Student projects

The deadline for half-time reports is Wednesday, 15th of March (using this form) and the half-time presentations will be scheduled in week 12.

Half-time presentations should be 20 minutes long (plus ~10 minutes for questions) and must cover the goals/objectives, expected contribution/novelty, results achieved so far, and a refined plan on how to proceed.

Please remember that the report must be approved by supervisors first, so a reasonable schedule is: send the report to supervisors around the 1st of March; around the 8th of March, you get feedback; then you have a week to address the comments.

Final presentations will be in week 22, which means the deadline for supervisor-approved reports will be early in week 21.

Information about MSc Thesis process

This page contains some information about the MSc thesis project (courses DT7001 and DT7002 and ET7002).

MSc thesis project information (the LaTeX template can be found here) and information about accessibility and the official front cover you should use

MSc thesis Introductory lecture: slides from 2022.10.03 and video recording (from a few years back, but mostly still applicable)

MSc thesis Course Description

MSc thesis Grading Criteria

DT7001 course syllabus and DT7002 course syllabus

Plagiarism course: https://academy.sitehost.iu.edu/index.html

Useful resources for your MSc Thesis process

A YouTube course from Lund University on academic writing (in English and in Swedish)

Teaching Technical Writing Using the Engineering Method course from Tufts University

How to write an abstract

A checklist based on common issues:

• The link to contemporary development within the area is often weak; theses rarely discuss international research sufficiently in-depth (use the references poorly)

• Clear research questions are often missing (having a clear research question that is answered in the report usually results in a much clearer evaluation and conclusions.)

• The text must argue for the research questions, hypothesis, methods and results; often, there is no argument to support the choice.

• Citations should not be used as a word in a sentence, e.g., "as [1] says ..."

• Make sure your figure/table captions are informative; all figures must be references in the text

• The abstract should be clear, specifying exactly what is to be investigated and why

• Results should be put in context to the previous works that they identified in the report

