Kurzinfos zu den Projektgruppen - SS 2019

Automated Human: Towards Motor Skill Learning by Self-Demonstration

1. Automated Human: Towards Motor Skill Learning by Self-Demonstration

In this project group, we investigate novel ways to control user`s movement. Learning motor skills such as playing piano, manual construction, or playing golf requires long-lasting repetitive training to get the desired level of performance. Currently, users learn these skills by repeating it with a teacher who basically corrects the motion of the user. This is done by correcting the posture and movement manually which is very time consuming. The goal of this project group is to build a system that is capable of replacing the teacher. The system should take over the control of the user and actuate him or her so that the desired movement is performed. The group will use an optical tracking system in combination with electrical muscle stimulation which will allow to gain fine grained control of the user´s muscles.

Jennifer Brings, M.Sc., Torsten Bandyszak, M.Sc., Prof. Dr. Klaus Pohl / AG Pohl

2. CPS-RU: Engineering kollaborativer cyber-physischer Systeme unter Berücksichtigung von Laufzeitunsicherheiten

Cyber-physische Systeme sind software-intensiv, eng in physische Prozesse integriert und interagieren mit dem Systemkontext über Sensoren und Aktuatoren. Überdies bilden sie Systemverbünde, in denen einzelne Systeme mit anderen Systemen kollaborieren. Der Systemverbund erlaubt es den beteiligten Systemen Ziele zu erreichen, Funktionen auszuführen und Verhalten aufzuweisen, zu denen die Einzelsysteme nicht in der Lage wären. Innerhalb derartiger Systemverbünde werden Informationen, beispielsweise über die wahrgenommene Umwelt, zwischen Kollaborationspartnern ausgetauscht. Die Erkennung, Dokumentation, Analyse und Handhabung von Uncertainty, die im Betrieb auftreten können, ist unerlässlich, da Uncertainties möglicherweise sicherheitsgefährdende Situationen verursachen.

Nils Schwenzfeier, M.Sc., Prof. Dr. Volker Gruhn / AG Gruhn

3. DeeVa: Deep Video Augmentation

Die Studierenden entwickeln eine Anwendung, die mit Machine Learning Produkte in einem Video erkennen soll und zusätzliche Informationen aus dem Netz bereitstellt.

Dr. Matteo Ceriotti, Prof. Dr. Pedro Marrón / AG Marrón

4. See'In: Seeing the Invisible Wireless Internet of Things

Have you ever had problems connecting to the Internet with your smartphone due to a weak or non-existing WiFi signal? What if you could see those wireless signals in the environment and utilise them, e.g., to find a more beneficial spot for either your phone or conversely the wireless router to provide a better signal coverage? Now think about the Internet of Things made of billions of wireless devices, whose functionality depends on wireless connectivity. This project group targets the visualisation of wireless waves emitted by IoT devices as if they were visible light ones, with the goal of supporting users in planning functional IoT systems.

After a series of lectures discussing the behaviour of wireless signals in real environment and tutorials in which the wireless modelling engine developed in prior research work will be introduced, the participants will present specific techniques of relevance for the project. In particular, the attendees will learn about how wireless signals propagate in an environment with obstacles as well as how 3D visualisation engines can be exploited to represent such information. Depending on the number of students, we will also explore the techniques that will allow to operate such visualisation on a smartphone and navigate in the wireless environment while the user moves around.

Afterwards, the participants will design, implement and experiment with a practical system in which the acquired knowledge will be exploited to prototype a visualisation engine able to represent the wireless signals in real-world scenarios.

Dr. Michael Striewe, Björn Zurmaar, M.Sc., Prof. Dr. Michael Goedicke / AG Goedicke

5. TextJACK – E-Assessment mittels automatischer Textanalyse und -generierung

In der PG sollen aktuelle Techniken der automatischen Textanalyse und -generierung ausprobiert und auf ihre Tauglichkeit für den Einsatz im E-Assessment hin untersucht werden. Ausgewählte Techniken sollen dann prototypisch im vorhandenen E-Assessment-System JACK umgesetzt werden, um konkrete neue Aufgabentypen und Feedbackmöglichkeiten zu ermöglichen.

Hagen Tarner, M.Sc., Prof. Dr. Fabian Beck / AG Beck

6. ViVaSD – Visualizing Variability in Software Dynamics

Modern profiling tools leverage visualization to make complex program behavior and performance information readable. Whereas they provide support for representing single executions, solutions to compare and summarize multiple runs are still lacking. Such solutions can potentially reveal misleading uncertainties in performance measurement, help identify performance regressions, or support a practical analysis of algorithmic complexity. In this project group, we will first assess the state of the art, both studying available tools as well as latest results discussed in the academic community. Based on the outcomes, we will then design and develop our own prototype that advances the state of the art regarding the visual comparison of executions and the visualization of performance variability. Challenges will be the efficient profiling of multiple program executions, the scalable encoding of the required information in a visual representation, and the integration of the approach into an interactive, Web-based framework.

Ausführliche Informationen zu den einzelnen Projektgruppen gibt es unter Moodle

Hinweis: Nicht-UDE-Studierende ohne UDE-Unikennung müssen sich für Moodle erst registrieren. Den Einschreibeschlüssel fragen Sie bitte bei Ansprechpartnern nach.