Robots in Industry

ANTONI GRAU, MARINA INDRI,

The Past, Present, and Future of a
LUCIA LO BELLO, and THILO SAUTER Growing Collaboration With Humans

R
obots have been part of analyzed, and the relationships with in- the loom of Jacquard (in 1801), Albert
automation systems for dustrial communications and software the Great (1204–1282), and Roger Ba-
a very long time, and in technologies are also discussed. Some con (1214–1294), to mention just a few.
public perception, they future directions for robotics are en- The automaton can be considered the
are often synonymous visaged, focusing on the contributions forerunner of modern industrial robots.
with automation and coming from new materials, sensors, The word “robot” was used for the
industrial revolution per actuators, and technologies. Open is- first time in 1921 when the Czech writ-
se. Fueled by Industry 4.0 and Internet sues are highlighted as well as the main er Karel Capek (1890–1938) released
of Things (IoT) concepts as well as by barriers that currently limit the deploy- in Prague his work, Rossum’s Universal
new software technologies, the field of ment of industrial robots in the small Robot (R.U.R.), depicting class fighting
robotics in industry is currently un- and medium enterprise (SME) world. in a society with automated workers.
dergoing a revolution on its own. This From that moment on, the term “ro-
article gives an overview of the evolu- Background and Motivation bot” has been used by science fiction
tion of robotics from its beginnings to Throughout history, humankind has writers, and in 1926, the movie Me-
recent trends like collaborative robot- been fascinated by machines and de- tropolis finally made it popular around
ics, autonomous robots, and human– vices able to imitate the functions and the world. Isaac Asimov first used
robot interaction. Particular attention movements of living beings. The an- the term “robotics” in science fiction
is devoted to the deep changes of the cient Greek civilization had the word books that inspired scientists and en-
last decades, from the traditional indus- autómatos to refer to such devices. The gineers to develop early industrial ro-
trial scenario based on isolated robotic first automaton was arguably built by bots. He was the leading promoter of
cells up to the most recent coworking Hero of Alexandria (85 AD), who made the word “robot.”
and collaborative robots. The role of ro- animated mechanisms that moved with In industrial practice, the fascina-
botics in the Industry 4.0 framework is hydraulic devices, pulleys, and levers, tion with human-like machines plays
mostly for ludic purposes. For many no significant role. Rather, robots have
Digital Object Identifier 10.1109/MIE.2020.3008136 centuries, various inventors created always been an element of automation,
Date of current version: 23 December 2020 automatons, from Leonardo da Vinci to their main tasks being to relieve human

VITRUVIAN MACHINE ©SHUTTERSTOCK.COM/GIOVANNI CANCEMI

BACKGROUND ©SHUTTERSTOCK.COM/CHEREZOFF

50 IEEE INDUSTRIAL ELECTRONICS MAGAZINE ■ MARCH 2021 1932-4529/21©2021IEEE

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
workers from heavy, dangerous, or but also with their surroundings; and standards relevant to industrial robot-
monotonous work and to improve the entire range of cloud and edge ics are displayed separately. The figure
product quality by increasing the pre- computing, permitting information ac- also includes the publication timeline
cision and repeatability of manufac- quisition and processing from the indi- of some iconic movies on robotics.
turing processes. A highly controver- vidual device up to the enterprise level These movies are science fiction art,
sial aspect of the use of robots is the (Figure 1). In this context, wireless com- but they coined public perception of
possibility to establish fully automat- munications play an eminent role as robotics over the decades.
ed production lines, leading to almost they support the mobility of robots as The present and short-term evolu-
personnel-free factories and serious well as the inclusion of sensors or actua- tion of industrial robotics can be seen
threats to the job market, especially tors around the actual robot [4]. in Figure 3. The installation of indus-
in the low-qualification segment. The purpose of this article is to give trial robots is growing, with a high
While robots traditionally have been a historical overview of the evolu- increment foreseen in 2020 and in the
operated as stand-alone machines in tion of industrial robotics from its ear- next two years. The largest growth
confined production cells, there is a re- ly stage to current developments and is expected in Asia (10.9%), followed
cent trend toward collaborative robot- future trends. It will also shed some by America (8.4%) and Europe (6.3%).
ics and human–robot interaction [1]. light on the communication and soft- China continues to lead the instal-
This trend is closely connected to the ware technologies required for modern lation of industrial robots, thus con-
Industry 4.0 idea and, more generally, to collaborative robotics. tributing to the big increase in Asia,
the concepts of the IoT and cooperating together with Japan and South Korea.
objects. It is fueled by recent develop- A Taxonomy of Industrial Robots The increasing market in car manu-
ments from the IT world and industrial Starting from the progenitors of ro- facturing and electronics promoted
electronics or even material science. bots, i.e., telemanipulators, Figure 2 the largest growth in the past year in
Crucial enabling technologies for these provides a timeline of the evolution of robot installations in Asia, together
trends are, e.g., smart sensors [2] for a industrial robotics together with some with the still-emerging production in
better perception of a robot’s environ- famous robots and the relevant mile- China. Figure 3(d) also reports the ro-
ment; industrial communications [3] for stones of technologies and sciences bot density ranking, i.e., the number of
improved real-time interaction and strictly related to robotics. The main deployed industrial robots per worker
coordination not only among robots achievements and communication in various countries.

Robotics Technology

Distributed Systems AI-Based Methods

Perception and Agents
Collaborative Robotics Learning
Brain–Machine Interface
Cognitive Robotics
Autonomy
Behavior
Assembly
Ethics
Internet of Robotic Things
Cloud Robotics
Elasticity
Ambient Intelligence
Industry 4.0 Robots as a Service
Embedded Mobile
Everyday Things Maintenance Service and Storage
5G Edge Computing
Digital Twin
M2M and Wireless Physical-World Multitenancy Productivity
Sensor Network Web

Cloud Computing
IoT Technology

FIGURE 1 – The technology fields influencing modern robotics. AI: artificial intelligence; M2M: machine to machine.

MARCH 2021 ■ IEEE INDUSTRIAL ELECTRONICS MAGAZINE 51

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
 Globally, the automotive industry
leads the number of units installed
in car and car supplies factories with
an attractive 2% increment compared

2020
5G to the previous year, while the also-
leading electrical/electronics indus-

Ex Machina
try shows a decrease of 14%. Worth

Industrial
Robot in Space

Internet
Humanoid
Payload 1,000 kg
noting is the increase of 32% of units
in the food industry, which includes

Light Weight
Cobots
beverages and tobacco production.

