Nof S.Y. Springer Handbook of Automation

Подождите немного. Документ загружается.

Construction Automation 61.2 Background 1065

once performed by workers. This practice has removed

workers from hazardous construction environments, but

those workers that need to remain in place are still

vulnerable to accidents. Automation efforts may fur-

ther reduce the possibility of accidents or near misses

by creating a sensor-based network that tracks the po-

sition of workers and equipment, thereby alerting the

worker and supervisor when a hazardous condition

arises.

Productivity improvement constitutes another rea-

son for automating construction operations. This im-

provement is critical since traditionally the construction

industry has been one of the worst industries with

regard to annual increase in productivity [61.13].

This concern, combined with ever-increasing costs,

high accident rates, late completions, and poor qual-

ity, has been the subject of dedicated studies on

construction productivity improvement [61.14]. Con-

struction is one of the largest product-based industries,

and contributes a major portion to the gross do-

mestic product of both developed and developing

countries [61.15, 16]. However, when compared with

manufacturing, construction is rather slow in terms

of technological progress [61.17]. There are also dif-

ferences associated with purchasing conditions, risk,

market environment, sales network, product unique-

ness, and project format [61.18]. Construction is often

considered an antiquated industry. There have not been

dramatic changes in basic construction methods in the

last 40years. The modest attention to research and

development by both the public and private construc-

tion sectors has undermined possible breakthroughs in

construction automation that may impact the overall

productivity of the industry. Furthermore, the assumed

endless availability of workers and the focus on cost

reduction and short-term efﬁciency have continued to

limit the rationale for automation efforts in construc-

tion [61.19].

61.2 Background

The historical development of construction automa-

tion has been marked by equipment inventions aimed

at performing speciﬁc tasks originally done by work-

ers, and by ground-breaking methodologies intended

for improving the systematic behavior of resources in

a construction project setting. Table 61.1 shows the his-

torical development of construction automation, from

the early stages of equipment inventions to the latest

trends inautomated projectcontrol and decisionsupport

systems. Since construction has been mostly an adopter

of innovation from other ﬁelds rather than a source of

innovation, every development indicated in the table is

shown in the period where it had a dramatic impact on

construction means, management, and methods, and not

necessarily when it was invented.

Table 61.1 includes chronological developments

related to technologies, means, and methods that

have had a dramatic impact on the way construction

operations are performed, managed, and conceived,

allowing them to be partially or fully automated.

The discipline of construction management studies

the practice of the managerial, technological, and

business-oriented characteristics of the construction

industry, whereas construction engineering analyzes

design, operational, and constructability aspects. This

differentiation has been reﬂected in the way several

automation efforts have been focused throughout the

last few decades. These efforts and periods have been

catalogued in different manners: based on the tax-

onomy of ﬁeld operations [61.3, 20, 21]; based on

geographical development [61.6, 7];basedonmatu-

rity [61.22]; and whether technologies are considered

hard or soft [61.23]. This chapter describes automa-

tion efforts from the perspective of their purpose in the

construction domain, that is, whether the automation

effort has been focused primarily on the construction

of highway or heavy structures, on buildings, or on

computer-supported integrated technologies that facili-

tate construction management decision making, which

can be applied to the design, procurement, assem-

bly, construction, maintenance, and management of any

type of facility.

Part G 61.2

1066 Part G Infrastructure and Service Automation

Table 61.1 Historical development of construction automation

Period Development

1100s Pulleys, levers

1400s Cranes

1500s Pile driver

1800s Elevators, steam shovels, internal combustion engine, power tools, reinforced concrete

1900s Slip-form construction

1910s Gantt charts, work breakdown structures (WBS)

1920s Dozers, engineering vehicles

1930s Prefabrication, hydraulic power, concrete pumps

1950s Project evaluation and review technique (PERT), computers

1960s Time-lapse studies, critical path method (CPM)

