|
Posted by Thierry Fraichard on September 15, 2005, 2:08 pm
----------------------
Post-doctoral proposal
----------------------
The Inria e-Motion research project-team at Inria Rhone-Alpes is opening a
Post-doctoral position in the field of:
Autonomous navigation in dynamic environments.
o Subject Description:
Mobivip is a French research project whose purpose is to study, evaluate
and experiment technologies revolving around novel forms of transportation
systems based on "Public Individual Vehicles" (see
http://www-sop.inria.fr/mobivip for more details). Public Individual
Vehicles are similar to their European counterparts, the Cybercars (see
http://www.cybercars.org ).
In this framework, the e-Motion group of Inria Rhone-Alpes addresses the
problem of automated driving in open and dynamic environments. An
environment is dynamic when it features moving objects and it is open when
full knowledge about it is not available beforehand (which is typically the
case of most real environments, in particular roadway-like environments).
The Inria e-Motion research project-team (formerly known as Sharp) has a
long history in the field of motion autonomy. It has been involved in a
number of French, European and International programmes and has
demonstrated in various occasions different autonomous driving skills with
real vehicles (parking, platooning, obstacle avoidance). e-Motion has
developed a number of techniques in fields as different as perception,
localisation, map building, motion planning, sensory-motor coupling and
reactive motion execution.
e-Motion is currently exploring a motion planning scheme called "Partial
Motion Planning" (PMP) whose purpose is to explicitly deal with the
constraints imposed by a dynamic environment, in particular the real-time
constraint imposed by the dynamicity of the environment: the motion planner
has a limited-time only to compute a motion.
Given the intrinsic complexity of motion planning, fulfilling such a
real-time constraint is in general impossible to guarantee (even with the
most efficient randomised motion planners existing today). One way to cope
cope with this issue is to compute a partial motion only, ie the best
partial motion given the time available. In a way, PMP can be seen as a
compromise between pure motion planning and reactive approaches.
PMP, like reactive approaches, raises a safety issue: what guarantee do we
have that the robotic system will never find itself in a dangerous
situation yielding a collision in the future (this issue is critical in
dynamic environments).
To solve this problem, we believe that the concept of "Inevitable Collision
States" (ICS) can be useful. An ICS is a state such that no matter what
the future trajectory of a robotic system is, it eventually collides with
an obstacles. This concept is general and permits to take into account the
dynamics of both the robotic system and the moving objects. Clearly, for
its own safety, a robotic system should never be in one of these ICS.
A first version of the PMP algorithm has been developed and implemented on
the experimental vehicle of the Inria research unit located near Paris (PhD
of Stephane Petti). The primary purpose of this Post-Doc position is to
integrate this algorithm onto the experimental vehicle of Inria Rhone-Alpes
and then to further the work on PMP and ICS so as to demonstrate the
validity of the overall concept on a real car-like vehicle.
o Requirements (educational and technical background, knowledge of
specific material):
The experimental component of this post-doc will be significant. It will
involve a lot of development on the e-Motion experimental platform, a
Cycab vehicle. Candidates should be motivated and definitely have a
taste for this kind of work. Former experience in real-time programming
and control theory is a plus.
Mobivip being a French programme, fluency in french is very much welcomed
(although not technically required).
o Starting date: as early as possible (December 2005 would be great).
o Duration: one year.
o Wages: approximately 1800 Euros per month, including standard health
coverage.
o Person to contact (name, email, postal address, URL):
- Dr. Thierry FRAICHARD
- thierry.fraichard@inria.fr
- Inria Rhone-Alpes
655 avenue de l'Europe
Montbonnot
38334 Saint Ismier Cedex
France
- http://emotion.inrialpes.fr/fraichard
o Application dossier: applicants should send a small (preferably
electronic) dossier featuring:
- motivation letter,
- complete Curriculum Vitae,
- academic background with the courses attended, the grades and the
rankings obtained (very important),
- list and pointers to selected publications and PhD.
- recommendation letters,
- and any additional information you might find useful.
o Project Web Server: http://emotion.inrialpes.fr
Thierry Fraichard, September 12, 2005
************** comp.robotics.research (moderated) **************
Summary: Academic, government & industry research in robotics.
Archives and information: http://www.metamech.com/crr
Charter: http://www.metamech.com/crr/Charter
Meta-discussions/information: crr-request@metamech.com
|
Yahoo! 
Google 
Windows Live 
del.icio.us 
digg 
Netscape 
XML site map