Robot
UWB

Other industries that have a large

Roomba
Cyber-Physical

I, Robot
ZigBee

by the Internet Robots Go

EtherCat

Systems
to Mars

growth are mainly relevant to agri -

Bluetooth

culture, mining, construction, and

ProfiNet

ASIMO
Powerlink

Industry 4.0

2000
WLAN

Remote Diagnosis

education and are included under the

AIBO
Bicentennial

The Matrix

FlexPicker
“All others” label in Figure 3(c). As
Man
ControlNet

IoT

with the general industry, the robot

DeviceNet

MQ1 Predator
industry devoted to robot production
Robot Control

Drone
PC-Based

foresees a huge increment for the

RoboCop
Computing

next years.

AdeptOne
ASi

Ubiquous
Profibus

WWW

Today, robots are extensively used in

Terminator

industry, being an essential element in

Direct-Driven

most manufacturing processes. Table 1

Robot
CAN

Microprocessor Motor-Driven

SCARA

displays the most typical robot opera-

1980
Robot

Blade Runner

tions in the industrial field, with some

PUMA

C3PO and
Star Wars
in Space
Modbus

Robots

R2D2

highlights. New application fields are

Stanford Arm
The Internet
Payload 60 kg

currently opening, for example, agricul-

Robot Control
Electromechanical

ture, construction, domestic and hazard-

Tentacle Arm
Logan’s
Actuators

ous environments, medicine, and health.

Run

Unimate in
Japan
ISO/OSI

2001: A Space
Odyssey

From Isolated to
Industrial Robot Versatran
Vision
Robot

Coworking Robots
Unimate

In the 1990s, industrial robots were

1960

already advanced mechatronic sys-

Colossus of
Ro-Man, Robot New York

tems, synergistically integrating me-

Behavior
Robot

First
Hydraulics
Actuators

chanical design, electronics, software,

Intelligence
Artificial

Cybernetics

and control, but with no real aware-

the Great
Tobor,

Monster
Servocontrol

ness of what was happening around

Neural Networks

Telemanipulator

them. They mainly operated in an iso-

Artificial

lated way. There were automated pro-

First
Three Laws
of Robotics

duction lines where robots seemed to

FIGURE 2 – A timeline of some remarkable aspects in robotics.
Doctor Satan
Mysterious

work together, but this was not quite

1940
Elektro

true. Each robot was actually an iso-

lated manufacturing cell with a spe-
cific manufacturing task, whereas the
Gordon
Flash

cells were connected to the rest of the

Metropolis

process by conveyor systems for work-

piece handling. Even in cases where
several robot arms worked together
on one workpiece (like in car assem-
1920

bly lines), those robots were not pro-

Communications

grammed independently but as one

Achievements
Technologies

Technology
Milestones
Industrial

Industrial
Famous

machine with predefined movements.

Movies

Robots

Robots
Robot
and

The lack of awareness, and hence the

inability of those robots to vary their
behavior in an autonomous way, im-
plied strong constraints to guaran-
tee safety. Consequently, classical

52 IEEE INDUSTRIAL ELECTRONICS MAGAZINE ■ MARCH 2021

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
 a.
p.
2%
+1
584
+10% 522
‘000 of Units

+6% 0%
465
422 421
400
304
254
221
178

2013 2014 2015 2016 2017 2018 2019* 2020* 2021* 2022*
(a)

China 154
Japan 55.2
United States 40.4
Rep. of Korea 37.8
Germany 26.7
Chinese Taipei 12.1
Italy 9.8
France 5.8
Mexico 5.7
Spain 5.3
India 4.8
Singapore 4.3
Canada 3.6
Thailand 3.3
Czech Republic 2.7
‘000 of Units

(b)

Automotive +2%
–14%
Electrical/Electronics
–1%
Metal and Machinery
Plastic and –6%
Chemical Products +32%
Food 2018 2017 2016
+48%
All Others
+44%
Unspecified
0 20,000 40,000 60,000 80,000 100,000 120,000 140,000
Units
Robots Installed per 10,000 Employees

(c)
831
774 Average Europe: 114
Average America: 99
Average Asia: 91

338 327 Average World: 99

247 240 221 217 200 188 182 175 174 172 168
165 154 146 140 140 135
Singapore

Rep. of Korea

Germany

Japan

Sweden

Denmark

Chinese Taipei

United States

Italy

Belgium

The Netherlands

Austria

Slovenia

Canada

Spain

Slovakia

France

Switzerland

Finland

China

Czech Rep.

(d)

FIGURE 3 – The evolution of industrial robot installation in absolute and relative numbers per (a) year and (b) country and (c) industrial field and
(d) per number of employees. (a) Robot installation evolution (*forecasted). (b) Robot installation per country. (c) Robot installation by industry
type. (d) Robot density per country. (Source: World Robotics 2019; used with permission.)

MARCH 2021 ■ IEEE INDUSTRIAL ELECTRONICS MAGAZINE 53

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
industrial manipulators were (and start/stop commands), but each robot- obstruct the workflow, thus leading to
still are) closed in working cells, with ic cell worked independently with local inefficiencies. 3) More dynamic moni-
doors equipped with safety devices control and local safety mechanisms. toring approaches would be needed to
that cause an immediate stop of the The communication system provided ensure that the workspace is adjusted
robots if opened [Figure 4(a)]. an interface to supervisory control and according to the actual status of the ro-
The 2000s were characterized by data acquisition systems, but it was not bot and the task that it is performing.
the first approaches allowing robots used for the actual real-time control of 4) The layout of collaborative robotic
and human operators to partially share the robot. cells should be enhanced, not only to
the same spaces [5], mainly through In recent years, collaborative ro- optimize the production workflow but
supervision solutions including the bots (cobots) are becoming part of the also to increase the operators’ feelings
prediction of human behavior [6] and most advanced manufacturing plants of safety and comfort.
the use of proper sensor systems [7]. to guarantee not only high levels of The most challenging issues refer
In that period, the necessity of open- safety but also flexibility in production. to the possibility of establishing a safe
ing the cages in which the robots were The introduction of cobots represents and efficient collaboration between hu-
enclosed to allow some initial form of a significant pillar of robotics in the In- mans and robots that were not original-
human–robot collaboration was also dustry 4.0 scenario, which is going to ly built as collaborative ones. A recent
addressed by the international ISO deeply change the manufacturing and article [12] investigates how to com-
10218 safety standards. The main is- production processes. The Internation- bine the benefits of high payload indus-
sues addressed by such standards refer al Federation of Robotics [9] noticed trial robots with human capabilities in
to the possibility of using safety-rated an increase of the ratio between col- a fenceless environment through the
soft limits as a means to define and re- laborative and traditional industrial ro- adoption of enabling technologies, like
duce the workspace of a manipulator as bots from 2.8% in 2017 to 3.4% in 2018. manual guidance techniques (based on
well as the adoption of devices that can The greater the diffusion of cobots a force/torque sensor directly attached
initiate the reduction of the robot ve- in industries, the greater the importance to the robot’s flange) and wearable
locity or its full stop through the robot and influence of human–robot collabo- devices [such as augmented reality
control system. Solutions based on in- ration (HRC) modalities. A recent in- (AR) glasses and smartwatches], for a
dustrial sensors, like the SafetyEYE by teresting survey on HRC in industrial multimodal interaction. In other solu-
Pilz [8] and ad hoc safety devices, were settings is available in [10], whereas tions available in the literature, safety
developed to allow the human operator a quite complete overview of HRC in- is achieved through a synergistic use of
to enter the robot workspace in a safe terfaces and interaction modalities is safe and unsafe sensors. For example,
manner [Figure 4(b)]. provided in [11]. There are, however, in [13], the developed dynamic safety
Safety concerns were also among main gaps that are still open. 1) Only architecture detects human motions by
the reasons why, in the past, industrial lightweight robots are used in most two separate systems. The primary one
communications were not used for the of the current HRC collaboration sce- is based on a generic human detection
coordination of robots. Classical field- narios, thus losing the original vision sensor system (e.g., Microsoft Kinects),
bus systems were sufficient for basic in- of robots as high-powered machines. while the secondary system is based on
formation and data exchanges (e.g., for 2) Safety functionalities can sometimes an actual safety sensor.

