For an updated and broader list of available thesis topics visit the RAI (Robots, Agents, Interaction) group page.
———————————————————————————————————————-
Please see below for a list of possible BSc or MSc topics. If you already have a clear vision on what your thesis project could look like – and you think it aligns well with our research interests – then please feel free to contact us to talk about your own ideas.
[wpex more=”Read more” less=”Read less”]
The following tags give a quick overview of the type of topic:
- #BSc suitable for a bachelor thesis
- #MSc suitable for a master thesis
- #Intership suitable for an internship
- #application writing a control program to solve a problem on hardware
- #hardware includes building new hardware
- #theory should extend the mathematical model or deliver proofs
- #simulation writing a control program to solve a problem in simulation
[/wpex]
Robot Vacuum Cleaners [joint project with PHILIPS!]
#MSc #Internship
(offered by Gerhard Weiss and Rico Möckel )
[wpex more=”Read more” less=”Read less”]
To invent the next generation of cleaning robots we’re interested in exploring the following interesting topics.
- Designing a robot that converses logically on relevant topics (cleaning) – Based on existing “chat engines” we’d like to see what’s possible.
- User Perception of intelligence and influencing factors – What should a robot do, and not do, to be perceived as intelligent? Of course this topics needs user studies and Leaning towards psychology.
- Designing low cost navigation based on sensor integration – Combining dead reckoning with existing ‘beacons’ like BT/WiFi signals, light, sound, temp, etc.
- Design of an adaptive cleaning strategy to increase the autonomy and cleaning efficiency while an RVC is used longer in a certain household situation. Using learning algorithms, the robot should learn to prevent getting stuck, or to go to the dirt areas
[/wpex]
Face-based Human-Robot Interaction [New!]
#MSc #Intership #robotics #application (offered by Stylianos Asteriadis)
[wpex more=”Read more” less=”Read less”]
Human-Machine Interaction, nowadays, is already taking advantage of the rapid developments in the area of Natural Interaction, making use of tools such as Computer Vision and Artificial Intelligence. Visual cues, related to facial analysis, can be a valuable source of information regarding a person’s emotional and/or cognitive state, while the recent explosion in computational power has paved the path for a new boost in machines’ capabilities to employ well-known and established algorithms like Active Shape Models and state-of-the-art variants for tracking facial features’ location, as well as for inferring expressivity features. Such instruments can be of great use for robust, accurate and real time Facial Expression Recognition (FER) and/or Visual Focus of Attention (VFoA) estimation in Assisted Living environments, allowing, thus, an AI-endowed robot to react promptly to human everyday needs.
The proposed work will focus on the development of cutting-edge interaction schemes in the brand new area of interaction between humans and robots, in non-verbal, contactless manners, just through visual cues conveying affective messages (emotions, cognition). Possible applications lay in the areas of medicine, education, game-play, etc.
Requirements: Good C++ programming skills, interest in research
[/wpex]
Community detection in airline/airport networks [New!]
#BSc #application #simulation (offered by Jerry Spanakis)
[wpex more=”Read more” less=”Read less”]
European airport network can be represented by a complex network where nodes are the airports and edges represent that there is a specific connection between the corresponding airports. The goal of the thesis is to explore several weighting schemes about the edges (available seats, available flights, combination of both e.t.c.) in order to be able to use network/graph analysis tools for extracting useful knowledge (i.e. detection of groups of airports or identification of network similarities between airlines (partners or rivals)).
Data will be real from the European airport schedules (already available).
Requirements: good programming skills, interest in visualization and networks
[/wpex]
Understanding affect during human-computer interactions [New!]
#MSc #application #theory (offered by Stylianos Asteriadis)
[wpex more=”Read more” less=”Read less”]
Context: Human-computer interaction is a rapidly evolving area of computer science, where the human is in a central position with regards to how interfaces, learning materials and games should be designed and presented to the final end-user, in order to match personal preferences. However, end-users usually have limited capabilities in adapting interactions with a machine
according to their current mood, emotional and cognitive states.
Focus: The goal of this thesis is to develop a tool for analyzing user (facial) reactions to certain events during HCI sessions. Existing computer vision algorithms will be tested and their output(s) will be correlated with Human-Machine Interaction logs. To that aim, existing datasets will be used. The final outcome of this thesis will be to infer visual patterns associated with interaction events and distinguish between generic and personalized affective features.
For inspiration and further material, click here
Requirements: Good C++ programming skills, interest in research
[/wpex]
Markovian modelling of brain signals [New!]
#BSc #application #theory #brain #HMM (offered by Kirill Tumanov)
[wpex more=”Read more” less=”Read less”]
In this thesis project you will be exploring how to design, build and evaluate a Hidden Markov Model (HMM), applied to brain signal analysis. Markovian approach assumes that there is a certain state regularity observed in the analized data. But what exactly is this pattern? So far HMMs were extremely successful for instance in speech and handwriting recognition. It has also been found useful in bioinformatics and brain signal temporal analysis.
