Sie sind auf Seite 1von 9

Robotics and Autonomous Systems 61 (2013) 12151223

Contents lists available at ScienceDirect

Robotics and Autonomous Systems


journal homepage: www.elsevier.com/locate/robot

Applied ontologies and standards for service robots


Tams Haidegger a,b, , Marcos Barreto c , Paulo Gonalves d , Maki K. Habib e ,
Sampath Kumar Veera Ragavan f , Howard Li g , Alberto Vaccarella h , Roberta Perrone h ,
Edson Prestes i
a
buda University, Budapest, Hungary
b
Austrian Center for Medical Innovation and Technology (ACMIT), Wiener Neustadt, Austria
c
Distributed Systems Laboratory (LaSiD), UFBA, Brazil
d
Polytechnic Institute of Castelo Branco / Technical University of Lisbon, Center of Intelligent Systems, IDMEC / LAETA, Portugal
e
The American University in Cairo, Cairo, Egypt
f
Monash University, Sunway campus, Selangor, Malaysia
g
Department of Electrical and Computer Engineering, University of New Brunswick, Canada
h
Department of Electronics, Information and Bioengineering (DEIB), NearLab Medical Robotics, Politecnico di Milano, Milan, Italy
i
Instituto de Informtica, UFRGS, Brazil

highlights
We present the ongoing work towards a core ontology dedicated to the field of robotics and automation.
We describe the role of ontologies in robotics at large.
We provide a thorough review of service robot ontologies.
We describe the existing standards for robots, along with the future trends in the domain.
We define the current issues and solutions possible with ontology development for humanrobot interaction.

article info abstract


Article history: Service robotics is an emerging application area for human-centered technologies. The rise of household
Available online 6 June 2013 and personal assistance robots forecasts a humanrobot collaborative society. One of the robotics com-
munitys major task is to streamline development trends, work on the harmonization of taxonomies and
Keywords:
ontologies, along with the standardization of terms, interfaces and technologies. It is important to keep
Service robotics
the scientific progress and public understanding synchronous, through efficient outreach and education.
Robot ontologies
Robot standards These efforts support the collaboration among research groups, and lead to widely accepted standards,
Humanrobot interaction beneficial for both manufacturers and users. This article describes the necessity of developing robotics on-
tologies and standards focusing on the past and current research efforts. In addition, the paper proposes a
roadmap for service robotics ontology development. The IEEE Robotics & Automation Society is sponsoring
the working group Ontologies for Robotics and Automation. The efforts of the Working group are presented
here, aiming to connect the cutting edge technology with the users of these servicesthe general public.
2013 Elsevier B.V. All rights reserved.

1. Introduction to rationalize production at manufacturing sites. Nowadays, robots


are becoming ubiquitous, reaching exceptional capabilities and ro-
Between the 1960s and the 90s, most robots and robotics in bustness; being able to support, accompany and nurse humans. In
general were related to industrial applications, mainly intending particular, robotic helpers are able to perform a wide range of tasks,
such as healthcare and personal assistance [13]. These kinds of
robots are commonly called service robots. Service robots typically
share the human environment and exhibit basic intelligent behav-
Corresponding author at: buda University, Budapest, Hungary. Tel.: +36 ior to accomplish assigned tasks. They are considered as a branch
13658550. in the evolution of robots and widely recognized as a dominant re-
E-mail addresses: haidegger@ieee.org (T. Haidegger), marcoseb@dcc.ufba.br
search field in the near future. Fig. 1 shows the evolution of the dif-
(M. Barreto), paulo.goncalves@ipcb.pt (P. Gonalves), maki@aucegypt.edu
(M.K. Habib), veera.ragavan@monash.edu (S.K.V. Ragavan), howard@unb.ca (H. Li), ferent fields of robotics, implying an increasing Degree of Autonomy
alberto.vaccarella@mail.polimi.it (A. Vaccarella), roberta.perrone@mail.polimi.it (DoA) and system complexity along with human-centered appli-
(R. Perrone), edson.prestes@ieee.org (E. Prestes). cations. Nowadays, industrial robots are becoming more versatile
0921-8890/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.robot.2013.05.008
1216 T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223

Fig. 1. Evolution of robotics. Current trends are leading towards more complex, more personalized systems and robot services. This implies flexible systems that are able
to perform tasks in an unconstrained, human-centered environment.

and flexible, partially due to the improved cooperation with hu- standardization, Section 6 describes the standards for robots, with
mans. a special focus on service robots. Section 7 describes the work in
The paradigm shift towards pervasive robotics leads to many progress by the IEEE ORA Working Group. Future work and con-
open questions and research directions relevant to ontological clusions are presented in Sections 8 and 9, respectively.
research and standardization [4,5]. Ontologies allow a clear dialog
between all stakeholders involved in the life-cycle of a robotic 2. Defining the domain of service robots
system, and enable the integration and efficient communication
of heterogeneous robotic systems. They can be used in the earlier Developing an ontology or a standard is not an easy task, es-
phase of an evolving domain to facilitate the communication and pecially when it encompasses a huge field, like service robotics.
knowledge exchange among groups from different areas, without Although the potential applications of these robots are well un-
really forcing them to align their research with the particular view derstood, it is not easy to provide a clear and precise definition for
of a given research group. service robots. There are several definitions and none of them are
On the other hand, standards are meant for industrial bodies to generally accepted due to the huge variety of intrinsic character-
provide guidelines for commercial system development and man- istics of service robots (in their form, structure and application
ufacturing. They are used to support legislation and to empower area) [7]. Below, we discuss some of them, pointing out possible
official bodies, enforcing liability. For that reason, they should be sources of confusion.
based on mature, established and widely accepted technologies. According to the International Federation of Robotics (IFR):
Ideally, standards could (and should) greatly rely on already
accepted and used ontologies. This is the reason why they are Definition 1. A service robot is a robot which operates semi or
discussed together in this paper, as two stages of an evolutionary fully autonomously to perform services useful to the well-being of
process. It is also important that the community keeps the pace humans and equipment, excluding manufacturing operations [7].
with the accelerating global Research and Development (R&D),
With this definition, an industrial robot, e.g., a robotic arm, could
employing self-regulation in development practices and research
easily be classified as a service robot if it is installed in a non-
ethics, even before consensual standards and regulations are born.
industrial environment to perform non-manufacturing task. For
This paper aims to give an introduction to the motivation and
instance, this arm could serve beer in a bar or be used to perform a
fundamental ideas of the IEEE Ontologies for Robotics and Automa-
surgery.
tion Working Group (ORA WG) [6], and an insight to the ongoing
To minimize ambiguity in classification, the International Stan-
work of its sub-groups. The sub-group stands to provide guidelines
dardization Organization (ISO) has been working on a standard def-
for creating ontologies that will help future research, then natu- inition under the scope of Technical Committee 184, Sub-Committee
rally transforming into standards for the field. Specifically, this ar- 2, Robots and Working Devices since 2007 [8]. The new version of
ticle focuses on the ORA WG Service Robotics sub-group (ORA SeR), the proposed definition (ISO 8373:2012) is:
discussing its goals, recent achievements and future challenges.
The current focus of the SeR group is to collect, categorize and Definition 2. A service robot is a robot that performs useful tasks
merge existing ontologies and standards, and initiate the develop- for humans or equipment, excluding industrial automation appli-
ment of new specific ones on top of the state of the art. cations [9].
The paper is structured as follows: an introduction to service
robots and their definitions is presented in Section 2. Section 3 This definition does not emphasize aspects related to autonomy
presents the rationale for ontologies and describes the motivation in order to classify a robot as a service robot, while in the ISO stan-
for using ontologies for service robots. Section 4 gives a thorough dard, autonomy is an inherent property of any robot.
literature review of ontologies used in robotics. When developing
robotics ontologies, human factors must be considered; the on- Definition 3. A robot is an actuated mechanism programmable in
tology development for humanrobot interaction is presented in two or more axes with a degree of autonomy, moving within its
Section 5. Since the development of robotics ontologies needs environment, to perform intended tasks.
T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223 1217