1970s Robotics, computer-aided design (CAD), discrete-event simulation

1980s 3-D CAD, 4-D CAD, massiﬁcation of personal computers, spreadsheets, relational databases, geographic

information systems (GIS), large-scale manipulators

1990s Internet, intranets, extranets, personal digital assistants (PDAs), global positioning systems (GPS),

barcodes, radiofrequency identiﬁcation systems (RFID), wireless communications, remote sensing, pre-

cision laser radars (LADARS), enterprise resource planning (ERP), object-oriented programming (OOP),

concurrent engineering, industry foundation classes (IFC), building information models (BIM), lean

construction (LC)

2000s Web-based project management, e-Work, parametric modeling, Wi-Fi, ultra wide band (UWB) for track-

ing and positioning, machine vision, mixed augmented reality, nanotechnology

61.3 Horizontal Construction Automation

Horizontal construction has been prone to automa-

tion due to the repetitive tasks, intensive labor, and

equipment involved in the operation. This type of

construction comprises linear projects, such as road

construction, paving, drilling, trench excavation, and

pipe laying.

One of the ﬁrst attemptsto fullyautomate the paving

process was the Road Robot [61.24]. The aim of this

project was to develop a self-navigating, self-steering

asphalt paver that would allow road engineers to im-

prove the quality of pavements, while also being more

environmentally friendly. The operation of the Road

Robot was divided among four subsystems: asphalt

materials logistics, traveling mechanism, road surface

geometry, and screeding. Although the Road Robot suc-

cessfully demonstrated the capabilities and advantages

of a fully automated asphalt paver, further development

appears to have been halted.

The computer-integrated road paving (CIRPAV)

prototype was another attempt to automate paving op-

erations [61.25]. The primary functions of the CIRPAV

system were to assist the operator in maintaining the

paver on its correct trajectory at the correct speed,

to automatically adjust the position and cross-slope

of the screed, and to record actual work performed

by the paver and transmit performance data to a re-

mote ground station, in order to maintain global quality

control at the site level. The CIRPAV system con-

sists of three main subsystems: the ground subsystem,

the onboard subsystem, and the positioning subsys-

tem. After several trials, the following improvements

were achieved, in contrast with the conventional as-

phalt paving process: the costs of establishing and

maintaining references for proﬁle control and equip-

ment operations were reduced from 10% of the total

cost of the work to below 5%; the ﬂuctuation of the

Part G 61.3

Construction Automation 61.3 Horizontal Construction Automation 1067

Fig. 61.1a,b Slip-form pavers: (a) screeding process,

(b) idle workers during operation [61.26]

layer thickness was decreased, with estimated sav-

ings of materials of about 5% of the total cost of

the work; and the quality of the ﬁnal pavement was

improved. The CIRPAV prototype was able to place

asphalt within ±5cm in both transversal and longitu-

Two

controllable

struts align

pipe

Laser

target

above

laser point

a) b) c)

Fig. 61.2a–c Construction excavation automation: (a) robotic excavator [61.27], (b) Lancaster University Computerized

Intelligent Excavator (LUCIE) [61.28],

dinal directions, and within ±0.5mm for the height

component.

The appearance of slip-form pavers semi-automated

concrete paving operations. Figure 61.1a shows an au-

tomated concrete screeding process.

However, these operations still require several la-

borers who remain idle most of the process time, as

seen in Fig.61.1b. Furthermore,slip-form machinesstill

depend on visual inspection and manual samples to per-

form quality controlof the concretemix. Inrecent years,

a prototype design of a fully autonomous robot for con-

crete paving was developed [61.30]. The Robopaver

prototype is a battery-operated robot consisting of sev-

eral different operations: placing prefabricated steel

reinforcement bar cages, placing and distributing con-

crete, vibrating, screeding, ﬁnal ﬁnishing, and curing.

Results from the simulation ofprototypical tests yielded

productivity improvements for Robopaver of 20% over

the traditional slip-form paver, while foreman utiliza-

tion achieved 99% with no operators involved, as

opposed to the traditional operation with 46% foreman

utilization and six laborers [61.31]. The construction

work zone for Robopaver was less prone to acci-

dents involving construction workers, while being more

productive.

Trench excavation and laying buried pipes are

among the most dangerous tasks in the construction

industry. While heavy construction equipment such

as cranes, loaders or backhoe excavators are used to

dig, hoist, and lower pieces of pipe into the trench,

workers guide the operation from inside the trench to

perform ﬁnal alignment and jointing. Excavating op-

erations, such as trenching, require precise control.

