an experimental prototype study, Automation in Construction 15 (3)
cations of Virtual Reality, 2007. (2006) 314–326.
[75] M. Aurel Schnabel, et al., From virtuality to reality and back, in: Conference on [104] X. Wang, P.S. Dunston, Groupware concepts for augmented reality mediated
International Association of Societies of Design Research 2007 (IASDR), 2007. human-to-human collaboration, in: The International Conference on Computing
[76] M. Schnabel, X. Wang, et al., Touching the untouchables: virtual-, augmented-, in Civil and Building Engineering, 2006.
and reality, in: International Conference on the Association for Computer- [105] X. Wang, P.S. Dunston, Mobile augmented reality for support of procedural
Aided Architectural Design Research in Asia, 2008. tasks, in: The International Conference on Computing in Civil and Building Engi-
[77] E. Scornavacca, S.J. Barnes, et al., Mobile business research published in neering, 2006.
2000–2004: emergence, current status, and future opportunities, Communica- [106] X. Wang, P.S. Dunston, Design, strategies, and issues towards an augmented
tions of the Association for Information Systems 17 (1) (2006) 635–646. reality-based construction training platform, Journal of Information Technology
[78] H. Seichter, Augmented reality and tangible interfaces in collaborative urban de- in Construction 12 (2007) 363–380.
sign, Computer Aided Architectural Design Futures, 2007. [107] X. Wang, P.S. Dunston, User perspectives on mixed reality tabletop visualization
[79] H. Seichter, M.A. Schnabel, Digital and tangible sensation: an augmented reality for face-to-face collaborative design review, Automation in Construction 17 (4)
urban design studio, in: International Conference on the Association for (2008) 399–412.
Computer-Aided Architectural Design Research in Asia, 2005. [108] X. Wang, P.S. Dunston, A user-centered taxonomy for specifying mixed reality
[80] D.H. Shin, P.S. Dunston, Identification of application areas for augmented reality systems for AEC industry, Journal of Information Technology in Construction
in industrial construction based on technology suitability, Automation in Con- 16 (2011) 493–508.
struction 17 (7) (2008) 882–894. [109] X. Wang, N. Gu, et al., An empirical study on designers' perceptions of augment-
[81] D.H. Shin, P.S. Dunston, Evaluation of augmented reality in steel column inspec- ed reality within an architectural firm, Journal of Information Technology in
tion, Automation in Construction 18 (2) (2009) 118–129. Construction 13 (2008) 536–552.
[82] D.H. Shin, P.S. Dunston, Technology development needs for advancing aug- [110] X. Wang, H. Jen, Designing virtual construction worksite layout in real environ-
mented reality-based inspection, Automation in Construction 19 (2) (2010) ment via augmented reality, in: International Symposium on Automation and
169–182. Robotics in Construction, 2006.
[83] D.H. Shin, W.-S. Jang, Utilization of ubiquitous computing for construction AR [111] X. Wang, N. Kwok, Inspecting interior built environments using augmented
technology, Automation in Construction 18 (8) (2009) 1063–1069. reality-based autonomous robot, in: International Symposium on Automation
[84] D.H. Shin, W. Jung, et al., Large scale calibration for augmented reality on con- and Robotics in Construction, 2007.
struction sites, in: International Conference on Construction Applications of Vir- [112] N. Webb, A. Brown, Augmented critique of lost or unbuilt works of architecture
tual Reality, 2007. using digitally mediated techniques, in: Conference on Education in Computer
[85] M.F. SIU, M. LU, Augmenting site photos with 3D as-built tunnel models for con- Aided Architectural Design in Europe (eCAADe 2011), 2011.
struction progress visualization, in: International Conference on Construction [113] M. White, P. Petridis, et al., Multimodal mixed reality interfaces for visualizing digital
Applications of Virtual Reality, 2009. heritage, International Journal of Architectural Computing 5 (2) (2007) 322–337.
[86] M.F. SIU, M. LU, Bored pile construction visualization by enhanced production- [114] C. Woodward, M. Hakkarainen, et al., Mixed reality for mobile construction, in:
line chart and augmented-reality photos, in: International Conference on Con- International Conference on Construction Applications of Virtual Reality, 2010.
struction Applications of Virtual Reality, 2010. [115] C. Woodward, J. Lahti, et al., Case Digitalo—a range of virtual and augmented re-
[87] R.S. Srinivasan, A.M. Malkawi, Real-time simulations using learning algorithms ality solutions in construction application, in: International Council for Research
for immersive data visualization in buildings, International Journal of Architec- and Innovation in Building and Construction, 2007.
tural Computing 3 (3) (2005) 265–280. [116] P.D. Xiangyu Wang, P.D. Phillip, S. Dunston, Potential of augmented reality as an as-
[88] S.A. Talmaki, S. Dong, et al., Geospatial databases and augmented reality visual- sistant viewer for computer-aided drawing, in: Journal of Computing in Civil Engi-
ization for improving safety in urban excavation operations, Construction neering, 20(6), American Society of Civil Engineers, ASCE, 2006, pp. 437–441.