Class 2 robots that operate closely with humans to alleviate


incommodity or to increase comfort, such as entertainment,
assisting the elderly, or carrying patients;
Class 3 robots that operate on humans, e.g., medical robots for
surgery, diagnosis, treatment and rehabilitation.

The above definitions and classifications show the complexity


involved in defining concepts. Much of the complexity comes
from the subjectivity in identifying the precise borderline between
systems assigned to particular tasks.1 For example, Definitions 1
and 2 explicitly exclude industrial applications. According to the
above classes, a service robot is a robot that should replace humans
at work in hazardous environments and tedious operations,
however, these operations and environments can be viewed as
associated to industrial robots in industrial environments.
To avoid these pitfalls, we need to use tools that allow formal
and unambiguous definitions. In this sense, ontology plays an im-
portant role, since an ontology permits an explicit and formal speci-
fication of a shared conceptualization [15].
Fig. 2 shows several applications related to service robotics. De-
pending on the intended use, similar systems can belong to differ-
Fig. 2. Categories of medical/non-medical personal service robots. Existing
ent categories. For instance, an exoskeleton can be classified as a
commercial product and better-known prototypes are listed to provide examples
to reduce ambiguity [12]. medical robot when it is used in a rehabilitation process or as a
non-medical robot when it is used in assistive tasks, according to
Definition 3 is precise in stating that a robot should be an ac- Definitions 6 and 7. Although both definitions belong to the same
tuated mechanism programmable in two or more axes. However, robot class, i.e., personal service robots, an exoskeleton can be used
it is still imprecise, since there is no consensus on DoA yet. That in military tasks [16], which contradicts the intended use of this
makes it difficult to determine what exactly is the minimum de- kind of robot. Currently, the categorization of a new system is
gree that makes an actuated mechanism with the above features solely the manufacturers responsibility. Understandably, there are
classified as a robot. The concept of autonomy is still being de- serious questions arising regarding the boundary issues of these
bated within the community. Some proposals have been formu- categories, which is now the focus of the international standard-
lated, such as the Autonomy Levels For Unmanned Systems (ALFUS) ization bodies [12].
framework developed by the National Institute of Standards and
Technology (NIST) [10,11]. Therefore, Definition 2 still has similar
problems of Definition 1. 3. Case for ontologies
In the ISO standard similarly to IFR service robots can be
categorized as personal and professional service robots according Ontology or subject of existence first appeared in philosoph-
to the tasks to be performed [9]: ical literature in the times of Aristotle. Ontologies can be viewed as
content theories that focus on properties and relationships among
Definition 4. A personal service robot is a service robot used for objects from a specific domain [17]. They act as a body of knowl-
a non-commercial task, usually by lay persons (e.g., a domestic edge that is based on a vocabulary used to describe the domain.
servant robot).
They have been applied to different fields, ranging from genet-
ics [18] to knowledge portals [19], as an approach to ensure a
Definition 5. A professional service robot is a service robot used
for a commercial task, usually operated by a properly trained consistent relationship between terms and their properties. Specif-
operator (e.g., a fire-fighting robot). ically in robotics, ontologies can be used to represent the domain
knowledge to be transferred and shared between different groups
Specifically for personal service robots, ISO TC 184/SC 2/WG of humans, robots and other devices [20,21].
7 workgroup for Personal care robot safety categorizes personal Ontologies are shared knowledge. So, it is important that they
service robots (also called personal care robots) as medical and non- are standardized for usage in a certain group of people such as
medical robots [12].
in IEEE Robotic and Automation Society. A review of literature on
Definition 6. Non-medical personal care robots are defined as ma- ontologies also reveals another startling fact. The word ontology
chines for performing aiding actions, and actions contributing di- refers to glossaries and data dictionaries; thesauri and taxonomies;
rectly towards improvement in the quality of life of humans, except schema and data models; formal ontologies and inference mod-
medical applications. els [22,23].
There are many definitions for Ontology. One possible definition
Definition 7. Medical robots are defined as medical electrical for Ontology is provided by Kalfoglou and Schorlemmer [24]:
equipment and systems which are used to diagnose, treat or re-
habilitate patients from medical conditions. Definition 8. Ontology is a tuple S , A, where S is the vocabulary
Moreover, service robots can be broken down into three classes (or signature) of the ontology and A is the set of ontological axioms
based on their relation to humans, since human-centered robotics specifying the intended domain vocabulary.
is becoming dominant [1,13,14]:
Class 1 robots that replace humans at work in dirty, hazardous
environment and tedious operations, such as vacuuming, 1 From the standardization point of view, a system is always handled according
rescuing people from fire or delivering medication; to its intended use.
1218 T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223

Table 1
Comparison of ontologies and models.
Feature Model Ontology

Deployment No need to be shared openly or shared within a Shared by all people in a domain, or across many domains.
small group (e.g., among developers/designers)a
Open/Closed Closed world, hence can be prescriptive or Open world assumption lead to a descriptive models only
descriptive
Transformable Can be transformed from one to another, Ontologies can only be mapped from one to another using additional
meta-models can be bridged between one and axioms and aligning them to create a shared ontology
another
Propagation of constraints Can be propagated both ways One way only
Implementation level At lower abstraction levels At computation-independent model levels
Possibility of integration Generally done using transformation and More difficult and done using techniques like lifting and bridging
Generation
Abstraction level Models are more concretehence they result in Ontology is more descriptive and abstract and hence it can be used for KR
actual implementation (e.g., can be transformed research.
into Code)
a
For example a nonlinear control model of a UAV is not an ontology.

