The Renaissance of Augmented Reality in Construction:
History, Present Status, and Future Directions
Ke Chen, and Fan Xue

This is the peer-reviewed post-print version of the paper:
Chen, K., & Xue, F. (2021). The renaissance of Augmented Reality in construction:
history, present status, and future directions. Smart and Sustainable Built Environment,
in press. Doi: 10.1108/SASBE-08-2020-0124
The final version of this paper is available at: https://doi.org/10.1108/SASBE-08-2020-0124
The use of this file must follow the Creative Commons Attribution Non-Commercial No
Derivatives License, as required by Emerald’s policy. Any other use requires prior
permissions of Emerald.

Abstract:
Purpose: Augmented Reality (AR) has become one of the most promising technologies in
construction since it can seamlessly connect the physical construction environment and virtual
contents. In view of the recent research efforts, this study attempts to summarize the latest
research achievements and inform future development of AR in construction.
Design/methodology/approach: The review was conducted in three steps. First, a keyword
search was adopted, and 546 papers were found from Scopus and Web of Science. Secondly,
each paper was screened based on the selection criteria, and a final set of 69 papers was
obtained. Thirdly, specific AR applications and the associated technical details were extracted
from the 69 papers for further analysis.
Findings: The review shows that: (1) design assessment, process monitoring, and maintenance
management and operation were the most frequently cited AR applications in the design,
construction, and operation stages, respectively; (2) information browser and tangible
interaction were more frequently adopted than collaborative interaction and hybrid interaction;

and (3) AR has been integrated with BIM, computer vision, and cloud computing for enhanced
functions.
Originality/value: The contributions of this study to the body of knowledge are twofold. First,
this study extends the understanding of AR applications in the construction setting. Second, this
study identifies possible improvements in the design and development of AR systems in order
to leverage their benefits to construction.
Keywords: augmented reality; construction management; physical environment; virtual
content; interaction; review

1

1. Introduction

2

Modern construction management in its broader sense is to make informed decisions

3

according to a wealth of information. Perhaps since the emergence of modern construction

4

management, the serious disconnection between the physical environment and virtual

5

information has become an intrinsic part of managerial problems in construction projects (Chen

6

et al., 2015). The problem has further escalated throughout the 2000-2010s and enlarged by the

7

increasing complexity in construction projects. In response, the industry started to embrace

8

advanced digital technologies in order to make accurate information readily available to both

9

managers and workers (Newman et al., 2020). Against this backdrop, Augmented Reality (AR),

10

among all digital technologies, has received attention from both researchers and practitioners

11

in construction.

12

AR creates a composite view of the virtual and reality by superimposing the digital

13

representations (e.g., image or model) of objects onto the physical environment (Azuma, 1997).

14

Such alignment of digital representations with people’s view of the real world enables the

15

interpretation of both the virtual and reality simultaneously. By doing this, AR could improve

16

the information perception process and thereby greatly facilitate decision-making. Researchers

17

have recognized that AR can support operational and managerial activities in various industries

18

(Henderson and Feiner, 2010). In construction, AR has many applications such as the

19

communication of design and planning ideas (Wang et al., 2014b), on-site information retrieval

20

(Yeh et al., 2012), and visualization of facility information for operation and maintenance (Baek

21

et al., 2019).

22

Given the existing research efforts, a literature review becomes necessary to evaluate

23

existing knowledge and uncover supplementary directions for future studies (Webster and

24

Watson, 2002). Some researchers subscribing to this point have reviewed the literature of AR

25

in construction. Chi et al. (2013) reviewed 101 studies before 2012, with a focus on technologies

26

influencing the development of AR applications (e.g., localization and user interface). Rankohi

27

and Waugh (2013) reviewed the relevant literature as of 2012 to discover facts about the stage,

28

sector, scope, and devices of AR applications. Wang et al. (2013) reviewed articles published

29

between 2005 and 2011 and found that no article introduced the industrial adoption of AR in

30

construction. Behzadan et al. (2015) investigated challenges faced by AR visualization in

31

infrastructure projects and provided solutions correspondingly. Recent years also see some

32

review articles regarding AR in construction, but nearly all of them focused on a few specific

33

perspectives. For instance, Li et al. (2018) and Moore and Gheisari (2019) reviewed the AR

34

literature solely on construction safety management. Calderon-Hernandez and Brioso (2018)

35

reviewed 48 papers regarding the simultaneous use of BIM and AR in the design and

36

construction stages. Elghaish et al. (2020) reviewed the use of immersive technologies

37

(including AR) and drones to support digital transformation in construction.

38

In view of the ever-updating technologies today, this paper attempts to uncover the recent

39

development and implementation of AR systems from the construction literature published after

40

2012. Three specific objectives will be addressed:

41

42

43

44

45

1. To revisit the research and development trends of AR systems in the construction
setting;
2. To synthesize how AR systems receive and react to users’ input, and how they present
information to users; and
3. To identify how AR has been integrated with other technologies in construction.

46

47

2. Overview of Augmented Reality

48

As a concept of providing the digitally interactive experience, AR is not new. Dating back

49

to 1901, Frank Baum, an American author, mentioned an electronic spectacle named as

50

“character marker” that enabled overlaps of data and real life. Decades later in the 1950s and

51

1960s, the concept of AR was firstly made true - such as “Sensorama” and “head-mounted

52

display” - by scientists from different disciplines. Attributed to the fast development of

53

technologies, today, when people talk about AR, they are mainly referring to AR systems. A

54

typical AR system must have three main features (Azuma, 1997). First, it must enable

55

interactions between the physical and virtual contents. Secondly, it enables the real-time

56

overlay of virtual contents onto the real world. Thirdly, it should be registered in three

57

dimensions. The realization of these features requires the use of various techniques, including

58

tracking techniques, display techniques, and interaction techniques. By using these

59

technologies, AR can enhance people’s interpretation of the real world (van Krevelen and

60

Poelman, 2010).

61

Tracking techniques are used to log and verify the position and orientation of users, and

62

thus play important role in the alignment and registration of virtual contents onto the physical

63

environment (Azuma et al., 2001). Depending on the specific application scenarios, different

64

types of tracking techniques can be adopted. For example, sensor-based and vision-based

65

methods generally verify the position and orientation by using various sensors (e.g., magnetic,

66

acoustic, and mechanical) and image processing methods, respectively. In addition, hybrid

67

tracking techniques are becoming increasingly popular for AR applications in many different

68

fields. A hybrid tracking technique takes advantage of both sensing and computer vision

69

techniques, which can generate a more robust estimation of user’s position and orientation than

70

using one single type of technique (Zhou et al., 2008).

