Theses & Semester Projects
The Institute for Dynamic Systems and Control offers the following projects to ETH students:
- Studies on Mechatronics (SM)
- Bachelor Theses (BT)
- Semester Projects (SP)
- Master Theses (MT)
- You should read the ETH Citation Etiquette and working accordingly.
- For more information, read the ETH information on plagiarism.
- Furthermore, you are urged to use the Download IDSC Thesis Template (ZIP, 55.7 MB) to write your report.
How to apply:
- Please review the available projects below
- Send an email to the project contact.
Safe Real-Time Online Learning for Autonomous Racing
Safely adapting to uncertain environments is a key requirement for robust autonomy. Robust Model Predictive Control (RMPC) optimizes decisions while explicitly accounting for uncertainty in the predictions, rendering it particularly well suited for safety-critical systems. In this thesis, you will develop and deploy a safe and efficient robust model predictive controller that can adapt its model under strict safety constraints.
Keywords
Autonomous Racing, Online Learning, Gaussian Process MPC, Robust Control
Labels
Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2026-07-10 , Earliest start: 2026-08-03
Applications limited to ETH Zurich
Organization Research Zeilinger
Hosts Lahr Amon
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Ultrasonic Transducer Development
This project focuses on the experimental characterization and model-based optimization of a novel polymer-based (PVDF) acoustic transducer for underwater communication. Unlike conventional ceramic transducers, the PVDF approach enables inherently broadband operation with reduced internal reflections and improved signal fidelity, supporting significantly higher data rates in challenging underwater environments.
Labels
Semester Project , Collaboration , Internship , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2026-07-08
Organization Research D'Andrea
Hosts Ramachandran Aswin
Topics Information, Computing and Communication Sciences , Engineering and Technology , Physics
Teaching Assistant: Few-Shot Adaptation of RL Policies on Real-World Impact Wrenches
This project investigates reinforcement learning for impact-wrench control. Reinforcement Learning (RL) is attractive here because it can learn a low-latency policy directly from interaction, but the impact dynamics, together with the relative scarcity of real-world data, make the problem hard. The project focuses on methods that target high performance from limited samples, such as offline RL, residual RL, and finetuning a simulation- or meta-trained prior, with the aim of quantifying on the physical tool what performance is reachable and how much additional data is needed to close the gap under real-world model mismatch.
Keywords
Reinforcement Learning, Applied RL
Labels
Machine Learning (PBL) , Student Assistant / HiWi , Robotics (PBL)
Description
Goal
Contact Details
More information
Published since: 2026-07-02 , Earliest start: 2026-07-01
Applications limited to Department of Mathematics , Department of Computer Science , Department of Information Technology and Electrical Engineering , Department of Mechanical and Process Engineering
Organization Center for Project-Based Learning D-ITET
Hosts Carron Andrea
Topics Information, Computing and Communication Sciences
Few-Shot Adaptation of RL Policies on Real-World Impact Wrenches
This project investigates reinforcement learning for impact-wrench control. Reinforcement Learning (RL) is attractive here because it can learn a low-latency policy directly from interaction, but the impact dynamics, together with the relative scarcity of real-world data, make the problem hard. The project focuses on methods that target high performance from limited samples, such as offline RL, residual RL, and finetuning a simulation- or meta-trained prior, with the aim of quantifying on the physical tool what performance is reachable and how much additional data is needed to close the gap under real-world model mismatch.
Keywords
Reinforcement Learning, Applied RL
Labels
Master Thesis , Machine Learning (PBL) , Robotics (PBL)
Description
Goal
Contact Details
More information
Published since: 2026-07-01 , Earliest start: 2026-07-01
Applications limited to Department of Mathematics , Department of Computer Science , Department of Information Technology and Electrical Engineering , Department of Mechanical and Process Engineering
Organization Center for Project-Based Learning D-ITET
Hosts Ghignone Edoardo
Topics Information, Computing and Communication Sciences
Structured Learning for MoE and Looped Transformers
This project investigates how sparse and low-rank structured learning can be extended to Mixture-of-Experts (MoE) models and looped transformers. Building on SALAAD, the goal is to study whether these architectures can be decomposed into shared low-rank components and sparse expert- or iteration-specific residuals. The project will analyze MoE expert specialization, routing behavior, and looped-transformer refinement dynamics, and evaluate whether such structure can improve interpretability, parameter efficiency, memory footprint, or inference-time computation without sacrificing task performance.