In a similar definition given in [25], the signature S is broken is that they represent general knowledge shared by a community,
down into three sets (1) a set of concepts (C), (2) a set of rela- i.e., they do not only stand for a particular view of a person of a
tions (R) and (3) the set of instances (I); thus, an ontology is a 4- small group of people. Other differences are mainly based on the
tuple C , R, I , A. Other ontology definitions that have n-tuple with open-world and closed-world assumptions. In Table 1, we list the
n = 5 [26] and n = 6 [27] can also be found in the literature. differences between a model and an ontology.
As evident from the numerous definitions proposed for ontol-
ogy, it has been applied informally and formally, as abstract and
4. Ontologies at work
concrete models [28]. With such diversity [22], contrasting notions
of ontologies ranging from being a concrete engineering artifact to
a branch of metaphysics concerned with nature and relations to 4.1. Benefit of creating ontologies
being, i.e., from a human-readable ontology such as a semantic
Wiki to a completely machine-processable and interpretable on- Since ontologies can efficiently represent the knowledge of a
tology such as Ontology Web Language (OWL).2 specific system (or domain), their analysis allows a better under-
Ontological engineering research received its impetus from in- standing of such systems. Without ontologies, it is very difficult to
dependent research work carried out mainly by different commu- establish a common vocabulary to represent all the knowledge in-
nities in computer science, Artificial Intelligence (AI) or semantic volved. A lack of these can cause inconsistencies and hide the in-
web. Recently, the focus of ontological development community teraction between the various elements of the system.
has moved from AI Labs to domain expertspeople who have accu- Object-oriented systems apply domain-specific ontologies to
mulated valuable functional knowledge of an area. In many fields, represent objects, attributes and procedures. AI systems use
(ranging from medicine to defense research), domain experts play ontologies to represent and to operate the knowledge by means of
an important role in developing standard ontologies. They help to inference. Expanding the scope, various knowledge-based systems
define a common vocabulary and basic domain concepts (including rely on ontologies to perform reasoning tasks that depends on
machine-interpretable definitions), which are then used to share large amounts of information [31]. For instance, the construction
information [29]. and use of a data based ontology for steam turbine diagnosis is
Ontology engineering for robotics is at a crossroads, since ex- presented in [32]; ontologies are applied to the development of a
perts from different fields (electro-mechanical, information tech- home automation system in [33].
nology, etc.) bring their own language, practices and methodology Ontologies also enable knowledge sharing and the construction
to create and store knowledge in order to create ontologies. Flouris
of a knowledge base describing specific situations within a domain.
et al. [30] state:
Focusing on robotics, ontologies are important to represent the
We believe that this lack of a standard terminology constitutes knowledge acquired by robots. They allow sharing the concepts
a major bottleneck for the ontology change community, causing coming from different sub-domains and the communication
an unnecessary confusion as well as misunderstandings. between robots with different roles and capabilities, or between
The right balance should be adopted by the robotics ontology humans and robots.
community between philosophically well founded heavyweight
ontologies using languages focusing on expressiveness and clarity 4.2. Review of robotic ontology projects
vs. lightweight ontologies focused on computer properties to guar-
antee machine-interpretable definitions.3 Ontologies have been benefiting robotics for the past years by
providing new approaches to robot design and operation. These
3.1. Models vs. ontologies have been implemented successfully for autonomous, industrial,
or service robots. For robot design and development, ontologies
Ontologies are different from models (model-based representa- have been using OWL to describe robots including their structural
tion and expert systems) for several reasons. The most important and operation capabilities for urban search and rescue [34,35].
Regarding operations, they could be [35]:
long range, human stairway-access and upper floor situational
2 http://www.w3.org/TR/owl-ref.
awareness;
3 A lightweight ontology is a semantic network using only typed binary relations,
contaminated area survey; site assessment;
and usually defined using a lightweight ontology language(e.g. XOL, RDF(S),
SHOE etc.) whereas a heavyweight ontology is usually domain specific, expressive
victim identification;
ontologies consisting of an extensive vocabulary and taxonomy. mitigation activities and so on.
T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223 1219

