ChapterPDF Available

Abstract and Figures

Infrastructure service robotics is a discipline studying robotic systems and methodology for buildings and civil infrastructure construction, inspection, and maintenance. The target could be buildings, estates, parks, bridges, power plants, power transmission lines, underground tunnels, sewage pipes, port facilities, etc. In this article, several new infrastructure service robots projects for construction services and deep tunnel inspection carried out in Singapore will be introduced. With new actuators, low cost sensors, and open source robotics software, infrastructure robots represent a new breed of intelligent systems that help the society to overcome manpower shortage and ageing workforce. These projects are examples of user-led and user-inspired robotics R&D effort led by government agencies, universities, and industrial alliance of local and overseas robotics and construction machinery manufacturers, start-up companies, and system integrators. The ultimate goal is to strengthen the robotics R&D capability in Singapore and to foster a robotics industry and the ecosystem that transform Singapore into a Smart Nation.
Content may be subject to copyright.
Innovations in Infrastructure Service
Robots
I-Ming Chen, Ehsan Asadi, Jiancheng Nie, Rui-Jun Yan,
Wei Chuan Law, Erdal Kayacan, Song Huat Yeo,
Kin Huat Low, Gerald Seet and Robert Tiong
Abstract Infrastructure service robotics is a discipline studying robotic systems
and methodology for buildings and civil infrastructure construction, inspection, and
maintenance. The target could be buildings, estates, parks, bridges, power plants,
power transmission lines, underground tunnels, sewage pipes, port facilities, etc. In
this article, several new infrastructure service robots projects for construction ser-
vices and deep tunnel inspection carried out in Singapore will be introduced. With
new actuators, low cost sensors, and open source robotics software, infrastructure
robots represent a new breed of intelligent systems that help the society to over-
come manpower shortage and ageing workforce. These projects are examples of
user-led and user-inspired robotics R&D effort led by government agencies, uni-
versities, and industrial alliance of local and overseas robotics and construction
machinery manufacturers, start-up companies, and system integrators. The ultimate
goal is to strengthen the robotics R&D capability in Singapore and to foster a
robotics industry and the ecosystem that transform Singapore into a Smart Nation.
Keywords Professional service robot Construction robot Large-diameter
tunnel inspection robot
I.-M. Chen ()E. Asadi J. Nie R.-J. Yan W.C. Law E. Kayacan S.H. Yeo
K.H. Low G. Seet
School of Mechanical and Aerospace Engineering, Nanyang
Technological University, Siangapore, Singapore
e-mail: michen@ntu.edu.sg
R. Tiong
School of Civil and Environment Engineering, Nanyang
Technological University, Singapore, Singapore
© CISM International Centre for Mechanical Sciences 2016
V. Parenti-Castelli and W. Schiehlen (eds.), ROMANSY 21 - Robot Design,
Dynamics and Control, CISM International Centre for Mechanical Sciences 569,
DOI 10.1007/978-3-319-33714-2_1
3
1 Introduction
Infrastructure service robotics is a discipline studying robotic systems and
methodology for buildings and civil infrastructure construction, inspection, and
maintenance. According to the classication by International Federation of
Robotics (IFR), such service robots include professional cleaning systems for
oors, windows, walls, tank, hulls and pipes, inspection and maintenance systems
for facilities, plants, tanks, tubes, and pipes, and construction and demolition sys-
tems for nuclear plants, building, civil/heavy structures, and road constructions.
Developing robotic technology for infrastructure services has the following
signicance:
(1) Economics and sustainabilityRobotic technology would be able to reduce
the reliance on unskilled workers and also skilled workers operating on
sophisticate construction machinery.
(2) ProductivityRobotic technology will streamline and further optimize cur-
rent construction process for shorter project period, and also assure quality
consistency of the construction project.
(3) Safety and healthRobotic technology will reduce the human exposure to
hazardous and inaccessible environment during construction.
Due to the recent advancement in robotic technology and cost down on key
robotic components, such as low cost 3D imaging sensors, high precision 3D laser
scanners, high density lightweight actuators, open source robotics software, AI and
cloud computing, standardized plug-and-play components, and robust wireless
communication and control, developing innovative robotic systems to service
public infrastructure and private estates become affordable. However, there are still
a number of technical challenges to overcome before these professional service
robotic systems making massive inroad into real world services. These challenges
include:
Complete understanding of existing human workow and processes and also the
limitations in the application domain
Performance evaluation, validation, and optimization of sub-systems and the
overall robotic system for eld operations
Designing robustness robotic manipulation, mobility and perception modules
for eld operations
Developing machine learning and intelligence for robotic systems that can adapt
to environment and task variations in the elds.
Designing suitable level of human-robot interaction and machine autonomy for
effective operations
In this paper, we introduce three new infrastructure service robot projects for
construction services and deep tunnel inspection carried out in Singapore. The basic
design consideration and technology developed and integrated into the system will
be presented. Preliminary results on these robots in lab trials will be described.
4 I.-M. Chen et al.
Finally, the paper will be concluded with some thoughts on further development
and commercialization of these professional service robots.