Keywords
Mixture-of-Experts, Looped Transformers, Sparse and Low-Rank Learning, Efficient Transformers, Model Compression
Labels
Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2026-06-30 , Earliest start: 2026-07-01 , Latest end: 2026-12-31
Organization Research Zeilinger
Hosts Ma Hao
Topics Mathematical Sciences , Information, Computing and Communication Sciences
Teaching a robot to master a board game using Reinforcement Learning
Updated Project Description This project investigates how physical robots can achieve human-level performance in board games through Reinforcement Learning (RL). We use a custom 2D gantry robot integrated with ROS2 to play the dexterity-based game KLASK. Policies are trained in high-fidelity simulation—currently transitioning from NVIDIA Isaac Gym to Isaac Lab—to exploit GPU-parallel learning on RTX 4090 hardware. A central challenge is bridging the sim-to-real gap, addressed through an actuator policy model and extensive domain randomization. Self-play with opponent pools accelerates strategy discovery and improves robustness against human unpredictability. The system supports both human-vs-robot and robot-vs-robot configurations, enabling rapid fine-tuning of simulation-trained agents in the real world. Our ultimate objective is to develop an agent that consistently outperforms human players, demonstrating a complete pipeline from simulation to real-world dominance in a competitive board game setting. The project offers opportunities to refine hardware/software pipelines, explore advanced RL methods, and push the boundaries of sim-to-real transfer in physically interactive tasks.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2026-06-25 , Earliest start: 2024-08-15
Organization Research D'Andrea
Hosts Ramachandran Aswin
Topics Engineering and Technology
The Way of Water: Development of a fleet of water-based drones for live performance
Read available topics in notion link The Way of Water (wow.ethz.ch) fleet comprises 24 holonomic USVs, performing synchronized choreographies to music. Each vehicle is equipped with water fountains, RGB lighting, and mist generators, and utilizes a Fossen-based MPC with a multi-rate EKF (IMU + RTK-GPS) for tracking preplanned trajectories with ~5 cm accuracy. Time‐synced via GPS‐PPS, the swarm communicates over a hybrid Wi-Fi/4G network to broadcast real-time “trigger primitive” commands with <20 ms latency.
Keywords
Water based rovers, Electronics development, Distributed Robotics, Control Systems, Game design
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2026-06-25
Organization Research D'Andrea
Hosts Ramachandran Aswin
Topics Arts , Information, Computing and Communication Sciences , Engineering and Technology
Embeddings-augmented Scenario Retrieval and Generation for Autonomous Driving [Industrial thesis]
The goal of this thesis is to build upon existing scenario processing, logging, and simulation infrastructure to develop an enhanced scenario catalogue for autonomous driving. The catalogue should extend the currently available structured metadata extracted from logs and scenario files with richer representations of driving situations, traffic participants, and scenario context
Labels
Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2026-06-15 , Earliest start: 2026-07-01
Applications limited to ETH Zurich
Organization Research Frazzoli
Hosts Fiaschi Lorenzo
Topics Information, Computing and Communication Sciences , Engineering and Technology
The Way of Blood – Conduct Your Thesis in a Start-Up
Normothermic ex vivo perfusion extends organ preservation and enables rescue of marginal donor organs. However, rupturing of red blood cells under mechanical, biochemical, and thermal stress remains a critical barrier, limiting preservation duration and organ viability.
Keywords
MedTech, Biomedical Engineering, Mechanical Engineering, Fluid Dynamics
Labels
Semester Project , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2026-06-05 , Earliest start: 2026-06-08 , Latest end: 2027-02-28
Organization Research Onder
Hosts Machacek David
Topics Engineering and Technology
perfusionX – Heart Perfusion
Over 700,000 people await heart transplants globally, but only ~5,000 become available annually. Normothermic ex vivo perfusion can potentially salvage 20–30% of discarded hearts.
Keywords
Ex vivo organ perfusion, Heart transplantation, Biomedical engineering, Mechanical engineering, Startup environment
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2026-06-05 , Earliest start: 2026-06-08 , Latest end: 2027-02-28
Organization Research Onder
Hosts Machacek David
Topics Engineering and Technology
Control Systems Development for the CubliEvo (Evolution)
The Cubli, a research platform from ETH Zurich’s Institute for Dynamic Systems and Control, balances on its corner, jumps up, and stabilizes itself using reaction wheels. Our goal is to release a fully open‐source “CubliEvo” by redesigning its mechatronic hardware, finalizing an embedded control system, and producing comprehensive documentation so that anyone can reproduce the platform.