• Social aspects of the work are rarely discussed

Current Proposals of Msc and Bsc Project

	Supervisors	OneLineSummary
A Reliable IoT Messaging Protocol Based on MQTT Standard	Mahdi Fazeli	In this project, we will modify the well-known IoT protocol, i.e., MQTT to consider a topic-based reliability strategy between the broker and subscribers.
AI for sustainability management and reporting	Eric Järpe Cristofer Englund	Connect AI models to sustainability / ESG (social, environmental, and governance) to enable automated processing related to EU sustainability regulations.
AI-driven Automotive Service Market Logistics	Sławomir Nowaczyk TBD	This project, in collaboration with Volvo Logistics, focuses on using state-of-the-art methods based on meta-learning to improve demand forecasting, inventory management and spare parts availability at Volvo dealers and warehouses.
Action Library for Robot Execution	Eren Erdal Aksoy	Action Library for Robot Execution
Adapt LoCoMotif to forklift data	Kunru Chen ...	LoCoMotif is a novel TSMD method able to discover motifs that have different lengths (variable-length motifs), exhibit slight temporal differences (time-warped motifs), and span multiple dimensions (multivariate motifs)
Adaptive Obfuscation Techniques for Privacy- Preserving Machine Learning in IoT Edge De- vices	Mahdi Fazeli	This master thesis project focuses on developing an adaptive obfuscation frame- work for protecting multi-modal data in resource-constrained IoT environments.
Analysing comments (NLP) for Malware Analysis	Pablo Picazo-Sanchez	Analysing the comments of users in the WebStore to look for malware patterns
Analysis of industrial time series	Hadi Fanaee	studying the recent advances in time series forecasting and their application in modelling time series of Alfa Laval's industrial machines
Analyzing Privacy Policies (NLP) -- Malware Analysis	Pablo Picazo-Sanchez	Analyzing Privacy Policies (NLP) -- Malware Analysis
Anomaly Detection for Time Series using Diffusion Approaches	Sławomir Nowaczyk & TBD	Development of Anomaly Detection techniques based on diffusion models (instead of autoencoders) for time series data
Automated Inference regarding Goals in Elite Football Data	Andreas Summrina Kunru Martin	Automated Inference regarding Goals in Elite Football Data
Automatic Idea Detection from social media for Controlling and Preventing Healthcare-Associated Infections (with funding opportunity)	Fabio Gama Mahmoud Rahat Peyman Mashhadi	This project aims to use advanced NLP tools to automatically detect interesting ideas by processing text available in the medical forums to address the Healthcare-associated infections problem in the hospitals
Be The Change: Video Analysis for Environmental Sustainability	Sławomir Nowaczyk Zeinab Shahbazi	Analysis of online climate change video contents and identification of video features rendering a video ‘effective’ using machine learning techniques
Blockchain for polls and elections	To be decided (contact Slawomir Nowaczyk for more details)	Today, there is no blockchain solution that meets the requirements for polls/elections; therefore, we would like to develop our own
Body posture alignment feedback using xAI	Cristofer Englund Fernando Alonso-Fernandez	Extracting the body alignment in different postures and giving feedback to reduce harm during physical exercise
Clock Glitch Attacks on Embedded IoT Devices: An FPGA-Based Exploration	Mahdi Fazeli	this thesis aims to provide a comprehensive understanding of the vulnerabilities and potential countermeasures associated with clock glitch attacks on FPGA based IoT devices
Concept Re-identification to Explain Online Continual Learning	Sepideh Pashami Nuwan Gunasekara	This project aims to apply techniques from recurrent concept drifts to explain the predictions of Online Continual Learning methods.
Conditional GAN for better embedding and generation of medical codes	Stefan Byttner Amira Soliman Kobra Etminani Atiye Sadat Hashemi	Synthetic data generation of Electronic Health Records with a focus on medical codes
Connected Safety Vest for Roadworkers	Oscar Amador Molina Alexey Vinel	Development and testing of an embedded system for the protection of Vulnerable Road Users
Data Heterogeneity in Federated Learning	Amira Soliman Sławomir Nowaczyk	Addressing the challenges of data imbalance in Federated Learning
Data analysis in collaboration with WirelessCar	Mahmoud Rahat Peyman Mashhadi Sławomir Nowaczyk	Data analysis in collaboration with WirelessCar
Deep Active Learning for LiDAR Point Cloud Segmentation	Abu Mohammed Raisuddin Eren Erdal Aksoy	Active Learning to improve data efficiency for LiDAR point Cloud Segmentation
Deep Decision Forest	Sławomir Nowaczyk	Designing a deep model that uses decision trees instead of artificial neurons
Deep stacked ensemble	Sławomir Nowaczyk Peyman Mashhadi	This project aims at training multiple parallel deep networks in such a way to learn different representation of data which will be suitable to frame these networks in stacked ensemble framework.
Developing a device for rapid water quality assessment	Ying Fu	develop a device with which a water sample may be analysed rapidly on the spot
Digital Twin - AFRY	TBD (Wojciech Mostowski)	Digital twin for consulting firm
Dynamic Churning-Based Logic Locking for Enhanced Hardware Security	Mahdi Fazeli	This project aims to explore and implement advanced logic locking techniques to improve the security of integrated circuits (ICs) against modern attacks.
EXIST: sEXism Identification in Social neTworks	Pablo Picazo-Sanchez	sEXism Identification in Social neTworks
Enhancing the Accuracy of CSI-Based Positioning in Massive MIMO Systems	Hazem Ali Ali Nada	CSI-Based Positioning in Massive MIMO Systems
Evaluating the Digital Tools for Promoting Sustainable Food Consumption	Azadeh Sarkheyli	To identify the key features and functionalities of sustainable food apps in Sweden
Evaluating the Effects of Social Media on Educational Sustainability in Sweden	Azadeh Sarkheyli	The research analyzes sentiment in social media data related to educational sustainability practices and outcomes in Sweden.
Evaluation of JAX in AI/ML software engineering	Veronica Gaspes Sławomir Nowaczyk	Analysis of the benefits of JAX (and/or similar solutions) in terms of performance, development time, module reusability, etc.
Evolving Kolmogorov-Arnold Networks	Mohammed Ghaith Altarabichi	This project aims to enhance the architecture of Kolmogorov-Arnold Networks (KANs) by optimizing key components such as loss functions, activation functions, initialization methods, and learning processes to improve their performance and interpretability.
Explainable AI and poverty prediction	Thorsteinn Rögnvaldsson Mattias Ohlsson	Provide explanations of AI data-driven poverty predictions in sub-saharan africa
Explainable AI by Training Introspection	Jens Lundström Peyman Mashhadi Amira Soliman Atiye Sadat Hashemi	Research and development of novel XAI methods based on training process information
Explainable Decision Forest	Sławomir Nowaczyk Hamid Sarmadi Sepideh Pashami	Designing an explainable decision forest classifier for fault detection
FLBench: A Comprehensive Experimental Evaluation of Federated Learning Frameworks	Sadi Alawadi Jens Lundström	Exploring Federated Learning Frameworks
Fair Conformal Prediction	Ece Calikus	Our goal is to design algorithms using conformal prediction framework that make fair predictions across various groups based on e.g., age, sex, income.
Fair representation learning of electronic health records	Ali Amirahmadi Ece Calikus Kobra Etminani	Fair representation learning of electronic health records
Fault detection using acoustic signals through anomaly detection	Elena Haller Peyman Mashhadi	Fault detection using acoustic signals through anomaly detection
Federated Learning Aggregation Strategies by Weight Exploration	Jens Lundström Amira Soliman Sadi Alawadi	Investigation of aggregation strategies for federated learning
Federated learning in automotive industry	Zahra Taghiyarrenani Sławomir Nowaczyk Sepideh Pashami	Federated learning in automotive industry
Foundation Models for Time Series Analysis	Zahra Taghiyarrenani Yuantao Fan Ali amirahmadi	Explore the use of large pre-trained time-series foundation models (TSFM) and design fine-tuning strategies on tasks of interest
Generating synthetic time series data in case of data scarcity	Alexander Galozy Peyman Mashhadi	Generating synthetic time series data in case of data scarcity
Generative Approach for Multivariate Signals	Kunru Chen Thorsteinn Rögnvaldsson Abdallah Alabdallah	The topic focuses on generative models (VAE) for CAN-bus data and investigating the representation learning capabilities of such techniques
Graph Neural Networks for Multivariate Time Series Data Analysis	Prayag Tiwari Guojun Liang	Our project aims to analyze multivariate time series through the innovative application of graph neural networks.
Graph Neural Networks for Traffic Flow Forecasting	Prayag Tiwari Sławomir Nowaczyk	The main goal of this project is to explore GNN for traffic flow forecasting
Graph Neural Networks for cardiovascular disease	Prayag Tiwari	The main goal of this project is to explore GNN for cardiovascular disease
Hardware Security Enhancement in Cyber-Physical Systems using Deep Learning-based Anomaly Detection	Mahdi Fazeli	In this project, we intend to employ a deep learning approach to detect anomalies in cyber-physical systems using data flow monitoring.
Human ground robot interaction	Cristofer Englund Martin Cooney Fernando Alonso-Fernandez	External communication from mobile robots to minimize conflicts with pedestrians
… further results

If you've added a project and it didn't show up, wait for cache to update, or press "refresh" button at top of the page! (refreshing the page in the browser is not always enough)

(make sure to give your project a name before clicking the button!)

Draft Proposals of Msc and Bsc Project (do not pick this unless you have checked with the supervisor!)