2 Mobile High-Rise Spray Painting Robot
2.1 Motivation
Despite the rapid technology evolution, construction services are mostly
labors-dependent and performed with conventional techniques and occasionally in
dangerous situations. Beside the low efciency in this sector, the needs for building,
construction and maintenance are growing rapidly all around the world while the
construction industry is facing a future shortage of skilled workers and wage
increases. It is evident that inefcient management of resources and use of unskilled
workers can result in a considerable decline in construction quality and produc-
tivity. Hence, there is a high demand for introducing novel robotic technologies that
can be applied to boost productivity by focusing on quality and time saving, as well
as to enhance safety and to reduce cost compared to the traditional method.
An overview of the relevant state of art on construction robots indicates a lack of
advanced and multi-purposes robots for construction painting; particularly for
interior nishing where only very few robots were designed and developed. The
feasibility analysis and economic impact of robotizing interior nishing services for
productivity improvement on the construction sites were initially studied by
Rosenfeld et al. (1993), and Warszawski and Rosenfeld (1994,1997). Later,
Kahane and Rosenfeld (2004) developed a method for evaluating the effects of
human-and-robot integration on automating a construction task and examined the
method using an interesting multi-purpose robot, named TAMIR, for block laying
and wall painting. The painting system consists of a commercial 6-DOF robot arm
mounted on a computer controlled 3-DOF mobile platform. The robot was designed
for research and development purposes, and it had an average reach access. Aris
et al. (2005) mainly investigated the problem of automating upright spray painting
for only ceiling nishing by designing a 3-DOF robot. The ceiling painter is derived
horizontally by making use of a single-phase induction motor and a chain-sprocket
mechanism and moved vertically via a zigzag ladder structure. The painting
workspace of the robot is signicantly small relative to the platform size. Another
study (Naticchia et al. 2007) introduced a scaled down interior painting set up for
laboratory use that consists of a 6-DOF manipulator to be mounted on a 2-DOF
Hexapod for horizontal movement. The research also studied reproducing colored
artworks by developing a multi-colored spraying tool. A roller-based interior wall
painting robot was proposed by Sorour et al. (2011), which includes a horizontally
moving platform (3-DOF), and a painting arm (2-DOF) with a roller brush, attached
to end-effector, which solely scans the walls vertically up to 2.7 m. As a very
low-cost robot, Keerthanaa et al. (2013) designed an airbrush spraying system for
Innovations in Infrastructure Service Robots 5
interior nishing of simple and small spaces. The robot was equipped with a
four-wheel platform, conveyor shafts, and chain-sprocket that allow transferring
spraying system vertically up to limited height.
To the best of our knowledge, none of the previously reported paint robots is
capable of delivering all desirable functionalities for interior nishing of the ceiling
and walls of high-rise warehouses within a stand-alone automated system. All the
existing indoor paint robots have low or average reach access and do not suit for
painting high rise ceilings and corners. Besides, most of the paint robots are merely
able to paint walls or to paint ceiling due to the insufcient mobility or autonomy of
robots. High ceiling painting is inevitable and common in building construction,
especially in industrial workshops. Traditional high ceiling painting is manually
done by means of ladders or hydraulic lifts and often results in unreliable painting
quality. It is hard to obtain consistent quality because of the unstable working
position and movement of the worker thus yielding a big difference in productivity
in accordance with the level of workers skill and experience. Moreover, the
high-place operation will create lethal dangers to painting workers. This project
aims to develop an advanced paint robot that enables automating the execution of
interior painting of high-rise ceilings and walls. A modular system is considered in
this work, and the robot is designed with ve primary subsystems: a 3-DOF mobile
robot, a 1-DOF long reach mechanism, a 6-DOF commercial robot arm, a spraying
system and a control and safety system. The proposed system conguration enables
free access to a large workspace for painting ceilings and walls up to 10 m height
and can improve the productivity due to the precise robot movement and even paint
distribution throughout the whole painting patches.
2.2 Overall System
This project aims to develop a stand-alone robot with a higher degree of mobility
compared to other existing indoor paint robots and with particular functionalities for
automating the entire process of interior nishing of high-rise spaces. Unlike the
other existing interior paint robots, the proposed robot includes a novel long reach
mechanism that enables paint delivery to high-rise walls and ceilings. Modular
design is considered in this work to reduce the design complexity at conceptual and
technical levels, and to accelerate the manufacturing cycle as well as to increase the
exibility of system integration and upgrade. By considering a modular system, the
robot is designed with ve primary subsystems: a 3-DOF mobile robot, a long
1-DOF reach mechanism, a 6-DOF commercial robot arm, a spraying system and a
computer-controlled system. Besides, the robot consists of several subsidiary
modules such as hydraulic outrigger stabilizers, the cable-hose suspension system
and diverse types of covers to protect the equipment and sensors against paint
pollution.
The proposed system conguration enables free access to a large workspace for
painting ceilings and walls up to 10 m height and sufcient maneuverability to paint
6 I.-M. Chen et al.
corners and non-planar surfaces. Using the proposed paint robot can improve the
productivity due to the precise robot movement and even paint distribution
throughout the whole spray pattern that ensures all painting patches receives a
certain amount of paint precisely and consistently.