Part G 61.3

1068 Part G Infrastructure and Service Automation

Previous experiments with robotic excavation have im-

plemented a conventional industrial robot ﬁtted with

a bucket as the end-effector [61.32,33]. More recently,

a prototype Komatsu PC05-7 hydraulic mini-excavator

was extensively modiﬁed to operate as an autonomous

robotic excavator [61.27]. Figure 61.2a,b shows modi-

ﬁed robotic excavators during trench-forming tasks.

Research in the telerobotic operation of hoisting

and placing pipes on trenches has shown promising

results for improving safety. Telerobotic systems are

mechanical devices that combine human and machine

intelligence to perform tasks remotely with the as-

sistance of various sensors, computers, man–machine

interface devices, and electronic controls. Building

upon a previous-generation pipe manipulator dubbed

PipeMan, further improvements were made by adding

a laser and video system in order for the operator to

control the entire device remotely, becoming a rugged

but simple man–machine interface for motion control

and control feedback, as shown in Fig.61.2c. Pipe in-

stallation tasks could be initiated and observed by the

operator with the help of wireless ﬁdelity (Wi-Fi)in-

terfaces to the electrohydraulic valves mounted on the

manipulator, as well as the video images transmitted

wireless to the ﬂat screen mounted to the side of the

cabin window [61.29].

61.4 Building Construction Automation

In the 1990s, despite numerous attempts to develop

highly automated machines and robotics for con-

struction ﬁeld operations in the previous decade, few

practical applications could be found on construction

sites [61.21]. InJapan, however,pushed bybuilding cor-

porations and manufacturers, the largest construction

ﬁrms of the time developed robots for building con-

struction. Among these ﬁrms (e.g., Takenaka, Shimizu,

Taisei, Kajima, Obayashi, and Kumagai-Gumi), a va-

riety of single-task robots were manufactured for

practical construction applications. These applications

mainly consisted of concrete ﬂoor ﬁnishing, exterior

wall spray painting, ceiling board installation, and

ﬁre prooﬁng. The goals of the deployment of these

single-task robots were mostly the improvement of pro-

ductivity, safety, and quality.

a) b) c)

Fig. 61.3a–c Automated building systems in Japan: (a) SMART [61.34], (b) Big Canopy [61.35], and (c) ABCS [61.36]

Japanese contractors also developed automated

building systems, which consisted of on-site construc-

tion factories that used ideas already tested in the

manufacturing and automobile industries, such as just-

in-time, material tracking or streamlining repetitive

operations [61.7].

The WASeda constructionrobot (WASCOR) project

entailed a building system carried out by assem-

bling factory-made interior units installed with frames,

boards, papers, and ﬁxtures, and using construction

robots [61.37]. The push-up method consisted of assem-

bling the roof ﬂoor ﬁrst, lifting it up with its supporting

columns using hydraulic jacks, thereby serving as the

working platform for the construction of the lower

ﬂoors. Every time a lower ﬂoor was completed, the

roof ﬂoor was jacked up [61.7]. Shimizu manufactur-

Part G 61.4

Construction Automation 61.4 Building Construction Automation 1069

ing system by advanced robotics technology (SMART)

was an integrated system that automated the erection

and welding of steel frames, laid concrete ﬂoor boards,

and installed exterior and interior wall panels and other

components. The automated building construction sys-

tem (ABCS) system consisted of a construction factory

placed on top of the building, lifting structural members

to the lower ﬂoors and welding the components with

robots. It had the capacity to build two ﬂoors at once.

Productivity shortcomings and cost considerations as-

sociated with the operation of the ABCS prompted the

development of the Big Canopy system. This system

featured four tower masts and a massive canopy at the

top, which lifted prefabricated material to the target

ﬂoor, where workers controlled the maneuvering of the

components with the use of joysticks. The T-up sys-

tem comprised a support, a base, and a manipulator.

