The GelSight Mini is an affordable, compact tactile sensor that provides high-resolution 2D and 3D surface mapping. Its plug-and-play design enables users to begin operations within five minutes of setup. Key features include:

• High-Performance Mapping: Delivers detailed 2D and 3D surface imagery, capturing fine textures and features.

• AI Integration: Compatible with existing computer vision techniques and software, facilitating AI training and research.

• Compact and Robust Design: Easily integrates into various robotic systems and withstands rigorous prototyping environments.

• Quick Gel Replacement: Allows toolless and rapid gel changes, enhancing efficiency during iterative testing.

The GelSight Mini is suitable for professionals, academics, and hobbyists interested in touch-based interfaces, robotics, and industrial scanning. Its user-friendly interface and affordability make it an accessible tool for advancing tactile sensing applications

Videos

• Research videos using the robot will be presented here. 

Publications

The DIGIT sensor is a compact, low-cost, and high-resolution vision-based tactile sensor designed for in-hand robotic manipulation tasks. It improves upon traditional tactile sensors by offering a smaller form factor, enhanced durability, and streamlined manufacturing, making it suitable for multi-fingered robotic hands. DIGIT utilizes an elastomer surface to measure contact forces via image deformation captured by an embedded camera, delivering precise tactile feedback. Its modular design allows easy replacement of components, supports task-specific elastomers, and ensures robustness under repeated use. With a cost of approximately $15 per unit in batch manufacturing, DIGIT provides an accessible and effective solution for tactile sensing in robotics.

Videos

• Research videos using the robot will be presented here. 

Publications

The LEAP Hand is a low-cost, efficient, and anthropomorphic robotic hand designed for dexterous manipulation and robot learning. The hand is robust, durable, and capable of exerting large torques over extended periods. With a novel kinematic structure that retains all degrees of freedom in any finger position, it supports a wide range of manipulation tasks, including grasping, teleoperation, and in-hand object rotation. The LEAP Hand is open-source, with detailed assembly instructions, simulation tools, and APIs, making it accessible and scalable for research and development.

Videos

• Research videos using the robot will be presented here. 

Publications

Short Bio

Dr. Christian Rauch is a Research Scientist at Bosch Corporate Research in the Robot Learning group. He received his Ph.D. degree in 2020 in Robotics and Autonomous Systems at the University of Edinburgh. He has over 10 years of experience in applying machine learning techniques to robotic perception and manipulation tasks in unstructured environments. In his current research, he focuses on the application of Foundation Models to robotic manipulation tasks in industrial and domestic contexts.

• Google Scholar
• GitHub
• LinkedIn

Research Interests​

Machine learning, Robot Learning, Robot Vision, Perception and Manipulation.

Contact

Dr. Christian Rauch
Research Group Leader at the Chair of Cyber-Physical-Systems
Montanuniversität Leoben
Franz-Josef-Straße 18, 
8700 Leoben, Austria 

Phone:  +43 3842 402 – 1901
Email: Christian.Rauch@unileoben.ac.at
Chat: https://unileoben.webex.com/meet/christian.rauch

Selected Publications

Student Assistant at the Montanuniversität Leoben

Short bio: Richard Marecek started at CPS in October  2024.

He is a bachelor student at Montanuniversität Leoben in the program Industrial Data Science. His research interests include industrial process modeling based on large sensor arrays and machine learning algorithms. 

Research Interests

• Industrial Process Modeling 
• Machine Learning
• Python Programming

Contact

Richard Marecek
Student Assistant at the Chair of Cyber-Physical-Systems
Montanuniversität Leoben
Franz-Josef-Straße 18,
8700 Leoben, Austria 

Email: richard.marecek@stud.unileoben.ac.at

Student Assistant at the Montanuniversität Leoben

Short bio: Rino Morina started at CPS in October  2024.

Rino Morina is a bachelor student at Montanuniversität Leoben in the program Industrial Data Science. His research interests include data modeling based on machine learning algorithms including neural networks. 

Research Interests

• Data Modeling 
• Machine Learning
• Python Programming

Contact

Rino Morina
Student Assistant at the Chair of Cyber-Physical-Systems
Montanuniversität Leoben
Franz-Josef-Straße 18,
8700 Leoben, Austria 

Email: rino.morina@stud.unileoben.ac.at

Student Assistant at the Montanuniversität Leoben

Short bio: Julia Schmelz started at CPS in October  2024.

Julia Schmelz is a master student at Montanuniversität Leoben in the program Industrial Data Science. Her research interests include data modeling based on machine learning algorithms including neural networks. 