71

Display techniques combine virtual contents and real-world environment and show both

72

simultaneously. Three types of display techniques are widely adopted, namely the handheld

73

display, the head-mounted display, and the projection-based display. The handheld display

74

employs mobile devices like a smartphone or a tablet (Wagner and Schmalstieg, 2003). These

75

mobile devices use video-see through methods, providing users with a video view of the

76

physical environment that is augmented by the corresponding virtual contents. The head-

77

mounted display is worn on the user’s head and can be a part of a helmet. Apart from video-see

78

through methods, the head-mounted display can use optical see-through methods that allow

79

users to view the real-world environment with their eyes and let them see the overlaid virtual

80

contents by holographic optical elements or half-silvered mirrors. Compared with the video-see

81

through methods, one major benefit of optical see-through methods is that they can create a

82

superior presentation of the physical environment (Zhou et al., 2008). The projection-based

83

display, also called the spatial display, does not require users to equip any devices, leading to

84

minimal intrusiveness. Projection-based displays directly show the virtual contents on the

85

surfaces of real-life objects and can naturally scale up to enable collaboration between a group

86

of people (Tonn et al., 2008).

87

Interaction techniques concern creating appropriate intuitive interactions between users

88

and AR systems, and can be classified as information browser, tangible interaction,

89

collaborative interaction, and hybrid interaction (Billinghurst et al., 2015). Information browser

90

refers to the most basic and straightforward interaction that views the visualized AR scene and

91

browses the information provided. For tangible interaction, objects in the real-world

92

environment could be deployed as the elements of the AR interface. For example, Kato et al.

93

(2000) proposed an AR application, in which the user can use a real paddle to select and place

94

the virtual objects in a living room environment. Collaborative interaction is designed and

95

developed for either co-located or face-to-face collaboration, both can improve physical

96

collaborative workspaces (Billinghurst and Kato, 2002). Multiple users in different roles can

97

use different display devices to look at the same object but be presented with different AR

98

experiences that are tailored to their needs. Hybrid interaction combines complementary

99

interfaces and allows users to interact with AR systems through various types of input such as

100

gesture and speech (Zhou et al., 2008).

101

The diversity in these three types of techniques produces many different types of AR

102

systems for corresponding application scenarios. Throughout a construction project life-cycle,

103

many operational and managerial tasks require enough information to interpret their complex

104

relations to the physical environment and objects (Shin and Dunston, 2008). Both commercial

105

and tailor-made AR systems have been adopted by different stakeholders in order to meet their

106

requirements of integrating the real world and virtual information, but there still remains much

107

room for further improvements (Wang et al., 2013). The understanding of extant efforts thus is

108

necessary before possible improvements can be made to leverage the benefits of AR to

109

construction.

110

111

3. Research methods

112

The review presented in this paper was carried out in three steps that have also been adopted

113

in many existing review studies (e.g., Chen et al., 2015; Mok et al., 2015). The first step is to

114

search the literature exhaustively. The databases for searching are Web of Science (WoS) and

115

Scopus. WoS core collection covers more than 21,100 high-quality journals (Clarivate

116

Analytics, 2019), and Scopus contains more than 23,452 journals with over 610 journals under

117

the fields of ‘architecture’, ‘construction and building’, and ‘civil and structural engineering’

118

(Elsevier, 2020). Searching both databases guarantees full coverage for the relevant literature

119

and thereby makes it possible to draw broad conclusions (He et al., 2017). The literature search

120

was conducted on 30th December 2019, using the query combination ‘(augmented reality)

121

AND (building OR construction OR civil OR infrastructure)’, i.e., the retrieved papers should

122

explicitly mention AR and a construction term. In addition, the review only considered peer-

123

reviewed journal papers since they tend to be more rigorous and mature than other types of

124

literature (Jesson et al., 2011), and the language of the paper was limited to ‘English’. Following

125

these rules, a total of 546 papers published from 2013 to 2019 were collected initially.

126

The screening of the title and keywords of the collected papers found that many of them

127

did not fit the review objectives. This is unsurprising because either building, construction, or

128

civil are general terms. Therefore, in the second step, one author filtered suitable literature

129

according to three criteria: (1) papers in areas irrelevant to construction, e.g., manufacturing

130

and surgery, were excluded; (2) papers not providing sufficient technical details of AR systems

131

were excluded; (3) papers only discussing the application potentials of AR without evaluation

132

(either in the laboratory or actual field) were excluded. Then, another author double-checked

133

the selected papers to decrease potential bias. Disparities were addressed by further discussions

134

until a joint agreement about the inclusion or exclusion of involved papers arrived. After

135

filtering, 69 papers were finally obtained for further analysis.

136

In the third step, the authors carefully read each paper and manually extract descriptive

137

information, namely, year of publication, project stage(s), and AR application(s). More

138

importantly, attention has been paid to the interaction between human and AR systems and the

139

integration of AR with other technologies.

140

141

4. Results

142

This section begins with an overall description of the literature (Section 4.1). Next, it

143

presents what the recent AR applications in construction are (Section 4.2), how users control

144

the AR systems and how the AR systems present information to users (Section 4.3), and

145

whether AR has been integrated with other technologies to leverage its full potential to

146

construction (Section 4.4).

147

4.1. General descriptions of the literature

148

As shown in Figure 1, the 69 papers came from 28 journals, and the top two journals were

149

Automation in Construction and Journal of Computing in Civil Engineering. Grouped into

150

“Others” were journals that published only one relevant paper each.

151

152

Figure 1. Sources of the 69 papers

153

Figure 2 shows the distribution of papers published between 2013 and 2019. Overall, the

154

number of papers continuously dropped since 2013, reaching its bottom of three in 2016 and

155

then bounced up in 2017. This U-shape curve could reveal a renaissance of AR research in

156

construction, which might be attributable to the increasingly mature AR technologies in recent

157

years (Gartner, 2018). A further investigation in the authors of the reviewed papers found that

158

Wang X., Kamat V.R., and Ayer S.K. were three notable researchers who published more than

159

five papers about AR in construction between 2013 and 2019. Moreover, 26.67% of the studies

160

published between 2017-2019 have the same author (at least one) of the studies between 2013-

161

2016. Such a result indicates that efforts in exploring the AR applications in construction are

162

coming from different researchers or research groups.

163

Figure 2 also shows the number of papers concerning different project stages. Findings can

164

be drawn by integrating Figure 2 and the results of Wang et al. (2013) and Rankohi and Waugh

165

(2013). In studies published before 2013, most of them used AR in the construction stage; less

166

in the operation stage. In studies published from 2013 to 2019, over half of them focused on the

167

construction stage, and the number of studies focusing on the operation stage has become nearly

168

two times as compared to the number of studies focusing on the design stage. These figures

169

indicate that the construction stage is still taking the leading position to embrace AR, and the

170

applicability of AR to project operation has been increasingly recognized.