2.3 Robotic System Realization
The mobile robot, the rst module, is designed to carry heavy payloads with
zero-turn maneuverability by utilizing a combination of two driving wheels and six
passive ones. The second module, long reach mechanism, comprises a dual mast
telescopic lift mechanism that is derived up to certain heights via deploying
hydraulic cylinders in conjugation with a series of cables, and it is centrally
embedded in the mobile platform. The third module is a six degree-of-freedom
robot arm, outtted with a spray-painting gun and time-of-ight camera on its
end-effector, to be located on a right spot on top of the long reach mechanism. The
commercial airless spraying system is selected for the fourth module, together with
an electrically actuated spray gun that allows high-quality interior nishing without
the need for the air compressor. All spraying subsystems are mounted on the mobile
robot that enables the free navigation of the robot within the construction site. To
accomplish the painting task freely in large spaces, having a high level of autonomy
and safety is vital. As the fth module, the mobile robot will be endowed with
several levels of autonomous capabilities for traveling through the construction site,
for levelling the mobile robot attitude, for adjusting the altitude of lift mechanism,
for planning the manipulator motion, for executing the spray painting task and for
safety management.
A preliminary design of the robot is depicted in Fig. 1, in which the upper part
covers of the robot and detailed design of subsidiary modules are not shown.
Figure 2demonstrates the manipulator arm currently set up in the lab to conduct
Fig. 1 Preliminary design of
high rise spray painting robot:
1mobile robot (3-DOF),
2spray painting system,
3dual mast telescopic
mechanism (1-DOF),
4control box,
5manipulator (6-DOF)
Innovations in Infrastructure Service Robots 7
preliminary experiments for developing motion planning and vision processing
algorithm. The robot arm is effectively outtted for a real paint operation by con-
sidering protective covers for both the arm and the camera coupled to the
end-effector as an eye-in-hand system.
3 Post-construction Quality Assessment Robot
3.1 Motivation
Post construction quality assessment of buildings is an indispensable procedure in
construction industry that is currently executed by manual inspectors. In a standard
daily operation, a large number of inspectors are needed to nish the wholly manual
assessment in the traditional way. Such a manual assessment procedure may import
several errors into inspection because of executing the operation in incorrect way or
the use of inaccurate inspection tools. What is more, the manual inspector may get
tired after some time and the inspection accuracy may decrease over time. In most
of the time, manual inspection has to be done during the daytime. Considering this
time consuming, tiresome and unexciting procedure, an automated post construc-
tion quality assessment robot system is proposed in Fig. 3. The proposed robot
system consists of a mobile robot, a 2D laser scanner, a thermal camera with a
heater and an inclinometer (Axon Robot is from CtrlWorks Pte Ltd). This robot
system can assess ve different types of defects on the oors and walls, such as
evenness, alignment, inclination, cracks and hollowness.
Fig. 2 Manipulator setup outtted with camera, spray gun and protective covers
8 I.-M. Chen et al.
3.2 Quality Assessment Methodology
In traditional assessment, cracks of grounds, walls and ceiling are visually
inspected. Evenness of ground and walls is inspected by using a 1.2 m spirit level,
and alignment of two walls is inspected by using a set square in Fig. 4. In the
assessment process of alignment, one edge of the setsquare tightly contacts with
Fig. 3 Automated
construction quality
assessment system
Fig. 4 Manual assessment tools and autonomous assessment sensors
Innovations in Infrastructure Service Robots 9
one wall (Ani et al. 2014), and the distance between the other one edge and the
other wall is calculated. Beyond doubt, measurement error is imported into the
operation during the assessment process, because it is a challenging task for the
inspectors to accurately keep the setsquare horizontal. To assess the hollowness, the
inspector needs to cover the whole ground with the end-effector of a CONQUAS
rod. Then, the ground with hollowness is inspected by distinguishing the scuff-
ing noise of the ground and the CONQUAS rod.
Considering this time consuming, tiresome and unexciting procedure of manual
assessment, our proposed robotic system can accurately assess all these defects in
real time. A Hokuyo UTM-03LX 2D laser scanner shown in Fig. 4is used to
localize the mobile robot and construct an environment map that can be used to
store the accurate locations of defects. A Sick LMS 500 2D laser scanner is used to
assess the evenness of ground and walls, and the alignment of two walls, which has
a maximum measurement distance of 80 m, a measurement accuracy of 6 mm and
a measurement range of 190°. The evenness is assessed by calculating the average
deviation between raw sensor data and its extracted line segment, which have been
mostly used to build a 2D environment map and localize a mobile robot (Yan et al.
2014,2015a,2015b). The alignment of walls is assessed by calculating the angle of
two extracted planes. The inclination of ground is assessed by directly obtaining
inclination angle in X- and Y-axis from a POSITAL inclinometer. To assess cracks
and hollowness, thermal camera and a heater are used to obtain thermal images and
RGB images. Then, these images are recognized by using support vector machine
(SVM) method.
3.3 Experimental Results
Assessment results of evenness, crack and hollowness are shown in Figs. 5and 6.