TABLE 1 – THE TYPICAL ROBOT OPERATIONS.

OPERATIONS HIGHLIGHTS
WELDING • Spot welding is one of the most common welding applications in
Arc welding, flux cored welding, laser welding, metal active gas welding, metal inert gas manufacturing.
welding, tungsten inert gas welding, orbital welding, oxyacetylene welding, other (plasma, • All arc-welding processes use an arc welding gun or torch to transmit
ultrasound) welding, resistance welding, shielded metal arc welding, spot welding, the welding current from a welding cable to the electrode.
submerged arc welding
MATERIAL HANDLING • There exists a huge variety of palletizing and material-handling robots
Collaborative operations, dispensing, injection molding, machine loading, machine tending, available in the market, with very different payloads and tools, like bag
material handling, packaging, palletizing, part transfer, pick and place, press tending grippers, suction, and magnetic grippers.
MACHINERY • Injection foundry was the first robotized task in 1960.
Cutting, deburring, drilling, foundry, grinding, material removal, milling, polishing, • The preferred technology for cutting metal and plastic is laser cutting. Among
refueling, routing, sanding, spindle, and waterjet the different laser types, the most used are gas, crystal, and fiber lasers.
DISPENSING • Automated painting applications require specialized equipment to
Painting and enameling, bonding/sealing, coating, gluing, thermal spray achieve accurate and consistent paint finish quality.
• Sealing robots have built-in additional fluid handling technologies and
numerous arm configurations to easily access any area of the part to seal.
OTHER ROBOT OPERATIONS • The introduction of robotized assembly lines can exponentially increase
3D laser vision, assembly, mounting, inserting, cleaning the production rate, consistency, and reliability.

54 IEEE INDUSTRIAL ELECTRONICS MAGAZINE ■ MARCH 2021

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
 Haptic technology also plays an im- robotic surgery in recent years [14] the virtual or remote environment by
portant role in robot safety and virtu- through the adoption of haptic devices using the robotic device, and haptic
alization of services. An experience of in various applications, from laparo- feedback allowed computer simulations
touch by applying forces, vibrations, or scopic and microsurgery to instrument of various tasks to relay realistic and
motions to the user can be created with positioning, needle insertion, palpation, tangible sensations to the user.
virtual objects in a computer simula- and tissue stiffness mapping. The use of Innovative solutions have been re-
tion to control virtual objects and to en- haptic devices in industry was tradition- cently proposed, e.g., a completely re-
hance the remote control of machines ally mainly restricted to teleoperation mote human–robot collaboration system
and devices. Haptics is transforming tasks, in which the user moved within in [15] that can flexibly work in different

(a)

8 Enabling Switches 6 Perimeter Guarding

6

1 Safety Light Curtains

6

1
8
1 4
4
3 5
4 Emergency Stop Devices
2 3 4

7 5 Mats and Area Guarding

2
5 Safety Edges and Bumpers
2 Safety Laser Scanners 7 Safety Programmable Safety Systems
7 Safety Monitoring Relays 3 Safety Switches and Operator Controls
(b)

FIGURE 4 – (a) No human operator can enter the cages of traditional robotic cells. (Source: ABB Group; used with permission.) (b) Synergetic use
of various safety devices allows human operators to enter the robot workspace. (Source: Valin/OMRON; used with permission.)

MARCH 2021 ■ IEEE INDUSTRIAL ELECTRONICS MAGAZINE 55

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
modes with the use of a collaborative from the main design principles of In- cooperate with other smart devices
robot and an industrial manipulator for dustry 4.0 [18], such as interoperabil- and factory workers as a unique team
hazardous tasks or through the use of ity, decentralization, real-time capabil- [20] or to act as a metasensor net-
vibrotactile rings. In [16], such a type ity, virtualization, service orientation, work supporting traditional automat-
of device is used to send acknowledg- and modularity. The distinction be- ed guided vehicles (AGVs). AMRs,
ments to the user during critical phas- tween industrial and service robotics cobots, enhanced manual stations, full
es of a collaborative assembly task. In will no longer be as sharp as in the past integration within the automated lines,
[17], a bilateral haptic collaboration is since the technologies traditionally ad- and mobile manipulators will be the
established using a soft gripper that is opted in the service robotics world are pillars of Industry 4.0 plants, as illus-
properly designed to guarantee a safe migrating into manufacturing plants to trated in Figure 5, although the most
interaction and a wearable interface to allow for the development of new kinds critical challenge in the industrial world
control the open/close motion of the of production lines [19]. is going to be the achievement of a
gripper and to feedback information The main elements of the production smooth transition from the current
about the important task parameters, line (i.e., the industrial manipulators) industrial standards.
e.g., the grasp tightness. Such a solu- are going to be replaced or placed side The role of the traditional AGVs is
tion has been successfully adopted in by side with cobots, whereas mobile ma- changing as well. Since the introduc-
a complex collaborative task in which nipulators (i.e., robotic arms on mobile tion of AGVs in 1953, technology has
a robot autonomously grasps a pipe on bases) are expected to render obsolete greatly evolved in those devices, lead-
which the human operator has to draw the classical idea of an industrial robot ing them to behave as autonomous
some circles before it is deposited in being strictly associated with a fixed robots, able to navigate and follow a
the final location by the robot. and caged manipulator. Nowadays, au- predefined, specific path for the ma-
tonomous mobile robots (AMRs) are terial flow pattern. The new ability to
Robotics in Industry 4.0 entering factories and taking on vari- plan trajectories and pathways allows
Robotics is going to play a key role in ous roles on the basis of the specific for the optimization of routes and the
the smart factories that will benefit requirements, e.g., to autonomously enhancement of goods transportation

Mobile agents and manipulators

enhance the flexibility of the
production process.