	Supervisors	OneLineSummary	Status
Analysis of ocular image synthesis for cross-spectral recognition	Kevin Hernández-Diaz Josef Bigun Fernando Alonso-Fernandez	analyze the performance of generative model for image-to-image translation of ocular images between different spectrum	Draft
Browser Extensions Updates	Pablo Picazo	Clustering and analyzing browser extensions by update frequency	Draft
Explainable AI for predictive maintenance in collaboration with Volvo	Mahmoud Rahat Peyman Mashhadi	Developing explainable models for predicting components failures of Volvo trucks	Draft
High Precision Power Use Measurement Device for Raspberry PI	Wojciech Mostowski Per Sandrup Joel Nyholm	Design and build a device from measuring power draw of a Raspberry PI with high precision to power profile software.	Draft
Human Value Detection	Pablo Picazo	Given a textual argument and a human value category, classify whether or not the argument draws on that category.	Draft
Identify boundary or, corner test cases for AD/ADAS functions using ML in MIL and HIL environment	Sławomir Nowaczyk Wojciech Mostowski	Identify boundary or, corner test cases for AD/ADAS functions using ML in MIL and HIL environment	Draft
Image Retrieval for Arguments	Pablo Picazo	Given a controversial topic, the task is to retrieve images (from web pages) for each stance (pro/con) that show support for that stance.	Draft
Mobile App Vulnerable Road Users	Oscar Amador Molina Alexey Vinel	Design, development, and evaluation of a mobile app for the protection of Vulnerable Road Users in a test track	Draft
No Signal Left to Chance	Pablo Picazo	Identify download patterns as a useful signal for analyzing browser extensions.	Draft
Optimizing Energy Consumption in Maritime Transportation with Machine Learning Methods (in collaboration with Cetasol)	Sławomir Nowaczyk Yuantao Fan Hadi Fanaee Or Mohamed Abuella	Develop machine learning methods for forecasting fuel consumption, path, and motion planning, with historical data from furries operation.	Draft
Project with HMS	Peyman Mashhadi Yuantao Fan	Few-shot Learning for Quality Inspection	Draft
Reinforcement Learning applied in mobile health application for hypertion	Farzaneh Etminani Sławomir Nowaczyk Alexander Galozy	Design a Reinforcement Learning applied in mobile health application for improving medication adherence in hypertensive patients	Draft
Representation Learning for Fault Detection and Prognosis	Yuantao Fan	Characterise the observed system using representation learning techniques, for fault detection and remaining useful life prediction	Draft

Older Proposals of Msc and Bsc Project

Those project proposals may still be valid, but contact supervisors before assuming so.