171

172

173

Figure 2. Number of publications from 2013 to 2019 (by project stage)
4.2. AR applications in construction

174

Figure 3 shows the specific applications of AR in different project stages. Based on

175

authors’ experience and relevant studies, various AR applications are grouped into three

176

categories for the design stage, and seven and six categories for the construction and operation

177

stages respectively. One may notice that several studies consider more than one application,

178

and the comparison of studies published before and after 2013 reveals that AR applications in

179

construction have become more diverse. In the design stage, AR has been popularly used for

180

assessing the designed drawings or models in the physical environment and communicating

181

design ideas to different stakeholders (Alsafouri and Ayer, 2019). In the construction stage,

182

progress monitoring, assembly instruction, and quality management were dominant AR

183

applications. This is consistent with the findings of Rankohi and Waugh (2013) and Wang et

184

al. (2013). However, research efforts have also been focusing on safety management,

185

positioning, and other applications, which were not widely investigated by studies published

186

before 2013. In the operation stage, maintenance management and operation was the main AR

187

application. This is because maintenance is one of the major tasks in the operation stage, which

188

requires a great amount of information to be available in-situ (Hou et al., 2014). AR can

189

seamlessly connect the physical facilities with their corresponding virtual information,

190

facilitating instant and informed decision making (Dong et al., 2013). By providing additional

191

context information, AR has also been used for disaster management and emergency response,

192

localization and navigation, and energy management (e.g., Golparvar-Fard and Ham, 2014;

193

Tsai and Yau, 2014).

194

Acknowledging the rising diversity of AR applications in construction, it is noticeable that

195

some AR applications are becoming more mature for actual projects. Jiao et al (2013)

196

implemented AR for monitoring the construction progress of the Shanghai Center. Zhou et al.

197

(2017) used AR to support quality management in a metro construction project in China.

198

Nevertheless, most of the AR applications in the operation stage are still under lab experiments.

199

In addition, applications such as personnel management and cost control wait for further

200

exploration.

201

202

203

Figure 3. Number of publications by target application
4.3. Interaction between users and AR systems

204

Based on the extracted data, a Sankey chart was drawn to summarize the AR-user

205

interactions in construction. The Sankey chart provides a graphical presentation, in which the

206

width of the series can clearly indicate the number of studies that mentioned the diverse options

207

of how users control the AR systems and how the AR systems present on-demand information

208

to users (see Figure 4). Over 65% of the reviewed studies adopted the handheld display, while

209

the head-mounted display is also a relatively popular choice. In contrast, the projection-based

210

display has not been widely used in construction. A possible explanation could be the

211

difficulties in setting up the projector and the projected texture might not be sufficiently bright

212

or visible in a complex construction environment.

213

214

Figure 4. Overview of AR-user interactions (The total number of display is larger than

215

69 because one paper used two types of displays)

216

Information browser dominates the interaction between users and AR systems in

217

construction. 14 studies used 2D-based input, and 13 studies used 3D-based input. 2D-based

218

input provides simple interaction, and thus are relatively easy to operate. In the AR system

219

developed by Espíndola et al. (2013), users were allowed to click the buttons or use the

220

keyboard to change their view. Other types of 2D-based interactions mentioned in the reviewed

221

studies include: (1) filtering the information to view; (2) navigating into details of virtual

222

contents; and (3) modifying the visualization style and format. These kinds of 2D-based

223

interactions could also be easily performed using traditional input devices (e.g., keyboard and

224

touchpad). 3D-based input benefits users by interacting with virtual construction components

225

and resources in a natural way. The input devices allow six degrees of freedom (DOF)

226

manipulation (e.g., translation and rotation in 3D) of virtual objects. Like 2D-based input, 3D-

227

based input supports information filtering and other functions. For instance, Soria et al. (2018)

228

developed an AR system for the management of underground facilities. In this system, users

229

can rotate the digital 3D model by swiping left and right and can also delete the model from

230

both the view and the database.

231

Following the information browser, tangible interaction has been adopted in 25 studies.

232

Tangible interaction is increasingly preferred in construction because it allows users to interact

233

with virtual contents by manipulating physical objects. This is attractive to construction since

234

stakeholders often require real-time information tied to dynamically changed objects and

235

locations (Chen and Lu, 2019). Thanks to the advances in AR devices, tangible interaction can

236

be more effectively deployed. For example, the Google Tango tablet used by Ratajczak et al.

237

(2019) was equipped with a time-of-flight (ToF) camera and an IR projector, both can help to

238

accurately measure the position of the Google Tango tablet with respect to its surrounding

239

environment. This type of mobile AR device can generate the 3D point clouds of physical

240

objects in a real-time manner, which can be further processed to obtain details of these objects

241

and link the generated information to them through AR systems.

242

The review study conducted by Wang et al. (2013) showed that multidisciplinary

243

collaboration would be one of the trends for AR applications in construction. In the reviewed

244

69 studies, seven of them deployed collaborative interaction and suggested that this kind of

245

interaction is suitable for information communication and exchange in construction. For

246

instance, the engineers can make annotations about existing cables and pipes overlaid with AR

247

visualizations, and the technicians can know the locations and other information of these cables

248

and pipes before they determine whether extra pipes and cables are required (Olbrich et al.,

249

2013). Hybrid interaction is also encouraged by many previous studies since it is more adaptive

250

to the changing environment and can relieve users from screen input or external devices (e.g.,

251

mouse, keyboard, and touchpad). However, only three recent studies reported success cases of

252

hybrid interaction in construction. In these studies, users were asked to wear the head-mounted

253

device (i.e., Microsoft HoloLens) and interact with the virtual contents by their gestures (Baek

254

et al., 2019; Mascareñas et al., 2019). Despite the diverse interaction mechanisms, no study has

255

used speech or eye movement as inputs for intuitive interaction, leaving a possibility for further

256

investigation.

257

4.4. Integration of AR and other technologies

258

The technical details of the reviewed studies allowed the authors to analyze whether the

259

AR systems were used alone or in conjunction with other technologies. A notable trend is that

260

many recent research works have integrated AR with building information modeling (33

261

papers), computer vision (22 papers), and cloud computing (12 papers). As shown in Figure 5,

262

three papers even integrated AR with all these three technologies to leverage the benefits of AR

263

to construction. Nevertheless, other technologies such as Auto-ID were only mentioned in very

264

few papers and thus were omitted in the following investigations.

265

266

Figure 5. Number of publications integrating AR with BIM, computer vision and cloud

267

computing

268

4.4.1. Building information modeling

269

Throughout the life-cycle of a construction project, the delivery of many operational and

270

management tasks could be improved if the corresponding workers and managers can access to

271

not only the geometric information (e.g., size, shape, etc.) but also the non-geometric

272

information (e.g., material, function, etc.) of construction elements (Jalal et al., 2020). In these

273

cases, the integration of BIM and AR provided an ideal solution to meet such need, through

274

which advantages of BIM and AR to construction were further leveraged. A BIM model is

275

basically a digital model that contains the physical and functional details of a building. AR can

276

superimpose the BIM model onto the physical project. The merge of the as-designed BIM

277

model and real-world environment provides a vivid presentation of geometric information for

278

operational and managerial tasks. Wang et al. (2014a) presented the BIM model in a handheld

279

AR device to guide on-site assembly tasks. The comparison between the as-designed BIM and

280

the as-built situation can significantly ease the progress monitoring. Zhou et al. (2017) used AR

281

and BIM to measure the segment displacement for quality management in tunneling

282

construction. They treated the BIM model as the baseline model and compared this model with

283

actual video through an image-matching program. Likewise, Kwon et al. (2014) used an AR

284

system to match the shape information of BIM objects and images taken on the construction

285

site, and the matching results can support defect detection.

