Transcript Applied ontologies and standards for service robots

FACUTALY OF SCINCES
Reasoning in ontology
By
MUSHRIK LUAY TAHA
Supervision
Assistant professor Dr. Yasser Fouad
WHAT IS ONTOLOGY ?
in the philosophical aspect A branch of metaphysics,
which focuses on patterns of existence
and assets and research entity in
the various entities, objects and their types
and their structure and properties,
and relied on the description and analysis.
This idea of the grease philosophical
Aristotle ( 322 - 384 BC )
What is ontology in Artificial
intelligence ?
• Formal description of a specific
domain by defining concepts and its
attribute and the relations among
that’s concepts
• A description of the knowledge in a
particular field and divided into
categories represent concepts and
each concept of the characteristics
and attributes.
Applied
ontologies and
standards for
service robots
Introduction
• Between the 1960s and the ’90s, most robots –
and robotics in general – were related to
industrial applications.
• robotic helpers are able to perform a wide range
of tasks, such as healthcare and personal
assistance.
• These kinds of robots are commonly called
service robots. Service robots typically share the
human environment and exhibit basic intelligent
behavior to accomplish assigned tasks.
Evolution of the fields of robotics
• the evolution of the different fields of robotics,
implying an increasing Degree of Autonomy
(DoA) and system complexity along with humancentered applications.
Ontologies in robotic systems
• Ontologies allow a clear dialog between all
stakeholders involved in the life-cycle of a
robotic system, and enable the integration and
efficient communication of heterogeneous
robotic systems.
• Developing an ontology or a standard is not an
easy task, especially in a huge field.
Definitions
• Definition . A service robot is a
robot that performs useful tasks
for humans or equipment,
excluding industrial automation
applications.
• Definition . A service robot is a
robot which operates semi or fully
autonomously
to
perform
services useful to the well-being
of humans and equipment,
excluding
manufacturing
operations
Definitions
• Definition . Medical robots are defined as medical
electrical equipment and systems which are used ‘‘to
diagnose, treat or rehabilitate’’ patients from medical
conditions.
• Definition . Non-medical personal care robots are
defined as machines for performing ‘‘aiding actions’’, and
actions contributing directly towards improvement in the
quality of life of humans, except medical applications
Categories of medical/non-medical
personal service robots.
service robots can be broken down
into three classes
• Class 1 robots that replace humans at work in
dirty, hazardous environment and tedious
operations.
• Class 2 robots that operate closely with humans
to alleviate incommodity or to increase comfort,
such as entertainment.
• Class 3 robots that operate on humans, e.g.,
medical robots for surgery, diagnosis, and
treatment.
Ontologies
• Ontologies can be viewed as content theories that focus
on properties and relationships among objects from a
specific domain .
• They act as a ‘‘body of knowledge’’ that is based on a
vocabulary used to describe the domain.
• Definition . Ontology is a tuple ⟨S, A⟩, where S is the
vocabulary (or signature) of the ontology and A is the set
of ontological axioms specifying the intended domain
vocabulary.
Models vs. ontologies
robotic ontology
• For robot design and development, ontologies
have been using OWL to describe robots including
their structural and operation capabilities.
• Robot Data Base [43] is a database that uses
semantic Web technologies to store information
about robot embodiments (physical components).
Based on this information, ontologies focused on
the physical characteristics of robots can be
created and populated
Human factor for ontology
development
• When building ontologies, the end-users must also be
considered. Today, Human–Robot Interaction (HRI) [51] is
a major concern for robotics researchers and
practitioners, due to the increasing need of robots
helping and cooperating with humans.
• the developer, a core ontology must be used to
represent specific knowledge (definition of the robot,
type of robots to use, components to use, and so on).
• Second, for the end-user, application domain ontology
needs to be defined for the robot’s work environment
The components of a robotic
system with core and domain
ontologies
Current robotics standards
• International Organization for Standardization: ISO
8373:2012
• American Society for Testing and Materials (ASTM):
ASTM E2521
• American Institute of Aeronautics and Astronautics
(AIAA): S-066- 1995
• Japanese Industrial Standards (JIS): JIS B 0144:2000
Roadmap and future work
• The IEEE ontology WG will focus on developing
ontologies and standards that will lower the
product costs while increasing sophistication of
service robot technology and software.
• Robotic systems end-users for some specific
application domain, will then use the specific
ontology, e.G., Service or industrial, to represent
specific knowledge of the environment.
Dynamic
Evaluation
of
Ontologies
Abstract
• The automatic construction of ontologies from texts is a
topic of continued and open research, their construction
requires both a study of human knowledge,
methodologies and tools to retrieve the text content.
• The field of ontology construction needs to go towards
more dynamic, more view of ontologies is to increase
intelligence in many applications such as information
retrieval, semantic indexing and semantic annotation.
• This paper propose an approach that improve the
process of automatic co-construction of ontologies from
a corporus.
Introduction
• Ontologies appeared as a key for automatic
handling of information at the semantic level
which makes them a central component in many
applications, and are called to play a crucial role
in the Semantic Web which is the next evolution
of the Web.
• It is a set of technologies designed to make the
content resources of the World Wide Web
accessible and useable by programs and
software agents
Methods for ontology
construction
• The ontological engineering is a subdomain of
knowledge engineering that studies the process
of ontology development.
• an ontology defines the terms and the relations
of the basic vocabulary of a domain and the
rules that show how combine terms and
relations so as to extend the vocabulary.
Methods for ontology building
from scratch
• They are based on the extraction of common
knowledge manually into the different sources,
then they use techniques of natural language
processing (NLP) and acquisition of knowledge to
generate new knowledge.