Information flows to/from

the central server and the cloud
and the local controller units.

try
dus
ar t In
Sm

Humans and robot share

spaces and cooperate
on the same process.

FIGURE 5 – A smart factory. No cages are present: cobots, AMRs, and humans share the same space. (Source: Adapted from FreePik.com.)

56 IEEE INDUSTRIAL ELECTRONICS MAGAZINE ■ MARCH 2021

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
in the plant while applying robust col- have already initiated a radical change Last but not least, the development of
lision-avoidance procedures. in the way industrial communication is the smart factories scenario is accompa-
All of these AGVs carry on-board viewed today [29]. 5G networks could nied by the introduction and application
intelligent sensors that make the robot be a further step toward providing a of the concept of robotics as a service.
react in front of any unexpected change ubiquitous communication infrastruc- More and more often, startup companies
in the environment. Those sensors are ture that can be used as a commodity. collaborate with factories, warehouses,
based mainly on radio-frequency iden- Moreover, 5G technology would native- and distribution centers, providing ser-
tification [21] and rotating laser and ly support mobile devices. vices instead of products. Most of these
computer vision [22]. Simultaneous Another aspect of Industry 4.0 is products are oriented to the integration
localization and mapping (SLAM) navi- the digitalization and virtualization of of smart sensors as well as technolo-
gation technology, which allows mo- services. This is highlighted by the gies that are more familiar to service
bile robots to locate themselves while concept of “digital twins,” where each robotics than industrial ones (e.g., the
building a map of the surrounding en- physical entity (such as a robot) has use of unmanned aerial vehicles to col-
vironment using sensors like lidar, cam- a virtual counterpart exhibiting all of lect data to be processed and integrated
eras, and odometry, is mature enough the properties and data of the real de- into large processes). Although the in-
to be deployed to AGVs. The collision- vice. This twin offers services that can creasing number of robots in manufac-
avoidance function of lidar can be be used by other virtual devices or turing plants could reduce the number
adopted for the intelligent multilevel by higher-level applications, e.g., for of human operators directly involved,
obstacle avoidance protection during production optimization or improving the growing automation of processes
motion [23]. SLAM is, right now, the collaboration. The digital twins are ex- has a positive effect on employment, on
navigation mode chosen by many AGV ecuted in some back office cloud en- the whole, thanks to the involvement of
manufacturers, and large e-commerce vironment providing sufficient com- various actors providing different ser-
businesses, such as Ali and Amazon, puting power. As depicted in Figure 6, vices and the reduction of the produc-
already use AGVs as storage robots. this results in a three-level service tion costs, leading to lower market pric-
In the Industry 4.0 scenario, advanced architecture typical for the Industry es, as already noticed in the automotive
data communication is fundamental for 4.0 idea. sector in Germany [30].
not only the safe operation of collabor-
ative robots [24], [25] but even more so
for AGVs and multirobot handling and
coordination [26]–[28] as they have to
cope with strict requirements of mobil- (3)
ity, reliability, and bounded latencies.
A proprietary technology available
today for supporting AGVs’ manage-
ment over wireless links is the Siemens
industrial wireless local area network
(iWLAN) that provides support to
real-time traffic in large industrial ar-
eas. Exploiting a time-division multiple
access-based scheme, iWLAN provides
(2)
deterministic access with controlled jit-
ter and roaming switchover time on the
order of 20–30 ms, thus allowing for the
real-time management of a number of
AGVs over large areas.
The need to transfer huge quantities
of information in a fast and reliable way
will demand more advanced communi-
cation methods. Current industrial com-
munication infrastructures may reach
their limits in terms of bandwidth, sup- (1)
ported nodes, and end-to-end response
times, thus drawing extensive research
interest to technologies that meet the
increasingly stringent requirements of
specific industrial applications. IoT and FIGURE 6 – The three-level hierarchy in Industry 4.0: the physical devices (1), their digital twins (2),
cyber-physical system (CPS) concepts and the service level (3). Communication is based on Internet technology and the IoT paradigm.