286

Apart from the geometric information, some of the reviewed studies used AR to make non-

287

geometric information readily available on site. This is extremely helpful for supporting

288

complex construction tasks. For example, Chen et al. (2016) integrated BIM into the AR system

289

so that the material information, rigging orders, and construction schedules can be

290

automatically fetched from BIM and augmented as a layer of information over the real-world

291

view of workers. Ratajczak et al. (2019) also suggested that the integration of BIM and AR

292

could provide context-specific information on tasks and relevant building components, and thus

293

enhance the quality of construction works. Regarding maintenance, some examples can be

294

found in Irizarry et al. (2014) and Soria et al. (2018) that integrated BIM and AR to facilitate

295

maintenance and repair operations for buildings and infrastructures. Their proposed systems

296

can simplify the information retrieval process, and facility managers and maintenance workers

297

can have the condition of a facility, maintenance requirements, and all other information they

298

needed for maintenance.

299

4.4.2. Computer vision

300

In its simplest definition, CV refers to training computers to generate and interpret explicit

301

information from images or videos (Ballard and Brown, 1982). CV has remarkably changed the

302

traditional construction management by enabling automatic activity recognition, object

303

tracking, and performance monitoring (Sherafat et al., 2020). The advances in CV technologies

304

and algorithms have also made them extremely suitable for AR applications since they can

305

make use of the images or videos taken by the built-in camera of an AR device to provide

306

functions such as visual tracking and registration.

307

CV technologies can benefit AR by facilitating both marker-based and marker-less tracking

308

and registration. Chi et al. (2013) found that marker-based tracking and registration methods

309

were generally adopted in studies published before 2013 but suggested that marker-less

310

methods should be more suitable for construction fields. However, among the reviewed 69

311

studies, marker-based methods were still used more frequently than marker-less methods.

312

Regular markers can be placed at different locations that are suitable for various construction

313

and operation activities (Portalés et al., 2018). When a marker is recognized, the AR device

314

would display the virtual content relevant to the location of the marker in space. Researchers

315

have also attempted to improve the marker-based tracking and registration methods so that they

316

can be applied to different scenarios. Ahn et al. (2019) proposed a two-step method, including

317

image segmentation for filtering out the markers and object detection for estimating the

318

coordinates, to perform geometric transformation for projection-based display. Mascareñas et

319

al. (2019) adopted various types of markers (e.g., barcode, QR code) in order to cope with the

320

common occurrence of facilities in nuclear infrastructure. Although marker-based methods are

321

relatively straightforward to use, they require markers being placed at suitable locations without

322

any occlusions.

323

The use of CV for marker-less tracking and registration relies on algorithms for extracting

324

features of the scenes and identifying the location and position of AR devices in the real-world

325

coordination system. Depending on the target scenes and their associated features, different

326

feature detectors and descriptors have been adopted in the reviewed studies. Examples include

327

the Speed Up Robust Features (SURF) used in Kim et al. (2018), and the Canny edge detector

328

used in Fazel and Izadi (2018). In addition, deep learning is expected to significantly improve

329

the marker-less tracking and registration for AR applications. For instance, the convolutional

330

neural network – a typical class of deep learning algorithms – has been demonstrated useful for

331

location tracking in AR-based facility management (Baek et al., 2019). Marker-less methods

332

do not need pre-installed markers and can provide accurate tracking and registration if robust

333

features are available. However, for a complex, dynamic environment like construction, the use

334

of CV for marker-less tracking and registration is vulnerable to drift and fact motion. In such

335

situations, a hybrid method that combines sensors and CV should be adopted to overcome

336

problems such as drift and occlusions (Hou et al., 2014).

337

With the tracking and registration enabled by CV, several value-adding AR services have

338

been developed. Bae et al. (2013) used CV to recognize buildings from images taken by the

339

camera of the AR device. Once a target building is recognized, the AR device can query relevant

340

information of that building and show the information to the field personnel for decision

341

making. Kim et al. (2017) proposed an AR-based safety management system in which multiple

342

objects in the images were tracked and their distances were continuously measured. This system

343

allowed managers to timely view the hazard information in an AR device, and make

344

corresponding safety instructions for the sake of risk prevention. Koch et al. (2014) and Baek

345

et al. (2019) used CV to perform indoor localization, based on which location-specific

346

information was represented in the AR device for indoor navigation and maintenance

347

operations. All these examples illustrated the integration of CV and AR can significantly ease

348

the information accessibility for users so that informed decisions can be made in a real-time

349

manner.

350

4.4.3. Cloud computing

351

Mobile AR devices are flexible and convenient for on-site personnel to use, but they

352

generally have limited memory and computing capacities for computation- and data-intensive

353

AR applications. This is extremely serious when AR is integrated with CV and BIM. Mobile

354

AR devices may not be able to run CV algorithms and render the BIM models. Therefore, cloud

355

computing has been used to facilitate the delivery of AR applications.

356

Cloud computing refers to “a model for enabling ubiquitous, convenient, on-demand

357

network access to a shared pool of configurable computing resources that can be rapidly

358

provisioned and released with minimal management effort or service provider interaction”

359

(Mell and Grance, 2011, p.2). In the reviewed studies, the integration of AR and cloud

360

computing has eased information exchange by storing the required information in a centralized

361

cloud platform. Because no information must be stored locally in the memory of an AR device,

362

wasteful duplications of information can be significantly reduced (Chu et al., 2018). This

363

benefit has been recognized in actual practices, Jiao et al. (2013) adopted a cloud-based system

364

to integrate AR and BIM in the 3D web environment and suggested that their proposed system

365

can be dynamically updated and adapted to different usage requirements. Additionally, Hou et

366

al. (2014) linked users to the object-oriented facility information that is stored on a cloud server,

367

and then augmented the information in the AR interface for real-time facility management.

368

By transferring the data processing workload to the cloud server, cloud computing can

369

relief the AR devices from heavy computational tasks and improve the efficiency of data

370

processing. In the cloud computing-based AR system developed by Olbrich et al. (2013), the

371

AR device was only responsible for simple computational tasks including the image acquisition

372

and visualization of the augmented images, and the cloud server was used to address all other

373

computational tasks such as processing the image and tracking the camera. Kim et al. (2017)

374

used a cloud server to conduct most of the processes, and the AR interface could be

375

synchronized with this cloud server by using HTML. Baek et al. (2019) allocated a computer

376

that had powerful graphics processing units as a cloud computing server in order to ensure a

377

fast deep learning-based localization. Acknowledging the benefits of cloud computing to AR

378

applications, it should be noted that the communication between cloud servers and AR devices

379

requires a stable network, which however might be extremely difficult to guarantee in a

380

dynamic construction site.

381

382

5. Future directions

383

In previous review studies, Chi et al. (2013), Rankohi and Waugh (2013), Wang et al.