In Fig. 5, the laser scanner is horizontally put on the ground of a constructed
testbed. In the construction of this testbed, wall 1 is an even wall and wall 2 is an
uneven wall. It can be seen from the comparison result of the average deviation for
two walls that the average deviation of wall 1 is larger than that of wall 2. In
Fig. 6a, the assessment result of a tile with cracks is shown by comparing the raw
images and the processed images with SVM. In Fig. 6b, even though the color
difference in the raw thermal image is not very clear, the process images can show
the hollowness result well. These promising results show that the proposed auto-
mated construction quality assessment robot system works well in the identication
of different types of defects.
10 I.-M. Chen et al.
Fig. 5 Assessment result of evenness
Fig. 6 Assessment result of
crack and Hollowness. aAn
example of crack assessment
bAn example of hollowness
assessment
Innovations in Infrastructure Service Robots 11
4 Deep Tunnel Sewerage System Inspection Robot
4.1 Motivation
The Deep Tunnel Sewerage System (DTSS) is a public wastewater utility based in
Singapore, a solution to meet Singapores long term needs for used water collec-
tion, treatment, reclamation and disposal. It has diameter of 6 m, and it is located
2050 m underground. In order to prolong the service life of the sewers, the
maintenance task such as inspecting the structural integrity of the tunnel is nec-
essary. Human cannot easily access to DTSS because of high water ow, the
presence of contagions or bio-hazardous materials, explosive gases and oxygen
deprivation (Walter et al. 2012), as well as the absence of light in DTSS. Also, the
tunnel is partially lled with water and muddy ground that is not easy for human to
walk across. Signicant study on pipeline and tunnel robot has been done due to its
ability to access underground spaces and to achieve easier and better inspection in
tunnels. However, most of the existing robots are designed for small or medium
pipelines and tunnels (Law et al. 2015). There is a need to develop a tunnel robot to
perform the maintenance of large-diameter tunnel.
4.2 Overall System
A generic locomotion mechanism for DTSS robot is designed that could cope with
various tunnel conditions from dry, muddy, and partially lled water. On the other
hand, a hoisting system will be congured for the launching, deploying and
retrieving of the inspection robot. The setup of the inspection system consists of a
number of subsystems, namely the auxiliary system and ground control station, the
hoisting and winch system, and the robotic platform as shown in Fig. 7.
On the surface, the auxiliary system provides electrical power to the winch and
the robotic platform as well as hydraulic power to actuate the A-frame. The control
station houses the control units for the robotic platform and monitors displaying
images from cameras and measurements from sensors.
The hoisting and winch system comprises an A-frame, a winch module and an
umbilical cable. The A-frame is for lowering and lifting the robot along the access
shaft without the needs of any other heavy-duty hoisting devices. The winch is
mounted on the platform of the A-frame for easy transport and operation. An
external 3-Phase AC power generator will supply the power to the winch module
and provides power supply for the tethered robotic platform. The electrical power is
transmitted through the multi-core umbilical cable carrying a Kevlar core, data optic
ber and multi-core copper wires.
The robotic platform has its own actuation and houses various sensors needed
for the inspection of the sewer tunnel. In particular, it will have lightings to illu-
minate the tunnel and cameras to capture the tunnel images. A laser proler and a
12 I.-M. Chen et al.
sonar proler are utilized to scan the tunnel above and below the water surface
respectively. A ballast control is used to vary weight of the robot to maintain the
maneuverability and stability of the robot in accordance to the internal condition of
the sewer.
4.3 System Designs
In Fig. 8, the semi-circular frame is designed for the mounting of 3 HD cameras and
4 LED spotlights, forming an inspection array radially at equal angle apart. It will
be mounted on the chassis of the robotic platform facing towards the inner cir-
cumference of the sewer, covering the surface of the sewers above the water level.
The base plate is the main support structure of the robotic platform. All the
mechanical parts will be mounted on the base plate, for instance, the front and rear
camera, the locomotion mechanism, ballast, etc. Two waterproof enclosures, xed
on the base plate, are used to house the electrical components such as controller and
drivers. The front facing camera is at a high position in the sewers, allowing an
unblocked eld of view of the sewer for navigation. Moreover, the robotic platform
has low center of gravity and has large base to increase the robot stability.
Figure 8also shows the situation when the robotic platform is deployed into the
tunnel through the vertical shaft. The locomotion mechanism in this design can be
tilted to certain angle that is perpendicular to the sewer surface before the robot
lands onto the sewer to avoid the possible presence of soft debris along the bottom
Fig. 7 Setup of inspection system
Innovations in Infrastructure Service Robots 13
of the tunnel and to acquire better traction force over the sewer surface. This is
achieved via a slider-crank mechanism that is used to transform translational motion
into rotational motion.
5 Conclusions and Discussion
The infrastructure service robot projects introduced in this paper represent a new
breed of service robots that are developed based on end-user input and more
cost-effective, more compact, and more versatile than before. With advancement in
sensors, actuators, and articial intelligence, such robots could conduct service
tasks autonomously with minimal human supervision. It is also possible to use such
service robots with human collaboratively. The robot can perform simple, mundane,
and large-scale tasks, whereas the human worker can conduct complex and
sophisticate task that could be too expensive for robot to do. In such way, there
could be robot shiftand human shiftco-existing in the work place. Such
practices could create a paradigm shift in productivity for the future society with
lots of professional service robots. Also there could be innovative workow
re-design due to such autonomous service robots along side with cloud services and
big data analytics. The integration of infrastructure service robots, big data collected
through the robots, and the value-add to the end users, as a complete service could
be the future form of infrastructure service robotics companies.