At the moment a standard Support Vector Machine (SVM) classification is used at UM when working with functional Near-InfraRed Spectroscopy (fNIRS) system for brain recording. SVM will be a banchmark for you to test the designed HMM performance against. Recent research indicated a high potential of HMM to replace SVM in this domain. You will be able to test if this is the case, while deepening your knowledge in Markovian approaches in general and the HMM in particular.
Requirements: Knowledge of Markov models, interest in time-series data analysis.
[/wpex]
Pattern Discover in Multivariate Time Series [New!]
#MSc #application #theory #simulation (offered by Jerry Spanakis)
[wpex more=”Read more” less=”Read less”]
Multivariate time series data are widely collected from various domains (industry, medical information, financial systems, social networks e.t.c.). Effective highlighting of patterns in multiple time series can be used for decision support, risk prevention, event detection e.t.c.
The goal of this thesis is to find (effective) ways of discovering sequences of events (patterns) that lead to specific events. The developed method(s) will be applied to real data about eating behavior and the goal will be to predict eating events.
Requirements: (unconditional) love for programming, interest in data mining
[/wpex]
Market basket analysis using product networks [Reserved]
#MSc #application #theory #simulation (offered by Jerry Spanakis)
[wpex more=”Read more” less=”Read less”]
Market basket analysis is one of the oldest research areas in data mining. The goal of this task is to reveal useful associations between customer transactions which then can be used for more effective marketing of products (collective offers, targeted placing or advertisements e.t.c.). Frequent itemsets for finding association rules like «80% of customers that bought cereal and sugar, also bought milk» are the baseline approach which has several drawbacks. Other approaches include graph based methods (like community detection e.t.c).
The goal of this thesis is to find (effective) ways of discovering association rules using both traditional and graph based and will be tested using real-life data concerning transactions but also data related to what people tend to eat together.
Requirements: (unconditional) love for programming, interest in data mining
[/wpex]
Application of deep learning to short text analysis [Reserved]
#MSc #application #theory #simulation (offered by Jerry Spanakis)
[wpex more=”Read more” less=”Read less”]
The explosion of WWW has marked a new age for the human society. The hug use of Social Media such as Facebook , LinkedIn and Twitter offers a place for people to share information in real time. The special characteristics of these (textual) data (besides their informality) are that they are shorter, nosier and sparser so that traditional data NLP/data mining techniques need to be revisited. Nevertheless, the analysis of their content (sentiment, emotion, clustering e.t.c.) is attracting more attention.
Deep learning has been recently applied to many NLP tasks yielding promising results. The goal of this thesis is to explore the effect of deep learning techniques (such as unsupervised word vector learning) to a corpus of short text data (in Dutch!) and find effective ways of analyzing these texts (clustering, polarity analysis e.t.c.).
Requirements: background in text retrieval, text/data mining (not necessary but will help), (unconditional) love for programming
[/wpex]
RoboEthics –- Exploring the intersection of robotics and ethics [Reserved]
#MSc #Intership #robotics #ethics (offered by Gerhard Weiss in cooperation with prof. dr. Tsjalling Swierstra )
[wpex more=”Read more” less=”Read less”]
Context: Robot technology is considered a core technology of the 21th century with tremendous impact on economy, society, and our everyday life. The rapid advancement of robotics, and especially intelligent and autonomous robotics, raises a number of ethical and moral questions (as sometimes thematized in science fiction movies such as “I, Robot”, “Terminator” and “Blade Runner”) — how to design appropriate robot behavior, will future robots be able to act ethically, can ethics be programmed, should next-next-generation robots have rights comparable to human rights, will humans have moral duties toward future robots, do robots influence our own ethical understanding, and so forth? This thesis deals with the intersection of robotics and ethics. The exact focus and addressed research question(s) is jointly determined by the supervisors and the student.
Requirements: Interest in ethical questions raised by robotic technology.
Further Information: Please mail to Gerhard Weiss
[/wpex]
Automated task assignment based on negotiation and contracting
#MSc #Intership #simulation (contact Gerhard Weiss)
[wpex more=”Read more” less=”Read less”]
Context: A problem of significant practical relevance is the so-called task assignment problem (TAP), that is, the problem of assigning tasks to executing agents so that certain performance criteria are fulfilled such as minimization of overall execution time, minimization of overall costs and maximization of the average execution quality. The TAP is non-trivial (actually, it is NP-hard in general), and challenging instantiations of the TAP can be found in numerous application domains, ranging from industrial manufacturing to networked computing.