From [34], operations are related to topological features (like ontology (support for the engineering process of autonomous sys-
tunnel, room, etc.), surface conditions (like sandy, dusty, etc.), vis- tems). Each sub-ontology is split into parts, to better represent con-
ibility, field and radiation conditions, etc. Both papers converge cepts and relationships. OASys also incorporates a methodology to
in defining similar structural robot capabilities; like size, locomo- generate conceptual models. A full UML-based implementation of
tion mechanism, sensors, humanrobot interaction and software/ OASys is also described, along with two testbed applications for
hardware processing. robot and process control.
To describe sensors and their specifications, several ontologies Ontologies are also used to describe spatial information and
have been developed, and were reviewed by Compton et al. [36]. navigational approaches from cognitive robotics, based on existing
In [37], the same previous group presents its sensor ontology that ontologies, such as DOLCE and BFO [47].
describes the capabilities and properties of sensors as well as de- Standard Ontology for Ubiquitous and Pervasive Applications
scribing sensors as compositions of their components. For exam- (SOUPA) [48] is a standard ontology that represents concepts re-
ple, a wind sensor can measure wind speed and direction, while lated to intelligent agents and a set of characteristics related to
having different performance requirements under various work- time, space, actions, policies, among other items. Context Broker
ing conditions. This sensor can also be added to a temperature Architecture (CoBrA) [49] exploits the SOUPA ontology to allow
sensor, to have a wind chill sensor, that outputs a value dependent
knowledge sharing, context reasoning, and user privacy protection.
on wind and temperature measures.
It also incorporates spatial and temporal features coming from ex-
In [38], the development of an ontology is presented that allows
isting ontologies expressed in the DARPA Agent Markup Language
for reasoning on the sensors needed to accomplish a given mission,
(DAML) [50].
e.g., a military mission. In this work, a package configuration (of
platform and sensors) is outputted for a given task, with precise
capability requirements. For a task that requires night operations 5. Human factor for ontology development
capability, and the ability to detect infrared (IR) radiation, what is
the best sensor to use? The reasoner then outputs the IR sensor, When building ontologies, the end-users must also be consid-
with the following capabilities (DayNight, IRRadiation) that best ered. Today, HumanRobot Interaction (HRI) [51] is a major concern
matches the task. The reasoner is based on matching the sensor for robotics researchers and practitioners, due to the increasing
and task capabilities defined in the ontology. need of robots helping and cooperating with humans [4]. People
Under the recent wave of open-source software for the robotics are expecting robots to become part of their lives e.g., the Roomba
community, two software tools that can be used to implement on- vacuum cleaner [52], the DaVinci surgical system [53], or advanced
tologies and put them to work in a robotic system were recently toys, such as the Sony Aibo [54].
proposed. Open Robot Ontology (ORO) [39] and a Knowledge Pro-
HRI poses new challenges such as safety issues and the degree
cessing Framework (KnowRob) [40] extract information from well-
of autonomy [12]. Current designs typically focus on how robots
known upper ontologies like SUMO and/or OpenCyc, and develop
and humans form teams [4]. In order to develop safe, precise and
specific knowledge databases for robots. Both approaches can be
easy to use HRI tools, knowledge should be properly and formally
used with the Robot Operating System (ROS) [41]. Robotic tasks and
defined. Knowledge representation will lead to robots that can use
robotic actions specific to home environments are implemented in
ontologies to perform tasks.
these ontologies. The ontologies are employed by the robot to rea-
son in the environment and also interact with humans. Specifically, Humans can interact with robots in several ways, but mainly as
KnowRob has the capability to reason from images captured by the a developer or as an end-user. First, for the developer, a core ontology
robots vision sensor, using reasoners and machine learning tools must be used to represent specific knowledge (definition of the
like Prolog and WEKAfrom the pre-defined ontology therein. robot, type of robots to use, components to use, and so on). Second,
In [42], an ontology to aid grasping and manipulation is pre- for the end-user, application domain ontology needs to be defined
sented. This ontology represents the functional and geometric part for the robots work environment, e.g., knowledge of the household
affordances of object to enhance a task based grasp. It is of special environment or the healthcare environment. Fig. 3 illustrates these
interest for a service robot, while interacting with humans. two cases.
Additionally, several ontologies have been proposed for de-
scribing the robot design, operational behavior, protocols for hu- 6. The need for standards in service robotics
manrobot interaction and also for robotic system software de-
velopment. For instance, RoboDB [43] is a database that uses
6.1. Benefit of standardization
semantic Web technologies to store information about robot em-
bodiments (physical components). Based on this information, on-
tologies focused on the physical characteristics of robots can be The evolution of service robots and their application in all
created and populated. RoboDB can also help with different knowl- dimensions of our daily life require standardization of robotics. The
edge representations used during the establishment of robot mid- total number of professional service robots sold in 2011 was over
dleware. 16,000, up from 15,000 in 2010 (medical robots grew by 38% while
In [44], the KnowLang formal language for knowledge represen- factory logistic systems grew by 13%). The sales value reached $3.6
tation is described as part of the ASCENS project [45]. This language billion. The sales forecast of service robots for professional use
is used in a multi-agent system (comprised of mobile and intel- (201215) projects an increase to about 94,000 units at a value
ligent robotic units) to specify different knowledge domains and of $16.3 billion, and to about 11 million units with a value of
reasoning primitives. ASCENS considers four knowledge domains, $4.8 billion for personal service robots [55]. The latter includes
while a specific ontology together with a logical framework is vacuum and floor cleaning, helpers for the elderly, lawn-mowing
used to describe concepts, objects and the relationships between and entertainment robots, etc.
them. A case study based on a swarm of MarXbot robot units is The development of service robots requires the R&D of context-
also presented. based interdisciplinary science and technology, relying on a com-
In [46], the Ontology for Autonomous Systems (OASys) is pre- mon understanding of knowledge and terminologies within the
sented. OASys is comprised by two sub-ontologies: ASys ontology field, such as mechanisms, modules, functions, sensors, and soft-
(conceptualization of autonomous systems) and ASys engineering ware, etc.
1220 T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223

Fig. 3. The components of a robotic system with core and domain ontologies, and interaction with humans.

Robot standards facilitate the design, manufacturing, produc- International Organization for Standardization: ISO 8373:2012,
tion, technology transfer, purchase and use of robots. In the ISO 9787:1999, ISO 11593:1996, ISO 14539:2000 for industrial
meantime, standards ensure a common legal framework. Stan- robotics (under revision to include service robots), safety issues
dardization activities are associated with manufacturers, are well defined in ISO 10218-1/2:2011;
researchers, users and authorities. Common interests and re- American Society for Testing and Materials (ASTM): ASTM E2521-
quirements are ensured through the development of standards. 07, for Urban Search and Rescue Robotic Operations;
Standardization is essential in the development of Robotic and American Institute of Aeronautics and Astronautics (AIAA): S-066-
Automation (R&A) systems. Lacking guidance and standardization 1995, for Space Automation and Robotics;
in robotics may cause slower development, or lead to divergence Japanese Industrial Standards (JIS): JIS B 0144:2000, JIS B
of development, causing frustration for not only consumers but 0185:2002, JIS B 0186:2003, JIS B 0187:2005, TR B 0010:1999,
also manufacturers. Standardization not only facilitates commer- for Assembly, Intelligent, Mobile, Service and Personal Robots.
cialization and knowledge transfer, but also guides R&D activities
In addition, several groups have been defining ontologies as
towards more focused solutions.
draft standards for service robots, such as the Robot Standards and
ISO is one of the pioneers creating robotics standards. ISO
Reference architecture (RoSta) (wiki.robot-standards.org). RoSta was
drafted standards for industrial robots in 1980s. Besides ISO,
a Coordination Action funded under the European Unions Sixth
the International Electrotechnical Commission (IEC), the Object
Framework Programme (FP6) that aimed at becoming the main
Management Group (OMG) and other organizations also started
international contact point for robot standards and reference ar-
developing standards for the emerging service robotics. According
chitectures for service robotics. Another example is the Ontology-
to the IEEE RAS Standing Committee for Standards, the objectives for
based unified robot knowledge for service robots in indoor
standardization include:
environments [56], developed in Korea within the project Devel-
Common measures and definitions in R&A; opment of Context Adaptive Cognition Technology.
Measurability and comparability of R&A technology;
Integrability, portability and reusability. 6.3. Further standardization efforts
A standard for R&A should:
One of the newest results is the ISO/FDIS 13482:2013 that will
Define precisely the concepts of the robotics domain; be published in August 2013, defining safety requirements for non-
Ensure a common understanding among various stakeholders medical personal care robots. At the ISO Technical Committee 184
(involved in any part of the life cycle of robotics systems); (TC on Automation Systems and Integration) meeting (Budapest,
Enable efficient and reliable data integration and information 2010), the SC 2 technical sub-committee on Robots and Robotic De-
transfer among robotics systems; vices assisted in the formation of a joint work group with the IEC/SC
Allow for unambiguous knowledge transfer among any group 62A sub-committee (Common Aspects of Electrical Equipment used
of humans, robots and other systems. in Medical Practice) under IEC administration to develop a new pro-
posal for a Collateral Standard within the IEC/ANSI 60601 family.
Existing standards focus on one specific problem each, such as Thus the JWG 9: Joint Work Group on Standard for Medical Robots
vocabulary, performance criteria and related test methods, me- was created with the involvement of several international experts
chanical interfaces, safety requirements, graphical user interface from the academia, government and industry [12]. The scope of the
for programming and operation or terminology and practice for standard is the safety and essential performance requirements for
field robotics. medical electrical equipment. In addition, it may include systems
employing robotic technology (i.e., medical robots). The objectives
6.2. Current robotics standards include:
Defining medical software systems and associated technical
Robotic systems need to meet hardware and functionality re-
requirements;
quirements. Hardware requirements are related to the physical
Streamlining the application of risk management;
components of the robotic system, such as sensors and actuators.
Clarifying the definition of essential performance;
Functionalities can be intrinsic and extrinsic. The first is related to
the software architecture of the robotic system. The second is re-
Identifying essential performance and mitigating risks;
lated to the tasks to be performed, while interacting with human
Defining safety factors for robotic equipment.
operators in the environment. All these are affected by collateral Seven meetings of the workgroup have been held: in Budapest
and particular standards, developed by international bodies. (2010), Sunnyvale (2011), Nremberg (2011), Orlando (2012),
Several standards exist for robotics in general. Many should Seoul (2012), Santa Clara (2013) and Shanghai (2013), while one
be considered, when working on the standardization of service more is planned for 2013, and three additional meetings for 2014.
robots: The latest events focused on the proper classification of medical
T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223 1221