Acknowledgments The research projects are supported by National Research Foundation of
Singapore under TDIR2015-01-02, TDIR2015-01-03, and TDIR2015-01-04. Team members,
Chen Qiu, Lijing Soh, Lili Liu, Varun Maruvanchery, Chin Leong Low, Burhan, and company
co-developers, Aitech and CtrlWorkds are acknowledged. Lead public agencies, Jurong Town
Corporation, Building Construction Authority, and Public Utility Board of Singapore are
acknowledged.
Fig. 8 Robotic platform for DTSS
14 I.-M. Chen et al.
References
Ani, A., Tawil, N., Johar, S., Razak, M., & Yahaya, H. (2014). Building condition assessment for
new houses: a case study in terrace houses. Journal Teknologi, 70(1), 4350.
Aris, I., Parvez Iqbal, A. K., Ramli A. R., & Shamsuddin, S. (2005). Design and development of a
programmable painting robot for houses buildings. Journal of Teknologi, Universiti Teknologi
Malaysia, 42(A): 2748.
Kahane, B., & Rosenfeld, Y. (2004). Balancing human-and-robot integration in building tasks.
Computer-Aided Civil and Infrastructure Engineering, 19, 393410.
Keerthanaa, P., Jeevitha, K., Navina, V., & Indira, G. (2013). Automatic wall painting robot.
International Journal of Innovative Research in Science, Engineering and Technology, 2(7),
30093023.
Law, W. C., Chen I. M., Yeo, S. H., Seet, G. L., & Low, K. H. (2015). A study of in-pipe robots
for maintenance of large-diameter sewerage tunnel. In Proceedings of 14th IFToMM World
Congress in Mechanism and Machine Science Conference, Taipei, Taiwan, 2530 Oct 2015.
Naticchia, B., Giretti, A., & Carbonari, A. (2007). Set up of an automated multi-color system for
interior wall painting. International Journal of Advanced Robotic Systems, 4, 407416.
Rosenfeld, Y., Warszawski, A., & Zajicek, U. (1993). Full-scale building with interior nishing
robot. International Journal of Automation in Construction, 2(4), 229240.
Sorour, M. T., Abdellatif, M. A., Ramadan, A. A., & Abo-Ismail, A. A. (2011). Development of
roller-based interior wall painting robot. World Academy of Science, Engineering and
Technology, International Journal of Mechanical, Aerospace, Industrial, Mechatronic and
manufacturing Engineering, 5(11), 17851792.
Walter, C., Saenz, J., Elkmann, N., Althoff, H., Kutzner, S., & Stuerze, T. (2012). Design
considerations of robotic system for cleaning and inspection of large-diameter sewers. Journal
of Field Robotics, 29(1), 186214.
Warszawski, A., & Rosenfeld, Y. (1994). Robot for interior nishing works in building
Feasibility analysis. ASCE Journal of Construction Engineering and Management, 120(1),
132151.
Warszawski, A., & Rosenfeld, Y. (1997). Economic analysis of robots employment in building.
Proceedings of the 14th International Symposium on Automation and Robotics in Construction
(ISARC) (pp. 177184). Pennsylvania, USA: Pittsburgh.
Yan, R. J., Wu, J., Yuan, Q., Yuan, C., Luo, L.-P., Shin, K.-S., et al. (2014). Natural corners-based
SLAM with partial compatibility algorithm. Proceedings of IMechE, Part I: Journal of systems
and Control Engineering, 228(8), 591611.
Yan, R. J., Wu, J., Shao, M.-L., Shin, K.-S., Lee, J.-Y., & Han, C.-S. (2015a). Mutually converted
arc-line segment-based SLAM with summing parameters. Procedings of IMechE, Part C:
Journal of Mechanical Engineering Science, 229(11), 20942114.
Yan, R. J., Wu, J., Lee, Y., & Han, C.-S. (2015b). Representation of 3d environment map using
b-spline surface with two mutually-perpendicular LRFs. Mathematical Problem in Engineer-
ing, 2015(690310), 114.
Innovations in Infrastructure Service Robots 15
Author Biography
I-Ming Chen received his Ph.D. in Mechanical Engineering
from California Institute of Technology, Pasadena, USA in 1994.
He is an internationally renowned robotics researcher and has
been with the School of Mechanical and Aerospace Engineering
of Nanyang Technological University (NTU) in Singapore since
1995. Currently he is the Director of Robotics Research Centre
in NTU.
His research interests are in collaborative robots, infrastructure
robots, wearable sensors, human-robot interaction, recongurable
automation, and parallel kinematics machines (PKM).
He is now serving on the editorial board of Mechanism and
Machine Theory, Robotica (Cambridge Univ Press), and Frontiers
of Mechanical Engineering (Springer-Verlag) as well as senior
editor of IEEE Transactions on Robotics. He was Technical Editor
of IEEE/ASME Transactions on Mechatronics from 2003 to 2009.