	Supervisors	OneLineSummary	Modification dateThis property is a special property in this wiki.
LiDAR Denoising	Eren Erdal Aksoy	In this project, the candidate is supposed to implement various filtering algorithm to denoise 3D LiDAR point cloud data.	27 October 2021 10:59:47
Virtual reality to support traffic safety	Cristofer Englund Lei Chen	The thesis is part of an ongoing project to develop drone-based lighting solutions for improving traffic safety and for encouraging travels to take bicycles.	19 October 2021 14:28:33
Lighting up the bicycle roads with drones	Cristofer Englund Lei Chen	Lighting up the bicycle roads with drones	19 October 2021 14:26:46
Automatic Idea Detection for controlling Healthcare-associated infections	Peyman Mashhadi Mahmoud Rahat Fabio Gama	This project aims to use advanced NLP tools to automatically detect interesting ideas by processing text available in the medical forums to address the Healthcare-associated infections problem in the hospitals	19 October 2021 08:57:23
Autonomous flying drone for vehicle classification	Cristofer Englund Fernando Alonso-Fernandez Martin Torstensson	Building an autonomous flying drone for vehicle classification	18 October 2021 11:38:05
Modeling patient trajectories using different representation learning techniques	Stefan Byttner Kobra Etminani Amira Soliman	Modeling Electronic Health Record (EHR) data and predict future events for specific patients	12 October 2021 08:23:46
Visual Transformers for 3D medical images Classification: use-case neurodegenerative disorders	Stefan Byttner Kobra Etminani Amira Soliman	Using visual transformers for predicting the diagnosis of multiple neurodegenerative brain disorders	12 October 2021 08:12:05
Advanced AI based anonymization of traffic video data	Fernando Alonso-Fernandez Yury Tarakanov (Viscando) Felix Rosberg (Berge)	Advanced AI based anonymization of traffic video data (with Viscando and Berge)	11 October 2021 12:59:13
Surface normal estimation by Spiral Codes	Josef Bigun	Estimating 3d surface normal from a single image	11 October 2021 11:30:50
Forecasting Industrial IoT Time Series @AlfaLaval	Hadi Fanaee	Forecasting industrial IoT Time Series	9 October 2021 23:45:15
Anomaly detection from IoT Time Series @AlfaLaval	Hadi Fanaee	Anomaly detection from IoT Time Series @AlfaLaval	9 October 2021 23:44:45
Deep neural network optimization for path prediction in vessels!	Reza Khoshkangini Enayat Rajabi	The purpose of this thesis is analyzing a ferry dataset to identify the most optimal path using deep-net.	6 October 2021 13:45:59
The effect of contextual information on fuel consumption using Explainable AI!	Reza Khoshkangini Enayat Rajabi	There are many factors that can minimize pollutions and maximize energy efficiency and fuel consumption in vessels.	6 October 2021 13:43:51
Predicting electricity generation capacity in solar and wind power plants based on meteorological data using machine learning algorithms	Reza Khoshkangini Ramin Sahba Amin Sahba	ML algorithms will be used to analyze meteorological data to predict the electricity generation capacity of solar and wind power plants. This project is a collaboration between Halmstad and Sam Houston University (USA)).	6 October 2021 12:58:35
Deep clustering for vehicle operation type	Reza Khoshkangini Peyman Mashhadi	In this project, deep clustering will be used on the logged vehicle data (LVD) to find the best representation of vehicles’ operation to explain the behavior of the vehicles over time.	6 October 2021 06:48:09
Semi-supervised deep learning model to optimally charge and discharge the batteries of electric cars and balance distribution of electrical energy in the power grid	Reza Khoshkangini Amin Sahba Ramin Sahba	A semi-supervised deep learning model will be developed to optimally charge and discharge the batteries of electric cars while they are connected to the power grid. (In collaboration with Sam Houston State University (USA)).	6 October 2021 06:44:05
Incorporate behaviour modelling into AGV safety performance stack	Björn Åstrand	Incorporate behaviour modelling into AGV safety performance stack	5 October 2021 20:35:17
Optimising Energy Consumption for Ferries in Collaboration with Cetasol	Yuantao Fan Peyman Mashhadi	This project aims at developing data-driven methods to understand ferry operations and optimise enegery consumption	5 October 2021 20:05:46
Deepfake Detection	Stefan Byttner Jens Lundström Peyman Mashhadi	Detecting deepfake images and videos using a diversified ensemble of deep models	5 October 2021 15:53:22
Intrusion detection and prevention for IIoT using Ensemble Deep Network	Reza Khoshkangini Mohamed Eldefrawy	In this project students will work on network data and try to detect malicious behavior using ensemble deep network.	5 October 2021 08:02:00
Deep Networks for Semantic Scene Understanding	Eren Erdal Aksoy	The candidate will implement a neural network to detect spatial relations between objects in the scene. For instance, the book is on the table or the spoon is in the cup.	4 October 2021 09:08:13
Deep Graph Networks for Future Graph Prediction	Eren Erdal Aksoy	In this project, the candidate is supposed to implement a deep graph network that receives a set of graphs as input and returns the predicted next upcoming graph(s).	4 October 2021 08:57:49
Music style transfer	Peyman Mashhadi Yuantao Fan	Develop a system that receives a piece of music in one genre and changes/transfers its style into another genre, using machine learning algorithms.	3 October 2021 15:34:06
Security analysis of IIoT connectivity protocols	Mohamed Eldefrawy Yousra Alkabani	Potential security vulnerabilities of IIoTs platform connectivity protocols, such as CoAP and MQTT will be studied.	29 September 2021 20:26:14
Project with Whole AB	Slawomir Nowaczyk / TBD	TBD	28 September 2021 08:55:26
Building a Knowledge-based AI Framework for Mobility	Enayat Rajabi Sławomir Nowaczyk	Leveraging new knowledge to improve the productivity of mobility services	22 September 2021 18:33:42
Effecient implementation of DL models on embedded platforms	Nesma Rezk Yuantao Fan Sławomir Nowaczyk	In this project, we optimize DL models to run efficiently on resource-bounded embedded platforms.	20 September 2021 15:47:17
Sandboxed scripting on embedded systems	TBD Please contact Slawomir and Hans-Erik Eldemark	Sandboxed scripting on embedded systems	7 January 2021 16:19:04
Analysis of multi-machine/multi-sensor data	Hadi Fanaee-T (www.fanaee.com) Mahmoud Rahat	Data mining on multi-machine/multi-sensor time series data	22 October 2020 13:36:50
Beyond 5G baseband processing on a multicore architecture	Süleyman Savas	Implementation and evaluation of beyond 5G baseband algorithms on an embedded (Epiphany) processor with 16 cores	21 October 2020 06:21:39
Optimisation in Heavy Duty Vehicles Configuration	Reza Khoshkangini	In this project we aim to optimise the vehicles setting based on their usage style. Here we focus more on fuel consumption.	19 October 2020 12:51:51
Transfer Learning by Selection of Invariant Features	Mohammed Ghaith Altarabichi Abdallah Alabdallah	The project aims to develop novel methods to identify invariant features to transfer across multiple domains.	15 October 2020 10:25:42
Hide-and-Seek Privacy Challenge (NeurIPS 2020)	Onur Dikmen	Building novel methods for privacy-preserving data sharing and/or re-identification	13 October 2020 10:31:00
Smart Alarm	Hadi Fanaee-T(www.fanaee.com) Mahmoud Rahat	Data-driven alarm prediction using sensor data	12 October 2020 11:47:57
Predicting the status of machines with vibration data	Hadi Fanaee-T (www.fanaee.com) Mahmoud Rahat	Predicting the status of Alfa Laval's separator machines with vibration data	12 October 2020 11:47:36
Reinforcement learning in Automation	Reza Khoshkangini	In this project we plan to use reinforcement learning in multi-agent systems to improve decision making. in automated systems.	9 October 2020 11:51:36
Vehicle Usage Modeling over Time	Reza Khoshkangini Abbas Orand	This project intents to explore the modeling of the usage of vehicles using unsupervised machine learning algorithms in different context which are logged over time.	9 October 2020 11:42:02
Anomaly Detection of the Activities of the Elderly Living in the Smart Home	Abbas Orand Reza Khoshkangini	In this project we detect the anomaly of the actives of the elderly people or those with some sorts of health problem.	9 October 2020 11:37:53
Optimisation Algorithm for Feature enginnering	Reza Khoshkangini	In this project we intend to design an optimisation system using artificial intelligence algorithms in order to select/extract the best features for developing a forecasting system in predictive maintenance.	9 October 2020 10:21:14
Zero-Shot Learning for Semantic Segmentation	Tiago Cortinhal Eren Erdal Aksoy	Zero-Shot Learning for Semantic Segmentation	9 October 2020 08:40:01
Automatic Generation of Realtime Machine Learning Architectures	Yousra Alkabani Hazem Ali	In this project, it is required to build a tool to generate a dataflow model and construct architectures for such algorithms, while minimizing latency or meeting a specific deadline under area and power constraints.	8 October 2020 16:14:16
Reversible GANs	Cristofer Englund Felix Rosberg	Reversible GANs	8 October 2020 12:30:33
Situation awareness in traffic	Cristofer Englund Björn Åstrand	Situation awareness in traffic	8 October 2020 12:29:24
Indoor localization for ground vehicles	Cristofer Englund	Indoor localization for ground vehicles	8 October 2020 12:27:56
Transfer Learning for Machine Diagnosis and Prognosis	Peyman Mashhadi Yuantao Fan Mohammed Ghaith Altarabichi	Study and develop deep adversarial neural networks (DANN) based methods to detect faults and predict failures in industrial equipment, under transfer learning scenarios.	6 October 2020 18:32:30
Reinforcement Learning with Adaptive Representation Learning	Alexander Galozy Peyman Mashhadi	This project targets finding representations that make the reinforcement learning more efficient in terms of finding an easier state to action mapping.	5 October 2020 13:38:39
Risk as a Service with Volvia	To be decided (contact Slawomir Nowaczyk for more details)	Risk as a Service	4 October 2020 15:09:34
Intelligent claim Process with Volvia	To be decided (contact Slawomir Nowaczyk for more details)	Intelligent claim Process	4 October 2020 15:07:58
Forecast energy consumption in buildings to help Mestro customers save energy	To be decided (contact Slawomir Nowaczyk for more details)	The thesis will be focused on forecasting the energy consumption in buildings (e.g. electricity consumption), with some optional “add-ons” where student will also develop...	4 October 2020 12:56:56
Feature-wise normalization for 3D medical images	Amira Soliman Stefan Byttner Kobra Etminani	Normalization of 3D medical imaging either as a data pre-processing or as feature-wise batch normalization during CNN model training	29 September 2020 13:40:00
… further results

