[Stellenangebote] Student Assistant Position - Robotics Platform

[Benner, Philipp]

Porject description

While the normal abiotic steel corrosion process is well understood,
microorganisms can corrode steels much faster though localized pitting
corrosion. Understanding the triggers, processes and kinetics behind
this still highly unpredictable “microbiologically influenced corrosion
(MIC)” can help lift some of the 2.5 trillion dollar global cost of
corrosion.

At BAM, we have developed a dynamic flow reactor set-up that allows the
detailed and controlled study of this MIC process on 2mm steel beads.
It contains 600 beads over five reactors: three biological replicates
and two control reactors, each containing 6 sections. At the start and
end of the experiment, these beads have to be individually weighed and
(ideally) imaged. This is the current experimental bottleneck that
would lend itself well to automation. Increased automation would lower
the threshold for experiments, allowing more experiments to be done
leading to better insights and statistics.

To improve the experiments and in particular measuring the corrosion
rates in this system, we need you to turn our in-house developed open
source "RoWaN" robotic platform into a tool to help with this corrosion
study. Ideally, the robot would pick up the corroded spheres, weigh
them, and image them, followed by statistical analysis of the results.

The challenges in this project are very diverse. While the core of the
platform is set up and can be run using Jupyter Notebooks, this
particular task requires the addition and modification of hardware,
software, and development of (image) analysis tools for the statistical
evaluation of the large number of samples. Plenty of challenges will be
encountered along the way and workarounds need to be implemented.
Collaboration with a diverse team including biophysicists, physicists,
laboratory and workshop staff and computational scientists will be a
daily occurrence.


Project challenges

The project involves setting up a system for automated carbon steel
ball corrosion analysis.

This project is split into five components:

1. Configuring and modifying the individual components (currently we
have a Mecademic robot arm, precision balance, OT-2 pipetting station,
multipurpose Arduino i/o device, computer vision-equipped microscopes,
and a few more small components),
2. Configuring an experimental path between these components for sample
handling and analysis, ideally through automated path optimization
methods,
3. Implementation of fault and error handling systems in case of
unexpected behaviour of the robotic system,
4. Analysis of the resulting images and weights of many hundreds of
samples using statistical tools.
5. A stretch goal is to incorporate (parts of) the washing and drying
steps for each sphere.

Candidate requirements

1. Comfortable in Python and jupyter lab/notebook, familiarity with
statistical tools and ideally also OpenCV and python communication
libraries.
2. Comfortable with the Git versioning system
3. Enjoys interdisciplinary and pragmatic problem solving
4. Ideally: experience with hands-on work in the lab and hardware /
engineering / designing
5. Ideally: a basic understanding of electronics
6. Ideally: a basic understanding of statistics
7. Ideally: an affinity for bacterial life-forms!


More details and a demonstration of the current robotics platform can
be provided upon request.


-- 
Bundesanstalt für Materialforschung und –prüfung (BAM)

Dr. Philipp Benner
Section S.3 eScience

Unter den Eichen 87
12205 Berlin
GERMANY

P: + 49 30 8104-3647
philipp.benner at bam.de
www.bam.de

________________________________________________________

BAM is a senior scientific and technical Federal institute with
responsibility to Federal Ministry for Economic Affairs and
Climate Action.