image_pdfimage_print

Data Science Summer School 2022, Leoben

Univ.-Prof. Dr. Elmar Rueckert presents in his two slots how machine learning approaches can be used in robotics. 

Materials for the Robotics Workshop

Clemens Fritze, B.Sc. B.Sc

Student Assistant at the Montanuniversität Leoben

Foto_20220621

Short bio: Clemens Fritze, B.Sc B.Sc started at CPS in July 2022. After a short break he joined the team again in November 2023.

Clemens Fritze is a master student at the Montanuniversität Leoben. Prior to his master program he studied Mechanical Engineering at the Montanuniversität Leoben, where he passed his Bachelor defense in May 2022. In 2018, Mr. Fritze finished a Bachelor study in Business Informatics at the university of applied science in Vienna (german FH Technikum Wien).

Research Interests

  • Automation
  • Robotics
  • IoT

Thesis

Master Thesis

Bachelor Thesis

Contact

Clemens Fritze, B.Sc B.Sc.
Student Assistent at the Chair of Cyber-Physical-Systems
Montanuniversität Leoben
Franz-Josef-Straße 18, 
8700 Leoben, Austria 

Email:   clemens.fritze@stud.unileoben.ac.at

M.Sc. Thesis: Daniel Wagermaier: Predictive Models for Copper Content and Energy Consumption in Electric Arc Furnace Operations

Supervisor: Univ.-Prof. Dr Elmar Rückert, Qoncept GmbH
Start date: 1st of August 2022

Theoretical difficulty: mid
Practical difficulty: low

Introduction

As direct observations and permanent measurements during steelmaking processes are not possible, modelling has become a powerful tool. The technique of fundamental-based metallurgical modelling is well-established and demonstrates its capabilities in a wide range of applications in modern steelmaking. Following the general trend, data-driven approaches are increasingly used today in various areas of metallurgical modelling, in addition to these  classical fundamental approaches. Depending on the field of application, fundamental-based and data-driven models both have their own advantages and disadvantages.

The overall goal of the present thesis is to combine both models in order to leverage the strengths of  these two different methods. The first step is to apply several different data-driven models and compare them to the metallurgical model to see how they perform differently. In the second phase, various ways of combining data-driven models with the metallurgical model should be investigated. For example, this could be done via a data-driven optimization of its tuning parameters or by replacing them with data-driven models. Also, adding a data-driven residual term to the metallurgical model could be possible. Based on these findings, the third part of the thesis should focus on online learning and methods of how to avoid an off-drifting of the model. The fourth and last section of the thesis should investigate various ways of detecting errors in new data. While point one and two are the main focus of the thesis, point three and four are considered to be optional.

Tentative Work Plan

The following concrete tasks will be focused on:

  • Literature research.
  • Training of different data-driven models in Python.
  • Performance comparison between data-driven models and the metallurgical model.
  • Combination of selected data-driven models and the metallurgical model in Python.
  • (Optional) Investigate different ways for online learning and live performance evaluation.
  • (Optional) Anomaly detection in new data.
  • Thesis writing.

Thesis Document

B.Sc. or M.Sc. Thesis/Project: Simultaneously predicting multiple driving strategies using probabilistic inference

Supervisors: Univ.-Prof. Dr Elmar Rückert, LUPA Electronics GmbH
Start date: ASAP from June 2022

 

Theoretical difficulty: high
Practical difficulty: low

Abstract

Wir Menschen sind in der Lage unter widrigen Bedingungen z.B. bei eingeschränkter Sicht, oder bei Störungen komplexe Vorgänge wahrzunehmen, vorherzusagen und innerhalb von wenigen Millisekunden zusammenhängende Entscheidungen zu treffen. Mit dem zunehmenden Grad der Automatisierung steigen auch die Anforderungen an künstliche Systeme. Immer komplexere und größere Datenmengen müssen verarbeitet werden um autonome Entscheidungen zu treffen. Mit gängigen KI Ansätzen stoßen wir aufgrund der konvergierenden Miniaturisierung an Grenzen, die z.B. im Bereich des autonomen Fahrens nicht ausreichen, um ein sicheres autonomes System zu entwickeln.

Ziel dieser Forschung ist es probabilistische Vorhersagemodelle in massiv parallelisierbaren neuronalen Netzen zu implementieren und mit diesen komplexe Entscheidungen Aufgrund erlernter interner Vorhersagemodelle zu treffen. Die neuronalen Modelle verarbeiten hoch dimensionale Daten moderner und innovativer taktiler und visueller Sensoren. Wir testen die neuronalen Vorhersage und Entscheidungsmodelle in humanoiden Roboteranwendungen in dynamischen Umgebungen.

Unser Ansatz beruht auf der Theorie der probabilistischen Informationsverarbeitung in neuronalen Netzen und unterscheidet sich somit grundlegend von den gängigen Methoden tiefer neuronaler Netze. Die zugrundeliegende Theorie ermöglicht weitreichende Modelleinblicke und erlaubt neben den Vorhersagen von Mittelwerten auch Unsicherheiten und Korrelationen. Diese zusätzlichen Vorhersagen sind entscheidend für verlässliche, erklärbare und robuste künstliche Systeme und sind eines der größten offenen Probleme in der künstlichen Intelligenz Forschung.

Tentative Work Plan

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

  • Literature research on graphical model inference of motion plans.
  • Toy Task implementation in Python. 
  • Implementation of  GMMs, PTSMs and combinations in Python.
  • Visualization and analysis of the prediction performance. Definition of suitable evaluation criteria.
  • (Optional) Implementation in a realistic driving simulator.
  • Analysis and evaluation of the generated data.