Ongoing Projects

	ThesisAuthor	OneLineSummary	Supervisors
Analyzing white blood cells in blood samples using deep learning techniques		To analyze white blood cell content in blood samples using deep learning techniques.	Mattias Ohlsson
Article Identification for Inventory List in a Warehouse Environment	Yang Gao	Article Identification for Inventory List in a Warehouse Environment	Björn Åstrand Saeed Gholami Shahbandi
Automatic Generation of Descriptive Features for Predicting Vehicle Faults	Vandan Revanur Ayodeji Olanrewaju Ayibiowu	Automatic Generation of Descriptive Features for Predicting Vehicle Faults	Mahmoud Rahat Reza Khosh
Chess playing humanoid robot by vision	Joseph T. Sachin	Chess playing humanoid robot by vision	Josef Bigun
Face and eye categorization and detection	Zhao Cui Albert Hoxha	To build a new database of face and eye images of different species and to evaluate holistic and local detection algorithms	Fernando Alonso-Fernandez
Forklift Trucks Usage Analysis		This project is about applying machine learning methods to have a better understanding for the usage of forklifts trucks in industrial application.	Kunru Chen Alexander Galozy
Human identification by handwriting of identity text		Identify a hand writer when repeated identity relevant text is available	Josef Bigun Fernando Alonso-Fernandez
Ice rink resurfacing system for selfdriving vehicles having spiral codes		ice rink resurfacing system for selfdriving vehicles having spiral codes	Josef Bigun
Interactive Anomaly Detection		Anomalies can be relevant or irrelevant to the end-user. The goal of this thesis is to propose a new interactive anomaly detection method to leverage the user-feedback and learn to suggest more relevant anomalies.	Mohamed-Rafik Bouguelia Onur Dikmen
Modelling Health Recommender System using Hybrid Techniques		The goal of this project is to develop a health recommender system using existing machine learning techniques.	Hassan Mashad Nemati Rebeen Hamad
OpticalFlowFeaturesForEventDetection	Mohammad Afrooz Mehr Maziar Haghpanah		Stefan Karlsson
Pallet Rack Identification in Warehouse	Anil Kumar Kothapalli	Development of an identification algorithm for Pallet Rack Cells in a warehouse. Data acquisition is performed by a mobile robot via fisheye cameras and/or 3D sensors.	Björn Åstrand Saeed Gholami Shahbandi
Robot Cooking	Chandrashekhar Shankarrao Nasurade Vamsi Krishna Nathani	Common sense for a robot to cook healthy food	Martin Cooney
Sensor fusion and machine learning for drone detection and classification		Sensor fusion and machine learning for drone detection and classification	Cristofer Englund Eren Erdal Aksoy Fernando Alonso-Fernandez
Smart sensor	Can Yang	Small smart sensors	Martin Cooney Håkan Petterson
Social touch for robots	Prateek	something with social robots	Martin Cooney
Traffic Estimation From Vehicle Data	Sowmya Tamidala	Estimate traffic density based on logged vehicle data	Sławomir Nowaczyk Iulian Carpatorea

