PhD and Postdoc positions available at AASS, Örebro University, Sweden

[Martin Magnusson]

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

                    PhD student & Postdoc Positions
            AASS Research Center, Örebro University - Sweden

                                Topics:
            Rich 3D and Semantic Mapping on Multiple Scales
     Reliability-Aware Long-Term Mapping, Modeling and Predictions
     Integrated Planning, Coordination and Control of Robot Fleets

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

Several fully funded PhD student (4 years) and postdoc positions
(2+2 years) are available starting in January 2017 at the AASS
Research Center, Örebro University, Sweden (Contact: Achim
J. Lilienthal, achim.lilienthal at oru.se).

The positions will be funded by a European robotics project on
robotics applied in intra-logistics environments.


Research Topics
===============

The open positions relate to the following cutting edge research topics.


Topic "Rich 3D and Semantic Mapping on Multiple Scales"
----------------------------------------------------------------
Robots performing real-world tasks in real-world environments require
environment information that goes well beyond geometry. This research
topic is concerned with adding semantic information to geometric
models on multiple scales.

On a large scale, 2D and 3D maps require high-level information,
e.g. what type of objects and locations are in the map, what types of
activities are performed in different areas (e.g. loading/unloading,
charging/refueling, stacking/piling) or at specific poses (e.g. pickup
and drop off locations).

On a smaller scale, higher-level information is also required for
manipulation tasks, where three-dimensional models of objects need to
be endowed with information about material types, and how they are
stacked relative to adjacent objects.


Topic "Reliability-Aware Long-Term Mapping, Modeling and Predictions"
----------------------------------------------------------------------
While relatively mature methods for robot mapping and localization
exist today, long-term operation poses several challenges that are yet
to be overcome in a systematic way.

This research topic is concerned with, e.g., developing metrics for
assessing the reliability (precision and uncertainty) of map-building
and localization, as well as the long-term updating of robot maps,
thus enabling persistent, reliable self-localization in changing
environments -- especially in environments with significant perceptual
aliasing, such as industrial warehouses.

By learning over time from typical motion patterns in the environment
where the system is deployed, it can better blend in with current
operations and make predictions about future states by combining
long-term mapping and modeling, which is also a requirement for
efficient and reliable long-term operation.


Topic "Integrated Planning, Coordination and Control of Robot Fleets"
---------------------------------------------------------------------
Fleets of autonomous robots are essential in many industrial
applications, such as automated warehouses, mines, and construction
sites. Fleet automation includes four important computational
problems, namely task planning, motion planning, coordination, and
control. By and large, the industry-standard approach to fleet
management relies on disjoint, partially automated solutions to these
problems. This leads to costly deployment effort and the need to
perform off-line "what-if" analysis in order to guarantee adherence to
external requirements (e.g., the absence of deadlocks).

This research topic aims to achieve a formal and practical framework
for jointly reasoning about the computational problems underlying
fleet management. The idea is to exploit centralized and/or
decentralized decision making modules that act upon a common
constraint-based representation of the fleet's state over
time. Crucially, the common representation will simultaneously act as
a means for different problem solvers to share information, and as a
means to model (and ensure adherence to) given requirements on fleet
behavior (e.g., temporal and spatial constraints, absence of
deadlocks).


Environment
===========

Successful applicants will join the AASS (Applied Autonomous Sensor
Systems) research unit (http://www.oru.se/aass) at Örebro University,
Sweden.

Salaries for PhD students at Örebro University are internationally
competitive.

With over 40 senior researchers and PhD students, AASS is one of the
largest robotics research groups in Sweden.  The research and human
environment at AASS is young and enthusiastic, and PhD students come
from many different countries and have different scientific and
cultural backgrounds.

AASS is involved in several international projects, thus providing
PhD students with opportunities to travel and work together with
people in other countries.

At AASS we perform multidisciplinary research in autonomous systems
with a focus on their perceptual and cognitive capabilities and their
integration.  In other words, we specialize in research that combines
field robotics -- i.e., real-world robot applications outside of lab
environments, in cooperation with industrial partners -- with
artificial intelligence for perception and reasoning.

We are internationally recognized for our work in key scientific areas
including planning and scheduling, 3D mapping and localization, hybrid
reasoning, and mobile robot olfaction.


Instructions for applicants
===========================

Important: Please indicate clearly in your application what your
preferred topic is. If you want to apply for several topics, please
give your priority. Please also indicate in the subject whether you
apply for a PhD-student or postdoc position.


*PhD student, Prerequisites and Application Process*

In addition to a strong interest in the topic, a solid theoretical
background and excellent programming skills, applicants should also
have the equivalent of a Master's degree in a relevant field (e.g.,
Computer Science, Applied Mathematics, Robotics, Physics). Experience
in Robotics and AI and is a plus. It is not necessary to be familiar
with Swedish, but fluency in written and spoken English is mandatory.

To apply for a position, please send the following documents in PDF
format to Prof. Achim J. Lilienthal (achim.lilienthal at oru.se):
- a motivation letter,
- scanned course transcripts and degree certificate,
- a full CV,
- at least two academic references (names and contact details).


*Postdoc, Prerequisites and Application Process*

Applicants should have a PhD in Robotics or a related field, and of
course dedication and a strong interest in the particular topic. In
addition, the successful candidates should have a solid theoretical
background and excellent programming skills. It is not necessary to be
familiar with Swedish, but fluency in written and spoken English
is mandatory. It is expected that the candidate has publications at
ICRA, IROS, RSS, TRO, IJRR, JFR or equivalent conferences and journals.

To apply for a position, please send the following documents in PDF
format to Prof. Achim J. Lilienthal (achim.lilienthal at oru.se):
- a one page research statement that describes the preferred research
   topic and your suitability for this topic,
- a full CV,
- at least two references (names and contact details),
- a full list of publications,
- copies of the three most relevant publications.



Contact: Prof. Achim J. Lilienthal (achim.lilienthal at oru.se)

Closing date: Applications can be sent immediately and will be
considered until the position is filled.