The construction process began with the erection of the

buildingcore and the base was constructedat the ground

Table 61.2 Automation and robotics research for building construction, excluding Japan

Project Institution Features

TAMIR Technion – Israel Institute of Technology, Israel Autonomous multipurpose interior

robot

Rebar manufacturing Technion – Israel Institute of Technology, Israel,

and University of Ljubljana, Slovenia

Assembly of rebar cages for beams

and columns

Building assembly robot University of Karlsruhe, Germany Automatic crane handling system

SHAMIR Technion – Israel Institute of Technology, Israel Autonomous multipurpose interior

robot

Mobile bricklaying robot University of Stuttgart, Germany Brickwork erection

Handling robot Technion – Israel Institute of Technology, Israel Conversion of existing cranes into

large semi-automatic manipulators

ROCCO University of Karlsruhe

and Technical University of Munich, Germany

Assembly system for computer-

integrated construction

Brick laying robot (BLR) Delft University of Technology, Netherlands Brick placing from a platform on

top of a variable-height telescopic

mast

RoboCrane National Institute of Standards and Technology,

USA

Welding system and steel placing

ROMA University Carlos III de Madrid, Spain Autonomous climbing for con-

struction inspection

Contour crafting University of Southern California

and Ohio University, USA

Additive fabrication technology for

surface-forming troweling to cre-

ate smooth and accurate planar and

free-form surfaces

Freeform construction Loughborough University, UK Megascale rapid manufacturing for

construction

level. Guide columns and hydraulic jacks elevate the

base to the target ﬂoor. Figure 61.3 shows several exam-

ples of automated building systems developed in Japan.

Besides in Japan, there has been signiﬁcant ad-

vancement in robotics research and development for

building construction in the past two decades. Some

of these research projects have been led by academic

institutions and government agencies, as shown in

Table 61.2.

Almost 30years after robotics applications in con-

struction started being researched, explored, and pro-

totyped in the early 1980s, these applications are

still considered atypical for the construction industry.

However, robotic application efforts are still under-

way, mainly for performing single tasks that are part

of tedious dangerous jobs that demand high qual-

ity and productivity. This trend goes along with the

faster pace of robotics development in other industries,

whose success has not been paralleled by the con-

Part G 61.4

1070 Part G Infrastructure and Service Automation

struction industry. The decision-making and situational

analysis complexities, coupled with the project-based

nature and short-term focus of the construction in-

dustry has hindered success. Nevertheless, for the

past decade, construction researchers and practition-

ers have shifted their efforts toward the development

of integrated automated construction systems aimed

at providing decision-makers with robust tools for

project management. These systems consist of a variety

of computer-supported applications that take advan-

tage of the increasing computing power available, as

well as accessibility to tracking and positioning tech-

nologies, such as global positioning system (GPS),

radiofrequency identiﬁcation (RFID), and Wi-Fi,which

have demonstrated dramatic improvements in mater-

ials and resources management, progress tracking, cost

monitoring, quality control, and equipment operator

training.

Automation effortshave beenimplemented through-

out the architecture, engineering, construction, and

facilitymanagement lifecycle.During thedesign phase,

four-dimensional (4-D) computer models, i. e., three-

dimensional (3-D) objects plus time dimension, allow

project information to be shared among project partic-

ipants, showing realistic views of objects and activities

and their sequence of assembly. Complex designs take

advantage of 4-D models by using animations to link

computer-aided design (CAD) elements with sched-

ule activities, thereby improving the clarity of design

and construction information, allowing early clash de-

tection, and helping track the construction progress.

Building information models (BIM) emerged to for-

mally address these design automation needs, modeling

representations of the actual elements and components

of a building. BIM is based on industry foundation

classes (IFCs) and architecture, engineering and con-

struction extensive markup language (aecXML), which

are data structures for representing project informa-

tion [61.38]. Further implications of BIM to design

automation include the possibility of estimating build-

ing costs, schedule progress, green building rating,

energy performance, safety plans, material availabil-

ity, and creating a credible baseline for project control,

among others.

At the other end of the project life time, facil-

ity managers continuously make decisions on whether

or not to conduct refurbishments, and prioritize be-

tween cost and quality. As the built environment

ages, these assessments are applicable to demolition

decisions. Automated condition assessment and refur-

bishment decision support systems are leveraging the

complexity of the building systems, such as technical,

technological, ecological, social, comfort, aesthetical,

etc., where every subsystem inﬂuences the total ef-

ﬁciency performance and where the interdependence

between subsystems plays a critical role [61.39]. Fur-

thermore, design changes for refurbishment projects

appear frequently due to a number of factors such as

the lack of suitable design data, insufﬁcient condition

data, inadequate information on building condition, and

ineffective communication between the client and con-

tractors [61.40].

