Advanced Seminar Embedded Systems and Internet of Things
We will announce the available topics and the application process on the 08th of September 2025. You will then be able to apply for a seminar topic between the 08th of September until the 5th of October (23:59 pm).
It is mandatory to attend all the lectures of our Advanced Seminar in presence to complete the course successfully. Virtual Attendance is not possible.
Application Process
Due to the high interest in our seminar topics we use an application process to assign the topics.
If you are interested in one of the topics below, please send your application together with your CV and your transcript of records to seminar.esi(at)xcit.tum.de. Express your interest and explain why you want to have that specific topic and why you think that you are most suitable for the topic. This allows us to choose the most suitable candidate for the desired topic to maximize the seminar's learning outcome and to avoid dropouts.
Additionally, you can indicate a second topic that you would like to take, such that we can still find a topic for you if your primary choice is not available.
Deadline: We encourage you to apply until the 05.10.2025. Afterwards we will assign the topics and notify all applicants. After this date, we will answer to requests within 3 days (until 08th of October), assuming that there is enough motivation for the given topic. Once you are given the topic, we will ask for your confirmation. You must confirm your participation until the 10th of October.
Note: We do not assign topics on a first-come-first-served basis. Even though we appreciate your interest if you have asked or applied early for a topic we can not guarantee that you get a seat. Generally we have 3-4 applicants per topic. Please think carefully if you are able to do the work required as we have to reject other students. Generally, email clients remember the people you have communicated with. You will be registered to the seminar course by the Advanced Seminar Manager after the Kick-Off meeting on 15th of October.
Kick-off meeting
This semester the seminar will be conducted in physical mode. This means that you must join the physical classes and presentation which you will find on the Moodle page. Additionally, you can schedule weekly meetings with your supervisor via Zoom or on campus. Lecture materials and videos will be available on Moodle.
The kick-off meeting will be on the 15th of October at 9:45 on Campus. We ask all successfully selected participants to be present in the kick-off meeting. Please notify us in case you can not make it to the meeting, otherwise we will assume that you are no longer interested and give your place to another applicant.
Topics
We will soon announce the topics
Topics
Exploring AI-Agent Driven ZK-Circuit Design
Zero-Knowledge (ZK) proofs are a powerful cryptographic tool, but transitioning them from theory to practical applications begins with a challenging step: arithmetization. This process transforms a program into an Arithmetic Circuit (also known as a ZK-Circuit), which can then be verified using ZK proofs.
However, designing these circuits is often complex, error-prone, and requires low-level operations, making it a significant barrier for developers. With the emergence of Large Language Models (LLMs) and General Purpose AI Agents, there is growing potential to automate and simplify this process. In this seminar project, the student will: Investigate how AI agents can assist in the design of ZK-Circuits. Explore existing tools and frameworks that integrate AI with ZK proof systems. Benchmark AI-assisted circuit generation against traditional methods and models. Evaluate the accuracy, efficiency, and usability of AI-driven approaches.
Literature:
[1] https://hackmd.io/@jake/plonk-arithmetization
[2] https://learn.0xparc.org/materials/circom/additional-learning-resources/r1cs%20explainer/
[3] https://tlu.tarilabs.com/cryptography/rank-1
[4] https://zcash.github.io/halo2/
[5] https://arxiv.org/abs/2407.01502
[6] https://dl.acm.org/doi/full/10.1145/3655615
Supervisor: Marco CalipariSecure Multiparty Computation for Autonomous Vehicles Collaborative Perception
Collaborative perception is a cutting-edge paradigm in autonomous systems designed to enhance the perception capabilities of individual vehicles through the exchange of perception data with other vehicles. However, this data sharing poses significant privacy risks for the vehicles involved. Secure Multiparty Computation (SMC) is a privacy-preserving technology that enables parties to compute functions jointly while keeping their individual inputs private and ensuring fairness. This seminar aims to explore the potential integration of SMC within the framework of collaborative perception. We will focus on the necessary requirements for successful implementation and conduct a feasibility analysis of achieving collaborative perception in a privacy-respecting manner.
Literature:
[1] Xiong, Jinbo, et al. "Toward lightweight, privacy-preserving cooperative object classification for connected autonomous vehicles." IEEE Internet of Things Journal 9.4 (2021): 2787-2801.
[2] T. Li, L. Lin, and S. Gong, AutoMPC: Efficient Multi-Party Computation for Secure and Privacy-Preserving Cooperative Control of Connected Autonomous Vehicles. 2019.
