[robotics-worldwide] Phd fellowships at the Italian Institute of Technology

Giorgio Metta giorgio.metta at iit.it
Tue Sep 9 07:37:45 PDT 2008


Dear all,
Here's a call for 4 Phd positions on robotics.

----------------------------------------------------------------------------
---
GENERAL BACKGROUND
The Robotics, Brain and Cognitive Sciences (RBCS) Department of the Italian
Institute of Technology (IIT) is offering fellowships in the area of
Humanoid Robotics and Cognition. These fellowships are within one of three
main stream of research carried out at RBCS addressing, besides Humanoid
Robotics and Cognition, also Brain Machine Interface and Human Behavior and
Biomechanics. The three Research Streams will be developed jointly at RBCS
department of IIT under the responsibility of Giulio Sandini and a team of
scientists coordinated by Giorgio Metta (Humanoid Robotics), Luciano Fadiga
(Brain Machine Interface), Thierry Pozzo (Human Behavior), Franco Bertora
(Brain Imaging) and Pietro Morasso (Robotic Rehabilitation). 

The themes specifically under the Humanoid Robotics and Cognition stream,
group the research activities targeting the humanoid platforms of the lab
among which iCub (the platform of the RobotCub project
http://www.robotcub.org) and “James” (a one-arm humanoid built to
investigate manipulation and object affordance). The research themes
proposed are examples of the planned activities in the wider area comprising
cognitive systems, visual and tactile perception, sensorimotor coordination,
advanced materials for actuation and sensing.

More specifically the three research themes proposed for the four
fellowships are (short abstract and scientist in charge are included at the
end of the message):
•	Theme 5.1 Humanoid robot tactile sensing system 
•	Theme 5.2 Neuromorphic sensors for humanoid robots
•	Theme 5.3. The role of actions for perception: improving perception
abilities of the humanoid iCub

Interested applicants should refer to one of the following website to
download instructions on how to apply and/or contact directly the scientists
in charge (below) for more information regarding the individual research
plans.

For the description of the topics offered under the “Brain Machine
Interface” and the “Human Behavior and Biomechanics” streams refer directly
to Annex-A of the application material that can be downloaded from the
websites

----------------------------------------------------------------------------
---
http://www.liralab.it/iit2009phd.htm  
http://www.iit.it/phd_positions

----------------------------------------------------------------------------
---
DEADLINE is October 3rd, 2008

----------------------------------------------------------------------------
---
RESEARCH TOPICS PROPOSED
Theme 5.1: Humanoid robot tactile sensing system
Tutor: Prof. Giorgio Metta – Prof. Valle
N. available positions: 1

Tactile sensing is the process of detecting and measuring a given property
of a contact event in a predetermined area and subsequent pre-processing of
the signals – before sending them to higher levels for perceptual
interpretation.  Generally, tactile sensing is associated with the force
measurement, but - in view of the above definition - the tactile sensing in
robots should also include detection and measurement of object information
that cannot be deciphered from the force measurement only e.g. temperature. 
The Ph.D. research activity will involve the development of tactile sensing
modules and hence of the skin parts for humanoid robots, their integration
with robot and subsequent use in the robot control loop for exploration and
manipulation tasks. The research activity will focus on a limited set of
candidate transduction technologies for contact sensing; piezoelectric
polymers, resistive, capacitive - being the most promising candidates for
measuring geometric and mechanical quantities. The possibility of having
transducers and conditioning electronics on same medium/substrate – with
transducer directly coupled with an electronic device (i.e. FET/TFT device)
– will be explored. While the transducers can be placed either on flexible
(e.g. organic) or rigid (e.g. crystalline silicon) substrate – based on
their location on the robot’s body; the conditioning electronics will be
implemented on the silicon substrate.  Besides improving the performance of
the sensing system, the approach is expected to provide a solution to wiring
complexity – a key problem in robotics. Such a marriage of transducers (e.g.
smart materials like piezoelectric polymers) with FET or TFT devices on
rigid (e.g. silicon), or flexible (e.g. organic or elastomeric) substrates,
would also improve the reliability. Sensors readout and smart processing
tasks will be implemented with dedicated electronics embedded in the tactile
sensing array. The networking of the modules of tactile sensing arrays, so
obtained, is expected to result in a scalable system.
Applicants should have a background in one or more of the following fields:
electronic engineering, microelectronics, computer science, sensors and
robotics. The candidates must have good writing and communication skills and
motivation to work in a highly competitive and multidisciplinary
environment.
For further details concerning the research project, please contact:
giorgio.metta at iit.it


