Student Projects
Student Projects by Associated Institutes
Spatially-Enhanced Linear-Attention Transformer for End-to-End Navigation
Transformer architectures have proven highly effective in sequence modelling, but their standard self-attention mechanism suffers from quadratic complexity in sequence length. The work by Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention (Katharopoulos et al., 2020) shows how self-attention can be reformulated as kernel feature maps with a linear complexity implementation. Separately, in navigation and robotics, end-to-end learning of navigation policies (e.g., via reinforcement learning) has shown strong promise; however, many architectures (such as standard RNNs) struggle to build spatial memory — that is, integrating observations from different viewpoints into a coherent spatial representation. For example, the recent work Spatially‑Enhanced Recurrent Memory for Long‑Range Mapless Navigation via End‑to‑End Reinforcement Learning (Yang et al., 2025) proposes Spatially-Enhanced Recurrent Units (SRUs) to address this. This motivates the question: can we combine the efficiency and scalability of linear-attention transformers with a spatial-enhancement mechanism tailored for navigation, thereby achieving end-to-end navigation with long-term spatial memory and efficient inference?
Labels
Semester Project , Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-10-28 , Earliest start: 2025-11-15 , Latest end: 2025-12-15
Organization Robotic Systems Lab
Hosts Yang Fan
Topics Information, Computing and Communication Sciences
Learning-Based Decoding for Event Cameras
Develop a learning-based decoder for event cameras to facilitate efficient real-world tasks.
Keywords
Event-based Vision, Streaming Algorithms, CPU, GPU, Encoder-Decoder Learning, Encoding, Decoding, Machine Learning
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-10-21
Organization Robotics and Perception
Hosts Zubic Nikola
Topics Mathematical Sciences , Engineering and Technology
Hand Gesture Recognition Using High-Density Surface Electromyography (HD-sEMG
The project aims to develop machine learning models for real-time hand gesture recognition using high-density surface electromyography (HD-sEMG) signals. This research focuses on improving neuroprosthetic control and rehabilitation by accurately decoding hand movements from HD-sEMG data.
Keywords
HD-sEMG, neuroprosthetics, biosignal monitoring, machine learning, hand gesture recognition, rehabilitation
Labels
ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2025-10-20
Organization Biomedical and Mobile Health Technology Lab
Hosts Qi Shuaixin
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Gait and Posture Analysis Using Multimodal Data from Smart Walkers
The project aims to develop machine learning models to analyze gait and posture using multimodal data collected from participants using a robotic walker equipped with multiple cameras and inertial sensors. This research focuses on detecting abnormalities in gait patterns and postural control, contributing to the development of rehabilitation and assistive technologies.
Keywords
gait analysis, posture monitoring, smart walker, motion capture, biomechanics, sensor fusion, rehabilitation, inertial sensors
Labels
ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2025-10-20
Organization Biomedical and Mobile Health Technology Lab
Hosts Qi Shuaixin
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Detection of Cardiovascular Abnormalities from Paper ECG Images Using Deep Learning
This project aims to develop a deep learning algorithm to detect and classify cardiovascular abnormalities directly from paper-based or image-scanned ECGs. By bypassing the need for full digitization, the objective is to enable rapid diagnosis and improve access to cardiovascular healthcare, particularly in resource-limited settings.
Keywords
ECG, cardiovascular disease, image classification, deep learning, global health, medical imaging, underserved populations
Labels
ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2025-10-20
Organization Biomedical and Mobile Health Technology Lab
Hosts Qi Shuaixin
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Instrumentation of ice hockey skates to determine on ice performance
This semester or master thesis aims to instrument ice hockey skates to enable direct measurement of three-dimensional forces exerted by competitive ice hockey players on the ice during training and test scenarios.
Keywords
sports engineering, portable force transducer, ice hockey, skate instrumentation
Labels
Semester Project , Master Thesis
Project Background
Your Task
Your Benefits
Your Profile
Contact Details
More information
Published since: 2025-10-17 , Earliest start: 2026-01-05 , Latest end: 2026-09-15
Applications limited to ETH Zurich
Organization Sensory-Motor Systems Lab
Hosts Wolf Peter
Topics Engineering and Technology
Smart Microcapsules for Biomedical Advances
This Master's thesis/semester project focuses on the microfluidic fabrication of microcapsules with multi-environmental responsiveness. The aim is to develop microcapsule-based microrobots capable of adapting to various environmental cues. We envision that these microrobots will be used for complex tasks in biomedical applications.
Keywords
Microfluidics, Microcapsules, Microrobotics, Responsive Polymers, Biomedical Engineering
Labels
Semester Project , Internship , Master Thesis , Student Assistant / HiWi , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2025-10-14 , Earliest start: 2026-01-19
Organization Multiscale Robotics Lab
Hosts Hu Minghan
Topics Medical and Health Sciences , Engineering and Technology , Chemistry
Intelligent Micromachines Made from Droplet-Based Factory
We invite applications for a Master's thesis / semester project that focuses on the fabrication of microrobots with custom shapes. Using our developed droplet printing technique, this project will explore how different microrobot shapes, created by different magnetic fields and materials, influence their control behaviors in blood vessels. This research aims to advance biomedical technologies, particularly in targeted drug delivery and minimally invasive procedures.
