1

B.Sc. or M.Sc. Thesis/Project: Dimensionality Reduction using Variational Autoencoder on Synchrotron XRD data

Supervisor: Univ.-Prof. Dr Elmar Rückert, Dr. Petra Spörk-Erdely (Chair of Physical Metallurgy and Metallic Materials)
Start date: ASAP from December 2021

Theoretical difficulty: mid
Practical difficulty: low

Abstract

In the context of this thesis, we propose to apply modern machine learning approaches such as variational autoencoder to visualize and reduce the complexity of X-ray diffraction (XRD) data collected on advanced γ-TiAl based alloys. By classifying XRD data collected during in situ experiments into known phases, we aim at disclosing phase transformation temperatures and selected properties of the individual phases, which are of interest with regard to the current alloy development. Furthermore, the capabilities of the applied machine learning approaches going beyond basic XRD data analysis will be explored.


Sketch of a synchron from synchrotron.org.au, illustrating the process of accelerating electrons at almost the speed of light.   

Illustration of a collected data sample which is a 2D X-ray diffraction of a nominal Ti-44Al-7Mo (in at.%) alloy collected.

Topic and Motivation

Intermetallic γ-TiAl based alloys are a promising class of structural materials for lightweight high-temperature applications. Following intensive research activities, they have recently entered service in the automotive and aircraft engine industries, e.g. as low-pressure turbine blades in environmentally-friendly combustion engine options [1].

During the past decades, the development of these complex multi-phase alloys has been strongly driven by the application of diffraction and scattering techniques [2]. These characterization techniques offer access to the atomic structure of materials and provide insights into a variety of microstructural parameters. High-energy X-rays, such as available at modern synchrotron radiation sources (i.e. large-scale research facilities for X-ray experiments), and recent advances in hardware technology nowadays allow to conduct so-called in situ experiments that reveal at a high temporal resolution the relationship between selected external conditions (e.g. thermal or mechanical load) and structural changes in the material. Various specimen environments can be adjusted to emulate technologically relevant or real-life conditions, addressing a multitude of research topics ranging from fundamental questions in the primary alloy design over process-related to application-related issues. Modern setups at synchrotron sources even allow the investigation of elaborate manufacturing, joining and repair processes in an in situ manner, producing insights that have been inaccessible by means of conventional methods so far.

Advanced materials characterization techniques such as described above are often characterized by an ever-growing data acquisition speed and storage capabilities. While enabling novel insights, they, thus, also pose a serious challenge to modern materials science. In situ synchrotron X-ray diffraction (XRD) experiments usually bring about large sets of two-dimensional diffraction data such as those shown in Figure 1. New procedures are needed to quickly assess and analyze this type of datasets.

Tentative Work Plan

To achieve our aim, the following concrete tasks will be focused on:

  • Literature research on state-of-the-art materials characterization methods.
  • Implementation of  deep convolutional neural networks (CNN) and Variational Autoencoder in Python/Tensorflow.
  • Application and evaluation of variational autoencoder on the CNN features.
  • Analysis and Evaluation of the provided synchrotron data.

References

[1] Clemens, S. Mayer, Design, processing, microstructure, properties, and applications of advanced intermetallic TiAl alloys, Advanced Engineering Materials 15 (2013) 191-215, doi: 10.1002/adem.201200231.

[2] Spoerk-Erdely, P. Staron, J. Liu, N. Kashaev, A. Stark, K. Hauschildt, E. Maawad, S. Mayer, H. Clemens, Exploring structural changes, manufacturing, joining, and repair of intermetallic γ-TiAl-based alloys: Recent progress enabled by in situ synchrotron X-ray techniques, Advanced Engineering Materials (2020) 2000947, doi: 10.1002/adem.202000947.




Internship Positions – Just contact us!

Please contact us via cps@unileoben.ac.at if you want to join us for an internship.

We support you when you apply for an internship grant. Below we list some relevant grant application details.

CEEPUS grant (European for undergrads and graduates)

Find details on the Central European Exchange Program for University Studies program at https://grants.at/en/ or at https://www.ceepus.info.

Country of origin

Albania; Bosnia and Herzegovina; Bulgaria; Croatia; Czech Republic; Hungary; Kosovo; Moldova – Republic of; Montenegro; North Macedonia; Poland; Romania; Serbia; Slovakia; Slovenia

Duration

1 to 10 months. Undergraduates need to stay at least 3 months. 

Grant benefit paid

For students and graduates without PhD EUR 1050 per month. For teachers with a PhD at most EUR 1150.

Application Deadlines and Online Application Link

Apply online at https://ceepus.info/

In principle, you can apply at any time for a scholarship. However, also your country of origin matters and there exist networks of several countries that have their own contingent. 