robots, and the definitions of automation regarding automated 8. Roadmap and future work
function vs. autonomy, considering the ALFUS concept [11].
The group works on the identification of additional risks and The global market for robots will increase more than 10% per
hazards, and describes the various levels of DoA. Medical robots year through 2016 to $20 billion. The US market is expected to
standardization guidelines (defined by the ECs Medical Device grow at 15% per year while the Chinese market will grow 17% per
Directive in Europe) need basic safety and essential performance year. The service robot market started becoming more significant
requirementsyet to be formulated. in the first decade of the 21st century. The IEEE ontology WG will
The Robots and Robotic Equipment (UL 1740) [57] is under devel- focus on developing ontologies and standards that will lower the
R&D and product costs while increasing sophistication of service
opment by the Underwriters Laboratory. This standard addresses
robot technology and software.
issues for both professional and service robots, intended for indoor
Based on the existing standards and the ontologies under de-
and outdoor applications. The robotic equipment can be used for
velopment, within and/or outside the ORA WG, it is expected to
assembly, parts transfer, automated material handling, inspection,
roll out an IEEE standard for service robotics by 2015. Following a
clean room applications and beyond.
hopeful consensus within the working group members, in the next
Relying on these standards, service robot technologies should phase, the Service Robots sub-group chairs will invite well-known
be developed to comply and assure safety in a humanrobot co- experts of the field, to review each step of the process and have a
worker scenario. Nevertheless, safety should be both assured to suitable and consensual ontology for Service Robotics. After each
humans, robots and the environment itself. round of experts review, the ontology will hopefully be improved,
and also will serve to update other overlapping sub-group ontolo-
gies, e.g., industrial and core ontologies.
7. The IEEE RAS efforts for standardization
The core ontology and the sub-domain ontologies will be used
by robotic systems developers to represent specific components
The IEEE RAS Ontologies for Robotics and Automation Working knowledge, and seamlessly integrate and/or remove components
Group (ORA WG) was created in 2011 with the aim to link exist- from the system. Also, robotic systems end-users for some specific
ing ISO, IEC, etc. standards and the current research efforts towards application domain, will then use the specific ontology, e.g., service
new regulatory frameworks [58]. The goal is to develop a standard or industrial, to represent specific knowledge of the environment.
for knowledge representation and reasoning in robotics and au-
tomation, together with the representation of concepts of an initial 9. Conclusion
set of application domains [6].
The ORA WG is comprised of over 50 members, representing It is well understood that further development and deployment
over twenty countries, a cross-section of industry, academia and of service robots require standardization, not only regarding
government. The group has four years to prepare a document hardware and interface development, but also in terms of safety,
that will be evaluated by invited experts, organizations and the liability and quality.
IEEESA Standards Board. There are four sub-groups, of which one Ontologies allow the description of the robots world, tasks and
is dedicated to service robotics (SeR). services through precise and unambiguous vocabularies. Robotics
The SeR group is developing ontologies, by defining service research must be standardized to facilitate dissemination of the
robotic concepts and the relationship among concepts such that results. Robotics research needs to be clear, precise and easy to use.
humans and robots can interact to perform tasks in any environ- There is an urgent need for knowledge representation standards
ment. These ontologies will be used for reasoning by robots to for robots and humans.
perform tasks. When finished, they will be merged with the ontolo- In this paper, we presented how standards and ontologies can
gies developed by other ORA sub-groups and serve as a common better describe our complex world for robotic applications. Ontol-
reference for the R&A community. The final ontology will ensure a ogy can be used to produce well-defined definitions without am-
common understanding among members of the community and biguities, and eventually may contribute to the development of
facilitate more efficient integration and data transfer. This is guar- standards. A general framework of knowledge representation for
anteed through the ontology engineering process used that re- robotic reasoning is defined in the scope of the ORA WG. Ontolo-
gies for robotic intelligence and knowledge should be defined as
moves ambiguities and incompatible relationships. ORA WG uses
a standard. The group undertakes tasks such as merging/aligning
two methods called METHONTOLOGY [59] and OntoClean [60].
partial ontologies, evaluating distributed ontologies, and extract-
METHONTOLOGY is for building ontologies either from scratch, by
ing concepts from high-level process descriptions.
reusing other ontologies, or by re-engineering existing ones. It has
Ontologies should define the general knowledge used by service
a checklist that allows for the identification of inconsistencies, in-
robots, and knowledge for specific applications. In the meanwhile,
completeness and redundancy. On the other hand, OntoClean helps
ontologies are not only for professionals who use robots at work,
to validate the ontology according to ontological adequacy of tax- but also for the general public, willing to use service robots for
onomic relationships. It is based on general ontological notions domestic tasks.
drawn from philosophy, like essence, identity, and unity.
Furthermore, the ORA ontology is described through mathe- Acknowledgments
matical formalism where each concept and relationship is well-
grounded mathematically, using logics. In this case, the group uses The IEEE-SC WG is supported by the IEEE Robotics and Au-
description logics expressed in OWL. OWL is used due to popularity tomation Society. This work is partially supported by the Strategic
in the ontology development community, and due to the number of Project, PEst-OE/EME/LA0022/2011, through FCT (under IDMEC-
tools and reasoning engines available. Moreover, it provides addi- IST, research group: IDMEC/LAETA/CSI), FCT project PTDC/EME-
tional vocabulary, and facilitates greater machine interoperability CRO/099333/2008 and EU-FP7-ICT-231143 project ECHORD. The
of web contents. In this sense, the ORA ontology will be available to authors acknowledge the EuRoSurge FP7-ICT-2011-7-288233
the R&A community to be used as a knowledge source for various project, the support of Brazilian CNPq and the Hungarian Etvs
applications. Scholarship.
1222 T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223

