|
|
Postgraduate Research Topics
| Title |
Multi-sensor Data Fusion for Cooperative Indoor Target Localization |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc/PhD |
| Description |
Multi-sensor data fusion is the theory, techniques and tools, which are used for combining sensor data, or data derived from sensory data, into a common representational format to improve the quality of the information, so that it is, in some sense, better than would be possible if the data sources were used individually. In this context, better may mean improved system performance, improved robustness, extended spatial and temporal coverage, shorter response time or reduced communication and computing. The objective of this project will be to design of fusion algorithm to fuse data coming from heterogeneous sources such as video and IR cameras and wireless radio-based devices to solve the problem of indoor target localization. Fusing radio, visual and audio can increase the effective instantaneous visibility of the environment, reduce the uncertainty, improve the accuracy and increase sensing reliability allowing for more accurate and reliable localization. |
| Qualification Requirements |
BSc with backgroup in probability theory |
| Title |
Sensor Management for Optimal
Static Sensor Placement and Mobile Sensor Deployment |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc/PhD |
| Description |
Sensor Management refers to
the process that plans and controls the use of a set of
sensors in a manner that synergistically improves the process
of data fusion and ultimately enhances perception and understanding.
Managing scarce and heterogeneous sensors involves making
decisions and compromises regarding alternate sensing strategies
under time and resource availability constraints. A special
case is when the surveillance system combines multiple static
and mobile sensors in order to increase their capability.
Strategic Sensor Placement (SSP) and Mobile Sensor Deployment
(MSD) become then two essential SM problems that must be
tackled. SSP addresses how to optimally place minimum number
of static sensors to maximize the coverage of the volume
of interest (VOI). MSD aims at maximizing the coverage and
minimizing the energy consumption in communication and mobilization
of a set of the mobile sensors. This project will study
how a set of static and mobile sensors can be properly managed
to cooperatively and adaptively monitor a VOI. |
| Qualification Requirements |
BSc with backgroup in multiagent systems and
optimization theory |
| Title |
Pervasive Multimodal Surveillance
System for In-Home Care |
| Focus Group |
Service Robotics Focus Group |
| Degree Type |
MSc/PhD |
| Description |
In-house pervasive multimodal
surveillance system can be used to detect and track human
activities. This automatic monitoring and recognition of
human activities is an important component of building situational
awareness. In order to guarantee the safety and security
of elders and people with special needs, the surveillance
system must be able to autonomously and in real-time recognize
what activity the human is engaged in, and whether the activity
is proceeding normally, or whether an abnormal event is
taking place (such as a failure to accomplish the goal of
the activity). Many systems for activity recognition in
the home domain have been proposed in the literature. However,
the majority of the systems proposed to date either make
use of pre-defined or user specified activities to recognize,
or require an off-line training phase to learn the activities.
These approaches are fairly restrictive, as they are unable
to handle cases where new activities are introduced in the
environment following the initial specification/training
phase, or where the user’s performance of daily activities
changes over time, as is frequently the case with aging
users. The objective of this project is to design and develop
a pervasive multimodal surveillance system for implementing
situation awareness in a home setting. |
| Qualification Requirements |
BSc with backgroup in machine vision |
| Title |
Multisensor Cooperation in Surveillance Systems |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc/PhD |
| Description |
In multisensor systems, sensors
cooperation is directed toward improving situation awareness
through joint gathering and sharing of information. The objective
of this project is to implement cooperative behaviors such
as cueing and hand-off between multiple sensors for enabling
continuous tracking of objects. Cueing and Hand-off will be
implemented using a set of static and mobile sensors. The
cueing or slaving is the process of using the detections (i.e.,
contact-level cueing) or tracks (i.e., track-level cueing)
from a sensor A (master) to point another sensor B (slave)
towards the same target or event. The handoff occurs when
sensor A has cued sensor B for transferring the surveillance
or the fire-control responsibility from A to B. |
| Qualification Requirements |
BSc with backgroup in MATLAB and probability
theory |
| Title |
Automated
Visual Surveillance System for Outdoor and Indoor Public Security |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc/PhD |
| Description |
Effective monitoring
of persistent and transient objects and events is a key to
the effective protection of any Volume Of Interest (VOI).