Research Interests

• Data Modeling 
• Machine Learning
• Python Programming

Contact

Julia Schmelz
Student Assistant at the Chair of Cyber-Physical-Systems
Montanuniversität Leoben
Franz-Josef-Straße 18,
8700 Leoben, Austria 

Email: julia-maria.schmelz@stud.unileoben.ac.at

This post is intention to help future international PhD students to ease their enrollment process.

Step 1: Find a mentor and prepare your documents

You need to find a topic, working title, individual curriculum and supervising team (supervisor and mentor) yourself to be able to fill out the form. According to our curriulum for the doctoral studies the mentor needs to have a venia docendi. 

If your degrees are not from a EU country, you need a full authentication or an Apostille of your documents (and the translations, if your documents are not in English), please inform yourself if you’d need either an authentication or Apostille for Austria! This is valid for both, Master’s documents as well as bachelor’s!

If you can not obtain an authentication/Apostille we also accept a translation of your 4 academical documents by an officially licenced translator in Austria instead.

Step 2: Applications form

You have also required to fulfil 20 ECTS, they have to be split in main and secondary subject:

• Courses that don’t have anything to do with your subjects (language courses etc.) you can do up to 4 ECTS

• Privatissima (talking to your prof, SE type) for a maximum of 4 ECTS

• There is no limit on IV courses

• And you can also have stuff (eg. summer school) from other unis, however you have to go through the equality process (they will check if it really is equal to the ects here)

These stated in the curriculum, which is in German.

Example course:

Step 3: Online registration for graduates of Non-Austrian University

Create a basic account in MUonline of the university. You have to create a student account even you have a staff account.

p/s: Sign out your staff account when you’re in the process of creating a basic account.

Detailed admission process at here.

Step 4: Waiting and make appointment to verify the real document

Self-made, dexterous 5-finger hand

We use a adult-sized robot hand for learning grasping and object manipulation skills. The hand is mounted on our FRANKA EMIKA Panda robot. 

The hand has 19 degrees-of-freedom and uses 8 smart actuators for precise control (actuators contained inside the unit).

Under actuated design aims to provide the right balance between fine control and conformance to the shape of the objects.

Webserver GUI for Control

Within a student project, an ESP32 based web-server was developed for controlling the hand, see the git repository. 

Videos

• Research videos using the robot will be presented here. 

Publications

Course Content

In the first week, advanced machine and deep learning methods like multi-layer-perceptrons, convolutional neural networks, variational autoencoder, transformers, simultaneous navigation and mapping approaches, and more will be presented.

These methods can be tested using interactive tools like for example using   https://playground.tensorflow.org. To deepen the knowledge, students will answer well-crafted scientific questions using latex handouts alone or in teams of two students in the lecture room. 

Additionally, Jupyter notebook files were prepared to implement advanced machine and deep learning approaches without installing any software. For all participants of the course user accounts will be created using our JupyterHub at https://jupyter.cps.unileoben.ac.at. The accounts will remain active till the end of the semester. 

Prerequisites & If you Miss Course Contents

During the first week, a laptop or tablet will be needed to use the interactive tools and the Jupyter notebooks. 

Webex Online Sessions of the 1st Week

Find here the link to the online stream during the first week in October, 2024: https://unileoben.webex.com/unileoben/j.php?MTID=m5492385776dd885ca5dde72e52563c61

When you miss some course contens

If you miss some course contents due to overlapping events, you can watch recordings of the sessions online. All recordings will be hosted via Moodle at https://moodle.unileoben.ac.at/course/view.php?id=3082.

Course Description

Modern machine learning methods and in particular deep learning methods are entering almost all areas of engineering. 

The integrated course enables the students to apply these methods in the application domains of their study.

For this purpose, current problems from the industry are investigated and the possibilities of machine and deep learning methods are tested.

Students gain a deep understanding of method implementations, how data must be prepared, which criteria are relevant for selecting learning methods, and how evaluations must be performed in order to interpret the results in a meaningful way.

Initially, the basics of learning methods are developed in 5-6 lecture units. Then, students select one of the listed industrial problems and work on it alone or as a team (with extended assignments). The project work is accompanied by weekly tutorials with tips and tricks. Finally, the project results are discussed in a written report and presented for a final 10-15min.

Grading is based on the quality of the code, the report, and the final short presentation.

Among others, one of the following industry problems can be chosen:

1. Application and comparison of deep neural networks for steel quality prediction in continuous casting plants with data from the ‘Stahl- und Walzwerk Marienhütte GmbH Graz’.