MARCH 2021 ■ IEEE INDUSTRIAL ELECTRONICS MAGAZINE 57

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
Future Directions and Open Issues are cheap, resilient, and based on exist- to detect the object’s position and its
The smart factories of the very near ing technology. approximate shape before contact.
future will see a high presence of in- The research goal in the field of new The adoption of robots in new ap-
dustrial robots, not only for large-scale materials is the replacement of the me- plication scenarios often relies on the
manufacturing as usual but also in ver- tallic and rigid robots with smooth, soft use of innovative grippers [44] that are
satile production processes, e.g., in SMEs, robots that could be friendlier when able to successfully perform assembly
whose productions are characterized interacting and collaborating with hu- and picking tasks involving “critical”
by a strong commitment to continu- mans. For instance, the magnetic liquid items, like small and flat objects, for
ously adapting to customer requests metal droplet introduced in [36] can be which suction cup grippers may fail
and meeting the market demands. In stretched in large scales both horizon- if the objects are too lightweight and
this scenario, the possibility of having tally and vertically. Such a remarkable fragile. Innovative solutions have been
both manipulators and mobile agents, stretching capacity is reversible, long recently proposed, e.g., in [45] and [46],
acting in a coordinated way, sharing the lasting, and can be repeated multiple where passive and epicyclical mecha-
same spaces, and collaborating with times. In [37], a team of researchers nisms are adopted to mimic the sliding
the human operators, is very appealing created smart and biodegradable ma- motion of the human thumb below the
[31]. Multirobot coordination address- terials for robots that can be broken object to be grasped. In the soft robot
es several well-known issues, the cor- down and do not pollute the environ- context, a soft gripper, made up of four
rect management of which firstly relies ment. The plastic is replaced by bio- prestressed actuators, was developed
on the proper software and hardware plastic made of food waste with a low- in [47] for food handling.
architecture. Starting from the popu- energy process, and the stiffness of the
lar Robotic Operating System (ROS), material is suitable for external robot New Wearable Machines
which enables the implementation of parts. With such materials, robot arms A growing sector in the industrial sce-
complex and robust robot behaviors and androids would resemble humans, nario is represented by exoskeletons,
across a wide variety of robotic plat- and their bodies would decompose at thanks to decades of research, ad-
forms, a new initiative of ROS–Industri- the end of their life cycle as if they were vancements in enabling technologies,
al has been launched as an open source flesh and blood persons. and big investments. Exoskeletons
project that extends the advanced ca- can be divided into two categories, pas-
pabilities of ROS software to manufac- New Sensors and Actuators sive and active. Passive suits are fully
turing [32]. Software is one aspect only, Sensorial capabilities are fundamental mechanical and have no motors. They
even if it is important; the evolution of for any robotic application. A recent improve ergonomics and effectively dis-
robotics will depend upon a wide range overview of the most common types tribute weight for their wearers. Their
of innovative technologies and require of sensors (e.g., visual, laser, tactile widespread adoption across big compa-
inputs from diverse fields. sensors, and so on) for industrial ro- nies in automotive, aerospace, logistics,
bots can be found in [38]. The grow- and construction constitutes an attrac-
New Materials ing adoption of collaborative robots tive market. Active robotic exoskel-
In the next years, the research prom- is pushing toward the introduction of etons are a more ambitious technology.
ises new materials for a new generation advanced tactile skin sensors to be at- They use motors for actuation, enabling
of robots, i.e., smart materials for soft tached to the robot’s surface to guar- them to provide significant lift assis-
robots, which will be able to add new antee the human operator’s safety. Ex- tance to workers, thus reducing work-
features and capacities to robotics. The amples can be found in the literature, force injuries.
new materials can be hard, as piezoma- e.g., in [39], but some commercial The industrial exoskeleton sector
terials [33]; flexible, as in alloys with devices are also available, such as the is still in its early age, and the market
shape memory; soft, as in dielectric Kuka collaborative robot series [40]. opportunity is very large. Some car
elastomers [34]; or even fluid, as in fer- Such solutions allow the detection manufacturers have started to include
rofluids and electrorheological fluids, of contact pressure, but they cannot industrial exoskeletons in production
which change their shape in front of predict a possible impact in advance. lines, like Hyundai, Ford, and BMW.
electrical fields [35]. The idea of deploy- To enhance safety, the most recent There are other niches for the growth
ing soft robots in industry is not new, trends are toward the development of of exoskeleton technology. In construc-
although the term has evolved with the proximity skin sensors able to detect tion, manufacturing, agriculture, and
latest developments in robotics. Soft no an object before any contact happens, other industries that are adopting ro-
longer means deformable and not built e.g., capacitive sensors as in [41] or botic structures, exosuits augment hu-
with rigid elements. The new concept robotic skin modules as in [42], allow- man motion to allow for more lifting
for soft refers to a new generation of ing for the measurement of proximity, strength and improved production on
robots with an almost muscular defor- contact, and force through an array repetitive tasks like squatting, bending,
mation, built with polymers similar to of optical sensors. A further solution or walking [48], as seen in Figure 7. The
bones, and with muscles and actuators has been recently proposed in [43], market potential for industrial exoskel-
similar to gas bladders. These materials employing time-of-flight sensors able etons is as enormous as the rewards for

58 IEEE INDUSTRIAL ELECTRONICS MAGAZINE ■ MARCH 2021

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
entrepreneurial solution providers that optimization would be desirable for the issues are still open. For example, the
can aggressively innovate and come to entire robotic cell, e.g., as in [54], and for use of AR devices, like smart glasses, in-
market with workable solutions deliver- the management of the complete pro- cluded in several of the most advanced
ing business value [49]. duction process as well. solutions, is not always welcome be-
The possibility of performing collab- cause the operator’s field of view is
Issues Related to New Technologies orative assembly tasks is going to open limited, and their situational awareness
Apart from hardware-related aspects, brand-new application scenarios, but it could decrease. Other solutions have
there are also data- or information-relat- also raises challenging safety and hu- been recently proposed, like the one in
ed key points that characterize the fac- man–robot interaction social aspects. [56], where a visual indicator system is
tories of the future, like the integration As discussed in [55] and some refer- developed to communicate the robot’s
of different actors at the various levels ences therein, for a fruitful cooperation, status to the human operator.
(from the software point of view up to the operator and the robot must un-
the handling of the whole production derstand the actions and intentions of Drivers and Barriers
process), sustainable energy consump- each other, according to the following The deployment of industrial robots
tion, safety issues, or social human– four functional specifications relevant still suffers from some barriers in an
robot interaction aspects. The most to the operator’s working experience: important part of the productive sec-
recent trends consider the factory on 1) flexibility (the operator should not tor, i.e., the SME world. In the European
the whole as a CPS [50], [51] in which be forced to follow a strict, pre- Union, 91% of all employment corre-
the robotic systems play an important defined sequence of operations but sponds to SMEs, as does 68.2% of all
role. Their tasks are going to be mod- should be allowed to change them jobs in manufacturing [9]. However,
eled and programmed considering the on the fly) such companies have not been widely
overall production goals in terms of ef- 2) intelligibility (the operator should be adapted to robotic manufacturing. The
ficiency and quality to achieve high per- capable of intuitively understand­­­­ main reasons why robotized solutions
formance so as to facilitate the adapta- ing the robot actions and intentions have not been adopted are discussed in
tion of the robot tasks to the frequent through some form of communication) [57] and summarized in Figure 8.
changes of the production process. In 3) adaptability (the robot should adapt Companies often face two alterna-
[52], a CPS approach is also adopted to to the operator’s actions without re- tives, i.e., either opt for current automa-
establish a safe human–robot collabo- quiring an operator-specific calibra- tion solutions, even if they are unsuited
ration in a shared workplace. tion process) for low-volume and low-cost produc-
A proper definition of the perfor- 4) transparency (operators should not tions, or employ workers who perform
mance indicators to be optimized can be forced to stay in a specific loca- the manufacturing process manually
also include energy consumption, which tion all the time to safely collaborate and thereby compete based on lower
is of growing importance. Energy con- with the robot). wages. Innovative robotic solutions
sumption reduction for a single robot is An integrated architecture partially solv- should provide ways to tackle these
important and is achievable in different ing these problems can be found in [55], barriers for SMEs, which are often
ways (see, e.g., [53] and the references together with an interesting overview of impacted by low capitalization prob-
therein). However, an overall energy other solutions in literature, but several lems, difficult access to finance, lack

(a) (b) (c)

FIGURE 7 – (a) A chairless exoskeleton (H-CEX). (Source: Hyundai Inc.; used with permission.) (b) A back support exosuit. (c) A whole-body suit.
(Source: EksoBionics; used with permission.)