Supervisor: Marco CalipariImage Processing Techniques for Early Camera-Based Collaborative Perception
This seminar explores image processing techniques to enable early collaborative perception in camera-based systems, particularly within autonomous vehicles. As vehicles and devices increasingly rely on shared visual data to enhance situational awareness, the ability to process and interpret images efficiently and accurately becomes critical. Fundamentals of image preprocessing for collaborative perception, including denoising, contrast enhancement, and geometric corrections. Feature extraction and matching across multiple camera feeds to enable spatial and temporal alignment. Multi-view fusion strategies, such as stitching and depth estimation, to create a unified perception model. Compression and transmission techniques optimized for low-latency sharing of visual data between agents. Real-time object detection and tracking in distributed camera networks. Challenges and solutions in synchronizing heterogeneous camera systems under varying environmental conditions.
Literature:
[1]: Hussain, Manzoor, Nazakat Ali, and Jang-Eui Hong. "Vision beyond the field-of-view: A collaborative perception system to improve safety of intelligent cyber-physical systems." Sensors 22.17 (2022): 6610.
[2]: Yushan Han, Hui Zhang, Huifang Li, Yi Jin, Congyan Lang, and Yidong Li. 2023. Collaborative perception in autonomous driving: methods, datasets, and
challenges. IEEE Intelligent Transportation Systems Magazine, 15, 6, 131–151.
Supervisior: Marco Calipari, Michael Kühr
From AR/MR Interaction to IoT Integration: Exploring UX Patterns and Interoperability Challenges
Mixed Reality (MR) and Augmented Reality (AR) enable novel ways to present information and interact with digital content through Head-Mounted-Displays (HMDs). Beyond gaming and entertainment, these technologies can improve industrial and general IoT applications by offering spatial, hand-free, context-aware interfaces.
This seminar should analyze interaction and UX patterns unique to AR/MR in general and review how recent approaches apply them to IoT applications. Special attention should be drawn to the interoperability of these approaches with heterogeneous devices.
Literature:
[1] Designs and Interactions for Near-Field Augmented Reality: A Scoping Review, Jacob Hobbs, Christopher Bull, 2025, (https://doi.org/10.3390/informatics12030077)
[2] Creating the Internet of Augmented Things: An Open-Source Framework to Make IoT Devices and Augmented and Mixed Reality Systems Talk to Each Other, Oscar Blanco-Novoa, Paula Graga-Lamas, Miguel A. Vilar-Montesinos Tiago M. Fernandez-Carames, (https://doi.org/10.3390/s20113328)
Supervisor: Roman BinkertIoT-Driven Building Management: Interoperability and Adaptive Energy Strategies in Smart Buildings
IoT technologies transform traditional building management systems by enabling real-time monitoring, automation, and optimization across multiple domains like energy, comfort, or safety. In this seminar topic, the student shall analyze how IoT systems are integrated into building management systems, especially for energy management and adaptive optimization, such as dynamic heating/cooling, smart lighting, occupancy-based control, and the integration of renewable energy. Special attention should be given to the interoperability challenges when integrating heterogeneous devices and how to overcome them.
Literature:
[1] A review on enhancing energy efficiency and adaptability through system integration for smart buildings, Um-e-Habiba, Ijaz Ahmed, Muhammad Asif, Hassan Haes Alhelou, Muhammad Khalid, 2024 https://doi.org/10.1016/j.jobe.2024.109354
[2] Towards an Interoperable Approach for Modelling and Managing Smart Building Data: The Case of the CESI Smart Building Demonstrator, Omar Doukari, Boubacar Seck, David Greenwood, Haibo Feng and Mohamad Kassem, 2022 https://doi.org/10.3390/buildings12030362
Supervisor: Roman BinkertAdversarial Attacks against Vision Transformer-based Computer Vision Models
For the development of automated driving, cameras are important sensors to sense the environment. While these sensor are used for safety-critical applications such as object detection or lane keeping, multiple adversarial attacks are known [1, 2]. Many of these attacks are effective against convolutional neural networks (CNNs) but recently a new model architecture, based on Vision Transformers (ViTs) [3], gains popularity. While ViTs are also known to be vulnerable to adversarial attacks [4], their research is significantly underrepresented. Within this seminar, the student should research different digital and physical adversarial attacks against ViT-based models and highlight differences and similarities compared to CNN-based attacks as well as compare the ViT-based attacks with themselves.
Literature:
[1] N. Akhtar, A. Mian, N. Kardan, and M. Shah, “Advances in Adversarial Attacks and Defenses in Computer Vision: A Survey,” IEEE Access, vol. 9, 2021.