Completed Msc and Bsc Project

	ThesisAuthor	OneLineSummary	Supervisors
"TROLL": a regenerating robot	Yinrong Ma	A robot which can detect faults on itself and try to mark or fix them	Martin Cooney Anita Sant'Anna
Activity monitoring for AAL	Jianyuan Ma Yinan Qiu	Tracking of more than one person in a smart environment using fixed sensors and a mobile robot	Anita Sant'Anna
Adaptive warning field system		Adaptive warning field system	Björn Åstrand
Analysis of Multi-Lingual Vehicle Service Histories	Iyanuoluwa Akanbi	Automatic translation and similarity evaluation of multi-lingual natural text descriptions of vehicle repair and maintenance operations	Sepideh Pashami Sławomir Nowaczyk
Assistance-seeking strategy for a flying robot during a healthcare emergency response	Jérémy Heyne	Assistance-seeking strategy for a flying robot during a healthcare emergency response	Anita Sant'Anna Yuantao Fan Martin Cooney
Classifying heart diseases based on heart random numbers		Heart signal has entropy and can be used to generate random numbers. The idea is that, given a bunch of random numbers, we should predict if the source suffer from a desease	Pablo Picazo
Collaborative Filtering Recommendation System Location Content-based		Analyze the content stored in Collaborative Filtering Recommendation System based on the location of the users	Pablo Picazo
Consensus clustering for categorizing orthogonal vehicle operations	Dirar Sweidan	Discovering multiple clustering solutions, compare them, and find out if there is a single best (consensus) clustering, or multiple consistent clustering solutions.	Mohamed-Rafik Bouguelia Sławomir Nowaczyk
Constrained dynamic path planning for truck and trailer	Imanol Mugarza	Constrained dynamic path planning for truck and trailer	Iulian Carpatorea Sławomir Nowaczyk Jennifer David
Courteous robot guide for visitors to an intelligent home	Jiamiao Guo Yu Zhao	Courteous robot guide for visitors to an intelligent home	Wagner de Morais Martin Cooney
Detecting Points of Interest for Robotic First Aid	Wolfgang Hotze	Detecting Points of Interest for Robotic First Aid	Anita Sant'Anna Martin Cooney
Detection and intention prediction of pedestrians in zebra crossings	Dimitrios Varytimidis	Detection and intention prediction of pedestrians in zebra crossings	Fernando Alonso-Fernandez Cristofer Englund Boris Duran
Evolutionary Behavior Trees for Multi-Agent Task-Oriented Environment	Milosz Mazur	Evolutionary generating Behavior Trees for use in multi-agent task-oriented environment.	Sławomir Nowaczyk
Exploration and Mapping of Warehouse Using Quadrotor Helicopters	Maytheewat Aramtattana Yuantao Fan	Implementation of a navigation method for a flying robot (Quadrotor). The robot is assigned to explore and map the warehouse.	Björn Åstrand Saeed Gholami Shahbandi
F1tenth		F1tenth competition	Sławomir Nowaczyk Cristofer Englund Wojciech Mostowski
Finding patterns/motifs in time series data	Felix Nilsson	Finding patterns/motifs in time series data, for autonomous clustering or outlier detection	Thorsteinn Rögnvaldsson Mohamed-Rafik Bouguelia
FirstResponse	Gloria	First response to emergency situation in a smart environment using a mobile robot	Anita Sant'Anna
Fuzz testing of network protocols	Filip Kågesson	Investigation how fuzz testing of network protocols could be implemented and provide rapid robustness testing	Wojciech Mostowski
Graphical Traffic Scenario Editor	Iulian Carpatorea	Develop an interactive graphical application to draw vehicle paths and their surrounding environment, for rapid prototyping of traffic scenarios in intelligent vehicle research.	Roland Philippsen
Improved face tracking driven by optical flow	Andreas Ranftl	Face Tracking Using Optical Flow	Stefan Karlsson Fernando Alonso-Fernandez Josef Bigun
Improving MEDication Adherence through Person Centered Care and Adaptive Interventions iMedA	Alexander Galozy	Improving MEDication Adherence through Person Centered Care and Adaptive Interventions iMedA	Sławomir Nowaczyk Anita Sant'Anna
Integrating a new rigid-body dynamics model library with an existing whole-body controller	Anton Jerey Thomas Holleis Marlene Mohr	Integrating a new rigid-body dynamics model library with an existing whole-body controller	Roland Philippsen
Investigating Robustness of DNNs	Matej Uličný	This master thesis project aims at characterizing sensitivity to classification of images (based on deep neural networks).	Jens Lundström Stefan Byttner
Label and Barcode Detection and Location in Large Field of View	Guanjie Meng Shabnam Darman	Wide angle images are logged during a warehouse exploration. Design of a detection and localization method for barcodes (and/or labels) in the scope, based on such an acquisition is desired.	Björn Åstrand Saeed Gholami Shahbandi
Mixed-Reality Robot Platform	Norbert Gruenwald	Build foundations for our mixed-reality platform by integrating and demonstrating an extensible system with one or more robots, a simulator, some offboard sensors, and simple teleoperation.	Roland Philippsen
Mobile Social Robot for Healthcare	Matthias Mayr	Pilot study about a small interactive mobile robots for therapy and healthcare in homes.	Roland Philippsen Magnus Clarin
Model Volvo Truck Lifetime Repair History	Anton Palmqvist	Finding good representations for data-driven description of Volvo truck's repair and maintenance history	Sławomir Nowaczyk Sepideh Pashami
RAQUEL Robot Assisted QUiz Espying of Learners	Sanjana Arunesh Abhilash Padisiva	RAQUEL Robot Assisted QUiz Espying Learners	Josef Bigun Martin Cooney Fernando Alonso Fernandez
RaspberryPiVolvoLogger	Anestis Zaganidis	RaspberryPi-based solution for logging CAN data on Volvo trucks	Sławomir Nowaczyk Yuantao Fan
Recurrent and Deep Learning for Machine Prognostics	Kunru Chen	Construct and optimise Recurrent Neural Networks for industrial applications on machine prognostics; Augmenting industrial data for supervised learning	Sławomir Nowaczyk Sepideh Pashami Yuantao Fan
Robot Artwork	Daniel Westerlund Sowmya Narasimman	Capability for a robot to paint to express human feelings	Martin Cooney Maria Luiza Recena Menezes
Robotic First aid response	Tianyi Zhang and Yuwei Zhao	A robot system which assesses a person's state of health as a first step toward autonomous robotic first aid/ems	Martin Cooney Anita Sant'Anna
Sailboat Motion Planning using the Level-Set Method	Lin Ge Yifei Li	Explore the use of Level-Set and Fast-Marching Methods to create time-optimal motions of a point in a plane subject to direction-dependent velocity.	Roland Philippsen
Smart Home Simulation	Solved by internal/external resources	Developing and evaluation of a smart home simulator and outlier detection methods.	Jens Lundström Antanas Verikas Sławomir Nowaczyk
Supervised/Unsupervised Electricity Customer Classification	Soniya Ghorbani	Consumer characterization framework based on knowledge discovery in smart meter data	Sławomir Nowaczyk Anita Sant'Anna Hassan Mashad Nemati
Thermal Detection of Subtle Human Cues for a Robot Magic Performance	(NOT AVAILABLE)	Thermal Detection of Subtle Human Cues for a Robot Magic Performance (NOT AVAILABLE HT22/VT23)	Martin Cooney
Vehicle Operation Classification	Karthik Bangalore Girijeswara	Classify modes of operation of Volvo vehicles based on on-board data	Sławomir Nowaczyk Yuantao Fan Mohamed-Rafik Bouguelia
Visual analysis for infotainment in car interiors	Josef Bigun Maycel Isaac Faraj	Visual analysis to steer infotainment in car interiors	Josef Bigun Stefan Karlsson Maycel Isaac Faraj