2. Predictive maintenance of bearing shells using frequency analysis in decision trees and deep neural networks based on acoustic measurement data.

3. Motion analysis and path planning for human-machine interaction in logistics tasks with mobile robots of the Chair of CPS.

4. Autonomous navigation and mapping with RGB-D cameras of the four-legged robot Unitree Go1 for excavation inspection in mining.

The project list is continuously extended.

Links and Resources

• MUOnline
• Link to the ML Book by Univ.-Prof. Dr. Elmar Rueckert
• Project submission template
• Handout on Neural Networks
• CPS software dashboard

Location & Time

• Location: HS 3 Studienzentrum
• Dates: 01.10.2024 – 07.10.2024, see the course schedule above.

• Location: Digital Science Center (Roseggerstraße 11, 8700 Leoben)
• Date: 22.01.2025, 10:15 – 15:15, final presentation

Kickoff meeting of project

All meetings will be conducted at CPS chair. The time please refer to the email, contact us if reschedule is needed.

Previous Knowledge Expected

Formal Prerequisite: Basic Python programming skills, Fundamentals of Statistics.

Recommended Prerequisites: Introduction to Machine Learning (“190.012” and “190.013”).

Slides

• 01.10.2024
  • Data Modelling Fundamentals
  • Course Organization & Grading
  • Jupyter Notebook on Python Basics and Plotting Tricks
  • Projects Teaser
• 02.10.2024
  • Introduction to Multi-Layer-Perceptrons and Deep Learning  [Video] [Slide]
  • Interactive Session on MLPs using https://playground.tensorflow.org  [Video]
  •  
• 03.10.2024
  • Introduction to Machine Learning
  • [Talk] Deep Representation Learning [Video][Slide]
  • [Talk] Adversarial Machine Learning
• 04.10.2024
  • How to git
  • [Talk] Intro to Robotics & SLAM & LLMs
  • Quiz
  • Jupyter Notebook on Path Planning & Object Classification
• 22.01.2024
  • Presentations

Learning objectives / qualifications

• Implement or independently adapt modern machine learning methods and in particular deep learning methods in Python.
• Analyze data of complex industrial problems, process (filter) the data, and divide it into training- and test data sets such that a meaningful interpretation is possible.
• Define criteria and metrics to evaluate evaluations and predictions and generate statistics.
• Develop, evaluate, and discuss meaningful experiments and evaluations.
• Identify and describe assumptions, problems, and ideas for improvement of practical learning problems.

Grading

Continuous assessment: During the lectures and the tutorials 0-20 bonus points can be collected through active participation.

Project assignments: Alone or in small groups (2-3 students) one of the listed projects has to be implemented. A written report in form of a git repository wiki page have to be submitted.
– For the implementation (Python Code) 0-40 Points can be obtained.
– For the wiki page report, 0-60 Points will be given.

Grading scheme: 0-49,9 Points (5), 50-65,9 Points (4), 66-79 Points (3), 80-91 Points (2), 92-100 Points (1).

With an overall score of up to 79%, an additional oral performance review MAY (!) also be required if the positive performance record is not clear. In this case, you will be informed as soon as the grades are released. If you have already received a grade via MU online, you will not be invited to another oral performance review.

Literature

Machine Learning and Data-modelling:

– Rueckert Elmar 2022. An Introduction to Probabilistic Machine Learning, https://cloud.cps.unileoben.ac.at/index.php/s/iDztK2ByLCLxWZA

– James-A. Goulet 2020. Probabilistic Machine Learning for Civil Engineers. MIT Press.

– Bishop 2006. Pattern Recognition and Machine Learning, Springer.

Learning method Programming in Python:

– Sebastian Raschka, YuxiH. Liu and Vahid Mirjalili 2022. Machine Learning with PyTorch and Scikit- Learn. Packt Publishing Ltd, UK.

– Matthieu Deru and Alassane Ndiaye 2020. Deep Learning mit TensorFlow, Keras und TensorFlow.js., Rheinwerk-verlag, DE. 

Problem specific Literature:

– B. Siciliano, L.Sciavicco 2009. Robotics: Modelling, Planning and Control, Springer.

– Kevin M. Lynch and FrankC. Park 2017. MODERN ROBOTICS, MECHANICS, PLANNING, AND CONTROL, Cambridge University Press.

– E.T. Turkogan 1996. Fundamentals of Steelmaking. Maney Publishing,UK.