Keywords
Microrobotics, 4D Printing, Soft Materials, Biomedical Devices
Labels
Semester Project , Master Thesis , Student Assistant / HiWi , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2025-10-14 , Earliest start: 2026-02-02
Organization Multiscale Robotics Lab
Hosts Hu Minghan
Topics Engineering and Technology , Chemistry
Wireless Motion and Health Monitoring with Triboelectric Nanogenerators using a Custom Readout Circuit
This project aims to design and realize a custom single-PCB wireless readout circuit optimized for triboelectric nanogenerators (TENGs) for motion and health monitoring. Unlike previous designs that relied on assembling multiple commercial off-the-shelf PCBs, this project will integrate the functionalities of signal conditioning, energy harvesting, and wireless data transmission onto one compact, low-power circuit board). This PCB will be used to facilitate wireless joint angle analysis and fatigue monitoring during repetitive human movements using machine learning and deep learning.
Keywords
TENG, energy harvesting, wireless circuits, PCB design, low-power electronics, sensors, movement, mobile health, wearables
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-10-14 , Earliest start: 2025-10-01 , Latest end: 2026-08-31
Organization Biomedical and Mobile Health Technology Lab
Hosts Otesteanu Corin, Dr
Topics Engineering and Technology
Machine learning for Motion Estimation and Fatigue Monitoring Using Triboelectric Nanogenerators
This project explores the feasibility of using triboelectric nanogenerators (TENGs) for joint angle analysis and fatigue monitoring during repetitive human movements using machine learning and deep learning.
Keywords
machine learning, artificial intelligence, generative AI, triboelectric nanogenerator, joint angle estimation, motion analysis, fatigue
Labels
Semester Project , Internship , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-10-14 , Earliest start: 2025-09-01 , Latest end: 2026-08-01
Organization Biomedical and Mobile Health Technology Lab
Hosts Otesteanu Corin, Dr
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Learning-based object orientation prediction for handovers
Humans are exceptional at handovers. Besides timing and spatial precision, they also have a high-level understanding of how the other person wants to use the object that is handed over. This information is needed to hand over an object, such that it can be used directly for a specific task. While robots can reason about grasp affordances, the integration of this information with perception and control is missing.
Keywords
Robot-Human Handover, Human-Robot-Interaction, Mobile Manipulation, Robotics
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-10-13 , Earliest start: 2025-02-01 , Latest end: 2025-12-31
Organization Robotic Systems Lab
Hosts Scheidemann Carmen , Tulbure Andreea
Topics Information, Computing and Communication Sciences , Engineering and Technology
Fine-tuning Policies in the Real World with Reinforcement Learning
Explore online fine-tuning in the real world of sub-optimal policies.
Keywords
online fine-tuning, reinforcement learning (RL), continual learning, drones, robotics
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-10-13 , Earliest start: 2025-06-01 , Latest end: 2026-04-30
Organization Robotics and Perception
Hosts Geles Ismail
Topics Information, Computing and Communication Sciences , Engineering and Technology
Inverse Reinforcement Learning from Expert Pilots
Use Inverse Reinforcement Learning (IRL) to learn reward functions from previous expert drone demonstrations.
Keywords
Inverse Reinforcement Learning, Drones, Robotics
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-10-13 , Earliest start: 2025-06-01 , Latest end: 2026-04-30
Organization Robotics and Perception
Hosts Geles Ismail
Topics Information, Computing and Communication Sciences , Engineering and Technology
HandoverNarrate: Language-Guided Task-Aware Motion Planning for Handovers with Legged Manipulators
This project addresses the challenge of task-oriented human-robot handovers, where a robot must transfer objects in a manner that directly facilitates the human’s next action. In our prior work, we demonstrated that robots can present objects appropriately for immediate human use by leveraging large language models (LLMs) to reason about task context. However, integrating task-specific physical constraints—such as ensuring a full mug remains upright during transport—into the motion planning process remains unsolved. In this project, we aim to extend our existing motion planning framework for legged manipulators by incorporating such constraints. We propose using LLMs to dynamically generate task-aware constraint formulations based on high-level task descriptions and object states. These constraints will then be used to adjust the cost function of the model predictive controller in real time, enabling more context-sensitive and physically appropriate handovers.
Keywords
language-guided motion planning, legged robotics, human-robot collaboration
Labels
Semester Project , Bachelor Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-10-13
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Tulbure Andreea
Topics Information, Computing and Communication Sciences
Smart Stretchable Piezoelectric Hydrogels: From Joint Motion to Bioelectric Stimulation
We are developing an intelligent composite material composed of piezoelectric nanoparticles embedded in a stretchable hydrogel matrix with a kirigami-inspired design. Mechanical deformation activates the piezoelectric nanoparticles, generating bioelectric signals (like Pikachu) that can stimulate cell differentiation. This self-powered system shows promise for biomedical applications.
Keywords
Flexible electrodes optimization, piezoelectric mesh, biomedical applications.