384

(2013), and Behzadan et al. (2015) described several research directions of AR in construction,

385

such as expanding the applications of AR from the construction stage to other project stages,

386

the use of mobile AR devices, the integration of AR and cloud computing, human-AR

387

interaction, and context-aware AR. The findings of this study reveal that much research efforts

388

have been devoted to these directions. However, the reviewed papers still highlighted several

389

technical challenges associated with interaction and information processing processes (e.g.,

390

Zaher et al., 2018; Chalhoub and Ayer, 2019). Therefore, efforts are needed to further upgrade

391

users’ experience of AR in construction. Specifically, four possible future directions of AR in

392

construction are described as follows.

393

First, natural interactions between users and AR systems are important for streamlining the

394

deployment of AR in construction. The design of natural interaction should ensure that users

395

can easily interact with virtual content without problems such as information overload and

396

attention disruption. In a dynamic construction workspace, if workers hold mobile AR devices

397

or wear head-held AR devices while walking within the construction site, they should not be

398

required to pay much attention to the control of AR devices and cannot be disturbed by the

399

information presented to them in order to avoid safety accidents. Therefore, user-centric

400

interaction design has been advocated since it fully considers how to help users interpret the

401

virtual contents overlaid onto the physical environment (Eitoku et al., 2006). Future studies can

402

explore the user-centric design of human-AR interaction according to the nature of cognition

403

on the user’s performance when merging multimodal cues in the real-world environment.

404

Moreover, future studies can propose systematic methods to evaluate the human-AR interaction

405

in construction. Both subjective metrics (e.g., cognitive load and fatigue) and objective metrics

406

(e.g., the time and accuracy of task delivery) should be used to comprehensively measure the

407

effectiveness and efficiency of the control mechanisms and information presentation of AR in

408

various application scenarios.

409

Secondly, two major issues associated with the integration of BIM and AR have been

410

highlighted. One is related to the reduction of the model complexity and the other one is

411

associated with the communication of BIM information to an AR system, both concerning how

412

the model can be correctly visualized without the loss of essential information. Irizarry et al.

413

(2013) introduced some manual complexity reduction techniques, which include the welding

414

of overlapping vertices, the elimination of unnecessary geometry, and the simplification of

415

mesh. Tools including Vizup and Autodesk 3ds Max have also been used for polygon

416

decimation (Singh and Delhi, 2018). Depending on the platforms for developing the BIM

417

models and AR applications, various data formats and the corresponding format transfer

418

methods have been used for BIM-AR integration. Some of these methods need one time of

419

conversion (e.g., “HOOPS-X3D” in Jiao et al. [2013] and “RVT-IFC+DAE” in [Williams et

420

al., 2015]), and others require a series of conversions (e.g., “IFC-OBJ/MTL-L3D” in [Meža et

421

al., 2014] and “RVT-MAX-FBX-FBX-WT3” in [Chu et al., 2018]). Acknowledging the

422

availability of various methods for different scenarios, it should be noted that a full conversion

423

has hardly been achieved, and manual adjustments of the converted BIM models are often

424

needed. Therefore, more robust conversion methods are expected to be developed and validated

425

in actual projects.

426

Thirdly, the augmentation of the real and virtual contents should be more compelling for

427

construction stakeholders. When stakeholders conduct construction activities, they often need

428

information about the surrounding environment in addition to the design drawings or other

429

readily-available documents (Tsai, 2014). However, most of the reviewed AR systems treated

430

the physical environment as background, without deriving any semantic information from it for

431

decision making. It is thus expected that, in future AR applications, the information presented

432

to user-ends should not be derived solely from the virtual content, but the semantic

433

understanding of the surrounding environment. A few studies have used computer vision

434

technologies to help AR systems to understand the semantic information of objects in the

435

construction and operation stages, but an extensive database of reality should be developed to

436

make AR responsive to providing context-aware information.

437

Finally, AR can be integrated with a new computing paradigm – edge computing – to

438

improve the quality of service for computation-intensive applications. Although previous

439

studies have adopted cloud computing to provide such function in construction, the use of cloud

440

computing has been facing challenges such as high latency, resource consumption, and unstable

441

network in construction sites. In contrast, edge computing is a distributed computing paradigm

442

that transfer data processing and analysis close to the edge of data sources or networks

443

(Satyanarayanan, 2017). Advantages of edge computing include shifting the storage and

444

computation load from center to edge, reducing ingress bandwidth into the cloud, enabling a

445

real-time response, reducing latency, and enhancing scalability (Shi et al., 2016). Such

446

advantages make edge computing suitable for the rapid delivery of AR services. In the

447

manufacturing industry, Fernández-Caramés et al. (2018) has integrated AR with edge

448

computing to provide dynamic on-demand information and analyzed the benefits of edge

449

computing over cloud computing. Nevertheless, one cannot simply apply Fernández-Caramés

450

et al. (2018)’s solution to construction since the working environment and application scenarios

451

in the manufacturing and construction are significantly different. Therefore, future studies can

452

focus on how to deploy the edge computing for different scenarios in construction and how to

453

integrate AR and edge computing to enable fast AR device communications for various users

454

in construction projects.

455

456

6. Limitations of the review method

457

Some limitations of the review method should be recognized. In the literature search stage,

458

the construction term used in the search query, e.g., building or construction, is relatively

459

general, which led to the fact that the initial set of collected papers contained many irrelevant

460

studies. Additional search criteria can help to narrow the results but will increase the risk of

461

omitting important references. The combination of database searches and snowballing might be

462

a more suitable method for future review studies. Moreover, the filtering process was done

463

manually in this study by reading the abstracts of all 546 papers. The efficiency of this process,

464

however, can be significantly improved by using advanced text-mining tools such as

465

RobotAnalyst (Przybyła et al., 2018).

466

467

7. Conclusions

468

By taking research work published from 2013 to 2019 as recent advances, this study

469

reviewed 69 journal articles on AR in construction. Several key findings have been drawn from

470

this review. First, a renaissance of AR research in constriction is observed, but most of the

471

reviewed studies still focused on the construction stage. Secondly, design assessment, process

472

monitoring, and maintenance management and operation were the most widely implemented

473

AR applications in the design, construction, and operation stages, respectively. Thirdly,

474

information browser and tangible interaction were the major interaction mechanisms of AR in

475

construction, and hybrid interaction was only enabled by advanced AR systems for a few

476

application scenarios. Fourthly, AR has been integrated with BIM, CV, and cloud computing

477

in order to leverage its benefits to construction. BIM and CV can significantly improve the

478

virtual and physical contents presented to users; cloud computing shares the majority of

479

computation-intensive tasks with AR devices for rapid response in more complex applications.

480

The review also identifies some unaddressed issues, including natural interactions between

481

users and AR systems, seamless conversion of BIM to AR, merging virtual contents with

482

information obtained from the physical environment, and effective strategies for computation-

483

and data-intensive AR applications. Therefore, four possible directions for future search can be

484

proposed: (1) AR systems should follow a user-centric interaction design so that users can enjoy

485

more intuitive interactions for improved user experience; (2) Development of robust conversion

486

methods for BIM and AR integration; (3) Technical progress in semantic understanding of the

487

physical objects and surrounding environment is expected to unleash the full power of AR to

488

construction; and (4) Future deployment of AR in construction could make use of edge

489

computing which provides the ubiquitous capability of heterogeneous computing for AR

490

applications.