LUPA-Electronics GmbH / Berlin

Sicheres Autonomes Fahren mit Probabilistischen Neuronalen Netzen

Contact

 

Wir Menschen sind in der Lage unter widrigen Bedingungen z.B. bei eingeschränkter Sicht, oder bei Störungen komplexe Vorgänge wahrzunehmen, vorherzusagen und innerhalb von wenigen Millisekunden zusammenhängende Entscheidungen zu treffen. Mit dem zunehmenden Grad der Automatisierung steigen auch die Anforderungen an künstliche Systeme. Immer komplexere und größere Datenmengen müssen verarbeitet werden um autonome Entscheidungen zu treffen. Mit gängigen KI Ansätzen stoßen wir aufgrund der konvergierenden Miniaturisierung an Grenzen, die z.B. im Bereich des autonomen Fahrens nicht ausreichen, um ein sicheres autonomes System zu entwickeln.

Ziel dieser Forschung ist es probabilistische Vorhersagemodelle in massiv parallelisierbaren neuronalen Netzen zu implementieren und mit diesen komplexe Entscheidungen Aufgrund erlernter interner Vorhersagemodelle zu treffen. Die neuronalen Modelle verarbeiten hoch dimensionale Daten moderner und innovativer taktiler und visueller Sensoren. Wir testen die neuronalen Vorhersage und Entscheidungsmodelle in humanoiden Roboteranwendungen in dynamischen Umgebungen.

Offende Projekte, Bachelor- und Masterarbeiten​

Einführungs VO zu Cyber-Pysical Systems & ML

Der Lehrstuhl für Cyber-Physical-Systems widmet sich anwendungsorientierter Grundlagenforschung in den Bereichen der künstlichen Intelligenz, der Digitalisierung von Industrieprozessen und der Robotik. Ein Focus liegt dabei auf der Modelierung von intelligenten menschlichen Lernprozessen mit dem Ziel effiziente Lernmethoden und Vorhersagemodelle für cyber-physikalische Systeme zu entwickeln.

Hier finden sie den Foliensatz zum Vortrag.

B.Sc. or M.Sc. Thesis/Project: Machine Learning for Predicting Yield Strengths with the Stahl- und Walzwerk Marienhütte GmbH, Graz

Theoretical difficulty: low
Practical difficulty: low

Abstract

In this thesis, the student has the unique opportunity to investigate supervised machine learning methods for predicting yield strengths using probabilistic regression models and deep learning approaches. The thesis is implemented with support of the MSC Software GmbH and the Stahl- und Walzwerk Marienhütte GmbH in Graz.

In the image above and below you see the production line at the Stahl- und Walzwerk Marienhütte GmbH in Graz.

To ensure the high quality standards, frequent ‘yield strength’ measurements are performed. These measurements have resulted in a large dataset which can now be analyzed and used to learn a prediction model.  First tests were promising and the thesis will be very likely a big success. 

The goal of this thesis is to analyze the data and to learn prediction models taking uncertainty estimates into account.

The models will be implemented and tested in Python.

Tentative Work Plan

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

  • Literature research on the underlying physical & chemical processes.
  • Data analysis, filtering, preprocessing, visualization of the existing data. 
  • Implementation of  deep neural networks (Variational Autoencoder), neural processes and GPs in Python. Baseline implementations are existing.
  • Visualization and analysis of the prediction performance. An outlier detection and warning system should be implemented.
  • (Optional) Implementation of neural time-series models like LSTMs.
  • Analysis and evaluation of the provided data.

1 PhD Position/4-Year Contract, May 2022, RefID: 2205WPC

1 position for a fully employed University Assistant at the Chair of Cyber-Physical-Systems at the earliest possible date or beginning on 1st of July 2022 in a full-time (100%) 4-year term of employment. Salary Group B1 to Uni-KV, monthly minimum charge excl. SZ.: € 3.058,60 for 40 hours per week (14 times a year), actual classification takes place according to accountable activity-specific previous experience.

We are looking for a motivated student interested in a PhD thesis on “Machine Learning of Robot Motor Skills”. The goal of the doctoral thesis is to develop probabilistic or deep neural networks for interactive learning of autonomous systems and industrial processes. This will involve processing complex sensor data such as from RGB-D cameras, tactile data and sensory data of the industrial machines for motion control of robotic systems. The developed models will be tested using realistic industrial applications for process modeling and with robotic systems within our AI Robot Lab.

What we offer

The opportunity to work on research ideas of exciting modern topics in artificial intelligence and robotics, to develop your own ideas, to be part of a young and newly formed team, to go on international research trips, and to receive targeted career guidance for a successful scientific career.

Job requirements

Completed master’s degree in computer science, physics, telematics, statistics, mathematics, electrical engineering, mechanics, robotics, mechanical engineering or an equivalent education in the sense of the desired qualification; Programming experience in one of the languages C++, Python or similar; Experience in robotics, with ROS, and reinforcement learning or in machine learning. Willingness and ability for scientific work in research including publications with the possibility to write a dissertation.

Desired additional qualifications

Experience with movement primitives, Keras, Tensorflow and graphical models. Basic knowledge of Linux is advantageous. Good English skills and willingness to travel for research and to give technical presentations.

Application

Application deadline: June 2nd, 2022

Online Application via: Montanuniversität Leoben Webpage (2205WPC)

The Montanuniversitaet Leoben intends to increase the number of women on its faculty and therefore specifically invites applications by women. Among equally qualified applicants women will receive preferential consideration.