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-10-09 , Earliest start: 2025-09-01 , Latest end: 2026-09-01
Organization Multiscale Robotics Lab
Hosts Chen Huimin
Topics Engineering and Technology
Efficient data processing and reporting in stroke neuro-rehabilitation
Project goal is to optimise existing and develop new algorithms into an efficient system for signal pre-processing, data storage, analysis and visualization in motor-neurorehabilitation. This data is generated by stroke patients wearing motion sensors during their therapy sessions. Key endpoint of the project is to display real-time and longitudinal therapy results, which can aid therapists and patients. The results of the project will help develop a more efficient therapy and is a key part of a larger project that seeks to develop an intelligent and closed-loop neural stimulation system for stroke rehabilitation.
Keywords
health biomarkers, data science, computer science, data visualization, data processing, real-time, internet of medical things, IoMT, healthcare internet of things, healthcare IoT
Labels
Semester Project , Internship , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-10-08 , Earliest start: 2023-05-09
Organization Rehabilitation Engineering Lab
Hosts Donegan Dane , Viskaitis Paulius
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology
How to Touch: Exploring Tactile Representations for Reinforcement Learning
Developing and benchmarking tactile representations for dexterous manipulation tasks using reinforcement learning.
Keywords
Reinforcement Learning, Dexterous Manipulation, Tactile Sensing
Labels
Semester Project , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-10-05 , Earliest start: 2025-11-01 , Latest end: 2026-06-01
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Bhardwaj Arjun , Zurbrügg René
Topics Information, Computing and Communication Sciences
Evaluating Engagement and Adherence to Robot-Assisted Rehabilitation: Can Large Language Model-Based Conversational Agents Support Stroke and TBI Patients During Unsupervised Therapy?
Stroke and traumatic brain injury (TBI) are major causes of long-term disability, often requiring intensive rehabilitation to restore motor function, particularly in the hands. While robotic-assisted therapy and mobile health (mHealth) applications offer promising tools for recovery, sustaining patient engagement outside clinical environments remains a significant challenge. This project explores the use of the RehabCoach app and ReHandyBot device in a hybrid rehabilitation program (i.e., two weeks in-clinic followed by six weeks of unsupervised home therapy). The focus is on collecting and analyzing conversational data between patients and the app’s large language model (LLM)-based conversational agent to evaluate how these interactions influence therapy adherence and engagement.
Keywords
Stroke rehabilitation, traumatic brain injury, mobile health, conversational AI, large language models, therapy adherence, unsupervised rehabilitation
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-10-02 , Earliest start: 2025-11-16 , Latest end: 2026-09-30
Organization Rehabilitation Engineering Lab
Hosts Retevoi Alexandra
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Behavioural and Cognitive Sciences
Smart Breathing for Stress Reduction – A Closed-Loop Mobile Application
This master thesis aims to develop a smartphone-based system for stress monitoring and reduction. The App, implemented in React Native (JavaScript) or another suitable framework, will use stress-related physiological parameters to deliver audio feedback that guides controlled breathing in real time.
Keywords
stress monitoring, breathing rate, App programming, mobile health, biofeedback
Labels
Master Thesis
Project Background
Your Task
Your Benefits
Your Profile
Contact Details
More information
Published since: 2025-10-02 , Earliest start: 2025-11-03 , Latest end: 2026-06-01
Organization Sensory-Motor Systems Lab
Hosts Padilla Neira Sara
Topics Medical and Health Sciences , Engineering and Technology
Integrating Robotic Hands with Embodied Intelligence for Industrial Mobile Manipulators
This thesis explores the integration of our newly developed robotic hands with embodied intelligence on a state-of-the-art industrial bi-manual mobile manipulator. The objective is to address typical warehouse and logistics scenarios by enabling generalizable, flexible manipulation skills that go beyond task-specific programming.
Keywords
Robotics, Bi-manual manipulation, Mobile manipulators, Embodied intelligence, Warehouse automation, Robotic hands
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-10-01 , Earliest start: 2025-10-13 , Latest end: 2026-04-30
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , University of Zurich , Wyss Translational Center Zurich
Organization Soft Robotics Lab
Hosts Blöchlinger Marc , Katzschmann Robert, Prof. Dr.
Topics Information, Computing and Communication Sciences , Engineering and Technology
Deep Learning of Residual Physics For Soft Robot Simulation
Incorporating state-of-the-art deep learning approaches to augment conventional soft robotic simulations for a fast, accurate and useful simulation for real soft robots.
Keywords
Soft Robotics, Machine Learning, Physical Modeling, Simulation
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-30 , Earliest start: 2025-03-01 , Latest end: 2026-03-01
Organization Soft Robotics Lab
Hosts Michelis Mike , Katzschmann Robert, Prof. Dr.
Topics Information, Computing and Communication Sciences , Engineering and Technology
Robotic Holder for Endoscopic Tools in Orthopedic Surgery
Robotic assistance has revolutionized minimally invasive surgery by enhancing precision, dexterity, and patient outcomes. While endoscopic robots, such as the da Vinci system, are commonly used for soft-tissue procedures, orthopedic applications remain limited to rigid platforms for positioning and drilling, lacking endoscopic access and haptic teleoperation capabilities. To address this gap, we envision a robotic system for endoscopic surgery consisting of two holders: one for the endoscope and one for surgical tools. Due to the complexity of the overall system, this master’s thesis focuses on the development of the tool holder. The holder will be designed to accommodate commonly used endoscopic tools: a grasper and dissector. It will function as the end-effector of a robotic arm, allowing precise positioning and manipulation of the tools. The complete system should be implemented in ROS 2 for control and simulation. The system performances will be evaluated on an experimental bench and demonstrated through phantom and Sawbones surgical scenarios. This project is conducted at Balgrist under the supervision of Prof. Dr. Fürnstahl.