491

492

Acknowledgments

493

This research was funded by the Fundamental Research Funds for the Central Universities, grant

494

number 2019kfyXJJS185.

495

496

References

497

Ahn, S., Han, S. and Al-Hussein, M. (2019), “2D drawing visualization framework for applying

498

projection-based augmented reality in a panelized construction manufacturing facility:

499

Proof of concept”, Journal of Computing in Civil Engineering, Vol. 33 No. 5, p. 04019032,

500

doi: 10.1061/(ASCE)CP.1943-5487.0000843.

501

Alsafouri, S. and Ayer, S.K. (2019), “Mobile augmented reality to influence design and

502

constructability review sessions”, Journal of Architectural Engineering, Vol. 25 No. 3, p.

503

04019016, doi: 10.1061/(ASCE)AE.1943-5568.0000362.

504

505

Azuma, R. (1997), “A survey of augmented reality”, Presence: Teleoperators & Virtual
Environments, Vol. 6 No. 4, pp. 355-385, doi: 10.1162/pres.1997.6.4.355.

506

Azuma, R., Baillot, Y., Behringer, R., Feiner, S., Julier, S. and MacIntyre, B. (2001), “Recent

507

advances in augmented reality”, IEEE Computer Graphics and Applications, Vol. 21 No.

508

6, pp. 34-47, doi: 10.1109/38.963459.

509

Bae, H., Golparvar-Fard, M. and White, J. (2013), “High-precision vision-based mobile

510

augmented reality system for context-aware architectural, engineering, construction and

511

facility management (AEC/FM) applications”, Visualization in Engineering, Vol. 1 No. 1,

512

p. 3, doi: 10.1186/2213-7459-1-3.

513

Baek, F., Ha, I. and Kim, H. (2019), “Augmented reality system for facility management using

514

image-based indoor localization”, Automation in Construction, Vol. 99, pp. 18-26, doi:

515

10.1016/j.autcon.2018.11.034.

516

517

Ballard, D.H. and Brown, C.M. (1982), Computer vision. Englewood Cliffs. J: Prentice Hall,
ISBN-10: 0-13-165316-4.

518

Behzadan, A.H., Dong, S. and Kamat, V.R. (2015), “Augmented reality visualization: A review

519

of civil infrastructure system applications”, Advanced Engineering Informatics, Vol. 29

520

No. 2, pp. 252-267, doi: 10.1016/j.aei.2015.03.005.

521

Billinghurst, M., Clark, A. and Lee, G. (2015), “A survey of augmented reality”, Foundations

522

and Trends® in Human–Computer Interaction, Vol. 8 No. 2-3, pp. 73-272, doi:

523

10.1561/1100000049.

524

525

Billinghurst, M. and Kato, H. (2002), “Collaborative augmented reality”, Communications of
the ACM, Vol. 45 No. 7, pp. 64-70, doi: 10.1145/514236.514265.

526

Calderon-Hernandez, C. and Brioso, X. (2018), “Lean, BIM and augmented reality applied in

527

the design and construction phase: A literature review”, International Journal of

528

Innovation, Management and Technology, Vol. 9 No. 1, pp. 60-63, doi:

529

10.18178/ijimt.2018.9.1.788.

530

Chalhoub, J. and Ayer, S.K. (2019), “Effect of varying task attributes on augmented reality

531

aided point layout”, Journal of Information Technology in Construction, Vol. 24, pp. 95-

532

111, url: http://www.itcon.org/2019/6.

533

Chen, K. and Lu, W. (2019), “Bridging BIM and building (BBB) for information management

534

in construction: The underlying mechanism and implementation”, Engineering,

535

Construction and Architectural Management, Vol. 26 No. 7, pp. 1518-1532, doi:

536

10.1108/ECAM-05-2018-0206.

537

Chen, K., Lu, W., Peng, Y., Rowlinson, S. and Huang, G.Q. (2015), “Bridging BIM and

538

building: From a literature review to an integrated conceptual framework”, International

539

Journal

540

10.1016/j.ijproman.2015.03.006.

of

Project

Management,

Vol.

33

No.

6,

pp.

1405-1416,

doi:

541

Chen, Y.C., Chi, H.L., Kang, S.C. and Hsieh, S.H. (2016), “Attention-based user interface

542

design for a tele-operated crane”, Journal of Computing in Civil Engineering, Vol. 30 No.

543

3, p. 04015030, doi: 10.1061/(ASCE)CP.1943-5487.0000489.

544

Chi, H.L., Kang, S.C. and Wang, X. (2013), “Research trends and opportunities of augmented

545

reality applications in architecture, engineering, and construction”, Automation in

546

Construction, Vol. 33, pp. 116-122, doi: 10.1016/j.autcon.2012.12.017.

547

Chu, M., Matthews, J. and Love, P.E. (2018), “Integrating mobile building information

548

modelling and augmented reality systems: An experimental study”, Automation in

549

Construction, Vol. 85, pp. 305-316, doi: 10.1016/j.autcon.2017.10.032.

550

Clarivate

Analytics

(2019),

Web

of

Science

Core

Collection,

available

551

https://clarivate.com/webofsciencegroup/solutions/web-of-science-core-collection/

552

(accessed 30 December 2019).

at:

553

Dong, S., Feng, C. and Kamat, V.R. (2013), “Sensitivity analysis of augmented reality-assisted

554

building damage reconnaissance using virtual prototyping”, Automation in Construction,

555

Vol. 33, pp. 24-36, doi: 10.1016/j.autcon.2012.09.005.

556

Eitoku, S., Tanikawa, T. and Suzuki, Y. (2006), “Display composed of water drops for filling

557

space with materialized virtual three-dimensional objects”, IEEE Virtual Reality

558

Conference, IEEE, pp. 159-166, doi: 10.1109/VR.2006.51.

559

Elghaish, F., Matarneh, S., Talebi, S., Kagioglou, M., Hosseini, M.R. and Abrishami, S. (2020),

560

“Toward digitalization in the construction industry with immersive and drones

561

technologies: a critical literature review”, Smart and Sustainable Built Environment, doi:

562

10.1108/SASBE-06-2020-0077.

563

Elsevier

(2017),

Scopus

Content

Coverage,

564

https://www.elsevier.com/solutions/scopus/how-scopus-works/content

565

December 2019).

available
(accessed

at:
30

566

Espíndola, D.B., Fumagalli, L., Garetti, M., Pereira, C.E., Botelho, S.S. and Henriques, R.V.

567

(2013), “A model-based approach for data integration to improve maintenance

568

management by mixed reality”, Computers in Industry, Vol. 64 No. 4, pp. 376-391, doi:

569

10.1016/j.compind.2013.01.002.

570

Fazel, A. and Izadi, A. (2018), “An interactive augmented reality tool for constructing free-

571

form modular surfaces”, Automation in Construction, Vol. 85, pp. 135-145, doi:

572

10.1016/j.autcon.2017.10.015.