Methods for cooperative
construction of ontologies
• Ontology must be a consensus and be accepted
by its user community. These methods therefore
adopt a collaborative approach for construction
including the intervention of persons located in
different places
Methods for re-engineering of
ontologies
• Re-engineering of ontologies is the process of
rebuilding ontologies and linking a conceptual
model of an ontology already implemented in
another being implemented.
Methods of learning ontologies
• They consist in improving the construction of
ontological components by introducing plug-ins
in the process of ontology development, these
plug-ins can be text and knowledge bases
Proposed approach : Dynamic
Evaluation of Ontologies (DEO)
• Propose an approach based on cooperative
learning (system-users) for building ontologies
dynamically DEO (dynamic evaluation of
ontologies). This system is based on a set of
agents that cooperate and interact to evaluate
the ontology.
The DEO System integrates four
main processes:
1) Extraction process of candidate terms
• We done an analysis of terminology using an NLP tool
to identify indices of language knowledge (terms, lexical
relations, semantic classes, etc.).
• Like NLP tool, we use TreeTager 12 to tag texts. Next,
we used WN 1.2( Wordnet 1.2) 13 terminology to
identify concepts and semantic relationships.
The DEO System integrates four
main processes:
2) Process of identifying concepts T
•
the concept is necessarily unique. In this phase WSDA
agent (Agent Word Sense Disambiguation) is created to
disambiguate the sense of an ambiguous term.
Ambiguity is the property of a word or a sequence of
words to have several senses
The DEO System integrates four
main processes:
3) Hierarchization process of concepts
• The agent SSA send the message (SSAi), by this he asked the
agent LA to send their brothers and their father.
• Then, the agent LA responds by message Answer (S S Ai, Cj,
Ci, CPw) where Cj is the brother of Ci and CPw is the nearest
common father in the hierarchy of WordNet for Ci and Cj.
• Next, the agent SSA saves tuples (Cj, Ci, CPw) in a file .Temp
and agent LA specifies the semantic relationship between
these concepts and so on for other concepts.
• Similarly, ontologists may be involved in this step to improve
and remove tuples and / or relationships.
The DEO System integrates four
main processes:
4) Process of creation of the formal ontology
• In this phase, the ontology must be saved in a formal
format (OWL file, XML, RDFS), in using platforms
KAON217 or editors like protg
The DEO System
Evaluation and Analysis
• we have chosen, in the context of
experimentation, datasets of different sizes in
the field of security-cryptography from the
corpus 20 news groups.
A part of the ontology.
Future work
• test DEO prototype In a dynamic environment
(dynamic corpus) Where the collections of
documents can be changed;
Organization Of The Repair And
Maintenance In Road Sector With
Ontologies And Multi-agent
Systems
Abstract
• This paper defines the problems and tasks of maintenance
and repair of road-building equipment (DST). An approach
that provides decision support (PDS) in the management of
maintenance and repair of roads and vehicles based on the
adapted and modernized method RCM2 is proposed.
• The structure of the system of organization of repairs and
maintenance in the road sector with the use of ontologies and
multi-agent systems is defined.
• One of the objectives of the reformation is to improve
management technologies, which include the best of modern
approaches for the organization of maintenance and repair of
equipment (MRO) to ensure maximum performance of the
road-repair vehicle production program.
Introduction
• Road management is very important for any
country
• Modernization of road infrastructure allows us to
migrate to a new system of management, which
is results-oriented and improves the efficiency of
the industry.
Decision on the organization of
MRO
• an approach that involves a more progressive
planning, preventive maintenance and repair, in
which repairs of equipment are made at regular
intervals is used.
Information Decision Support
System
• an approach to support decision making (BPD) in the
management of road maintenance and repair techniques
is proposed.
• An Information Decision Support System (IDSS) is
designed to perform the following functions: planning
maintenance and repair of road vehicles, prioritization of
road maintenance and repair of machines
• Modules offered by IDSS knowledge base (ontology)
include the database system and the functional modules
Information Decision Support
System
• The system database and ontology are regularly updated.
Ontology is regularly updated with new knowledge of
professionals who are involved into the process control
system maintenance and repair of road vehicles.
• Ontology was developed to improve the efficiency of
decision making and solving the problem of road building
organization in the process of organizing MRO. Ontology
provides a complete knowledge of the process of doing a
repair of road vehicles
• How???? Developing ontology solves the problem of
sharing and reuse of knowledge by different users
involved in the management of the quality of the repair
of road vehicles
Information Decision Support
System
Fragment of ontology
Multi-agent model of MRO system
for road construction machines
• For monitoring, organizing an interactive and simplified
process of managing objects with similar structures
makes use of agents, which are grouped in multi-agent
systems (MAS).
Multi-agent model of MRO system
for road construction machines
1) the agent manager machine analyzes the failures of
internal nodes and determines the priority of repairing
them, and
2) these tasks are distributed among the agents executing,
and
3) each agent, the Executive complete their task,
sometimes dividing it into sub-tasks, and
4) to provide an overall result, composition and
integration of partial results corresponding to the
selected tasks is performed, and
5) the agent-manager machine is used to determine the
priority for repair of road-building machinery
Multi-agent model of MRO system
for road construction machines
Conclusion
• Ontology is A description of the knowledge in a particular
field and divided into categories represent concepts and each
concept of the characteristics and attributes This concept
has been used in a wide applications in artificial intelligence
including intelligent robotics and software and the Semantic
Web Application of the concept ontology in robotics to help
rather than models to achieve the biggest benefit and the
possibility of learning of the machine (robot) Or the creation
of intelligent systems include intelligent agents designed to
draw new knowledge And their use in important applications
in the daily life of the Iron priority operations and decision
support .
THANKS