MARCH 2021 ■ IEEE INDUSTRIAL ELECTRONICS MAGAZINE 59

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
 She received the Best Paper Award in
Too-Small Batch Sizes for Automation Factory Automation at Emerging Tech-
nologies and Factory Automation 2013.
No Process for Automation Lucia Lo Bello ([email protected])
earned her Ph.D. degree in 1998. She
Too Expensive
is a tenured associate professor at the
No Personnel for Running the Robot University of Catania, Catania, Italy. Her
research interests include industrial
Robots Too Heavy and Unflexible
and automotive communications, real-
Programming Too Time-Consuming time embedded systems, and wireless
networks for robotics. She has pub-
Reservations About Robot Technology lished more than 170 technical papers.
Bad Experiences With Known Firms She is responsible for the University of
Catania research contracts with STMi-
0 10 20 30 40 50 60 70
crolectronics on wireless robotics com-
munications as well as several research
FIGURE 8 – Some reasons why SMEs do not use robots.
projects funded by the Italian Ministry
of Research and by the European Com-
of awareness of the benefits of robotic for SMEs, which have the possibility of mission. She is a Senior Member of IEEE
solutions, low technical competence achieving a new competitiveness thanks and received the Industrial Electronics
outside of the core business, and low to the most innovative robotic solutions. Society Early Career Award in 2008.
capability for long-term investment. Thilo Sauter (thilo.sauter@tuwien
The constant decrease of robotic Biographies .ac.at) earned his Ph.D. degree in elec-
technology costs plays an important Antoni Grau ([email protected]) trical engineering from Technische
role in the adoption of automation solu- earned his Ph.D. degree in computer sci- Universität (TU) Wien in 1999. He is a
tions by SMEs. Financial incentives for ence from the Technical University of professor of automation technology at
those companies, like leasing solutions Catalonia (UPC), Barcelona, in 1997. He TU Wien, Vienna, Austria, and was the
and the refurbishment of robots, can is currently a professor with the Depart- founding director of the Center for Inte-
also be a way to engage SMEs in ro- ment of Automatic Control, UPC, Barce- grated Sensor Systems at Danube Uni-
botic technology. lona, Spain, giving lectures on computer versity Krems, Austria. His expertise
Today, there exist solutions offering vision, digital signal processing, and includes smart sensors and automation
financially attractive lightweight robots robotics at the School of Informatics of networks with a focus on real-time, se-
that can be easily moved from one in- Barcelona and the Industrial Engineering curity, interconnection, and integration
dustrial process to another depending School. His research interests include issues. He is an IEEE Fellow and a Senior
on the production necessity by sim- computer vision, pattern recognition, Administrative Committee member of
ply reprogramming them every time it autonomous mobile robots, factory au- the IEEE Industrial Electronics Society.
is necessary. The final goal is to have tomation, and education on sustainable
user-friendly robotic solutions that do development. He has chaired several References
not require workers to have technical international conferences. He has been [1] A. Grau, M. Indri, L. Lo Bello and T. Sauter, “In-
dustrial robotics in factory automation: From
knowledge of robots or machine learn- working on international research proj- the early stage to the Internet of Things,” in
ing. In the short term, good prospects ects on autonomous robotics in smart Proc. IECON 2017–43rd Annu. Conf. IEEE Ind.
Electron. Soc., Beijing, 2017, pp. 6159–6164. doi:
are envisaged for industrial robotic cities and education in robotics. 10.1109/IECON.2017.8217070.
technology. Energy efficiency and new Marina Indri (marina.indri@polito [2] P. Zheng et al., “Smart manufacturing sys-
tems for Industry 4.0: Conceptual framework,
materials can attract the use of robots. .it) earned her Ph.D. degree from Po- scenarios, and future perspectives,” Front.
The fast production of customized ele- litecnico di Torino, Italy, in 1995. She Mech. Eng., vol. 13, no. 2, pp. 137–150, 2018. doi:
10.1007/s11465-018-0499-5.
ments at competitive prices is also a has been an associate professor in [3] S. Vitturi, C. Zunino, and T. Sauter, “Industrial
good incentive to use this technology. Robotics and Automatic Control at Po- communication systems and their future chal-
lenges: Next-generation Ethernet, IIoT, and
Some markets forecast an increas- litecnico di Torino, Turin, Italy, since 5G,” Proc. IEEE, vol. 107, no. 6, pp. 944–961,
ing demand, such as the metal and ma- 2001. She serves as associate editor of June 2019. doi: 10.1109/JPROC.2019.2913443.
[4] X. Li, I. Lille, R. Falcon, A. Nayak, and I. Stoj-
chinery industry, the rubber and plastic IEEE Transactions on Industrial Informat- menovic, “Servicing wireless sensor net-
industry, and the food and beverage ics and technical editor of IEEE/ASME works by mobile robots,” IEEE Commun. Mag.,
vol. 50, no. 7, pp. 147–154, 2012. doi: 10.1109/
industry. Also, the electrical/electronics Transactions on Mechatronics. She has MCOM.2012.6231291.
industry will have big demands. All of been chairing the IEEE Industrial Elec- [5] D. Stengel et al., “An approach for safe and ef-
ficient human–robot collaboration,” in Proc.
these elements anticipate a significant tronics Society Technical Committee 6th Int. Conf. Safety Ind. Automat. Syst., 2010.
increment of robotized manufacturing on Factory Automation since December [Online]. Available: https://www.researchgate
.net/publication/268001665_An_Approach_
processes in the short and midterm, 2018. Her research interests are in the for_ S a fe _ a nd _ E f f ic ie nt _ H u m a n - Rob ot _
not only for large companies but also industrial and mobile robotics areas. Collaboration