61.5 Techniques and Guidelines

for Construction Management Automation

The management of construction projects deals with

planning and scheduling, material procurement, cost

control, safety, performance tracking, and design–

construction coordination, among other issues.

61.5.1 Planning and Scheduling

Automation

Planning and scheduling consists of sequencing pro-

cesses, activities, and tasks according to time, space,

and resources constraints, specifying the duration of

such tasks and the relationships between them. Tradi-

tionally, scheduling and planning were done through

simple bar charts. In the 1960s, the critical path method

(CPM) emerged, followed by discrete systems mod-

els [61.23]. Today, several software applications have

been developed toautomate theCPM, suchas Microsoft

Project, Primavera SureTrak, and Primavera P3. These

applications allow users to link different activities, allo-

cate resources, and optimize the schedule. In addition,

they provide a user-friendly representation that can be

used in different aggregation levels. A further step in

the automation is taken by specialized software appli-

cations that include sophisticated methods of resource

leveling and scheduling optimization considering un-

certainty. Four-dimensional CAD emerged in the 1980s

to associate time to spatial elements, thereby allowing

visual representation of sequences and communication.

To generate a 4-D simulation, a 3-D model is created

and a manual or semi-automated process of linking

Part G 61.5

Construction Automation 61.5 Techniques and Guidelines for Construction Management Automation 1071

3-D objects to tasks in time takes place. Visualization

of construction sequences is at the task level, which

means that the active task is visualized when the asso-

ciated construction object is highlighted. One drawback

of this approach is that activities that are not associated

to objects, such as milestones, cannot be visualized on

the simulation. In addition, since the links have to be

updated every time the 3-D model changes, it is a very

time-consuming process. The new trend of scheduling

automation is to incorporate BIM into the schedule,

including spatial, resource utilization, and productivity

information [61.38]. BIM consists of a set of parametric

building elements that have associated several proper-

ties and relationships between elements, which store

information on geometry, material, etc. Using BIM for

scheduling, the model is linked to the schedule, not

an object is linked to a task. This approach not only

saves time and is more accurate, but also allows the

evaluation of resources and material consumption over

time [61.41].

61.5.2 Construction Cost Management

Automation

Construction cost management deals with the relation

between the owner’s budget, the project estimates, and

the actual cost of the project. Several attempts to auto-

mate cost estimatingstarted inthe 1980s,when software

applications emerged to link construction quantities to

cost databases or to standard industry databases. Later,

this applications evolved to automate quantity take-offs

from CAD drawings and to transfer such estimates to

job cost management applications that allow automated

tracking of contract amounts, subcontracts, purchase or-

ders, quantity totals, billings, and payments. The new

trend in cost construction automation is to incorporate

a cost estimate application to BIM, becoming helpful in

conceptual estimates because those estimates are calcu-

lated based on project characteristics and project type

(i.e., ofﬁce building, school, number of ﬂoors, parking

spaces, number of ofﬁces, etc.). Those quantities are not

available in CAD models because they do not deﬁne ob-

ject types. On the other hand, BIM allows early quantity

extraction and cost estimates. Later in design, BIM al-

lows users to extract the quantities of components, area

and volume spaces, and material quantities. In terms of

a detailed estimate, the accuracy depends on the detail

of the model. An important advantage of using BIM for

estimating is that the estimate is automatically updated

when the model changes; in addition, it helps reduce

bid costs because it lowers uncertainty related to mater-

ial quantities. There are three different ways how BIM

can be used to aid the cost management process:

1. To export quantities to estimating software

2. To link BIM objects to estimating software, which

allows manual inclusion of costs that are not associ-

ated to a speciﬁc object

3. As a quantity take-off tool [61.38].

61.5.3 Construction Performance

Management Automation

Current research efforts to ﬁnd appropriate methods for

performance monitoring and control mainly fall into

three categories: (1) a series of automated techniques

for detection of building entities: radiofrequency iden-

tiﬁcation (RFID), global positioning systems (GPS),

laser scanners, Wi-Fi, and ultra wideband (UWB) po-

sitioning technologies, and visual sensing; (2) a series

of visualization techniques used to represent discrep-

ancies between as-planned and as-built; such methods

could help in comparing real-world data with the ﬁve-

dimensional (5-D) as-planneddata (i.e., 3-D model plus

sequencing and estimated cost described in BIM); for

example, it could help automatically detect discrep-

ancies and identify those elements that are ahead of

or behind schedule, identify those elements that are

on budget or present overruns, and assist in taking

corrective control actions; and (3) enterprise resource

planning (ERP) and computer-supported frameworks

that allow the integration of workﬂow information for

automated project performance control.

Radiofrequency Identiﬁcation (RFID)

In RFID technology, radiofrequency is used to capture

and transmit data from a tag embedded or attached to

construction entities. RFID is helpful for material track-

ing due to its largerdata storage capabilities, being more

rugged, not requiring line of sight, and being faster in

collecting data about batch of components [61.42,43].

The technology also has the advantage that it can be

combined with other tracking technologies (i.e., 3-D

laser systems or barcoding) that can complement and

enhance the tracking system. This method might be

troublesome for tracking numerous and dissimilar types

of construction objects and personnel, especially if the

RFID chips are not given the appropriate amount of

time to be recognized by RFID readers. Other disad-

vantages include the need for previously identifying

and tagging the entity to be tracked. This greatly lim-

its tracking capabilities and makes it unreasonable for

Part G 61.5

1072 Part G Infrastructure and Service Automation

tracking ever-changing entities, such as personnel and

cast-in-place materials on a construction site.

Global Positioning System (GPS)

GPS is a system that can provide 3-D position and

orientation coordinates. The GPS system can be ap-

plied in two ways: differential GPS based on range

measurements, andkinematic GPS based on phasemea-

surements. Differential GPS can be used to measure

locations at a metric orsubmetric accuracy. Inkinematic

GPS, positions are then computed with centimeter ac-

curacy [61.42]. As for differential GPS, the kinematic

computation can be performed either in postprocess-

ing or in real-time, which is called real-time kinematic

GPS. As a satellite-based technology, standard GPS

needs a line of sight between the receiver and the

satellite; therefore, it cannot normally operate indoors.

Recent developments enable GPS to operate indoors

by adding cellular, laser or other technology [61.44].

Research on the use of GPS in the construction in-

dustry has been characterized by the positioning of

equipment and by construction metrology in ﬁeld oper-

ations [61.25,45,46]. The latter has been complemented

with deployment of laser radar imaging of construction

sites. The GPS system is limited by several factors. It

can only be applied for outdoor use and needs a GPS re-

ceiver to be attached to the entity that is being tracked.

Since the number of materials involved in a project is

usually signiﬁcantly greater than the number of pieces

of equipment involved in installing them, it is, in most

cases, unfeasible to attach a GPS receiver to each piece

of material.

Laser Scanning

Laser scanning is a terrestrial laser imaging system that

creates a highly accurate 3-D image of a surface for

use mainly in computer-aided design (CAD) [61.47].

Typical 3-D laser scanners can create highly accurate

three-dimensional images of objects from quickly cap-

tured 3-D and 2-D data of the objects. Laser scanning

is especially applicable in construction sites, where the

high amount of objects and materials make it difﬁcult to

gather accurate spatial data. Although data acquisition

is fast, postprocessing tasks, such as georeferencing and

3-D model generation, can take several times longer,

accounting for 80% of the time needed to complete

the task, depending on the level of detail of the 3-D

model [61.48]. It also has issues with noise, which

needs to be removed during the segmentation process.

This segmentation process is still not realized automat-

ically and takes considerable processing time. Other

disadvantages include the high cost and current lack of

automation.