Focus:This project deals with the question how task assignment can be automated in applications where the executing agents are computational entities such as software programs or robots. Thereby the project focuses on TAP scenarios which show the following basic characteristics: the tasks can be divided into sub-tasks; the (sub-)tasks differ in the demands their completion raises; and the agents differ in their abilities to complete (sub-)tasks. The goal is to develop, implement and experimentally investigate a flexible and robust solution for this type of TAP scenarios which follows the idea that the executing agents autonomously decide and agree on the assignment of the (sub-)tasks on the basis bilateral negotiations and contracts. The project has available approaches to automated task assignment as its starting point and builds on related state-of-the-art work from the fields of artificial intelligence and multi-agent systems.
Further Information: Please get in touch with Gerhard Weiss.
[/wpex]
Automated concurrent negotiation
#MSc #Intership #simulation (contact Gerhard Weiss)
[wpex more=”Read more” less=”Read less”]
Context: The progressive establishment of e-commerce, e-business, e-trading and e-markets has led to a great practical interest in automated negotiations among software agents (think of “electronic buyers and sellers” of a good). This, in turn, has triggered various theoretical and empirical research efforts worldwide on all aspects of automated negotiation.
Focus: Whereas automated bilateral (one-to-one) negotiations have been intensively studied and is pretty well understood today, automated concurrent (one-to-many and many-to-many) negotiation is still an open challenge. This thesis concentrates, in the context of concurrent negotiation, on the core question what negotiation strategy an agent should apply in order to achieve the best possible outcome. This question is difficult to answer, especially if there are further constraints and conditions such as limited negotiation time, limited knowledge about the negotiating opponents, and the possibility to withdraw offers.
Basic requirements:: Good programming skills; interest in experimental research (simulation studies).
Further Information: Please get in touch with Gerhard Weiss or Siqi Chen.
[/wpex]
How to Trade Energy in a Smart Way? [Reserved]
#MSc #Intership #smart energy #automated trading #simulation (offered by Karl Tuyls and Gerhard Weiss)
[wpex more=”Read more” less=”Read less”]
Context: The energy market gets more and more decentralized and dynamic. New sources of energy (wind and solar) play an increasingly important role, energy needs to be transported from producers to consumers over possibly large distances, and energy supply and consumption is to be traded in real time. Handling all this complex market in a smart way is one of the key challenges facing modern societies and economies.
Focus: The focus of this MA thesis is on exploring efficient and effective strategies for trading energy in nontrivial energy supply-demand scenarios as can be simulated via PowerTAC (http://www.powertac.org/). PowerTAC is an advanced state-of-the-art simulator for energy trading that also used for running yearly international “power trading competitions”.
Requirements: Good programming skills; interest in experimental research; ability to use PowerTAC; and, optionally, interest in participating in the PowerTAC trading competition.
Further Information: For further information, please get in touch with Gerhard Weiss
[/wpex]
Analysis of the Social Networks in Chess!
#BSc #simulation (offered by Bijan Ranjbar-Sahraei)
[wpex more=”Read more” less=”Read less”]
Analysis of social networks has emerged as a key technique in modern sciences ranging from sociology, biology and economics to communication studies and information science. Application of social network analysis include but are not limited to studying evolution of behaviors using Facebook and Twitter data, analysis of food webs and fraud detection!
In this project we look into chess which is one of the oldest and yet very complex games ever invented…
Want to know more! Drop by in my office or send me an email. Then, we’ll discuss it more in detail!
Requirements: Knowing Chess!; Good Python programming skills; Interest in Data Mining and Social Network analysis;
[/wpex]
Outdated Thesis Titles
[wpex more=”Read more” less=”Read less”]
Applications of Tree Learning Search
#MSc #Monte-Carlo Tree Search #Decision Tree Induction #Robotics (contact Kurt Driessens)
Beating the unbeatable game Arimaa
#MSc #Nearest Neighbour #Monte Carlo Tree Search (contact Kurt Driessens and/or Mark Winands)
Unmasking Batman
#MSc #Autoencoders #Deep Learning #Noise removal (contact Kurt Driessens)
BookFace –- Smart Webpages for Textbooks (>>Facebook for Textbooks<<)
#MSc #Intership #web design (offered by Gerhard Weiss)
Population Reconstruction Using Visual Techniques
#BSc #application #simulation (offered by Bijan Ranjbar)
Toward Soft Heterogeneity in Robotic Swarms
#BSc #MSc #application #theory #simulation (contact Bijan Ranjbar)
Emergence of Social Networks in Presence of Competition Between Individuals
#BSc #application #theory #simulation (contact Bijan Ranjbar)
Network-based Entity Resolution
#MSc #application #theory #simulation (offered by Bijan Ranjbar)
Visualization of Large-Scale Genealogical Datasets
#BSc #application #theory #simulation (offered by Bijan Ranjbar)
Applications of Tree Learning Search (2)
#BSc #Monte-Carlo Tree Search #Decision Tree Induction #AI Competition (contact Kurt Driessens)
Predicting Pseudo-Random Behavior in Professional Sports
#BSc #Data Mining #Machine Learning #Sequences (contact Kurt Driessens)
[/wpex]