60 IEEE INDUSTRIAL ELECTRONICS MAGAZINE ■ MARCH 2021

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.
[6] F. G. Pereira, R. F. Vassallo, and E. O. T. Salles, “Hu- [25] A. Khan, B. Rinner, and A. Cavallaro, “Coopera- [43] S. Tsuji and T. Kohama, “Proximity skin sen-
man-robot interaction and cooperation through tive robots to observe moving targets: Review,” sor using time-of-flight sensor for human col-
people detection and gesture recognition,” J. Con- IEEE Trans. Cybern., vol. 48, no. 1, pp. 187–198, laborative robot,” IEEE Sensors J., vol. 19, no.
trol, Automat. Elect. Syst., vol. 24, no. 3, pp. 187–198, 2018. doi: 10.1109/TCYB.2016.2628161. 14, pp. 5859–5864, 2019. doi: 10.1109/JSEN.
2013. doi: 10.1007/s40313-013-0040-3. [26] G. Patti, L. Leonardi, and L. Lo Bello, “A novel 2019.2905848.
[7] J. Krueger, B. Nickolay, O. Schulz, “Image- MAC protocol for low datarate cooperative [44] M. Graña, M. Alonso, and A. Izaguirre, “A pan-
based 3D-surveillance in man-robot-coopera- mobile robot teams,” Electronics, vol. 9, no. 2, oramic survey on grasping research trends and
tion,” in Proc. INDIN ’04 2nd IEEE Int. Conf. Ind. p. 235, Feb. 2020. doi: 10.3390/electronics topics,” Cybern. Syst., vol. 50, no. 1, pp. 40–57,
Informat., 2004, pp. 411–420. 9020235. 2019. doi: 10.1080/01969722.2018.1558013.
[8] SafetyEYE Camera System by Pilz. Accessed: [27] R. Doriya, S. Mishra, and S. Gupta, “A brief sur- [45] F. Lévesque, B. Sauvet, P. Cardou, and C. Gos-
June 5, 2020. [Online]. Available: https://www vey and analysis of multi-robot communica- selin, “A model-based scooping grasp for the
.pilz.com/en - US/e shop/0 010 6 0 022070 42/ tion and coordination,” in Proc. Int. Conf. Com- autonomous picking of unknown objects with
SafetyEYE-Safe-camera-system put., Commun. Automat., Greater Noida, India, a two-fingered gripper,” Robot. Autonom. Syst.,
[9] International Federation of Robotics. Accessed: 2015, pp. 1014–1021. doi: 10.1109/CCAA.2015 vol. 106, pp. 14–25, Aug. 2018. doi: 10.1016/j.
June 5, 2020. [Online]. Available: https://ifr.org .7148524. robot.2018.04.003.
[10] V. Villani, F. Pini, F. Leali, and C. Secchi, “Survey [28] H. Huang, A. V. Savkin, M. Ding, and C. Huang, “Mo- [46] V. Babin, D. St-Onge, and C. Gosselin, “Stable
on human–robot collaboration in industrial set- bile robots in wireless sensor networks: A sur- and repeatable grasping of flat objects on hard
tings: Safety, intuitive interfaces and applica- vey on tasks,” Comput. Netw., vol. 148, pp. 1–19, surfaces using passive and epicyclic mecha-
tions,” Mechatronics, vol. 55, pp. 248–266, Nov. Jan. 2019. doi: 10.1016/j.comnet.2018.10.018. nisms,” Robot. Comput.-Integr. Manuf., vol. 55,
2018. doi: 10.1016/j.mechatronics.2018.02.009. [29] M. Wollschlager, T. Sauter, and J. Jasperneite, no. Part A, pp. 1–10, 2019. doi: 10.1016/j.rcim.
[11] A. Ajoudani, A. M. Zanchettin, S. Ivaldi, A. Albu- “The future of industrial communication – 2018.06.002.
Schäffer, K. Kosuge, and O. Khatib, “Progress Automation networks in the era of the Inter- [47] Z. Wang, Y. Torigoe, and S. Hirai, “A pre-
and prospects of the human–robot collabora- net of Things and Industry 4.0,” IEEE Ind. Elec- stressed soft gripper: Design, modeling, fabri-
tion,” Autonom. Robot., vol. 42, no. 5, pp. 957–975, tron. Mag., vol. 11, no. 1, pp. 17–27, 2017. doi: cation, and tests for food handling,” IEEE Ro-
2018. doi: 10.1007/s10514-017-9677-2. 10.1109/MIE.2017.2649104. bot. Automat. Lett., vol. 2, no. 4, pp. 1909–1916,
[12] G. Michalos et al., “Seamless human robot col- [30] T. Gregory, A. Salomons, and U. Zierahn, “Rac- Oct. 2017. doi: 10.1109/LRA.2017.2714141.
laborative assembly – An automotive case study,” ing with or against the machine? Evidence [48] J. Thilmany, “Exoskeletons for construction
Mechatronics, vol. 55, pp. 194–211, Nov. 2018. doi: from Europe,” Centre for European Economic workers are marching on-site,” Constructible,
10.1016/j.mechatronics.2018.08.006. Research (ZEW), Mannheim, Germany, Dis- Feb. 27, 2019. Accessed: June 5, 2020. [On-
[13] T. Salmi, I. Marstio, T. Malm and J. Montonen, cussion Paper No. 16-053, 2017. line]. Available: https://constructible.trimble
“Advanced safety solutions for human-robot- [31] J. Xu, C. Zhang, Z. Liu, and Y. Pei, “A review .com/construction-industry/exoskeletons-for
cooperation,” in Proc. ISR 2016: 47st Int. Symp. on significant technologies related to the -construction-workers-are-marching-on-site
Robot., Munich, June 2016, pp. 610–615. robot-guided intelligent bolt assembly under [49] D. Kara, “Industrial exoskeletons: New systems,
[14] N. Enayati, E. De Momi, and G. Ferrigno, “Haptics complex or uncertain working conditions,” improved technologies, increasing adop -
in robot-assisted surgery: Challenges and ben- IEEE Access, vol. 7, pp. 136,752–136,776, Sept. tion.” The Robot Report, Dec. 6, 2018. Accessed:
efits,” IEEE Rev. Biomed. Eng., vol. 9, pp. 49–65, 2019. doi: 10.1109/ACCESS.2019.2941918. June 5, 2020. [Online]. Available: https://www
Mar. 2016. doi: 10.1109/RBME.2016.2538080. [32] ROS-Industrial. Accessed: June 5, 2020. [On- .therobotreport.com/industrial-exoskeletons/
[15] H. Liu and L. Wang, “Remote human–robot col- line]. Available: https://rosindustrial.org/ [50] P. Leitão, S. Karnouskos, L. Ribeiro, J. Lee,
laboration: A cyber–physical system applica- [33] J. Curie and P. Curie, “Contractions et dilata- T. Strasser, and A. W. Colombo, “Smart agents
tion for hazard manufacturing environment,” tions produites par des tensions dans les cris- in industrial cyber–physical systems,” Proc.
J. Manuf. Syst., vol. 54, pp. 24–34, Nov. 2019. doi: taux hémièdres à faces inclines,” Comptes ren- IEEE, vol. 104, no. 5, pp. 1086–1101, May 2016.
10.1016/j.jmsy.2019.11.001. dus des hebdomadaires, vol. 93, pp. 1.137–1.140, doi: 10.1109/JPROC.2016.2521931.