Wi-Fi and UWB

Wireless local area networks (WLAN) featuring Wi-Fi,

also technically known as 802.11 for a 2.4GHz radio

band, have been considered as a key opportunity for the

construction industry sector [61.49]. The main function

of the Wi-Fi system is to integrate hardware compo-

nents, such as application server, positioning engine

server, and ﬁnder client, with the CAD drawings of the

facility. Integrated through a web interface, the user is

allowed to obtain material information from the site and

trace it on the ﬁnder client’s graphical interface in real

time. Presently, positioning accuracy is up to a meter

with current Wi-Fi technology, although it may become

3cm with the use of ultra-wide-band [61.50,51]. Wi-Fi

and UWB applications in construction suffer from sev-

eral disadvantages. The necessity for tagging is one of

the deﬁciencies of this system. Another limitation is

the necessity for measurement of infrastructure, which

means that , ideally, the use of a total station is required

in order to obtain accurate results. This increases the

time needed for the setup of the system, as well as the

cost of the tracking method.

Vision Tracking

Vision tracking is a method that blends video cameras

and computer algorithms to perform a variety of mea-

surement tasks. Traditional vision-based tracking can

provide real-time visual information of construction job

sites. This technology is unobtrusive, which means that

there is no need for tagging or installing sensors to the

tracked entities. Additional advantages include the sim-

plicity, commercial availability, and low cost associated

with video equipment and the ability to signiﬁcantlyau-

tomate the tracking process [61.52]. Disadvantages of

this technology are associated with its limitations due

to ﬁeld of view, visibility, and occlusion, and the inabil-

ity to track indoor elements using peripheral cameras.

There are also further complications of differentiating

between interesting and irrelevant entities at the site.

e-Work

Research on the automation of preconstruction work-

ﬂows has been documented with the application of

e-Work methods in steel reinforcement [61.53], and in

construction materials in general [61.54]. This work

is built upon previous research in the manufacturing

domain, where e-Work is composed of collaborative,

computer-supported activities and communications-

Part G 61.5

Construction Automation 61.6 Application Examples 1073

supported operations in highly distributed organiza-

tions, thereby investigating fundamental design princi-

ples for the effectiveness of these activities [61.55,56].

See Chap.88 on Collaborative e-Work, e-Business, and

e-Service.

61.5.4 Design–Construction Coordination

Automation

Design–construction coordination consists of synchro-

nizing all designs, ﬁeld conditions, special conditions,

trades, and systems, in order to deliver the project ac-

cording to the design intent. One of the most important

tasks is clash detection. Currently, most clash detec-

tion is done manually by overlying drawings in order

to try to detect clashes within the different systems, and

some contractors use CAD applications to overlay lay-

ers to detect conﬂicts visually; both approaches are very

slow and prone to errors. The new trends in clash de-

tection are based on BIM tools, which allow automatic

clash detection from two perspectives: 3-D geometry

conﬂict detection and object relationship and interac-

tion conﬂict detection [61.38,57]. As all objects are in

the model, clash detection can be performed at different

levels of detail as needed. Finally, detected conﬂicts can

be corrected very quickly before the next round of clash

detection takes place, making the updating of drawings

simpler.

61.6 Application Examples

Four particular application examples have been selected

because of their current impact on construction au-

tomation. The ﬁrst three examples are state of the art,

whereas the last one is still in the research stage. The

ﬁrst example is related to earthmoving automation; the

second example involves planning and scheduling au-

tomation; the third example describes the automation of

a construction cost estimation process; the last exam-

ple, still in the research stage, features the automation

of construction progress monitoring.

61.6.1 Grade Control System for Dozers

One of the latest inventions toward the accomplishment

of automation in construction is the technology incor-

porated into the new line of Caterpillar crawler tractors

or bulldozers. Even though these machines are not

factory-equipped with all the technical automated op-

tions, these units are preinstalled with all the accessories

to make them compatible with these new technologies.

The Caterpillar factory standard technology Accu-

Grade [61.58] is an integrated system that provides

accessibility for the implementation of more advanced

technologies into these machines; for example, the in-

tegration of GPS, laser grading, cross slope, etc. This

system has beneﬁts in terms of the productivity, ac-

curacy, and overall effectiveness of the equipment use

by the operator. It includes the automatic blade con-

trol function which ultimately eliminates the operator

from the accuracy-demanding task of blade positioning,

especially in cases where accuracy is very important.