Keywords
Robotics ; Endoscopic surgery ; Haptics ; Teleoperation ; Mechanical design ; ROS2 ; Robotic control ; Simulation ; CAD design ; Manufacturing ; Experimental design and validation
Labels
Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-30 , Earliest start: 2025-09-30
Organization Multiscale Robotics Lab
Hosts Gantenbein Valentin
Topics Medical and Health Sciences , Engineering and Technology
Causal Machine Learning and Data Fusion with Experts in the Loop for Spinal Cord Injury (SCI)
Causl Discovery aims to find causal relations from data, being increasingly important in various fields such as health science. Despite the growing amount of work on applying causal discovery methods with expert knowledge to areas of interest, few of them inspect the uncertainty of expert knowledge (what if the expert goes wrong?). This is highly important since in scientific fields, causal discovery with expert knowledge should be cautious and an approach taking expert uncertainty into account will be more robust to potential bias induced by individuals. Therefore, we aim to develop an iterative causal discovery method with experts in the loop to enable continual interaction and calibration between experts and data. Besides, fusing datasets from different sources is essential for holistic discovery and reasoning. This project will also focus on developing methods of machine learning and data fusion over distinct contexts under the scope of SCI. Based on the qualifications of the candidates, we can arrange a subsidy/allowance to cover traveling or living costs.
Keywords
Causal Discovery, Expert Knowledge, Iterative Algorithm, Data Fusion, Spinal Cord Injury
Labels
Master Thesis
Project Background
Your Task
Your Benefits
Your Profile
Contact Details
More information
Published since: 2025-09-29 , Earliest start: 2026-01-01 , Latest end: 2026-07-31
Organization Sensory-Motor Systems Lab
Hosts Paez Diego, Dr. , Paez Diego, Dr.
Topics Medical and Health Sciences , Mathematical Sciences , Information, Computing and Communication Sciences
Multi-Agent Local Collision Avoidance on Rough Terrain
Collaborative object transportation is a fundamental aspect of human interaction, particularly in scenarios where a single individual cannot manipulate an object due to its size or weight. Such tasks are common in practical domains, including construction sites and warehouse operations, where coordinated motion and spatial awareness are required to avoid collisions with environmental obstacles.
Keywords
Multi-Agent, Collision Avoidance, Rough Terrain
Labels
Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-25 , Earliest start: 2025-10-06
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Bray Francesca
Topics Engineering and Technology
Integrating Event-based Vision Capabilities into LLMs
This project explores integrating event-based vision into LLMs through custom layers and knowledge distillation, enabling efficient processing of sparse, asynchronous data for dynamic scene understanding in real-world applications.
Keywords
Event-based vision, large language models, multimodal AI, asynchronous data, sparse representation, high-dynamic range, fast motion, event cameras, knowledge distillation, autonomous navigation, human-robot interaction.
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-23
Organization Robotics and Perception
Hosts Zubic Nikola
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Event Representation Learning for Control with Visual Distractors
This project develops event-based representation learning methods for control tasks in environments with visual distractors, leveraging sparse, high-temporal-resolution event data to improve robustness and efficiency over traditional frame-based approaches.
Keywords
Event-based vision, representation learning, control policies, reinforcement learning, visual distractors, high-dynamic range, motion blur, sparse data, robust perception, locomotion, navigation.
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-23
Organization Robotics and Perception
Hosts Zubic Nikola
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Serial Manufacturing Platform for Magnetic Endovascular Devices
Electromagnetic navigation systems enable steerable endovascular devices to access previously unreachable brain regions for stroke and aneurysm treatment. This project develops a manufacturing setup for scalable production of these devices, followed by fabrication and testing to advance next-generation neurovascular therapies.
Keywords
Mechanical Engineering, Medical Robotics, Mechanical Design
Labels
Semester Project , Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-22 , Earliest start: 2025-09-22 , Latest end: 2026-04-01
Organization Multiscale Robotics Lab
Hosts Gantenbein Valentin
Topics Engineering and Technology
Improving Badminton Skills for Competitive Play in Legged Manipulators
We develop learning-based control methods to enable agile and coordinated robot badminton. Building on our recent Science Robotics work, where a quadrupedal manipulator intercepted shuttlecocks at speeds up to 12.06 m/s and played collaboratively with an human, this project aims to advance from collaborative play to competitive gameplay through more effective and strategic returns. The student will work hands-on with our robot, focusing on deploying reinforcement learning policies on hardware, integrating them with high-frequency control loops, and addressing sim-to-real challenges.
Keywords
Learning-based control; Autonomous badminton playing; Sim-to-real transfer
Labels
Master Thesis , ETH Zurich (ETHZ)
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-09-17 , Earliest start: 2025-09-30
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Robotic Systems Lab
Hosts An Tianxu
Topics Engineering and Technology
Codesign of shape and control: A study in autonomous perching
We aim to co-design a controller and the shape of a glider for a perching maneuver, involving deployment on the real system.