 X. Wang et al. / Automation in Construction 32 (2013) 1–13 13

[117] N. Yabuki, Z. Li, Cooperative reinforcing bar arrangement and checking by using aug- [128] F.J. Ferrin, Survey of helmet tracking technologies, in: SPIE Proceedings, Large
mented reality, in: Proceedings of the 4th International Conference on Cooperative Screen Projection, Avionic and Helmet Mounted Displays, 1456, 1991, pp. 86–94.
Design, Visualization, and Engineering, Springer-Verlag, Shanghai, China, 2007. [129] K. Meyer, H.L. Applewhite, F.A. Biocca, A Survey of Position-Trackers, in: Presence:
[118] N. Yabuki, K. Miyashita, et al., Evaluation of invisible height for landscape pres- Teleoperators and Virtual Environments, MIT Press 1 (2) (1992) 173–200.
ervation using augmented reality, in: Interational Conference on Construction [130] R.T. Azuma, A Survey of Augmented Reality, in: Presence: Teleoperators and Vir-
Applications of Virtual Reality, 2009. tual Environments, MIT Press 6 (4) (1997) 355–385.
[119] N. Yabuki, K. Miyashita, et al., An invisible height evaluation system for building [131] F. Zhou, et al., Trends in augmented reality tracking, interaction and display: A
height regulation to preserve good landscapes using augmented reality, Auto- review of ten years of ISMAR, in: the 7th IEEE/ACM International Symposium
mation in Construction 20 (3) (2011) 228–235. on Mixed and Augmented Reality, 2008, pp. 193–202.
[120] K.C. Yeh, M.H. Tsai, et al., A projective mobile device for presenting location- [132] D.W.F. Van Krevelen, R. Poelman, A Survey of Augmented RealityTechnologies,
based building information in a construction site, in: International Conference Applications and Limitations, in: The International Journal of Virtual Reality 9
on Construction Applications of Virtual Reality, 2010. (2) (2010) 1–20.
[121] R. Rice, The Augmented Reality Hype Cycle. url: http://www.sprxmobile.com/ [133] K. Krippendorf, Content Analysis: An Introduction to its Methodology, 2nd edi-
the-augmented-reality-hype-cycle/, Accessed in Jan 2011. tion Sage, Thousand Oaks, CA, 2004.
[122] K. Hinckley, R. Pausch, J. Goble, N. Kassell, A Survey of Design Issues in Spatial [134] P. Mistry, P. Maes, SixthSense: A Wearable Gestural Interface, in: Proc. ACM
Input, in: the 4th Annual Symposium on User Interface Software and Technolo- SIGGRAPH ASIA, 2009.
gy, 1994, pp. 213–222. [135] ISMAR 2009: International Symposium of Mixed and Augmented Reality 2009,
[123] J. Rozier, K. Karahalios, J. Donath, Hear & There: An Augmented Reality System of Orlando, Florida, USA. Url: http://www.interaction design.org/calendar/ismar_
Linked Audio, in: the International Conference on Auditory Display, 2000. 2009_conference.html.
[124] J.L. Gabbard, A Taxonomy of Usability Characteristics in Virtual Environments, [136] R. Rice, The Augmented Reality Hype Cycle. Internet article, url: http://www.
MS Thesis for Computer Science and Applications, December, Virginia Polytech- sprxmobile.com/the-augmented-reality-hype-cycle/ 2009, Accessed in Jan 2011.
nic Institute and State University (1997). [137] J. Fenn, M. Raskino, Mastering the Hype-Cycle: How to chose the right innova-
[125] F.P. Brooks Jr., M. Ouh-Young, J.J. Batter, P.J. Kilpatrick, Project grope: haptic dis- tion at the right time, Harvard Business Press, 2008.
plays for scientific visualization, Computer Graphics 24 (4) (1990) 177–185. [138] C. Bolchini, F.A. Schreiber, L. Tanca, A methodology for a Very Small Data Base
[126] K.B. Shimoga, A survey of perceptual feedback issue in dexterous telemanipulation: design, Information Systems 32 (1) (2007) 61–82.
part I. finger force feedback, in: Virtual Reality Annual International Symposium [139] A.K. Dey, Understanding and Using Context, Personal and Ubiquitous Computing
VRAIS, 1993, pp. 263–270. 5 (1) (2001) 4–7.
[127] R. Holloway, A. Lastra, Virtual Environments: A Survey of the Technology, in: [140] F. Zhu, M.W. Mutka, L.M. Ni, Service discovery in pervasive computing environ-
SIGGRAPH'95 Course 8 (A), 1995, pp. 1–40. ments, Pervasive Computing 4 (4) (2005) 81–90.

View publication stats