[2] A. Guesmi, M. A. Hanif, B. Ouni, and M. Shafique, “Physical Adversarial Attacks for Camera-Based Smart Systems: Current Trends, Categorization, Applications, Research Challenges, and Future Outlook,” IEEE Access, vol. 11, 2023.
[3] A. Dosovitskiy, et al., ”An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale.” International Conference on Learning Representations, 2021
[4] Z. Wei, J. Chen, M. Goldblum, Z. Wu, T. Goldstein, Y. G. Jiang, ”Towards transferable adversarial attacks on vision transformers,” In Proceedings of the AAAI Conference on Artificial Intelligence, vol. 36, no. 3, pp. 2668-2676, 2022.
Supervisor: Michael KührAutomotive Threat Analysis and Risk Assessment for AI-based Autonomous Vehicle Sensors
For the development of automated driving, cameras are important sensors to sense the environment. Such safety-critical sensors are crucial to operate highly autonomous vehicles but show severe vulnerabilities. To evaluate the risk of automotive cyber security risks, the International Organization for Standardization (ISO) defined threat analysis and risk assessment (TARA) methods in the automotive standard ISO 21434 [3]. Sensor data, such as camera images, are nowadays often used with AI-based algorithms. This opens new doors for both digital and physical attacks [1, 2]. The goal of this seminar is a critical analysis of the TARA methods given in ISO 21434 and the proposal of adaptions and/or extensions to cover also AI-related attacks against automotive sensors by using existing analysis [4] and/or systematically research new ones.
Literature:
[1] N. Akhtar, A. Mian, N. Kardan, and M. Shah, “Advances in Adversarial Attacks and Defenses in Computer Vision: A Survey,” IEEE Access, vol. 9, 2021.
[2] A. Guesmi, M. A. Hanif, B. Ouni, and M. Shafique, “Physical Adversarial Attacks for Camera-Based Smart Systems: Current Trends, Categorization, Applications, Research Challenges, and Future Outlook,” IEEE Access, vol. 11, 2023.
[3] International Organization for Standardization, “ISO/SAE 21434:2021(E): Road vehicles – Cybersecurity engineering,” International Organization for Standardization, CH-1214 Vernier, Geneva, Aug. 2021.
[4] K. Grosse and A. Alahi, “A qualitative AI security risk assessment of autonomous vehicles,” Transp. Res. Part C Emerg. Technol., vol. 169, no. 104797, 2024.
Supervisor: Michael KührKey Management in International Rail Traffic Management Systems
Key management in international rail traffic management systems represents one of the most critical cybersecurity challenges facing modern railway infrastructure. This research area focuses on the secure generation, distribution, storage, and lifecycle management of cryptographic keys that protect communications between train control systems operating across national borders.
The European Rail Traffic Management System (ERTMS) serves as the primary example of international rail interoperability, where the European Train Control System (ETCS) enables trains to operate seamlessly across different countries using unified signalling and communication protocols. Central to ERTMS security is the Key Management Centre (KMC), which generates and distributes 3DES keys to On-Board Units (OBUs) in trains and Radio Block Centres (RBCs) along railway lines. These cryptographic keys enable message authentication through the EuroRadio protocol, ensuring that only authorized entities can communicate within the rail network.
The current key management approach faces significant operational challenges. Tens of thousands of keys must be physically distributed using portable media devices such as USB drives and CDs, creating a prohibitively high management burden and introducing security vulnerabilities. This manual distribution process becomes particularly complex for international operations, where trains crossing borders require access to keys from multiple national KMCs. The existing system requires extensive pre-planning and coordination between infrastructure managers from different countries, often leading to delays and increased operational costs.
As part of this assignment, you will deep-dive the different key generation/negotiation/transportation/storage schemes used by different national railway (infrastructure) operators, concluding your research with a comparison of the different schemes and standardization proposals and highlighting shared strength/weaknesses.
Literature:
Thomas, R. J., Ordean, M., Chothia, T., & de Ruiter, J. (2017). “TRAKS: A Universal Key Management Scheme for ERTMS.”
European Union Agency for Railways. (2020). “Railway Cybersecurity Report.”
Prof. Dr. Katzenbeisser. "Why railway is safe but not secure (Talk)" - 37c3 Chaos Communication Congress
Franeková et al.. "Approaches to a solution of key management system for cryptography communications within railway applications"
Supervisor: Maximilian Lüdecke