References [33] P.A. Valiente-Rocha, A. Lozano-Tello, Ontology-based expert system for


home automation controlling, in: Proc. of the 23rd Intl. Conf. on Industrial
Engineering and Other Applications of Applied Intelligent Systems Part I,
[1] M.K. Habib, Service robots and humanitarian demining, in: A. Lazinica
IEA/AIE10, Springer-Verlag, Berlin, Heidelberg, 2010, pp. 661670.
(Ed.), Mobile Robotics: Toward New Applications, ARS-pro literature Verlag
Publishers, 2012, pp. 449480. [34] R. Chatterjee, F. Matsuno, Robot description ontology and disaster scene
[2] M.K. Habib, Real time mapping and dynamic navigation for mobile robots, description ontology: analysis of necessity and scope in rescue infrastructure
International Journal of Advanced Robotic Systems 4 (2007) 323338. context, Advanced Robotics 19 (2005) 839859.
[3] M.K. Habib, Y. Baudoin, Rescue and hazardous intervention robotics, [35] C. Schlenoff, E. Messina, A robot ontology for urban search and rescue, in:
International Journal of Industrial Robot 39 (2012) 423427. Proc. of the 2005 ACM Workshop on Research in Knowledge Representation
[4] J.L. Burke, R.R. Murphy, E. Rogers, V.J. Lumelsky, J. Scholtz, Final report for for Autonomous Systems, pp. 2734.
the DARPA/NSF interdisciplinary study on humanrobot interaction, IEEE [36] M. Compton, C. Henson, H. Neuhaus, L. Lefort, A. Sheth, A survey of the
Transactions on Systems, Man, and Cybernetics 34 (2004) 103112. semantic specification of sensors, in: Proc. of the 2nd Intl. Workshop on
[5] S. Thrun, Toward a framework for humanrobot interaction, Human Semantic Sensor Networks, at 8th Intl. Semantic Web Conference, 2009,
Computer Interaction 19 (2004) 924. pp. 2734.
[6] C. Schlenoff, E. Prestes, R. Madhavan, P.J.S. Gonalves, H. Li, S. Balakirsky, [37] M. Compton, H. Neuhaus, K. Taylor, K.N. Tran, Reasoning about sensors and
T. Kramer, E. Miguelanez, An IEEE standard ontology for robotics and compositions, in: Intl. Workshop on Semantic Sensor Networks, pp. 3348.
automation, in: Proc. of the 2012 IEEE/RSJ Intl. Conf. on Intelligent Robots and [38] A. Preece, M. Gomez, G. de Mel, W. Vasconcelos, D. Sleeman, S. Colley, G.
Systems, Vilamoura, pp. 13371342. Pearson, T. Pham, T. La Porta, Matching sensors to missions using a knowledge-
[7] Intl. Federation of Robotics (IFR), service robot, provisional definition of service based approach, in: SPIE Defense and Security Symposium, International
robots, visited April 27, 2012. http://www.ifr.org/service-robots/. Society for Optics and Photonics, 2008, 698109.
[8] Y. Ota, Safety Standardization Activities on Service Robots, ItalyJapan Sympo-
[39] S. Lemaignan, M. Ros, R., R.L. Alami, M. Beetz, Oro, a knowledge management
sium, Tokyo, 2007. http://www.robocasa.net/workshop/2007/pdf/toyota.pdf.
module for cognitive architectures in robotics, in: Proc. of the 2010 IEEE/RSJ
[9] ISO 8373:2012, Robots and Robotic DevicesVocabulary, 2012. Intl. Conf. on Intelligent Robots and Systems.
http://www.iso.org.
[40] M. Tenorth, M. Beetz, KnowRob knowledge processing for autonomous
[10] H.M. Huang, J. Albus, E. Messina, Toward a generic model for autonomy levels
personal robots, in: Proc. of the 2009 IEEE/RSJ Intl. Conf. on Intelligent Robots
for unmanned systems (ALFUS), in: Proc. of the 2003 Performance Metrics for
and Systems, pp. 42614266.
Intelligent Systems (PerMIS) Workshop, pp. 713.
[11] H.M. Huang, K. Pavek, B. Novak, J. Albus, E. Messina, A framework for autonomy [41] Robot operating system, visited April 27, 2013. http://www.ros.org.
levels for unmanned systems (ALFUS), in: Proc. of the 2005 AUVSIs Unmanned [42] K. Varadarajan, M. Vincze, Affordance based part recognition for grasping and
Systems, North America Symposium, Baltimore, MD, pp. 19. manipulation, in: Proc. of the ICRA Workshop on Autonomous Grasping, 2011.
[12] G.S. Virk, T. Haidegger, Classification guidelines for personal care [43] A. Juarez, C. Bartneck, L. Feijs, Using semantic web technologies to
robotsmedical and non-medical applications, in: Proc. of the 2012 IEEE describe robotic embodiments, in: Proc. of the 6th ACM/IEEE Intl. Conf. on
IROS Workshop on Safety in HumanRobot Coexistence & Interaction, HumanRobot Interaction, 2011, pp. 425432.
pp. 3336. [44] E. Vassev, M. Hinchey, Towards a formal language for knowledge representa-
[13] L. Dignan, Google aims to bridge android, cloud computing with robotics, tion in autonomic service-component ensembles, in: Proc. of the 3rd Intl. Conf.
ZD net May (2011). www.zdnet.com/blog/btl/google-aims-to-bridge-android- on Data Mining and Intelligent Information Technology Applications, 2011,
cloud-computing-with-robotics/48693. pp. 228235.
[14] D. Misener, Why robot brains could be the killer app of cloud computing, The
[45] ASCENS, visited April 27, 2012. http://www.ascens-ist.eu/home.
Globe and Mail (2012).
[15] T. Gruber, Toward principles for the design of ontologies used for knowledge [46] J. Bermejo-Alonso, R. Sanz, M. Rodriguez, C. Hernandez, An ontological
sharing, Journal of HumanComputer Studies 43 (1995) 907928. framework for autonomous systems modelling, International Journal on