Professor Chen is a Fellow of IEEE and Fellow of ASME, General Chairman of 2017 IEEE
International Conference on Robotics and Automation (ICRA 2017) in Singapore.
16 I.-M. Chen et al.
... M Obile service robots are being implemented in service and maintenance industries to mitigate the increasing shortage and rising costs of trained manpower [1]. Menial and repetitive tasks such as cleaning, inspection maintenance and elderly care have been delegated to mobile service robots over the years with the improvements in robotic technology and the need of quick responses for such tasks [2]. ...
Article
Full-text available
Doors are part of the building infrastructure that mobile robots have to pass through to reach zones on the other side. If robots were to clear these obstacles, they would require human assistance, advanced end-effectors, and complex control systems, making it challenging for robots. Therefore, a robot deployed in an environment should be capable of minimizing the passing through doors as well as path distance to improve overall efficiency. This paper proposes a novel Door-Density-Aware (DDA) path planning method. A vision-based door-detecting framework based on YOLOv8 has been developed to tag the door locations in a robot’s navigation map. The proposed DDA path planner uses a door-tagged map to plan an efficient path considering the cost of moving through doors and the path distance. Genetic Algorithm (GA) and Gray Wolf Optimization (GWO) have been considered for solving this optimization problem. According to the experimental results, the proposed method can effectively detect and tag doors in the navigation map and plan efficient paths. The proposed DDA path planner with GA produced the path plan with the least cost compared to other approaches.
... Finally, Fig. 9 (a) shows the results of binarization of the fuzzy enhanced image and Fig. 9 (b) displays the extracted wall are that should be painted. Again, the information about the extracted wall is forwarded to the robot's operating system in order to process the coordinates of the wall [42][43][44][45][46][47][48]. ...
... These robots can be guided to the painting area using a navigation system such as those discussed above. Subsequently, the robots controlled by logical programming can start spraying [137,138] or rolling the paint brush [139,140]. Since the installation of glass curtains and windows in high-rise buildings is also challenging work, a glazing robot (an integration of robot arm and navigating robot) has been designed with vacuum suction pads that can lift a 1400 kg glass panel steadily [141]. ...
Article
Full-text available
Building construction has developed from the use of primitive tools to that of machinery, with a tendency toward automation. Automation of processes and robotics can improve efficiency, accuracy and safety in construction. On the other hand, structural prefabrication for construction is increasingly being adopted worldwide to enhance productivity and to alleviate the environmental impact of conventional construction processes. The combination and application of automation and prefabrication technologies may therefore introduce new developments to the construction industry. This paper provides a comprehensive review of the use of automation technology for structural prefabrication and construction, including recent developments, challenges and future trends. Five stages following the sequence of construction are proposed: design, construction management, robotic manufacturing, autonomous transportation and automatic structural assembly. The paper concludes that the widespread use of automation technology is preferable to structural prefabrication for construction, and that the design for robotic construction introduced through connection innovations may be beneficial as a means of avoiding complex operations and thus improving the efficiency of robotic assembly processes.
... This monitoring phase can be automated to detect the appearance of crack, have a return to the quality of the surface, and check the adhesion between the printed layers. Several systems have been developed to automate the inspection process in some specific environments, such as subway tunnel [12], flexible pavement surfaces [13], [14], bridge decks [15] and building [16], [17]. In [18], a hexapod robot, named Hex-piderix, with a stereo camera has been used to inspect the cracking on surfaces like walls or roofs of a building. ...
Article
Full-text available
The growing environmental concerns have emerged the necessity of sustainable waste management of construction and demolition (C&D) wastes. This review explores the advancements in artificial intelligence (AI) and robotics to automate C&D waste sorting. A comprehensive examination of this domain is conducted by struc-turing the paper around six research questions. Current trends and potential future directions are revealed by performing methodology and data analysis involving bibliometric and scientometric studies. Notably, recent research emphasises circular economy, AI, and robotics, underscoring the importance to enhance AI for precise categorisation. The scarcity of publicly available datasets is a central challenge in the C&D waste domain, that hinders effective AI applications. However, data augmentation, data synthesis, generative AI, and transfer learning have been identified as crucial techniques to enhance dataset quality and categorization accuracy. While AI draws significant attention in the C&D waste domain, this review shows a lack of AI-enabled robotics systems due to the complex nature of waste sorting and collection. In summary, this study's findings highlight the need for new methods and techniques integrating multisensory fusion, unsupervised machine learning and robotics intelligence to continuously learn and adapt to new waste streams and materials, making them highly efficient in sustainable waste management.