573

Fernández-Caramés, T., Fraga-Lamas, P., Suárez-Albela, M. and Vilar-Montesinos, M. (2018),

574

“A fog computing and cloudlet based augmented reality system for the industry 4.0

575

shipyard”, Sensors, Vol. 18 No. 6, p. 1798, doi: 10.3390/s18061798.

576

Gartner

(2018),

Hype

Cycle

for

Emerging

Technologies

2018,

available

at:

577

https://www.gartner.com/smarterwithgartner/5-trends-emerge-in-gartner-hype-cycle-for-

578

emerging-technologies-2018/ (accessed 30 December 2019).

579

Golparvar-Fard, M. and Ham, Y. (2014), “Automated diagnostics and visualization of potential

580

energy performance problems in existing buildings using energy performance augmented

581

reality models”, Journal of Computing in Civil Engineering, Vol. 28 No. 1, pp. 17-29, doi:

582

10.1061/(ASCE)CP.1943-5487.0000311.

583

He, Q., Wang, G., Luo, L., Shi, Q., Xie, J. and Meng, X. (2017), “Mapping the managerial areas

584

of Building Information Modeling (BIM) using scientometric analysis”, International

585

Journal

586

10.1016/j.ijproman.2016.08.001

of

Project

Management,

Vol.

35

No.

4,

pp.

670-685,

doi:

587

Henderson, S. and Feiner, S. (2010), “Exploring the benefits of augmented reality

588

documentation for maintenance and repair”, IEEE Transactions on Visualization and

589

Computer Graphics, Vol. 17 No. 10, pp. 1355-1368, doi: 10.1109/TVCG.2010.245.

590

Hou, L., Wang, Y., Wang, X., Maynard, N., Cameron, I.T., Zhang, S. and Jiao, Y. (2014),

591

“Combining photogrammetry and augmented reality towards an integrated facility

592

management system for the oil industry”, Proceedings of the IEEE, Vol. 102 No. 2, pp.

593

204-220, doi: 10.1109/JPROC.2013.2295327.

594

Irizarry, J., Gheisari, M., Williams, G. and Roper, K. (2014), “Ambient intelligence

595

environments for accessing building information”, Facilities, Vol. 32 No. 3/4, pp. 120-

596

138, doi: 10.1108/F-05-2012-0034.

597

Irizarry, J., Gheisari, M., Williams, G. and Walker, B.N. (2013), “InfoSPOT: A mobile

598

Augmented Reality method for accessing building information through a situation

599

awareness approach”, Automation in Construction, Vol. 33, pp. 11-23, doi:

600

10.1016/j.autcon.2012.09.002.

601

Jalal, M.P., Roushan, T.Y., Noorzai, E. and Alizadeh, M. (2020), “A BIM-based construction

602

claim management model for early identification and visualization of claims”, Smart and

603

Sustainable Built Environment, doi: 10.1108/SASBE-10-2019-0141.

604

Jesson, J., Matheson, L. and Lacey, F.M. (2011), Doing Your Literature Review: Traditional

605

and Systematic Techniques, Sage Publications, London, ISBN-10: 9781848601543.

606

Jiao, Y., Zhang, S., Li, Y., Wang, Y. and Yang, B. (2013), “Towards cloud augmented reality

607

for construction application by BIM and SNS integration”, Automation in Construction,

608

Vol. 33, pp. 37-47, doi: 10.1016/j.autcon.2012.09.018.

609

Kato, H., Billinghurst, M., Poupyrev, I., Imamoto, K. and Tachibana, K. (2000), “Virtual object

610

manipulation on a table-top AR environment”, Proceedings IEEE and ACM International

611

Symposium on Augmented Reality, pp. 111-119, doi: 10.1109/ISAR.2000.880934.

612

Kim, H.S., Kim, S.K., Borrmann, A. and Kang, L.S. (2018), “Improvement of realism of 4D

613

objects using augmented reality objects and actual images of a construction site”, KSCE

614

Journal of Civil Engineering, Vol. 22 No. 8, pp. 2735-2746, doi: 10.1007/s12205-017-

615

0734-3.

616

Kim, K., Kim, H. and Kim, H. (2017), “Image-based construction hazard avoidance system

617

using augmented reality in wearable device”, Automation in Construction, Vol. 83, pp.

618

390-403, doi: 10.1016/j.autcon.2017.06.014.

619

Koch, C., Neges, M., König, M. and Abramovici, M. (2014), “Natural markers for augmented

620

reality-based indoor navigation and facility maintenance”, Automation in Construction,

621

Vol. 48, pp. 18-30, doi: 10.1016/j.autcon.2014.08.009.

622

Kwon, O.S., Park, C.S. and Lim, C.R. (2014), “A defect management system for reinforced

623

concrete work utilizing BIM, image-matching and augmented reality”, Automation in

624

Construction, Vol. 46, pp. 74-81, doi: 10.1016/j.autcon.2014.05.005.

625

Li, X., Yi, W., Chi, H.L., Wang, X. and Chan, A.P. (2018), “A critical review of virtual and

626

augmented reality (VR/AR) applications in construction safety”, Automation in

627

Construction, Vol. 86, pp. 150-162, doi: 10.1016/j.autcon.2017.11.003.

628

Mascarenas, D., Harden, T., Morales, J., Boardman, B., Sosebee, E., Blackhart, C., ... Dasari,

629

S. (2019), “Augmented reality for enabling smart nuclear infrastructure”, Frontiers in

630

Built Environment, Vol. 5, p. 82, doi: 10.3389/fbuil.2019.00082.

631

632

Mell, P. and Grance, T. (2011), The NIST definition of cloud computing, National Institute of
Standards and Technology.

633

Meža, S., Turk, Ž. and Dolenc, M. (2014), “Component based engineering of a mobile BIM-

634

based augmented reality system”, Automation in Construction, Vol. 42, pp. 1-12, doi:

635

10.1016/j.autcon.2014.02.011.

636

Mok, K.Y., Shen, G.Q. and Yang, J. (2015), “Stakeholder management studies in mega

637

construction projects: A review and future directions”, International Journal of Project

638

Management, Vol. 33 No. 2, pp. 446-457, doi: 10.1016/j.ijproman.2014.08.007.

639

640

Moore, H.F. and Gheisari, M. (2019), “A Review of virtual and mixed reality applications in
construction safety literature”, Safety, Vol. 5 No. 3, p. 51, doi: 10.3390/safety5030051.

641

Newman, C., Edwards, D., Martek, I., Lai, J., Thwala, W.D. and Rillie, I. (2020), “Industry 4.0

642

deployment in the construction industry: a bibliometric literature review and UK-based

643

case study”, Smart and Sustainable Built Environment, doi: 10.1108/SASBE-02-2020-

644

0016.

645

Olbrich, M., Graf, H., Kahn, S., Engelke, T., Keil, J., Riess, P., ... Picinbono, G. (2013),

646

“Augmented reality supporting user-centric building information management”, The

647

Visual Computer, Vol. 29 No. 10, pp. 1093-1105, doi: 10.1007/s00371-013-0840-2.

