PhDs in the Department of Computer Sciences

Research in the Department of Computer Science is challenging. Below we provide a brief insight into the variety of topics covered by the research work of our doctoral students and list who supervises them at H-BRS. Further links lead to research institutes, cooperation partners, publications, etc.(selection, last update April 2024)
 

Saugata Biswas, IVC
Online remote training for assembly, operation and maintenance is advantageous in the industry to save time and money, especially when corporations have locations on several continents (e.g. automotive industry, oil companies, etc.).Currently, remote training experiences are restricted by limited camera perspectives and the lack of suitable navigation techniques for camera view depending on the task.In this PhD project, PhD student Saugata Biswas is developing an autonomous camera viewpoint management system using a robotic arm in a multi-camera training scenario.This PhD will focus on improving the online training experience of remote trainees.PhD student Saugata Biswas has been a scholarship holder of the Institute for Visual Computing (IVC) since 2019.
Supervision:  Prof. Dr. Ernst Kruijff

Ahmad Drak, TREE
Ahmad Drak is developing a flying robotic system that is capable of efficiently exploring the ever-changing environment in which it moves. The result is a wealth of useful information that the system is designed to learn and maximize. This firstly shortens the time the robot needs to explore its environment and secondly reduces the energy consumption of the robotic system. PhD student Ahmad Drak has been a fellow of the Institute for Technology, Resource Efficiency and Energy Efficiency (TREE) since 2018. Further information.
Supervision: Prof. Dr. Alexander Asteroth

Martin Pluisch, IVC
Computer scientist Martin Pluisch is working on augmented reality environments in his dissertation. He is researching the use of stimuli via various sensory channels, for example vibrations or sounds, to provide users of AR glasses with additional information alongside what they are seeing. The dynamic environment as well as technical and human limitations are challenging.
Supervision: Prof. Dr. Ernst Kruijff

Michal Stolarz, A2S
Michal Stolarz works on assistance robots that are used in the social sector, for example in the care of the elderly or in therapy for children. These robots must first learn to react appropriately to human behavior and facial expressions. To do this, the computer scientist uses the Interactive Reinforcement Learning (IRL) method in simulations and user studies so that the robot learns from both the carer and the person being cared for.
Supervision: Prof. Dr. Teena Chakkalayil Hassan

Christoph Pomrehn, IVC
Raman and infrared spectroscopy are two different optical measurement methods. They are used in conjunction with a microscope to examine the smallest microscopic samples. The samples are examined with regard to their surface properties, but also with regard to their material composition. Under certain conditions, the images generated (hyperspectral images) can contain complementary information about the same sample. As part of this research topic, Christoph Pomrehn is developing strategies to extract this information from the generated images and make it accessible for computer-aided analysis in selected applications. Christoph Pomrehn has been a scholarship holder of the Department of Computer Science since 2018.
Supervision: Prof. Dr. Rainer Herpers

Sven Schneider, A2S
In contrast to industrial robots, we expect domestic robots to be able to perform gripping movements with their arms depending on the situation. However, they first have to learn this ability with the help of models from various disciplines (e.g. mechanics or control engineering). PhD student Sven Schneider provides household robots with this interdisciplinary knowledge by developing application-specific languages. If you like, he is an interpreter and language teacher for domestic robots.
Supervision: Prof. Dr. Nico Hochgeschwender

Bastian Stahl, ISF
As a computer scientist, Bastian Stahl is investigating how he can improve the reliability and safety of 3D time-of-flight cameras. These camera systems emit a modulated light signal and use the phase shift between the transmitted and received signal in the individual pixels to generate a depth image. They are used, among other things, in the field of occupational safety and are intended to prevent accidents in the interaction between man and machine.
Supervision: Prof. Dr. Robert Lange, Prof. Dr. Norbert Jung

Katharina Stollenwerk, IVC
Lower back pain is an important issue in modern Western societies. Wearable devices can be used to track postural changes through posture training, measure the shape of the spine and reconstruct spinal curvature. PhD student Katharina Stollenwerk's research focuses on the systematic evaluation of (specific) sensor-based wearables and the reconstruction of spinal curvature as well as the analysis of the recorded data. This improves posture training in back training with reliable and objective measurements of the shape of the spine. Trainers and trainees thus gain a better understanding of their actions and concepts.
Supervision: Prof. Dr. André Hinkenjann

Santosh Thoduka, A2S
Robots are usually programmed to perform tasks by following a list of actions such as moving, looking, picking, etc. When something unexpected happens, a robot often cannot handle the situation because 1) it did not recognize that something went wrong and 2) it was not programmed for the new situation. Recognizing such situations allows robots to decide whether to continue with the task, inform a human about the problem or try to solve the problem themselves. Santosh Thoduka's work focuses on using the robot's camera to detect unexpected situations that may occur while the robot is performing a task. PhD student Santosh Thoduka has been a fellow of the Graduate Institute since 2018.

Further information
Supervision: Prof. Dr. Paul Plöger

Christina Trepkowski, IVC
Augmented reality glasses are data glasses in whose field of vision all conceivable information is visually superimposed at the same time. This information is intended to improve awareness of certain situations through correct perception, interpretation and assessment of the environment. However, current AR glasses have a disadvantage: their field of vision is so small that the information displayed can obscure critical information from the environment, distract the wearer or overwhelm them with too much information. Like Alexander Marquardt, Christina Trepkowski is also working on converting some of the visual digital information into audio and vibration stimuli. As a psychologist, her focus is on evaluating, comparing and optimizing these new methods by developing and using procedures to measure the situational awareness of people wearing glasses. Christina Trepkowski has been a fellow of the Institute of Visual Computing (IVC) in the 3DMi working group since 2018.
Supervision: Prof. Dr. Ernst Kruijff

Mohammad Wasil, A2S
Autonomous systems and robots operate in dynamic environments that change gradually or radically over time. To perform tasks in such ever-changing environments, the robot should continuously learn and adapt its capabilities. Continuous learning (CL) is a machine learning method that learns from continuous streams of data. CL aims to balance the trade-off between the capabilities to be maintained and those to be enhanced, a characteristic that current machine learning struggles with. This PhD thesis focuses on using the physical interaction of a robot to utilize spatio-temporal information about the environment to obtain more robust data to deal with the continuous learning problem. 
Supervision: Prof. Dr. Sebastian Houben

Youssef-Mahmoud Youssef
Distributed robotic systems are used in many industrial applications, e.g. in intelligent warehouses, in different areas of logistics or in the disposal of nuclear and hazardous waste. To ensure the reliability of these complex systems, it is necessary to develop intelligent, explainable models that efficiently describe the different behaviors of the robots. PhD student Youssef-Mahmoud Youssef is investigating the fault detection and diagnosis of distributed robotic systems using explainable hypotheses. Youssef Mahmoud Youssef has been a scholarship holder of the Department of Computer Science since 2019.
Supervision: Prof. Dr. Martin Müller