Keywords
optimal control, shape design, global optimization, MPC, reinforcement learning, fluid dynamics, glider, UAV
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-16 , Earliest start: 2025-10-01 , Latest end: 2027-03-31
Organization Robotics and Perception
Hosts Reiter Rudolf
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Radiometric Calibration for LiDAR Inertial Odometry
This project aims to adapt radiometric calibration techniques to robotics-grade LiDAR sensors to correct inconsistencies in intensity values caused by distance and incidence-angle effects. By producing more consistent LiDAR intensity images, the project seeks to enhance the robustness and accuracy of LiDAR Inertial Odometry (LIO) for localization and mapping.
Keywords
Robotics, LiDAR, Odometry, Calibration, State Estimation, Localization, Mapping
Labels
Semester Project
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-09-15 , Earliest start: 2025-09-21 , Latest end: 2026-04-30
Organization Autonomous Systems Lab
Hosts Pfreundschuh Patrick
Topics Information, Computing and Communication Sciences , Engineering and Technology
Custom Wireless Readout Circuit for Triboelectric Nanogenerator (TENG) Sensing and Energy Harvesting
This project aims to design and realize a custom single-PCB wireless readout circuit optimized for triboelectric nanogenerators (TENGs). Unlike previous designs that relied on assembling multiple commercial off-the-shelf PCBs, this project will integrate the functionalities of signal conditioning, energy harvesting, and wireless data transmission onto one compact, low-power circuit board.
Keywords
TENG, energy harvesting, wireless circuits, PCB design, low-power electronics, sensors
Labels
Semester Project
Description
Goal
Contact Details
More information
Published since: 2025-09-12 , Earliest start: 2025-10-01 , Latest end: 2026-08-31
Organization Biomedical and Mobile Health Technology Lab
Hosts Zada Muhammad
Topics Engineering and Technology
Development of a robotic hand actuated using soft HASEL actuators
Development of a forearm to actuate a tendon driven humanoid robotic hand using soft electrostatic HASEL actuator and demonstration of its capabilities with a Franka robotic arm.
Keywords
Electrostatic actuator, soft actuator, Robotic Hands, Soft Robotics
Labels
Semester Project
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-12 , Earliest start: 2025-09-22 , Latest end: 2025-12-01
Organization Soft Robotics Lab
Hosts Kazemipour Amirhossein , Hinchet Ronan , Katzschmann Robert, Prof. Dr.
Topics Engineering and Technology
Learning Precise Force Estimation Under Dynamic Loads
Force estimation on robotic arms is valuable for performing mobile manipulation tasks, such as pick-and-place or playing badminton. However, at high speeds, traditional methods such as momentum-based observers break down due to compounding errors in the dynamic model. Recent works have proposed using LSTMs to apply either a residual to a momentum based observer, or outright replace it with a learned model. However, the problem of doing this at high velocities remains relatively unexplored. In this project, you will develop a data-driven model for estimating forces during dynamic trajectories. You will investigate different model architectures and learning approaches.
Keywords
force estimation, machine learning, robotics
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-11 , Earliest start: 2025-09-14
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Fischer Oliver
Topics Engineering and Technology
Vision-Based Agile Aerial Transportation
Develop a vision-based aerial transportation system with reinforcement / imitation learning.
Keywords
aerial transportation, reinforcement learning (RL), drones, robotics
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-11 , Earliest start: 2025-05-01 , Latest end: 2026-02-28
Organization Robotics and Perception
Hosts Geles Ismail
Topics Information, Computing and Communication Sciences , Engineering and Technology
Simulation of cellular-force-microscopy-integrated lab-on-a-chip devices for in-situ biohybrid microrobot characterization
Microfluidic devices can be employed in biological research as lab-on-a-chip (LoC) and organ-on-a-chip (OoC) systems. These platforms enable precise in-situ cell manipulation within a highly controlled environment and have proven beneficial for the large-scale production of biohybrid magnetic microrobots. Integrating the LoC devices with the cellular force microscope (CFM) allows for real-time monitoring and in-situ mechanical characterization of biohybrid magnetic microrobots during their production and incubation. The goal of this project is to design and simulate an intelligent LoC/OoC device integrated with CFM for in-situ monitoring and characterization during microrobot production and drug testing.
Keywords
Lab-on-a-chip, organ-on-a-chip, magnetic microrobot, cellular force microscope, simulation
Labels
Semester Project , Bachelor Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-09 , Earliest start: 2025-09-09 , Latest end: 2026-03-31
Organization Multiscale Robotics Lab
Hosts Zhu Jiawei
Topics Engineering and Technology
Label-Efficient Representation Learning for EEG Time Series
This master`s thesis aims to explore self-supervised learning and transformer-based sequence models to develop label-efficient EEG representations.
Keywords
Deep Learning; Label-efficient Learning; Transformers; Sequence Models
Labels
Master Thesis
Project Background
Your Task
Your Benefits
Your Profile
Contact Details
More information
Published since: 2025-09-06
Organization Sensory-Motor Systems Lab
Hosts Dey Sharmita
Topics Information, Computing and Communication Sciences
Gaussian Splatting Super Resolution
Radiance fields like 3D Gaussian Splatting enable fast, high-fidelity scene reconstruction from real-time data. However, reconstruction quality improves in noticeable stages—e.g., coarse at 500 epochs, detailed at 5000. This project explores whether AI-based upscaling and denoising techniques can be used to predict and accelerate these improvements during training. By anticipating reconstruction quality transitions, we aim to reduce training time while maintaining visual fidelity—advancing the efficiency of radiance field learning.