Article
Full-text available
This article presents an integrated concept of an aerial robot used for predictive maintenance in the construction sector. The latter can be remotely controlled, allowing the localization of cracks on wall surfaces and the adaptive deposit of the material for in situ repairs. The use of an aerial robot is motivated by fast intervention, allowing time and cost minimizing of overhead repairs without the need for scaffolding. It is composed of a flying mobile platform positioned in stationary mode to guide a soft continuum arm that allows to reach the area of cracks with different access points. Indeed, some constructions have complex geometries that present problems for access using rigid mechanical arms. The aerial robot uses visual sensors to automatically identify and localize cracks in walls, based on deep learning convolutional neural networks. A centerline representing the structural feature of the crack is computed. The soft continuum manipulator is used to guide the continuous deposit of the putty material to fill the microscopic crack. For this purpose, an inverse kinematic model-based control of the soft arm is developed, allowing to estimate the length of the bending tubes. The latter are then used as inputs for a neural network to predict the desired input pressure to bend the actuated soft tubes. A set of experiments was carried out on cracks located on flat and oblique surfaces, to evaluate the actual performances of the predictive maintenance mechatronic robot.
Article
This paper presents the use of the ZED depth sensor in a robot-based painting application. The use of a stereo depth sensor is a very important factor in robotic applications, since it is both the initial and the essential step in a sequence of robotic operations, where the goal is to detect and extract the useful surface and objects or the obstacle on a wall that is not intended for painting. The ZED depth sensor was used for surface recording and navigation of our painting robot. Later, wall extraction was performed using simple image processing and morphological operations in a surface extraction algorithm. The client’s instruction was to use well-known, simple, and proven image processing operations in the algorithm to ensure both reliable and smooth operation of the robot’s vision system in an industrial environment. The experiments showed that the developed algorithm detects and extracts the wall successfully under various depth measurement conditions.
Article
Full-text available
This paper presents the use of the ZED depth sensor in a robot-based painting application. The use of a stereo depth sensor is a very important factor in robotic applications, since it is both the initial and the essential step in a sequence of robotic operations, where the goal is to detect and extract the useful surface and objects or the obstacle on a wall that is not intended for painting. The ZED depth sensor was used for surface recording and navigation of our painting robot. Later, wall extraction was performed using simple image processing and morphological operations in a surface extraction algorithm. The client’s instruction was to use well-known, simple, and proven image processing operations in the algorithm to ensure both reliable and smooth operation of the robot’s vision system in an industrial environment. The experiments showed that the developed algorithm detects and extracts the wall successfully under various depth measurement conditions.
Article
In this paper, the design optimisation and localisation with sparse sensing array of range-based sensors is presented. This optimisation focuses on an optimised sparse sensing array for localisation and autonomous navigation in tunnel environments with right rectangular prism geometry and the autonomous ascent and descent in vertical cylindrical shafts. A optimisation is conceived to optimise sensor configuration to traverse both tunnels and shafts with constraints from the field of the view of the camera system and the mechanical layout. The optimised sensor configuration solved using a genetic algorithm performed on average 21 times better than configurations that were placed heuristically. The optimised design and the localisation algorithm is implemented on SWIRL: Surveyor with Intelligent Rotating Lens. The proposed localisation method with sparse sensing results in at least four-fold reductions in the weight and eight-fold reduction in power consumption of the sensing system compared to existing ones. SWIRL was autonomously flown in a series of experiments, consisting of tests in shaft and tunnels in both indoor mock-up and field environment. Lastly, in an endurance test in a mock-up shaft, SWIRL achieved over 35 minutes of autonomous flight using the proposed sensing placement and scheme, making extended aerial robot navigation and inspection of deep hazardous tunnels possible.
Article
Full-text available
Nowadays robots are widely used in many applications such as military, medical application, factories, entertainment, automobile industries etc. However, the application of robot is still not widely implemented in construction industry. In construction industry, robots are designed to increase speed and improve the accuracy of construction field operations. It can also be used to do hazardous and dangerous jobs in construction. For example, currently house painting is done manually. This process can be simplified using a special dedicated robot. It is very difficult and troublesome for human being to work in an upright position, especially for painting, cleaning and screwing in the ceiling for a long time. Painting in an upright position is also very dangerous for the eyes. To overcome this difficulty, a programmable painter robotic system is proposed, designed and developed. This paper describes all the processes that are involved in designing and constructing the proposed painter robot. The system is divided into two main parts namely hardware and software. In hardware part, mechanical design, fabrication, electrical and electronics system are described and in software part, control algorithm is explained. The testing results indicate that the performance of the painter robot is better compared with that of using manual painting technique.
Article
Full-text available
This paper proposes a map representation method of three dimensional (3D) environment by using B-spline surfaces, which are first used to describe large environment in 3D map construction research. Initially, a 3D point cloud map is constructed based on extracted line segments with two mutually-perpendicular 2D laser range finders (LRFs). Then two types of accumulated data sets are separated from the point cloud map according to different types of robot movements, continuous translation and continuous rotation. To express the environment more accurately, B-spline surface with covariance matrix is proposed to be extracted from each data set. Due to the random movements, there must be overlap between extracted B-spline surfaces. However, merging of two overlapping B-spline surfaces with different distribution directions of their control points is a complex problem, and which is not well addressed by far. In our proposed method, each surface is divided into overlap and non-overlap. Then generated sample points with propagated uncertainties from one overlap and their projection points located on the other overlap are merged using the product of Gaussian probability density functions. Based on this merged data set, a new surface is extracted to represent the environment instead of the two overlaps. Finally, proposed methods are validated by using the experimental result of an accurate representation of an indoor environment with B-spline surfaces.