[16] A. Casalino, C. Messeri, M. Pozzi, A. M. Zanchettin, Dec. 1881. [51] A. Bonci, M. Pirani, and S. Longhi, “Robotics
P. Rocco, and D. Prattichizzo, “Operator aware- [34] Z. Xing, J. Zhang, D. McCoul, Y. Cui, L. Sun, 4.0: Performance improvement made easy,” in
ness in human–robot collaboration through and J. Zhao, “A super-lightweight and soft Proc. 22nd IEEE Int. Conf. Emerg. Technol. Facto-
wearable vibrotactile feedback,” IEEE Robot. Au- manipulator driven by dielectric elastomers,” ry Automat., ETFA 2017, Limassol, Cyprus, 2017,
tomat. Lett., vol. 3, no. 4, pp. 4289–4296, Oct. 2018. Soft Robot., vol. 7, no. 4, pp. 512–520, 2020. doi: pp. 1–8. doi: 10.1109/ETFA.2017.8247682.
doi: 10.1109/LRA.2018.2865034. 10.1089/soro.2018.0134. [52] F. Lévesque, B. Sauvet, P. Cardou, and C. Gosse-
[17] G. Salvietti, M. Z. Iqbal, and D. Prattichizzo, “Bi- [35] T. Albrecht, C. Bührer, M. Fähnle, K. Maier, D. lin, “A model-based scooping grasp for the au-
lateral haptic collaboration for human-robot co- Platzek, and J. Reske, “First observation of fer- tonomous picking of unknown objects with a
operative tasks,” IEEE Robot. Automat. Lett., vol. romagnetism and ferromagnetic domains in a two-fingered gripper,” Robotics Autonom. Syst.,
5, no. 2, pp. 3517–3524, Apr. 2020. doi: 10.1109/ liquid metal,” Appl. Phys. A, Mater. Sci. Process., vol. 106, pp. 14–25, Aug. 2018. doi: 10.1016/
LRA.2020.2975715. vol. 65, no. 2, pp. 215–220, 1997. doi: 10.1007/ j.robot.2018.04.003.
[18] R. Drath and A. Horch, “Industrie 4.0: Hit or hype? s003390050569. [53] A. Fenucci, M. Indri, and F. Romanelli, “An
[industry forum],” in IEEE Ind. Electron. Mag., [36] L. Hu, H. Wang, X. Wang, X. Liu, J. Guo, and J. off-line robot motion planning approach for
vol. 8, no. 2, pp. 56–58, June 2014. doi: 10.1109/ Liu, “Magnetic liquid metals manipulated in the the reduction of the energy consumption,” in
MIE.2014.2312079. three-dimensional free space,” ACS Appl. Ma- Proc. 21st IEEE Int. Conf. Emerg. Technol. Fac-
[19] M. Indri, L. Lachello, I. Lazzero, F. Sibona, and ter. Interfaces 2019, vol. 11, no. 8, pp. 8685–8692, tory Automat. (ETFA), Berlin, 2016, pp. 1–8. doi:
S. Trapani, “Smart sensors applications for Feb. 15, 2019. doi: 10.1021/acsami.8b22699. 10.1109/ETFA.2016.773359c5.
a new paradigm of a production line,” Sen- [37] “Smart materials,” IIT Centrals Research Labs, [54] L. Bukata, P. Šůcha, Z. Hanzálek, and P. Burget,
sors, vol. 19, no. 3, p. 650, 2019. doi: 10.3390/ ­Genoa, Italy. Accessed: June 5, 2020. [Online]. “Energy optimization of robotic cells,” IEEE
s19030650. Available: https://www.iit.it/research/lines/smart Trans. Ind. Informat., vol. 13, no. 1, pp. 92–102,
[20] M. Indri, A. Grau, and M. Ruderman, “Guest -materials Feb. 2017. doi: 10.1109/TII.2016.2626472.
editorial special section on recent trends [38] P. Li and X. Liu, “Common sensors in industrial [55] K. Darvish, F. Wanderlingh, B. Bruno, E. Simet-
and developments in Industry 4.0 motivated robots: A review,” J. Phys.: Conf. Ser., vol. 1267, ti, F. Mastrogiovanni, and G. Casalino, “Flexible
robotic solutions,” IEEE Trans. Ind. Informat., pp. 1–10, Apr. 2019. human–robot cooperation models for assisted
vol. 14, no. 4, pp. 1677–1680, 2018. doi: 10.1109/TII [39] A. Cirillo, F. Ficuciello, C. Natale, S. Pirozzi, and shop-floor tasks,” Mechatronics, vol. 51, pp. 97–
.2018.2809000. L. Villani, “A conformable force/tactile skin for 114, May 2018. doi: 10.1016/j.mechatronics.
[21] J. Mehami, M. Nawi, and R. Y. Zhong, “Smart au- physical human–robot interaction,” in IEEE Ro- 2018.03.006.
tomated guided vehicles for manufacturing in bot. Automat. Lett., vol. 1, no. 1, pp. 41–48, Jan. [56] G. Tang, P. Webb, and J. Thrower, “The develop-
the context of industry,” Procedia Manuf., vol. 26, 2016. doi: 10.1109/LRA.2015.2505061. ment and evaluation of Robot Light Skin: A novel
pp. 1077–1086, 2017. doi: 10.1016/j.promfg [40] “KUKA Collaborative Robot Series,” Robot- robot signalling system to improve communica-
.2018.07.144. Worx, Marion, OH. Accessed: June 5, 2020. tion in industrial human–robot collaboration,”
[22] “What are AGV?” Atria Innovation, Zaragoza, [Online]. Available: https://www.robots.com/ Robot. Comput.-Integ. Manuf., vol. 56, pp. 85–94,
Spain, Nov. 13, 2018. Accessed: June 5, 2020. [On- series/kuka-collaborative-robot-series 2019. doi: 10.1016/j.rcim.2018.08.005.
line]. Available: https://www.atriainnovation [41] F. Xia, B. Bahreyni, and F. Campi, “Multi-func- [57] G. Papadopoulos, S. Rikama, P. Alajaasko, Z.
.com/en/what-are-agvs/ tional capacitive proximity sensing system for Salah-Eddine, A. Airaksinen, and H. Luomaran-
[23] S. Allen, “Robots with laser and vision sys- industrial safety applications,” in Proc. IEEE tai, “Statistics on small and medium sized enter-
tems conquer new industrial terrain,” Vision SENSORS, 2016, pp. 1–3. doi: 10.1109/ICSENS prises,” Eurostat Statistics Explained, 2015. Ac-
Spectra, Sept. 2016. Accessed: Dec. 10, 2020. .2016.7808496. cessed: June 5, 2020. [Online]. Available: https://
[Online]. Available: https://www.photonics [42] D. Hughes, J. Lammie, and N. Correll, “A ro- ec.europa.eu/eurostat/statistics-explained/
.com/Article.aspx?AID=61118 botic skin for collision avoidance and affective index.php/Statistics_on_small_and_medium
[24] A. Koubaa and K. Abdelmajid, Eds., Coopera- touch recognition,” in IEEE Robot. Automat. -sized_enterprises
tive Robots and Sensor Networks 2014. Berlin: Lett., vol. 3, no. 3, pp. 1386–1393, 2018. doi:
Springer-Verlag, 2014. 10.1109/LRA.2018.2799743. 

MARCH 2021 ■ IEEE INDUSTRIAL ELECTRONICS MAGAZINE 61

Authorized licensed use limited to: Carleton University. Downloaded on May 25,2021 at 17:06:12 UTC from IEEE Xplore. Restrictions apply.