Postdoctoral positions in collective animal behaviour: experiments and computations

Two postdoctoral positions are available in the Collective Behaviour
Group, at the Centre for Statistical Mechanics and Complexity - CNR
Rome. The positions are funded by the IIT project ART-SWARM, focusing
on the experimental and theoretical study of collective behaviour in
bird flocks and insect swarms, and its potential applications to
artificial systems.


Collective animal behaviour is a fascinating phenomenon. How does global co-ordination emerge in a flock of thousands starlings swirling at dusk? How does a school of sardines organize in a circular pattern? What is the evolutionary function of collective behaviour? Is it purely defensive, or has it some other purpose, perhaps social? INFM-CNR is the leader node of a European project - STARFLAG - dedicated to the study of collective animal behaviour.

The candidates will work under the supervision of Irene Giardina and
Andrea Cavagna. Information on our collective behaviour research can
be found at:

http://www.smc.infm.it/index. php?option=com_content&view= category&layout=blog&id=45& Itemid=103

A more thorough description of the candidates profiles and of the
project's aims can be found below.

Each position is for 1+1 year, starting as early as March 2010 and not
later than October 2010. Salary will be in line with Marie Curie (EC)
standards. Applicants should send CV, publications list, research
interests, and at least two recommendation letters to:

Dr Irene Giardina (subject: postdoc ART-SWARM)
irene.giardina@roma1.infn.it

In order to receive full consideration applications should arrive
within February 28, 2010.


With best regards,
 
Irene Giardina
Andrea Cavagna

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

Both candidates must have a strong interest in collective phenomena in
the physical and/or biological sciences. Although the two postdocs
will work in a highly integrated fashion, the two positions have
different and complementary scientific profiles:


Position 1: EXPERIMENTAL STUDY OF 3d COLLECTIVE ANIMAL BEHAVIOUR

The postdoc will be part of an experimental team of 3 people; ideally
(but not necessarily) he/she will be the team leader. Work will
include: setting up a new experimental apparatus for 3d swarm
reconstruction; calibration and testing; preparatory field
observations; field data-taking; data analysis.

* Prerequisites: background in either experimental physics, or
experimental biology, or engineering; good computer skills. 

* Bonuses (by no means necessary):
- field work
- Unix/Linux knowledge
- camera/video equipment
- practical stereoscopy
- electronics
- mechanics lab equipment

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

Position 2: COMPUTATIONAL METHODS FOR 3d COLLECTIVE ANIMAL BEHAVIOUR

The postdoc will work on the computational tools needed to perform the
3d reconstruction, i.e. to transform the experimental digital images
in a 3d data set. He/she will also work on dynamical tracking, in
order to produce the full individual trajectories. Finally, the 3d
data will be analyzed looking for new biological patterns.

* Prerequisites: background in either statistical physics or computer
science; strong programming experience in C++; excellent Unix/Linux
knowledge; basic script programming experience (Python/Pearl/...).

* Bonuses (by no means necessary):
- computer vision
- 3d reconstruction
- image processing
- optimization
- montecarlo methods
- numerical simulations
- html/php/sql
- openCV

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

Project ART-SWARM
>From self-organized animal groups to distributed artificial
swarms: exporting natural behavioral rules to mobile robotics


The study of self-organization and collective behavior encompasses
fields as diverse as statistical physics, ethology, mathematical
biology, control theory, and cooperative robotics. Three-dimensional
animal aggregations, as bird flocks, fish schools and insect swarms,
provide wonderful examples of emergent self-organization. The major
issue, both for theoretical studies and for technological
applications, is to understand how self-organization emerges within a
system with distributed intelligence. Several multi-agent models of
flocking and swarming exist, which produce collective behavior
starting from simple rules followed by the individuals. Yet, due to
the lack of 3D large-scale data, these models are hardly tested
against quantitative observations. Moreover, the rules of interaction
among the agents are guessed on the basis of common sense, rather than
being quantitatively modelled on empirical observations.  This is a
severe limit. In biological groups individual strategies are selected
by evolution to achieve functioning and overall efficiency at
collective level. Thus, empirically based information on these
strategies would not only lead to more appropriate models, but also
help to design optimal control strategies in artificial systems.

This project has the following objectives:

1. Observe. We will perform quantitative field studies of bird flocks
and insect swarms. Using innovative techniques in computer vision, we
will reconstruct individual 3D positions and dynamical trajectories in
cohesive aggregations of thousands of animals.

2. Understand. By analyzing the data, our aim is to unveil the laws of
self-organization and collective behavior in 3D animal
aggregations. Spatial and dynamical correlations among the individuals
will provide a full characterization of the rules of interaction in
the animal groups considered.

3. Discover. Dealing with several species and phyla, endowed with
specific individual abilities and facing different collective tasks,
we will investigate the crucial link between sensory/cognitive
functions and behavioral strategies, and determine how individual
cognition regulates group coordination.

4. Model. We will exploit the insight gained from empirical data as an
input to develop new 3D models of animal collective behavior. The
tools will be multi-agent theory and mathematical biology. Models
output will be quantitatively tested against 3D data.

5. Export. We will design new schemes of distributed control
quantitatively modelled on 3D animal behavior. Target applications
will be cooperative mobile robotics for environmental monitoring, and
nanorobotics for medical applications.


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