[16] Military exoskeletons uncovered: Ironman suits a concrete possibility, 2013. Advances in Intelligent Systems 3 (2010) 211225.
http://www.army-technology.com/features/featuremilitary-exoskeletons- [47] J. Bateman, S. Farrar, Modelling Models of Robot Navigation Using Formal
uncovered-ironman-suits-a-concrete-possibility. Spatial Ontology, in: Proc. of Spatial Cognition Conference, Lecture Notes in
[17] B. Chandrasekaran, J.R. Josephson, V.R. Benjamins, What are ontologies, and Computer Science, Vol. 3343, 2005, pp. 366389.
why do we need them? IEEE Intelligent Systems 14 (1999) 2026. [48] H. Chen, T. Finin, A. Joshi, The SOUPA ontology for pervasive computing,
[18] M. Ashburner, C. Ball, J. Blake, D. Botstein, H. Butler, J. Cherry, A. Davis, K. in: Ontologies for Agents: Theory and Experiences, BirkHauser, 2005,
Dolinski, S. Dwight, J. Eppig, Gene ontology: tool for the unification of biology, pp. 366389.
Nature Genetics 25 (2000) 2529. [49] H. Chen, T. Finin, A. Joshi, An ontology for context-aware pervasive computing
[19] S. Staab, A. Maedche, Knowledge portalsontologies at work, Al Magazine 22 environments, Journal on Ontologies for Distributed Systems, Knowledge
(2001) 6375. Engineering Review 18 (2003) 197207.
[20] P. Jannin, X. Morandi, Surgical models for computer-assisted neurosurgery, [50] DAML, visited April 27, 2012. http://www.daml.org/.
Neuroimage 37 (2007) 783791.
[51] M.A. Goodrich, A.C. Schultz, Humanrobot interaction: a survey, Foundations
[21] D. Neumuth, F. Loebe, H. Herre, T. Neumuth, Modeling surgical processes: a and Trends in HumanComputer Interaction 1 (2007) 203275.
four-level translational approach, Artificial Intelligence in Medicine 51 (2011)
[52] Roomba, visited April 27, 2012. http://www.irobot.com/en/us/robots/home/
147160.
[22] P. Giaretta, Ontologies and knowledge bases towards a terminological roomba.aspx.
clarification, Towards Very Large Knowledge Bases: Knowledge Building & [53] DaVinci, visited April 27, 2012. http://www.intuitivesurgical.com/products/
Knowledge Sharing (1995) 25. davinci_surgical_system/.
[23] W. Pidcock, What are the differences between a vocabulary, a taxonomy, a [54] S. Kock, Industrial robotics trends and opportunities in a changing world,
thesaurus, an ontology, and a meta-model?, Web Page URL: ROBIO08 Plenary Talk, 2008.
www.metamodel.com/article.php (2004). http://www.robotics.it-chiba.ac.jp/wang/ROBIO2008/pdfs/lenaryTalk-
[24] Y. Kalfoglou, M. Schorlemmer, Ontology mapping: the state of the art, The Dr_Kock.pdf.
Knowledge Engineering Review 18 (2003) 131. [55] Service robot statistics, visited April 27, 2013.
[25] J. de Bruijn, F. Martin-Recuerda, D. Manov, M. Ehrig, State-of-the-art-survey www.ifr.org/service-robots/statistics.
on ontology merging and aligning, SEKT Project deliverable D 4 (2004) 211.
[56] G.H. Lim, I.H. Suh, H. Suh, Ontology-based unified robot knowledge for service
[26] M. Li, X.Y. Du, S. Wang, Learning ontology from relational database, in:
robots in indoor environments, IEEE Transactions on Systems, Man and
Proc. of the 2005 Intl. Conf. on Machine Learning and Cybernetics, vol. 6,
Cybernetics, Part A: Systems and Humans 41 (2011) 502509.
pp. 34103415.
[57] UL 1740, Robots and Robotic Equipment, Underwriters Laboratory, 2012.
[27] I. Serra, R. Girardi, A process for extracting non-taxonomic relationships of
http://www.ul.com/global/eng/pages/offerings/industries/
ontologies from text, Intelligent Information Management 3 (2011) 119124.
powerandcontrols/robots/.
[28] N. Guarino, Understanding, building and using ontologies, International
Journal of HumanComputer Studies 46 (1997) 293310. [58] C. Schlenoff, E. Prestes, R. Madhavan, P.J.S. Goncalves, H. Li, S. Balakirsky,
[29] N.F. Noy, D.L. McGuinness, et al., Ontology development 101: a guide to T. Kramer, E. Miguelanez, An ieee standard ontology for robotics and
creating your first ontology, 2001. automation, in: A. Chibani, G.P. Zarri, Y. Amirat (Eds.), Bridges between the
[30] G. Flouris, D. Manakanatas, H. Kondylakis, D. Plexousakis, G. Antoniou, Methodological and Practical Work of the Robotics and Cognitive Systems
Ontology change: classification and survey, The Knowledge Engineering CommunitiesFrom Sensors to Concepts, in: Intelligent Systems Reference
Review 23 (2008) 117152. Library, SpringerVerlag Berlin Heidelberg, 2012, pp. 107113 (in press).
[31] M.A. Musen, Modern architectures for intelligent systems: reusable ontologies [59] M. Fernndez, A. Gmez-Prez, N. Juristo, METHONTOLOGY: from ontological
and problem-solving methods, in: 1998 AMIA Annual Symposium, pp. 4652. art towards ontological engineering, in: Ontological Engineering, AAAI Spring
[32] S. Klai, M.T. Khadir, Data based ontology construction coupled to expert system Symposium, vol. 6, pp. 3340.
for steam turbine aided diagnostic, in: Proc. of the 3rd Intl. Conf. on Innovation [60] N. Guarino, C.A. Welty, An overview of OntoClean, in: S. Staab, R. Studer (Eds.),
and Information and Communication Technology, ISIICT09, British Computer Handbook on Ontologies, Intl. Handbooks on Information Systems, Springer
Society, Swindon, UK, 2009, pp. 1616. Berlin Heidelberg, 2009, pp. 201220.
T. Haidegger et al. / Robotics and Autonomous Systems 61 (2013) 12151223 1223