Theme 5.2: Neuromorphic sensors for humanoid robots
Tutor: Dr. Chiara Bartolozzi
N. of available positions: 1

Biological sensory systems vastly outperform conventional digital systems in
almost all aspects of perception tasks, where the system must process noisy
and ambiguous stimuli to produce appropriate behavioral responses. Digital
systems require vast amounts of resources to extract relevant information
from sensors, but still fail to produce appropriate responses for
interacting with the real world in real time. Part of the reason for this
might be the fundamental differences in handling sensory data in biological
systems and machines. “Frame-based” time sampling and quantization artifacts
present in conventional sensors are particularly problematic for robust and
reliable performance. On the contrary biological sensory systems make use of
continuous time, stimulus-driven, asynchronous, distributed, collective, and
adaptive principles, that make their interpretation of the external world
reliable and robust. The goal of this project is to introduce in the field
of robotic vision the principles of biological sensory systems design.
Specifically we aim at combining the design of novel data-driven
biologically inspired sensory devices with the development of new
asynchronous event-driven computational paradigms, with structure and
morphology that are matched to the requirements of the robots body and its
application domain. The candidate shall work on testing of existing
asynchronous vision sensors and on the design of new sensors with
non-uniform morphology, using analog real-time low-power VLSI neuromorphic
circuits. The candidate will participate in the whole project development by
also tackling the problem of the realization of supporting data driven
asynchronous computational paradigms for machine-vision methodologies that
are radically different from conventional ones, and test the developed
vision system performance on advanced humanoid robotic platforms.
Requirements:
Applicants should have a strong interest in bio-inspired hardware
engineering and the ability to work independently. Good skills in
programming C and Matlab. Fundamental notions of microelectronics;
Background in neuroscience.
For further details concerning the research project, please contact:
chiara.bartolozzi at iit.it


Theme 5.3: The role of actions for perception: improving perception
abilities of the humanoid iCub
Tutor: Prof. Giorgio Metta – Dr. Lorenzo Natale
N. available positions: 2

In modern robotic systems perception is too often inadequate and simplified.
This seriously affects the ability of robots to cope with unpredictability
and successfully interact in the real world. Artificial perception is a
complex task, in which, despite great efforts in computer science, robotics
and artificial intelligence, only partial successes have been achieved.
The study of perception in humans shows that the brain takes advantage of
the integration of the wealth of information available from the different
sensory modalities, including information about the incipient action
generation. Populations of multimodal neurons are responsible on one side
for controlling eye/head, arm, and grasping movements and, on the other, to
interpret actions performed by others, for the recognition of objects, and
to support goal-directed attention processes. The representation of the
world in the brain happens through the activation of multiple pathways and
involves the use of sound, vision, touch, and proprioception mixed with
activations describing the current motor context and the intended goal. This
requires a new way of looking at the problem of learning to extract relevant
information at each stage of processing. Outstanding questions regards: How
do we represent the space around us? What visual features are important?
What is the relationship between sound, vision and touch? How motor
information structures the recognition process? Of particular interest is
the determination of the link between object-related information and the use
of this information for the control of manipulation.
In this project we are interested in studying methods for improving the
perceptual abilities of the humanoid robot iCub (http://www.robotcub.org),
with the goal of exploring ways of using and integrating sensory information
originating from the interaction between the robot and the environment. This
project requires the investigation and implementation on the robot of
several different elements ranging from sensory perception, to motor
control, and machine learning: for example, the realization of explorative
behaviors to extract information about the environment and objects. This
might include grasping and manipulation, but also simpler strategies like
pushing, prodding and squeezing objects. One of the goals here is to build a
representation of objects that goes beyond the level of the single sensory
modality (e.g. vision) but merges the visual appearance of an object with
the haptic sensation it produces when grasped, or the sound it produces when
touched. To reach this level of competency the robot would require the
control of attention and self-recognition.
We seek candidates with a strong background in computer science and
engineering, who are also interested in the study of perception and modeling
of biological systems. Knowledge of electronics and mechanics is not
required but it might be useful considered that the successful applicant
will be working directly on a real humanoid robot.
For further details concerning the research project, please contact:
giorgio.metta at iit.itlorenzo.natale at iit.it 

----------------------------------------------------------------------------
---

Giorgio Metta <giorgio.metta at iit.it>

Italian Institute of Technology
Via Morego, 30
16163 Genoa, Italy
Ph: +39 010 7178-1411                     
Fax: +39 010 7170-817

URL: http://pasa.liralab.it 





More information about the robotics-worldwide mailing list