Internal Drafts

	OneLineSummary	ThesisAuthor	Supervisors
A decision support system for reducing false alarms in ICU	Developing a clinical decision support system using machine learning and biomedical signal analysis techniques for an ICU setting.		Sławomir Nowaczyk Awais Ashfaq
Acumen Robot Model Series	Build a series of increasingly sophisticated robot models in Acumen, to (1) explore mathematical formulations and (2) create tutorials and didactic examples.		Roland Philippsen Walid Taha
Algorithm development for realtime route planning	Project at Sigma Technology		Thorsteinn Rögnvaldsson
Analysing Engine Performance based on Vehicle Data	Estimate engine perfromance based on data logged on-board Volvo vehicles and using it for diagnostics, e.g. detection of cylinder heads in need of replacement		Sławomir Nowaczyk Magnus Svensson
Anomaly Detection for Predictive Maintenance with Elvaco	Anomaly Detection for Predictive Maintenance with Elvaco		Mohamed-Rafik Bouguelia Yuantao Fan
Anomaly detection in district heating data - with the Elvaco company	Project around predictive maintenance in district heating		Yuantao Fan Mohamed-Rafik Bouguelia
Clustering of battery usage pattern for Electric buses	Clustering of battery usage pattern for Electric buses		Sepideh Pashami Yuantao Fan
Convolutional Neural Network (CNN) responses when the number of classes increase	Convolutional Neural Network (CNN) features behaviour in the context of textures		Josef Bigun Fernando Alonso-Fernandez
Developing an Intelligent Data-Driven Chatbot for Enhanced Information Retrieval and Interaction	Design, implement, and evaluate a chatbot system capable of interacting with users and extracting pertinent information from data		Yuantao Fan Sławomir Nowaczyk
Development of surveillance methods for sterilizers	Development of surveillance methods for sterilizers		Thorsteinn Rögnvaldsson Sławomir Nowaczyk Stefan Byttner
Evaluation of Open Source Robot Simulators for Smart Mobility Applications	Can open source robot simulators serve as starting point for cloud services that support automotive R&D and V&V?		Roland Philippsen Saeed Gholami Shahbandi Christian Berger (Chalmers)
Explainable Anomaly Detection	Explainable anomaly detection in time series data utilising causal inference	Afroj Divan; Athulya Ashok	Yuantao Fan
Gait analysis using wearable sensors in Parkinson's disease	The project aims to develop a machine learning tool for the assessment of Parkinsonian gait in a natural environment		Taha Khan
Merging Clothoids with B-Splines	Develop an approach to create natural clothoidal lane-change maneuvers for automobiles on lanes that are specified using B-splines.		Roland Philippsen
MultiScale Microscopy Detailed	Master Thesis Project	Amir Etbaeitabari Mekuria Eyayu	Stefan Karlsson Josef Bigun
Obstacle Identification from 3D Data for AGVs in a Warehouse Environment	Obstacle Identification from 3D Data for AGVs in a Warehouse Environment		Björn Åstrand Saeed Gholami Shahbandi
Path Planning	Path and Motion Planning for a Ferry		Sławomir Nowaczyk Yuantao Fan Hadi Fanaee Or Mohamed Abuella
Predicting Energy Consumption for Heavy-Duty Vehicles (in collaboration with Volvo)	Develop machine learning methods to forecast energy consumption for heavy-duty vehicles		Yuantao Fan Mahmoud Rahat
Representation of Complex Data Types for Machine Learning	Finding ways to represent complex data types (e.g. histograms) present in Logged Vehicle Database databse for machine learning-based fault prediction		Sławomir Nowaczyk Sepideh Pashami
Semantic Analysis of 2D Maps With a Metric-Topological Approach	Semantic Analysis of 2D Maps With a Metric-Topological Approach.		Björn Åstrand Saeed Gholami Shahbandi
Simulating Crowds for Traffic Safety Research	Integrate crowd simulation into a mixed-reality platform for development and testing of advanced automotive safety systems.		Roland Philippsen
Thesis with Jayway	Thesis with Jayway		TBD
Thesis with Medius	Three thesis topics with Medius		TBD Please contact Slawomir Nowaczyk
Time series anomaly detection for heavy-duty vehicles	Detecting anomalies in multivariate time series data collected from vehicle operations		Yuantao Fan Hamid Sarmadi

Introduction lecture presenting the process and expectations on MSc project will take place on Monday, 14 October 2019 at 15:15.

The next opportunity for MSc presentations is on Monday, 23 September 2019 at 13:15 (contact Slawomir if you intend to make the presentation, so that we know how many to plan for). Also, make sure to send the supervisor-approved reports to the examiners a week before, so by 16 September 2019. After that we will have the final(!) opportunity sometime in late December/early January.

Final presentations should be 15 minutes long, and must cover the goals/objectives, final contribution/novelty, the results and conclusions from the work.

Final presentation is scheduled on Wednesday, 29 May 2019 [turns out Thursday is a holiday] (which means the examiners need to receive your reports by Thursday, 23 May 2019 at noon).

There will be a chance to re-do half-time presentations, for those who were not ready in March, around middle of May.
Half-time presentations should be 20 minutes long, and must cover the goals/objectives, expected contribution/novelty, results achieved so far, and a refined plan on how to proceed.

Topic selections are due on 27th of October 15:00 (use this GoogleForms link).

For students who started their MSc in 2018, the final opportunity to present their thesis will be on Friday, 13th of December 2019, at 16:00 (room F506). Deadline (strict!) for submitting reports is Wednesday, 11th of December, at noon.

Half-time seminar will be on 19th of February (preliminary time: 9-15, depending on number of projects that are ready in time). This means you should send the reports to the examiners on the 18th of February before lunch (this form). Please note that it's a bit earlier than I've indicated during the introductory lecture, as we've decided it makes more sense to provide this feedback sooner rather than later. The final seminar will be at the end of May (reports due in the middle of May).

Start report is due (approved by supervisors!) on 6th of December 23:59, and presentations will be done on 9th and 11th of December (you are expected to attend & listen both days).
You should prepare 10 minutes presentation, focusing on problem formulation, novelty & contribution, literature review and project plan.

The second chance for half-time presentations will be on Thursday, 12th of March (preliminary time: 13-16, depending on number of projects that are ready in time). This means you should send the reports to the examiners on the 11th of March before lunch (this form). The final seminar will be at the end of May (reports due in the middle of May).

Half-time presentations should be 20 minutes long (plus ~10 minutes for questions), and must cover the goals/objectives, expected contribution/novelty, results achieved so far, and a refined plan on how to proceed.

The final presentation is scheduled on Thursday, 28 May 2020 (which means the examiners and opponent need to receive your supervisor-approved reports by Thursday, 21 May 2020 at noon -- use this form to send your report to supervisors, and email it to the opponent).
Make sure to have at least two iterations of feedback on the report, so you should send the initial draft to your supervisors at latest in the first week of May. If you don't make it, the next opportunity will be at the end of August/beginning of September.

Final presentations should be 15 minutes long and must cover the goals/objectives, final contribution/novelty, the results and conclusions from the work. It will be followed by 15 minutes of questions/discussion with the opponent (selected by the supervisor) and the examiners. Be advised that 15 minutes is very short, so you should carefully select what do you talk about, focusing on the most important aspects.

The second opportunity for final presentations will be on Thursday, 3 September 2020 (done online, on Zoom)... which means the examiners and opponent need to receive your supervisor-approved reports by Friday, 28 August 2020 at noon -- use this Google form to send your report to examiners, and email it to the opponent.
Make sure to have at least two iterations of feedback on the report, so you should send the initial draft to your supervisors at latest in the first week of August (take into account any vacation plans!). If you don't make it, the next opportunity will be in December/January.