Keywords
Embedded Control, Nonlinear Dynamics, State Estimator, Cascaded Controller, Mechatronic Validation, Open-Source Documentation
Labels
ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2026-05-14 , Earliest start: 2024-09-01
Organization Research D'Andrea
Hosts Ramachandran Aswin
Topics Information, Computing and Communication Sciences , Engineering and Technology
Strategic Interactions of Future Mobility Systems
Mobility is typically self-optimized for a particular region to accommodate internal travel needs. However, as soon as one considers multiple, interacting regions (e.g., urban areas interacting with agglomerations, and agglomerations interacting with rural areas), important coordination issues occur, including scheduling mismatches, fleet allocations, and congestion peaks. In short, a mobility system composed of self-optimized mobility systems seems to often operate suboptimally. In this project, we will investigate the idea of strategic interactions of future mobility stakeholders across heterogeneous regions, such as urban areas, agglomerations, and rural areas, leveraging techniques from network design, optimization, game theory, and policy making.
Keywords
Optimization, Game theory, Multi-agent interactions, Transportation systems, Robotics
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2026-04-28 , Earliest start: 2026-04-28 , Latest end: 2027-02-28
Applications limited to ETH Zurich
Organization Research Frazzoli
Hosts He Mingjia
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Scaling Urban Simulation with Machine Learning
Urban transport policy decisions, from congestion pricing to new transit lines, rely on agent-based simulation to forecast their impact before implementation. MATSim, one of the most widely adopted open-source frameworks for large-scale transport simulation, is used by city governments and research institutions worldwide to model the daily mobility of millions of agents and inform real policy decisions. However, at metropolitan scale, simulation runtimes remain a fundamental bottleneck, limiting the number of scenarios that planners can feasibly evaluate.
Keywords
MATSim, Agent-Based Simulation, Machine Learning, Data-Driven Optimisation, High-Performance Computing, Transport AI
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2026-04-28 , Earliest start: 2026-04-29 , Latest end: 2027-08-31
Organization Research Frazzoli
Hosts He Mingjia
Topics Information, Computing and Communication Sciences , Engineering and Technology
SALAAD Beyond LLMs: Structured Sparse and Low-Rank Training for Multimodal Foundation Models
SALAAD is a structured training paradigm that introduces sparse and low-rank constraints during optimization to promote inherently compressible model representations without sacrificing task performance. While prior work has primarily focused on autoregressive language models, its applicability to broader model families remains unclear. In this project, we investigate whether SALAAD can generalize as a unified structured training framework across architectures and modalities, with a particular focus on multimodal foundation models such as vision-language and vision-language-action systems. These models introduce architectural heterogeneity and complex cross-modal interactions, posing new challenges for structured optimization. By combining insights from optimization, deep learning systems, and multimodal modeling, we aim to empirically and systematically evaluate the effectiveness of SALAAD in inducing compressible representations in such settings. Our findings seek to advance the understanding of structured training and its role in enabling efficient and elastic deployment of modern AI systems.
Keywords
Large Language Models, Foundation Models, Multimodal AI, Vision-Language Models, Vision-Language-Action Models, Deep Learning, Machine Learning, Artificial Intelligence, AI Systems, Representation Learning
Labels
Semester Project , Master Thesis , Other specific labels , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2026-04-07 , Earliest start: 2026-04-07 , Latest end: 2026-12-31
Organization Research Zeilinger
Hosts Ma Hao
Topics Information, Computing and Communication Sciences , Engineering and Technology
Direct Projects
The projects from Prof. Chris Onder's group are hosted on the student projects page.
The projects from Prof. Melanie Zeilinger's group are hosted on the student projects page.
Custom Projects
From time to time, project supervisors will develop custom student research projects to fit with a student's particular interests or skills.
If you are interested in doing a custom student research project, please email the project supervisor of your choice directly. We recommend that you carefully review their area of research before you contact them.
Please note that the decision of whether to develop a custom student project is at the full discretion of the project supervisor.