Ernst Mach Grant (Worldwide for PhDs and Seniors)

Find details on the program at https://grants.at/en/ or at https://oead.at/en/to-austria/grants-and-scholarships/ernst-mach-grant.

Country of origin

Worldwide.

Duration

1 to 9 months.

Grant benefit paid

For PhD students EUR 1050 and for senior researcher EUR 1150 per month

Application Deadlines and Online Application Link

Apply online at http://www.scholarships.at/

Max. age is 35 years. Forthcoming closing date for applicants is 01.02.2022. However, note that due to the pandemic applications may be processed at any time. 




Artificial intelligence as a new research area

News on UNI Leoben website: Artificial intelligence as a new research area


This article provides an insight into Univ.-Prof. Rückert’s thoughts on Artificial intelligence, Robots, CPS future and more. 

Brief, but interesting reading can be accessed here.




Holidays in Austria

Feiertage in Österreich / Holidays in Austria

Here is a quick informative list of important Holidays. 
For other information of such nature, visit the main website of our University in Leoben.




Development of an accurate low-cost sensor glove for learning grasping skills

Video

Link to the file

You may use this video for research and teaching purposes. Please cite the Chair of Cyber-Physical-Systems or the corresponding research paper. 

Publications

2021

Denz, R.; Demirci, R.; Cansev, E.; Bliek, A.; Beckerle, P.; Rueckert, E.; Rottmann, N.

A high-accuracy, low-budget Sensor Glove for Trajectory Model Learning Inproceedings

In: International Conference on Advanced Robotics , pp. 7, 2021.

Links | BibTeX

A high-accuracy, low-budget Sensor Glove for Trajectory Model Learning




spirit OF STYRIA N°08

Univ.-Prof. Rückert in Current Issue of Styria’s elegant Magazine spirit N°08




Links

All our readers can read through the latest edition of Spirit of Styria Magazine.

Three full pages (pg. 50 – 52) are dedicated to Univ.-Prof. Rückert.

Access via the following link.




Univ.-Prof. Rückert about Artificial Intelligence in the new STADT MAGAZIN 2021

Print Media Article in Leoben’s STADT MAGAZIN 2021


Further Links and Description

Article in Stadt Magazin comes with title: Artificial Intelligence. It features Univ.-Prof. Dipl.-Ing. Dr.techn. Elmar Rueckert in the tenth edition of 10/21 print.

Article is available for online reading. Every reader can find it under the following link. Alternatively, reader can access via article picture. Simply click on the picture displayed in this post.

Finally, the section dedicated to Univ.-Prof. Rueckert is captured on the tenth page of Stadt Magazin 10/21.




Innovative Research Discussion

Meeting on the 13th of October 2021

Location: Chair of CPS

Date & Time: 13th October. 2021, 1pm to 2pm

Participants: Univ.-Prof. Dr. Elmar Rueckert, Linus Nwankwo, M.Sc.

 

Agenda

  1. Review of the previous meeting action points
  2. Discussion of the research progress and possible improvement

Topic 1: Review of the SLAM state-of-the-art

  1. Review the current literature to identify the areas of possible improvement
  2. Investigate the possible integration of RGB-D sensors with thermal sensors for indoor or outdoor SLAM
  3. Evaluate the impact of semantic segmentation and pose estimation on the quality of the map in dynamic environment

Topic 2: Existing Algorithms Performance Evaluation

  1. Implement a loop closure trigger for possible use with any of the SLAM algorithm
  2.  Implement the following SLAM algorithms and evaluate the performances in terms of optimality, accurate pose estimation, computational cost, etc.
  • RTABMap
  • Hector SLAM
  • GMapping

Next Steps

Implement own SLAM algorithm robust against:

  • Odometry Lost
  • Variation in weather and illumination
  • Weak dynamic object detection

Next Meeting

Yet to be scheduled




CPS and Univ.-Prof. Dr. Elmar Rückert in latest Edition of Triple M Magazine 2021

Print Media Article in Triple M Montanuniversität 2021

Links

Online Triple M Magazine’s latest Editions are available here.




Meeting notes 08.10.2021

VISA D Gainful Employment

Prof. Elmar told me to contact the Austrian embassy in Moscow to ask about the “VISA D Gainful Employment”, what are the required documents, and when can I get the visa.

This visa type should allow me to start working immediately without any delay.

RL Simulator

  • Game simulator on Gitlab written in C/C++
    • I should get access to the repository
    • Read the code of the simulator
    • Start applying a basic RL agent on the player
    • with time we should improve the RL algorithms
    • The main idea is to transfer the learned policy in a nonheuristic manner to new environments with different parameters.
    • Start reading about preference learning.
  • Future possibilities:
    • Apply our algorithm on other environments.
    • Apply our algorithm on physical systems.