Surveillance is defined as systematic observation of aerospace,
surface or subsurface areas, places, persons, or things, by
visual, aural, electronic, photographic, or other means. This
systematic observation includes the timely detection, localization,
recognition and identification of objects and events, their
relationships, activities, and plans, in a given VOI in order
to determine whether they are behaving normally or if there
is any deviation from their expected behavior. Growing demand
for high-level security and safety in commercial, law enforcement,
and military applications has led to active research to build
intelligent automated surveillance system. The objective of
this project is to design and implement a visual surveillance
system for outdoor and indoor security in public places. |
| Qualification Requirements |
BSc with backgroup in machine vision |
| Title |
Cooperative Sensor and Actor Network |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc |
| Description |
Cooperative sensor and actor network
(CSANET) that encompasses a set of heterogeneous sensing agents,
acting agents, situation awareness agents, resource management
agents and decision support/making agents. These agents are
endowed with know-how capability for solving problems in an
autonomous way and a know-how-to-cooperate capability by which
the agents can share common interests and interact with each
other. These spatially distributed agents, when properly managed,
can sense collaboratively and continuously a volume of interest
and physically manipulate and interact with it. This project
aims at designing and implementing a sensor and actor network
for distributed monitoring such as oil pipeline monitoring,
infrastructure health monitoring, boarder security monitoring,
traffic monitoring or air pollution monitoring. |
| Qualification Requirements |
BSc with backgroup in multiagent systems, web
programming, Java and XML. |
| Title |
Sensor Web for GUC PV Station |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc |
| Description |
The objective of this project is
to design and implement a sensor web for GUC PV station. This
system will allow researcher to access archive or live data
from sensors (cameras, temperature sensors, wind speed, humidity
sensors, etc.) from anywhere and at anytime. It can also provide
services to perform various levels of data processing and
models which describe how to create desired research products/algorithms
from available sensor data. The system will contain a catalog
server on which research products/algorithms such as data
classification/clustering are registered and are easily discoverable,
accessible, modifiable and extensible via web service chains.
|
| Qualification Requirements |
BSc with backgroup in multiagent systems, web
programming, Java and XML. |
| Title |
Communication Relay for Unmanned
Aerial Vehicles in Autonomous Search and Rescue Mission |
| Focus Group |
Cooperative Intelligent Systems Focus Group |
| Degree Type |
MSc/PhD |
| Description |
The objective of this project is
to simulate search and rescue scenarios where autonomous unmanned
aerial vehicles are deployed to locate multiple rescue targets.
When a target is found, the swarm of micro-aerial vehicles
(MAVs) self-organizes to utilize their range-limited communication
capabilities for setting up a communication relay network
between the target and the base. This solution is appropriate
for real-world situations where rescue targets are trapped
on intraversable terrain with a limited radius of communication. |
| Qualification Requirements |
BSc with backgroup in multiagent systems, Java
or NetLogo. |
| Title |
A Bio-Inspired Coordination Mechanism
for a Swarm of Miniature Robots |
| Focus Group |
Cooperative Intelligent Systems Focus Group |
| Degree Type |
MSc/PhD |
| Description |
Coordination addresses the interdependency
management among the cooperative or competitive entities of
the system in order to achieve their goals. The objective
of this project is to design a bio-inspired coordination algorithm
that can be used to manage a group of tiny robots in a search
and rescue mission. The algorithm will be based on understanding
the crowd dynamics during emergency evacuation. Simulating
this behavior can be useful in predicting occupants' distribution
and the whole evacuation process. This simulation model can
be applied in fire protection design, emergency preparation
and planning an evacuation strategy from public areas (such
as buildings, stadiums, subways, train stations, shopping
malls, airports, etc.). In this project, this model will be
applied to design the coordination mechanism of swarm robot
system. NetLogo will be used as a simulation environment for
complex systems. NetLogo
will be used as a simulation environment for complex systems. |
| Qualification Requirements |
BSc with backgroup in NetLogo. |
Current Postgraduate Researchers
| Title |
Cooperative Behaviors in
Multi-robot Systems |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc |
| Description |
Cooperative Multi-robot System (MRS)
is a group of robots that are designed aiming to perform
some collective behavior. The MRS is gaining great interest
because of the following reasons:
• Some tasks may be quite complex for a single robot to do
or even it might be impossible.
• Having small, simple robots will be easier and cheaper to
implement than having only one single powerful robot.
• Increasing the system reliability because having only one
robot may work as a bottleneck for the whole system
especially in critical times. But when having multiple
robots doing a task and one fails, others could still do the
job.