Keywords
Gaussian Splatting, Computer Vision, Reconstruction
Labels
Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-05 , Earliest start: 2025-11-01 , Latest end: 2026-07-31
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Patil Vaishakh , Wilder-Smith Max
Topics Information, Computing and Communication Sciences
Informed Exploration in Reinforcement Learning via Sampling-Based Planning
Robust locomotion across diverse terrains remains one of the most challenging and exciting problems in legged robotics. While deep reinforcement learning (RL) has enabled impressive quadrupedal behaviors, generalization to unseen or complex terrains often suffers from low sample efficiency. To address this, researchers typically rely on labor-intensive curriculum design, where training terrains are manually structured to progress from easy to difficult, while trying to avoid catastrophic forgetting. These staged curricula allow the RL agent to master simple environments before tackling more complex ones. However, even with such manually designed curricula, RL training can take several days. This is partly because the agent does not retain knowledge about where it has succeeded or failed, nor does it use this experience to inform future exploration. This project aims to develop an automated approach to guide the RL exploration by leveraging sampling-based planning techniques. We formulate the training process as a graph expansion problem, wherein the terrain space is incrementally explored to maximize gains in locomotion robustness and performance. Can the Voronoi bias inherent in sampling-based planning steer the agent toward more effective and sample-efficient training trajectories? If successful, this approach could enable the use of real-world terrain scans for locomotion training, reducing dependence on expert-designed environments and substantially accelerating the training process. The project brings together concepts from sampling-based planning, adaptive curricula, and locomotion learning.
Keywords
Curriculum Learning, Quadruped Locomotion, Sampling-Based Planning, Reinforcement Learning
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-05
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , University of Zurich
Organization Robotic Systems Lab
Hosts Klemm Victor , Qu Kaixian
Topics Information, Computing and Communication Sciences
Event-based Object Segmentation for Vision-Guided Autonomous Systems
This project develops an end-to-end pipeline for object segmentation using event camera data, leveraging their microsecond latency, high dynamic range, and sparse asynchronous output to overcome the limitations of frame-based vision. By combining advanced deep learning architectures with rigorous benchmarking, the approach aims to deliver accurate, temporally consistent, and robust segmentation for real-time applications in navigation and control.
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-05 , Earliest start: 2025-09-05 , Latest end: 2026-12-01
Organization Robotics and Perception
Hosts Pellerito Roberto
Topics Engineering and Technology
Novel Learning Paradigms for Low-latency and Efficient Vision
Design and implement efficient event-based networks to achieve low latency inference.
Keywords
Computer Vision, Event Cameras
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-05 , Earliest start: 2024-12-12
Organization Robotics and Perception
Hosts Pellerito Roberto , Messikommer Nico
Topics Information, Computing and Communication Sciences
Vision Language Action models for Drones
This project explores generative modeling of drone flight paths conditioned on natural language commands and spatial constraints, aiming to produce plausible 3D trajectories for training reinforcement learning policies. We investigate model architectures, data sources, and trajectory extraction methods to ensure generated paths are both physically feasible and stylistically aligned with textual descriptions.
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-05 , Earliest start: 2025-09-05 , Latest end: 2026-12-01
Organization Robotics and Perception
Hosts Pellerito Roberto
Topics Information, Computing and Communication Sciences , Engineering and Technology
Event-Based Tactile Sensor for Humanoid Robotic Hands
Humanoid robots require tactile sensing to achieve robust dexterous manipulation beyond the limits of vision-based perception. This project develops an event-based tactile sensor to provide low-power, high-bandwidth force estimation from material deformation, with the goal of integrating it into a human-scale robotic hand.
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-05 , Earliest start: 2026-01-05 , Latest end: 2026-12-01
Organization Robotics and Perception
Hosts Pellerito Roberto
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Neural Vision for Celestial Landings (in collaboration with European Space Agency)
In this project, you will investigate the use of event-based cameras for vision-based landing on celestial bodies such as Mars or the Moon.
Keywords
event-based camera, vision-based navigation
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-05 , Earliest start: 2025-09-01 , Latest end: 2026-09-01
Applications limited to University of Zurich , ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , Institute of Robotics and Intelligent Systems D-MAVT , IBM Research Zurich Lab , Empa
Organization Robotics and Perception
Hosts Pellerito Roberto
Topics Engineering and Technology
Learning Perception-Aware Navigation Utilizing MPC Layers
In this project, we will use uncertainty-aware localization and an MPC layer within an RL policy to navigate challenging environments.
Keywords
model predictive control, robotics, computer vision, reinforcement learning
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-04 , Earliest start: 2025-10-01 , Latest end: 2026-09-30
Organization Robotics and Perception
Hosts Reiter Rudolf
Topics Information, Computing and Communication Sciences , Engineering and Technology
M4A: Multi-Modal Multi-Mission Autonomy
In modern forestry, selective thinning is used to manage forests more sustainably than traditional clear-cut techniques. Robots can play a key role in this process by providing the foresters with high quality information. But how do we create large scale maps of a forest? As the robots must be mobile, battery life is a limiting factor. Thus, how can we use maps of previous missions for safe robot navigation?