AccuGrade GPS control system isone ofthe most effec-

tive tools to havebeen integrated into thesemachineries,

essentially because it creates 3-D model replicas of the

ground contours designed by engineers and then auto-

matically matches these contours with the actual work

done by the machine. This machine eliminates the use

of an operator’s ability to match, as accurately as pos-

sible, the ground’s elevation to the design, making it an

automatic procedurecalculated accurately by amachine

and guided by an educated operator. Job site safety is

improved, since it reduces the need for personnel to

guide the machine, such as ﬂaggers, stakers, checkers,

surveyors, etc. Another feature that the AccuGrade sys-

tem has integrated is the safety interlock, which locks

the blade in a position when the system is not active.

In order to accomplish the overall goal of automat-

ing the earthwork activities of a construction project,

this system provides a tool to create complex 3-D de-

signs and incorporate them into the machine’s GPS

system. Thistool ismost commonly used for more com-

plex earthwork projects, such as golf courses or speciﬁc

projects requiring constant slope degrees differences.

Also, the ability to create ﬂat, single, and dual-planar

designs for less demanding projects, such as parking

lots or building pads, is a great advantage that the GPS

control system provides in order to achieve greater au-

tomation, accuracy, and cost reductions.

Figure 61.4 shows the sloping designs that the GPS

grade control system will integrate into the computer

of the machine in order to automate the earthmoving

necessary to achieve desired specs.

Part G 61.6

1074 Part G Infrastructure and Service Automation

a) b) c)

Flat slope

Single slope

Dual slope

Fig. 61.4a–c Grade control system for CAT Dozer: (a) 3-D sloping designs, (b) in-cab computer, (c) earthmoving

operation [61.58]

Using AccuGrade, the dozer achieves quicker com-

pletion times and minimizes the need for staking, string

lines, and grade checkers, apart from automatically per-

forming thejob without havingto worry aboutaccuracy,

material waste, and several other time-consuming fac-

tors that would regularly be involved in an earthmoving

project. According to studies done by Caterpillar, pro-

ductivity is increased by 50% and surveying costs are

reduced by 90% [61.58].

There are some off-board components and some

ofﬁce components involved in the operation of the

GPS system. The off-board component includes GPS

satellites which send positioning information to the ma-

chine, a GPS base station located within radio range

of the machine that is used to transfer information

from the satellite, and a GPS receiver installed in the

machine for information receipt by the machine. In

addition, there are ofﬁce components involved, includ-

ing 3-D design software and ofﬁce software used to

design and convert any kind of design into a format

that the machine can use to create its design mod-

els.

Fig. 61.5 CAD model, bill of quantities, and construction sched-

ule [61.41]

61.6.2 Planning and Scheduling

Automation

This example shows the use of BIM as an aid in the

scheduling process [61.41]. The project was the con-

struction of the Süddeutscher Verlag Corporate Head-

quarters in Munich, Germany, a 28-storey, high-rise

ofﬁce building with an area of 78500m

. The con-

struction period was 36months. In order to manage

the schedule of the project, 4-D CAD and BIM were

used. The CAD model consisted of 40 000 objects

and a schedule that included 800 tasks, 600 of which

could be visualized in the 4-D simulation, as shown

in Fig.61.5.

Linking CAD objects and tasks took 4days. When

there was a change in the CAD model, the task linking

update took another 4days. Linking BIM and the time

schedule took 0.5days, while changes on the model just

needed to reload the data and apply the relationships,

a process that took just a few minutes. After comparing

the quantities generated by the BIM tool and the quan-

tities generated manually, the BIM quantities proved to

be more accurate.

61.6.3 Construction Cost Estimating

Automation

This example demonstrates BIM capabilities as a con-

ceptual cost estimating automation tool [61.38]. The

project, Hillwood Commercial in Dallas, USA, is a six-

storey, mixed-use building with an area of 12500m

The building was constructed under a design–build de-

livery method. The design team started modelingduring

the conceptual stage to evaluate different alternatives to

be presented to the client, using DProﬁler, a parametric

BIM tool that assigns cost information to the objects.

DProﬁler is integrated into a commercial cost database,

Part G 61.6