648

Portalés, C., Casas, S., Gimeno, J., Fernández, M., Poza, M. (2018), “From the paper to the

649

tablet: On the design of an AR-based tool for the inspection of pre-fab buildings.

650

preliminary results of the SIRAE project”, Sensors, Vol. 18 No. 4, p. 1262, doi:

651

10.3390/s18041262.

652

Przybyła, P., Brockmeier, A.J., Kontonatsios, G., Le Pogam, M.A., McNaught, J., von Elm, E.,

653

... Ananiadou, S. (2018), “Prioritising references for systematic reviews with

654

RobotAnalyst: a user study”, Research Synthesis Methods, Vol. 9 No. 3, pp. 470-488, doi:

655

10.1002/jrsm.1311.

656

Rankohi, S. and Waugh, L. (2013), “Review and analysis of augmented reality literature for

657

construction industry”, Visualization in Engineering, Vol. 1 No. 1, p. 9, doi:

658

10.1186/2213-7459-1-9.

659

Ratajczak, J., Riedl, M. and Matt, D.T. (2019), “BIM-based and AR application combined with

660

location-based management system for the improvement of the construction

661

performance”, Buildings, Vol. 9 No. 5, p. 118, doi: 10.3390/buildings9050118.

662

663

Satyanarayanan, M. (2017), “The emergence of edge computing”, Computer, Vol. 50 No. 1,
pp. 30-39, doi: 10.1109/MC.2017.9.

664

Sherafat, B., Ahn, C.R., Akhavian, R., Behzadan, A.H., Golparvar-Fard, M., Kim, H., ... Azar,

665

E.R. (2020), “Automated methods for activity recognition of construction workers and

666

equipment: State-of-the-art review”, Journal of Construction Engineering and

667

Management, Vol. 146 No. 6, p. 03120002, doi: 10.1061/(ASCE)CO.1943-7862.0001843.

668

Shi, W., Cao, J., Zhang, Q., Li, Y. and Xu, L. (2016), “Edge computing: Vision and challenges”,

669

IEEE

Internet

of

Things

670

10.1109/JIOT.2016.2579198.

Journal,

Vol.

3

No.

5,

pp.

637-646,

doi:

671

Shin, D.H. and Dunston, P.S. (2008), “Identification of application areas for Augmented Reality

672

in industrial construction based on technology suitability”, Automation in Construction,

673

Vol. 17 No. 7, pp. 882-894, doi: 10.1016/j.autcon.2008.02.012.

674

Singh, A.R. and Delhi, V.S.K. (2018), “User behaviour in AR-BIM-based site layout planning”,

675

International Journal of Product Lifecycle Management, Vol. 11 No. 3, pp. 221-244, doi:

676

10.1504/IJPLM.2018.094715.

677

Soria, G., Alvarado, L.O. and Feito, F.R. (2018), “Augmented and virtual reality for

678

underground facilities management”, Journal of Computing and Information Science in

679

Engineering, Vol. 18 No. 4, p. 041008, doi: 10.1115/1.4040460.

680

Tsai, M.K. (2014), “Streamlining information representation during construction accidents”,

681

KSCE Journal of Civil Engineering, Vol. 18 No. 7, pp. 1945-1954, doi: 10.1007/s12205-

682

014-0240-9.

683

Tsai, M.K. and Yau, N.J. (2014), “Using mobile disaster response system in bridge

684

management”, Journal of Civil Engineering and Management, Vol. 20 No. 5, pp. 737-745,

685

doi: 10.3846/13923730.2013.802731.

686

Tonn, C., Petzold, F., Bimber, O., Grundhöfer, A. and Donath, D. (2008), “Spatial augmented

687

reality for architecture—designing and planning with and within existing buildings”,

688

International Journal of Architectural Computing, Vol. 6 No. 1, pp. 41-58, doi:

689

10.1260/147807708784640126.

690

van Krevelen, D.W.F. and Poelman, R. (2010), “A survey of augmented reality technologies,

691

applications and limitations”, The International Journal of Virtual Reality, Vol. 9 No. 2,

692

pp. 1-20, doi: 10.20870/IJVR.2010.9.2.2767.

693

Wagner, D. and Schmalstieg, D. (2003), “First steps towards handheld augmented reality”,

694

Proceedings of the Seventh IEEE International Symposium on Wearable Computers,

695

IEEE, pp. 127-135, doi: 10.1109/ISWC.2003.1241402.

696

Wang, X., Kim, M.J., Love, P.E. and Kang, S.C. (2013), “Augmented Reality in built

697

environment: Classification and implications for future research”, Automation in

698

Construction, Vol. 32, pp. 1-13, doi: 10.1016/j.autcon.2012.11.021.

699

Wang, X., Truijens, M., Hou, L., Wang, Y. and Zhou, Y. (2014a), “Integrating Augmented

700

Reality with Building Information Modeling: Onsite construction process controlling for

701

liquefied natural gas industry”, Automation in Construction, Vol. 40, pp. 96-105, doi:

702

10.1016/j.autcon.2013.12.003.

703

Wang, J., Wang, X., Shou, W. and Xu, B. (2014b), “Integrating BIM and augmented reality for

704

interactive architectural visualization”, Construction Innovation, Vol. 14 No. 4, pp. 453-

705

476, doi: 10.1108/CI-03-2014-0019.

706

Webster, J. and Watson, R.T. (2002), “Analyzing the past to prepare for the future: Writing a

707

literature

review”,

708

https://www.jstor.org/stable/4132319.

MIS

Quarterly,

Vol.

26

No.

2,

pp.

xiii-xxiii,

url:

709

Williams, G., Gheisari, M., Chen, P.J. and Irizarry, J. (2015), “BIM2MAR: an efficient BIM

710

translation to mobile augmented reality applications”, Journal of Management in

711

Engineering, Vol. 31 No. 1, p. A4014009, doi: 10.1061/(ASCE)ME.1943-5479.0000315.

712

Yeh, K.C., Tsai, M.H. and Kang, S.C. (2012), “On-site building information retrieval by using

713

projection-based augmented reality”, Journal of Computing in Civil Engineering, Vol. 26

714

No. 3, pp. 342-355, doi: 10.1061/(ASCE)CP.1943-5487.0000156.

715

Zaher, M., Greenwood, D. and Marzouk, M. (2018), "Mobile augmented reality applications

716

for construction projects", Construction Innovation, Vol. 18 No. 2, pp. 152-166, doi:

717

10.1108/CI-02-2017-0013.

718

Zhou, F., Duh, H.B.L. and Billinghurst, M. (2008), “Trends in augmented reality tracking,

719

interaction and display: A review of ten years of ISMAR”, Proceedings of the 7th

720

IEEE/ACM International Symposium on Mixed and Augmented Reality, pp. 193-202, doi:

721

10.1109/ISMAR.2008.4637362.

722

Zhou, Y., Luo, H. and Yang, Y. (2017), “Implementation of augmented reality for segment

723

displacement inspection during tunneling construction”, Automation in Construction, Vol.

724

82, pp. 112-121, doi: 10.1016/j.autcon.2017.02.007.