Keywords
Robotics, Drones, MAV, Exploration, Multi Session, Place Recognition, Multi Sensor, SLAM, Forestry, Digital Twin
Labels
Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-04 , Earliest start: 2025-09-14
Applications limited to ETH Zurich
Organization Mobile Robotics Lab
Hosts Freissmuth Leonard
Topics Information, Computing and Communication Sciences
Vision-Based World Models for Real-Time Robot Control
This project aims to use vision-based world models as a basis for model-based reinforcement learning, aiming to achieve a generalizable approach for drone navigation.
Keywords
Robotics, Autonomous Systems, Computer Vision, Foundation Models, Reinforcement Learning
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-04 , Earliest start: 2025-05-01 , Latest end: 2026-02-28
Organization Robotics and Perception
Hosts Reiter Rudolf
Topics Information, Computing and Communication Sciences
Vision-Based Reinforcement Learning in the Real World
We aim to learn vision-based policies in the real world using embedded optimization layers within reinforcement learning.
Keywords
Robotics, Model Predictive Control, Computer Vision, Reinforcement Learning
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-04 , Earliest start: 2025-05-01 , Latest end: 2026-02-01
Organization Robotics and Perception
Hosts Reiter Rudolf
Topics Information, Computing and Communication Sciences , Engineering and Technology
Real-time control of neural stimulation for stroke patients.
Real-time analysis of movement kinematics can benefit multiple different strategies in rehabilitation after stroke, including allowing closed-loop brain stimulation. Use of inertial measurement units (IMUs) allows detection of movement and extraction of kinematic features, but application in real-time remains challenging. This project will develop algorithms for real-time movement data analysis and feature extraction in typical rehabilitation tasks and general real-life movements. In turn, these algorithms will be applied to control novel brain stimulation approaches in stroke neurorehabilitation.
Keywords
Inertial measurement unit, IMU, movement tracking, machine learning, real-time, signal processing
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-09-02 , Earliest start: 2023-05-09
Organization Rehabilitation Engineering Lab
Hosts Viskaitis Paulius , Donegan Dane
Topics Engineering and Technology
Event-Guided 3D Gaussian Splatting for HDR Reconstruction and Relighting
This project investigates how asynchronous event streams can enhance 3D Gaussian Splatting for high dynamic range (HDR) reconstruction and photorealistic relighting. The aim is to achieve faithful scene reconstructions under challenging low-light conditions by fusing event data with sparse or degraded frames.
Keywords
Event cameras, 3D Gaussian Splatting, 3D reconstruction, low-light vision, high dynamic range, event-based vision
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-01 , Earliest start: 2025-09-01
Organization Robotics and Perception
Hosts Zou Rong
Topics Information, Computing and Communication Sciences , Engineering and Technology
Self-locking cuff electrodes for the nerve stimulation
Nerve cuff electrodes are designed for reliable recording and stimulation of peripheral nerves, as illustrated in Figure A. This project aims to develop a wireless, self-locking cuff electrode tailored specifically for nerve stimulation, as depicted in Figure B&C.
Keywords
Flexible electrodes design, FEM simulation, Biomedical applications
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-01 , Earliest start: 2025-09-01 , Latest end: 2026-09-01
Organization Multiscale Robotics Lab
Hosts Chen Huimin
Topics Engineering and Technology
Reflection and Ghosting Removal with Event Streams
This project explores leveraging event-camera signals to suppress reflections and ghosting in images captured through glass or other reflective surfaces. By exploiting temporal and motion differences between scene content and reflection artifacts, the goal is to disentangle true scene information from distracting overlays.
Keywords
Event cameras, reflection removal, ghosting suppression, image restoration, computational photography, event-based vision
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-01 , Earliest start: 2025-09-01
Organization Robotics and Perception
Hosts Zou Rong
Topics Information, Computing and Communication Sciences , Engineering and Technology
Reinforcement Learning for Magnetic Levitation
This project explores the control of freely levitating objects in space using electromagnetic navigation using reinforcement learning. By generating controlled magnetic fields, we can wirelessly manipulate the position and orientation of the levitating magnetic object. A fully functional experimental platform is already available, providing a rich testbed for exploring advanced control strategies.
Keywords
reinforcement learning
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-01 , Earliest start: 2025-09-02 , Latest end: 2026-12-31
Organization Multiscale Robotics Lab
Hosts Zughaibi Jasan
Topics Engineering and Technology
Event-based Occlusion Suppression for Robust Detection and VO in Adverse Weather
This project explores the effect of dynamic occlusions on the performance of event-based detection and VO algorithms. It involves the development of low-latency suppression strategies that mitigate such effects, restoring robust perception in adverse weather.
Keywords
Event cameras, occlusion suppression, visual odometry, object detection, adverse weather, event-based vision
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-09-01 , Earliest start: 2025-09-01
Organization Robotics and Perception
Hosts Zou Rong
Topics Information, Computing and Communication Sciences , Engineering and Technology
Meta-model-based-RL for adaptive flight control
This research project aims to develop and evaluate a meta model-based reinforcement learning (RL) framework for addressing variable dynamics in flight control.