Article
Full-text available
This study determines the types of defects often detected in residential buildings based on the criteria set by the Construction Industry Standard (CIS) 7: 2006-Quality Assessment System for Building Construction Work. Twenty-two terraced two-story houses located in Selangor, Malaysia were visually inspected and assessed in terms of building condition, and results were reported based on the Condition Survey Protocol 1 Matrix. Assessment findings were consolidated with those of defective groups based on the criteria of CIS 7: 2006 to determine the defect type, building component, and construction field in which defects often occur. Results show that most of the inspected houses are dilapidated even though they were recently completed. The most severe building defects are detected in the architecture. Furthermore, floors and walls are major contributors to building defects. Mainly as a result of poor workmanship, the most common defects involve finishing, alignment and evenness, and joint and gap. Thus, this study proposes a method to ensure high-quality workmanship.
Article
Full-text available
This paper presents a mutually converted arc-line segment-based simultaneous localization and mapping (SLAM) algorithm by distinguishing what we call the summing parameters from other types. These redefined parameters are a combination of the coordinate values of the measuring points. Unlike most traditional features-based SLAM algorithms that only update the same type of features with a covariance matrix, our algorithm can match and update different types of features, such as the arc and line. For each separated data set from every new scan, the necessary information of the measured points is stored by the small constant number of the summing parameters. The arc and line segments are extracted according to the different limit values but based on the same parameters, from which their covariance matrix can also be computed. If one stored segment matches a new extracted segment successfully, two segments can be merged as one whether the features are the same type or not. The mergence is achieved by only summing the corresponding summing parameters of the two segments. Three SLAM experiments in three different indoor environments were done to demonstrate the robustness, accuracy and effectiveness of the proposed method. The data set of the MIT CSAIL Building was used to validate that our method has good adaptability.
Article
Full-text available
This article presents natural corner-based simultaneous localization and mapping (SLAM) using a new data association algorithm that achieves partial compatibility in a real unknown environment. In the proposed corners’ extraction algorithm, both the end points of an extracted line segment far away from the other segments and the intersection point of the two closer line segments are considered as corners. In data association, a partial compatibility algorithm obtaining a robust matching result with low computational complexity is proposed. This method divides all the extracted corners at every step into several groups. In each group, the local best matching vector between the extracted corners and the stored ones is found by joint compatibility, while the nearest feature for every new extracted corner is checked by individual compatibility. All these groups with the local best matching vector and the nearest feature candidate of each new extracted corner are combined, and its joint compatibility is c
Conference Paper
Full-text available
Abstract—This paper describes the development of an autonomous robot for painting the interior walls of buildings. The robot consists of a painting arm with an end effector roller that scans the walls vertically and a mobile platform to give horizontal feed to paint the whole area of the wall. The painting arm has a planar twolink mechanism with two joints. Joints are driven from a stepping motor through a ball screw-nut mechanism. Four ultrasonic sensors are attached to the mobile platform and used to maintain a certain distance from the facing
Article
The paper presents a complete and a simplified model for the economic analysis of robots employment on building sites. Four generic types of building robots are analyzed with the aid of the model and of the available productivity data, which have been published to-date by developers and users of building robots. General conclusions are presented with respect to the feasibility of employment of these four types of robots on building sites: (1) exterior finishing robots, (2) horizontal finishing robots, (3) interior finishing robots, and (4) assembling-handling robots.
Article
The paper presents an economic analysis of robotized execution of interior building tasks. Four tasks—painting, plastering, tiling, and masonry—have been examined in full-scale with “TAMIR”—Technion Autonomous Multipurpose Interior Robot. This experimental preprototype building robot has six degrees of freedom, an average reach of 1.70 m, and is mounted on an automatically guided mobile three-wheeled carriage. The robotized mode of execution of these four tasks is described, along with time and cost data, and logistical considerations. The labor input and cost of these activities in robotized versus manual performance are quantitatively analyzed under different site and employment conditions. Suggestions for improvements in the robotized system, give rise to make the robotized option more attractive in the future. Conclusions with respect to managerial and economic implications of robotized construction are also presented.
Article
In the last two decades, a wide range of robotic systems for interior construction tasks was being examined (mainly by research and development institutions around the world). A main parameter that differentiates between these systems is their level of autonomy. Reducing the level of autonomy of a given robotic system necessitates increasing auxiliary human work. The efficiency of the system as a whole (i.e., robot/s + operator/s + worker/s) must be examined in order to determine the proper balance between robots and human beings. This article presents the development of a methodology for evaluating a robotized construction task by examining and comparing various ways to share the work between robots and human operators and workers. The methodology is demonstrated by implementing it on two interior finishing tasks—block laying and wall painting. Testing these tasks included full-scale experiments. The detailed analyses of these two applications show the breakeven points of profitable robotic systems, depending on the combination of their own cost, the hourly cost of construction workers, and the expected hourly cost of construction-robot operators in a given labor market. The results lead to a better understanding of possible human-and-robot integration. In particular, this article provides researchers and developers of construction robotics with a systematic approach for evaluation and validation of various ideas prior to investing resources in actual prototypes.