Tams Haidegger received his M.Sc. degrees from the Sampath Kumar Veera Ragavan is currently a Senior Lec-
Budapest University of Technology and Economics in Elec- turer in Mechatronics Engineering at Monash University,
trical and Biomedical Engineering in 2006 and 2008, Sunway Campus. Before joining Monash, he worked for
respectively. His Ph.D. (2011) was based on a neurosurgi- several multinational companies in various capacities. He
cal robot he developed when he was a visiting scholar at has more than 17 years of industrial experience in the de-
the Johns Hopkins University. Currently, he is an Adjunct sign and development of factory automation systems, fluid
Professor at the buda University, serving as the deputy power automation, software for consumer electronics and
director of the newly founded ABC Center for Intelligent industrial automation and has executed several projects
Robotics. Also he is a Research Area Manager at the Aus- from concept to commissioning. He is also a Chartered En-
trian Center of Medical Innovation and Technology and gineer from the Engineering council (UK). He serves as a
running a spinoff company focusing on objective medical consultant in the field of Mechatronics, Telematics and In-
hand hygiene control. Tams is an active member of various professional organiza- dustrial Automation.
tions, including IEEE RAS, EMBC, SPIE and MICCAI. He is a national delegate to an
ISO/IEC standardization committee focusing on safety for medical robots. Howard Li is an Associate Professor in the Department of
Electrical and Computer Engineering, University of New
Brunswick, Canada. He is a registered Professional En-
Marcos E. Barreto received a Ph.D in Computer Science gineer in the Province of Ontario, Canada. He is a se-
from the Federal University of Rio Grande do Sul (UFRGS), nior member of IEEE. He received the Ph.D. degree from
Brazil, 2010. From 1995 to 2010, he did work at the Par- the University of Waterloo, Canada. He worked with At-
allel and Distributed Processing Group (GPPD), at UFRGS, lantis Systems International, Defence Research and Devel-
conducting research in cluster and grid computing, parallel opment Canada, and Applied AI Systems Inc. to develop
programming libraries and performance evaluation. From unmanned ground vehicles, unmanned aerial vehicles, au-
1998 to 2010, he taught several courses on distributed sys- tonomous underwater vehicles and mobile robots for both
tems, parallel programming, databases and Web develop- domestic and military applications. His research interests
ment in several universities and colleges in the south of include linear control, nonlinear control, intelligent control, distributed control,
Brazil. He was also a Project Manager at HP Brazil, from unmanned vehicles, mechatronics, robotics, multi-agent systems, artificial intelli-
2000 to 2002. Since 2010, he has been an Associate Profes- gence, motion planning, and simultaneous localization and mapping.
sor at the Federal University of Bahia (UFBA) and a researcher at the Distributed
Systems Laboratory (LaSiD), leading and conducting research in cloud comput-
Alberto Vaccarella graduated cum laude in Biomedical
ing and dependability applied to e-science support, parallel programming and
Engineering at the Politecnico di Milano in December 2009
robotics.
with the master thesis: Development and validation of an
integrated sensors and actuators system for neurosurgical
Paulo J.S. Gonalves was born in was born in Covilh, Por- procedures. He is currently a Ph.D. candidate at NearLab,
tugal, in 1972. He received the M.Sc. and Ph.D. degrees in Bioengineering Department of Politecnico di Milano,
Mechanical Engineering from the Technical University of where he is dealing with the design and implementation
Lisbon, Portugal, in 1998 and 2005, respectively. He is cur- of software applications for CAS. Since January 2010 he has
rently an Associate Professor with the Electrotechnical and collaborated at the open source project Image- Guided
Industrial Engineering Department, School of Technology, Surgery ToolKit (IGSTK). He is involved in two European
Polytechnic Institute of Castelo Branco, Portugal. He is projects: ACTIVE (FP7-ICT-270460) and EuRoSurge (FP7-
also a Researcher at Center of Intelligent Systems (CSI) - ICT-288233). His scientific interests cover sensors architecture, surgical navigation
IDMEC - LAETA. He has authored or coauthored more systems and medical robotics.
than 70 papers and articles published in journals and con-
ference proceedings. His main research interests include Roberta Perrone (Milan, 1988) lives and studies in Milan.
robotics, computer vision, and in particular visual servoing, chiefly as it applies to She holds a biomedical engineering master degree in elec-
healthcare. His research interests also include image processing, and pattern recog- tronics biotechnologies from Politecnico di Milano in 2013
nition. with the thesis on Ontological modeling of Neurosurgery:
application to automatic classification of temporal and ex-
tratemporal lobe epilepsies. She is working on the Euro-
Maki K. Habib (IEEE Senior member) is a Doctor of En- pean project EuRoSurge, where she is implementing the
gineering Science (DES) in Intelligent and Autonomous ontological part for the scenario that includes the reach-
Robots (University of Tsukuba, Japan 1990). He did work ing of a certain point by an industrial robot observed by
at different places, such as, RIKEN (Japan), RISO Labo- two trackers.
ratories (Japan), EPFL (Switzerland), GMD-Japan (Japan),
Monash University (AustraliaMalaysia), KAIST (South
Korea), Saga University (Japan). He is currently a Full Edson Prestes received his B.Sc. degree in Computer Sci-
Professor of Robotics and Mechatronics at the Ameri- ence (CS) from the Federal University of Par (UFPa), Brazil,
can University in Cairo. In addition, he is the Director of in 1996 and M.Sc. and Ph.D. in CS from the Federal Univer-
Robotics, Control and Smart Systems master/Ph.D. pro- sity of Rio Grande do Sul (UFRGS), Brazil, in 1999 and 2003,
grams. He did and continues to work as a Consultant and respectively. Currently, he is Associate Professor and Head
Technical Advisor in the field of Mechatronics, Robotics and Industrial Automation of the Theoretical Informatics Department at Informatics
with highly recognized organizations and companies. Prof. Habib has edited seven Institute (UFRGS), Vice Chair of working group funded by
books and published more than 210 papers published in internationally recognized IEEE RAS entitled Standard for Ontologies for Robotics and
Books, Journals and Conferences. His main area of research focuses on intelligent Automation (IEEE RAS WG ORA), and a Member of IEEE
and service robots, robots and sensors for risky Environments and humanitarian RAS Special Interest Group on Humanitarian Technology
demining, human adaptive and friendly mechatronics, telecooperation, distributed (IEEE RAS SIGHT; member of IEEE, IEEE RAS (Robotics and
teleoperation and collaborative control, intelligent sensors systems, wireless sen- Automation Society) and IEEE SA (Standards Association). His interest fields are mo-
sor networks and ambient intelligence, biomechanics and biomimetic robots, intel- bile robots (SLAM, path planning, integrated exploration, and so on), ontologies and
ligent control, robotics and industrial automation. R&A for humanitarian applications.

Das könnte Ihnen auch gefallen