Introduction lecture presenting the process and expectations on MSc project will take place on Thursday, 15 October 2020 at 10:00 (sharp) through Zoom. The link should be available in your schedules.

Topic selection: provide a ranking of three preferred topics by Wednesday, 28th of October, 18:00 using this GoogleForm

A start report is due (approved by supervisors!) on 10th of December 23:59 (using this link).
Schedule for presentations in week 51 (you are expected to attend & listen to all) is available here (please observe some changes are still expected, mainly based on supervisors' availability, so please check it regularly).
You should prepare 10 minutes presentation, focusing on problem formulation, novelty & contribution, literature review and project plan.
Presentations Zoom link

The next deadline for start reports, for those who didn't make it this time, is 15th of January 23:59 (using this link).

For those who were not ready with halftime reports in March, there will be another opportunity on Wednesday, 12 May 2021 (the deadline for submitting reports is 9th of May, same Google Form)
Half-time presentations should be 20 minutes long (plus ~10 minutes for questions) and must precisely specify the goals/objectives, clearly explain the expected contribution/novelty, showcase the results achieved so far, and present a refined plan on how to proceed further.

The final seminar will be in about two weeks, preliminarily on 2nd, 3rd and/or 4th of June, depending on the number of projects that are ready (reports due on 25th of May at 12:00 noon, using this form).

The second opportunity for final presentations of MSc theses will be on Friday, 20 August 2021 (done online, on Zoom)... which means the examiners and opponent need to receive your supervisor-approved reports one week earlier -- use this Google form to send your report to examiners, and email it to the opponent.
Make sure to have at least two iterations of feedback on the report, so you should send the initial draft to your supervisors at the latest in the first week of August (take into account any vacation plans!). If you don't make it, the next opportunity will be in December/January.

Final presentations should be 15 minutes long and must cover the goals/objectives, final contribution/novelty, the results and conclusions from the work. It will be followed by 20-30 minutes of questions/discussion with the opponent (selected by the supervisor) and the examiners. Be advised that 15 minutes is very short, so you should carefully select what do you talk about, focusing on the most important aspects.

The next and final opportunity for half-time presentations will be in late August. Submit your reports on Google Form and email Slawomir to schedule the presentation.

Introduction lecture presenting the process and expectations on the MSc project will take place on Wednesday, 13 October 2021 at 15:15 in D315.
Please note that it will not appear in your schedule, since the course has not started yet. Before the lecture, please watch the video recording and read updated slides.
This way, in the physical meeting we can focus on discussion and answering any questions you might have.

For the selection of thesis topics, you should provide a ranking of three preferred topics by Wednesday, 27 October 2021, 18:00 using GoogleForm.

A start report is due (approved by supervisors!) on 12th of December 18:00 using this form. This means that you should send a reasonably complete draft to your supervisors before the end of November, so that they have time to provide feedback, and you have time to incorporate this feedback into your final report...
Presentations will take place in week 50 (you are expected to attend & listen to all, or at least most, of them).
You should prepare 10 minutes presentation, focusing on problem formulation, novelty & contribution, literature review and project plan.

For those students who started in January, the initial report is due (approved by supervisors!) on 30th of January at 16:00 using this form. This means that you should send a reasonably complete draft to your supervisors quite soon, so that they have time to provide feedback, and you have time to incorporate this feedback into your report...
Presentations will take place in week 5, probably on Friday, 4 February 2022.
You should prepare 10 minutes presentation, focusing on problem formulation, novelty & contribution, literature review and project plan.

Dear students, you should have received an email with a link to Google Spreadsheet with thesis topics assignment. It was sent to all those who submitted thesis topic selections using GoogleForm. If you have not received it, please let me (Slawomir) know ASAP.

Half-time reports are due tomorrow. However, I have been hearing that many of you have problems making the deadline. Thus, we've decided to offer an (additional) second opportunity -- in two weeks. If you can have a good quality report ready tomorrow you should still submit it, since timeliness is an important part of MSc project.

If you cannot, on the other hand, your next deadline for sending the reports to the examiners is Friday, 01 April 2022, at lunchtime (using this form).
Presentations, in that case, will be sometime in week 14.

Half-time presentations will be in week 12, probably between Wednesday, 23 March and Friday, 25 March 2022 (depending on number of projects that are ready in time).
This means the deadline for sending the reports to the examiners is the Friday, 18 March 2022, lunchtime (using this form).
The final seminar will be at the end of May (reports due in the middle of May).

The schedule for half-time presentations has now been updated with (preliminary) presentation times for those who submitted their reports on the second deadline: https://docs.google.com/spreadsheets/d/1FK-OjzhQdIbfMkeyvGLneJzSV6znejgG9fG19Wv1VXU/edit#gid=223334201

Half-time presentations should be 20 minutes long (plus ~10 minutes for questions), and must cover the goals/objectives, expected contribution/novelty, results achieved so far, and a refined plan on how to proceed.

Final presentations of MSc theses will be primarily on Wednesday, 1 June and Thursday, 2 June... which means the examiners and opponent need to receive your supervisor-approved reports one week earlier -- use this Google form (deadline: Wednesday, 25 May 2021 at 12:00 noon) to send your report to examiners, and email it to the opponent. Make sure to have at least two iterations of feedback on the report, so you should send the initial draft to your supervisors at the latest in the second week of May. If you don't make it, the next opportunity will be in late August/early September.

Final presentations should be 15 minutes long and must cover the goals/objectives, final contribution/novelty, the results and conclusions from the work. It will be followed by 25-30 minutes of questions/discussion with the opponent (selected by the supervisor) and the examiners. Be advised that 15 minutes is very short, so you should carefully select what do you talk about, focusing on the most important aspects.

You should pick your topic (select three, in order of preference) by Wednesday, 26th of October, 18:00. Submit your choice on this GoogleForm.
The introduction lecture presenting the process and expectations of the MSc project took place on Monday, 3 October 2022. Slides are available below.

The start reports are due on December 8th at 21:00 (using this form); presentations will take place in week 50.
Please remember that the report must be approved by supervisors first, so a reasonable schedule is: on the 25th of November send the report to supervisors; around the 2nd of December, you get feedback; then you have a week to address the comments.

For those who begin the Thesis course in January, the start report is due on Wednesday, 25th of January (make sure to also account for supervisor approval and revision time).