• Effective and robust cooperation among the robots can also
synergistically improve the performance of the system and
can endow it with higher-level faculties, such as group
formation, distributed search, dynamic task allocation,
communication relaying, cooperative target detection and
tracking and shared situation awareness.
The objective of this thesis is to study different
cooperative behaviors that can be developed in MRS. A formal
model of cooperation will be developed and its efficacy will
be demonstrated through the implementation of different
cooperative behaviors such as group formation, communication
relaying and cooperative target detection and tracking. |
| Researcher |
Eng. Marwa Shalaby |
| Title |
Self-organizing Group Formation in
Multi-robot Systems |
| Focus Group |
Cooperative Intelligent Systems Focus Group |
| Degree Type |
MSc |
| Description |
Self-organizing formation is an
emergent process of making whole forms by local interactions
of distributed simple autonomous robots without global
information at all and without depending on the initial
position and orientation of the robots. These simple
autonomous robots store local information and guiding rules
needed for the colonial self-organization. Additional
information is cooperatively generated as the organization
proceeds following external stimuli. The outcome is an
adaptable complex system that can perform many tasks, learn
and change itself accordingly. In this thesis,
self-organization will be investigated as a control paradigm
for massively distributed systems. Self-organizing group
formation algorithms will be designed and tested in
simulation environment as well as in a real test-bed that
encompasses a number of mobile robots with different
capabilities. |
| Researcher |
Eng. Ahmed Wagdy |
| Title |
Bayesian Approach to Multisensor
Data Fusion with Pre-and Post-Filtering |
| Focus Group |
Environment Perception Focus Group |
| Degree Type |
MSc |
| Description |
Data provided by sensors is always affected by some
level of uncertainty or lack of certainty in the measurements.
Combining data from several sources using multisensor data
fusion algorithms exploits the data redundancy to reduce this
uncertainty. This paper proposes an approach to multisensor
data fusion that relies on combining a modified Bayesian fusion
algorithm with Kalman filtering. Three different approaches
namely: Pre-Filtering, Post-Filtering and Pre-Post-Filtering are
described based on how filtering is applied to the sensor data, to
the fused data or both. A case study of estimating the position
of a mobile robot using optical encoder and Hall-effect sensor is
presented. Experimental study shows that combining fusion with
filtering improves least square error and variance of fused data
in both centralized and decentralized data fusion architectures. |
| Researcher |
Eng. Waleed A. Hafiz |
| Title |
Team-theortic Approach to Cooperative Multirobot Systems |
| Focus Group |
Cooperative Intelligent Systems Focus Group |
| Degree Type |
MSc |
| Description |
This thesis presents a team-theoretic approach to cooperative multirobot systems. The cooperative behaviors between heterogeneous agents are designed to assist in humanitarian demining mission. Belief-Desire-Intention model is used to control the actions of the agents and the relation between the team members are governed by the TeamLog formalism that determines the team member?s commitments to each other. The multi-agent system outputs a clustered mine map by using genetic algorithm for clustering. Then this clustered map is passed to a number of deminers to successfully remove the landmines. |
| Researcher |
Eng. Asser ElGindy |
|
External
Collaborators :
|
-
Dr. Abder Rezak Benaskeur, Defence
Scientist, DRDC Valcartier Canada
-
Dr. Hengameh Irandoust, Defence
Scientist, DRDC Valcartier Canada
-
Prof.
Dr. Miguel Angel Salichs, Director of RoboticsLab,
Department of Systems Engineering and Automation,
Carlos III University of Madrid, Spain
-
Prof.
Dr. Mohamed Kamel, Director of Pattern
Analysis and Machine Intelligence (PAMI) research
group, Department of Electrical and Computer Engineering,
University of Waterloo, Canada
-
Prof.
Dr. Fakhreddine Karray, Associate Director of
Pattern Analysis
and Machine Intelligence (PAMI) research group,
Department of Electrical and Computer Engineering,
University of Waterloo, Canada
-
Prof.
Dr. Howard Li, Associate Professor, Department
of Electrical and Computer Engineering, University
of New Brunswick, Fredericton, New Brunswick, Canada
-
Prof.
Dr. Hossam Fahreem, Faculty of Computers and Information, Ain Shams University, Egypt
-
Dr. Mohamed Salem, Faculty of Computers and Information, Ain Shams University, Egypt
|
|
Former
Researchers:
|
|
|
|
|