Keywords
Model-based Reinforcement Learning, Meta Learning, Drones, Robotics
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-31 , Earliest start: 2025-05-01 , Latest end: 2026-04-30
Organization Robotics and Perception
Hosts Geles Ismail
Topics Information, Computing and Communication Sciences , Engineering and Technology
Advancing Vibrotactile Sensory Stimulation for Brain Imaging
Design of comfortable textiles and stimulation protocol for a pneumatic, MRI-compatible vibrotactile stimulator
Keywords
touch sensation, tactile illusions, neurorehabilitation, MRI
Labels
Semester Project , Internship , Master Thesis
Project Background
Your Task
Your Benefits
Your Profile
Contact Details
More information
Published since: 2025-08-28 , Earliest start: 2025-09-15 , Latest end: 2026-06-26
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Sensory-Motor Systems Lab
Hosts Wolf Peter
Topics Medical and Health Sciences , Engineering and Technology
Spiking Architectures for Advanced Event-Based Temporal Reasoning
This thesis explores novel spiking neural network architectures that leverage event-based vision data and emergent neural synchronization to enable efficient and robust temporal reasoning for dynamic scene understanding and sequential tasks.
Keywords
Spiking Neural Networks (SNNs), Event-Based Vision, Neuromorphic Computing, Temporal Reasoning, Event Cameras, Neural Synchronization, Dynamic Scene Understanding, Gesture Recognition, Robot Navigation, Computational Efficiency, Biological Plausibility
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-27
Organization Robotics and Perception
Hosts Zubic Nikola
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Rethinking RNNs for Neuromorphic Computing and Event-based Vision
This thesis develops hardware-optimized recurrent neural network architectures with novel parallelization and kernel-level strategies to efficiently process event-based vision data for real-time neuromorphic and GPU-based applications.
Keywords
Recurrent Neural Networks (RNNs), LSTMs, GRUs, Event-Based Vision, Neuromorphic Computing, Hardware-Aware Optimization, Parallelization, Custom Kernels, GPU Acceleration, Sparse Data Processing, Real-Time Inference, Event Cameras
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-27
Organization Robotics and Perception
Hosts Zubic Nikola
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Design data acquisition solution for smart clothing
The aim of this project is to develop and improve wearable electronics solutions for data acquisition from textile-based sensors used in our smart clothing.
Keywords
smart clothing, wearable technology, textile sensor, fitness tracking, sports medicine, PCB, electronics, computer science
Labels
Semester Project , Internship , Bachelor Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-27 , Earliest start: 2025-07-01 , Latest end: 2025-12-31
Organization Biomedical and Mobile Health Technology Lab
Hosts Ahmadizadeh Chakaveh
Topics Information, Computing and Communication Sciences , Engineering and Technology
Better Scaling Laws for Neuromorphic Systems
This project explores and extends the novel "deep state-space models" framework by leveraging their transfer function representations.
Keywords
deep learning, state space models, transfer function, parameterizations, S4 model, Fourier Transform, Convolution, Neuromorphic Systems, Neuromorphic, Sequence Modeling, Event-based Vision
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-26
Applications limited to University of Zurich , ETH Zurich
Organization Robotics and Perception
Hosts Zubic Nikola
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Reinforcement Learning for Whole-body Robot Manipulation
Reinforcement learning is deployed to make the robot interact with and move large objects. Additionally, sampling-based methods can be used to improve training speed. The scientific selection, evaluation, and comparison of learning methods and control architectures provides valuable insights into advanced robotic manipulation.
Keywords
Robotics, Reinforcement Learning, Manipulation, Humanoid,
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-08-25 , Earliest start: 2025-09-01 , Latest end: 2026-08-31
Organization Robotic Systems Lab
Hosts Bruedigam Jan
Topics Information, Computing and Communication Sciences , Engineering and Technology
Yell at your robot: Learning navigation with ANYmal
Help advance the next generation of quadrupedal robots by combining Vision-Language-Action (VLA) models with a Forward Dynamics Model for embodiment-aware navigation. You will develop and test a system that predicts safe, physically feasible actions, validated in simulation and real hardware. This is a unique chance to work on cutting-edge embodied AI in collaboration with Stanford’s ASL.
Keywords
Robotic Navigation, Vision-Lanaguage-Action Models, Foundational Models, Forward Dynamics Models, Robot Learning, Machine Learning, Deep Learning
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-08-21
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , Swiss National Science Foundation , University of Zurich
Organization Robotic Systems Lab
Hosts Roth Pascal , Frey Jonas
Topics Information, Computing and Communication Sciences
Understanding your environment from images for RL-Navigation
Tackle the challenge of learning how quadrupeds understand terrain and risk by training a traversability module grounded in real-world data from ANYmal. The project combines dataset-driven learning with reinforcement learning to build an embodiment-aware navigation system. Work with Stanford’s ASL on a project that blends semantics, dynamics, and robotics.
Keywords
Robot Learning, Reinforcement Learning, Deep Learning, Navigation, Robot Navigation
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-08-21
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , IBM Research Zurich Lab , Swiss National Science Foundation
Organization Robotic Systems Lab
Hosts Roth Pascal , Frey Jonas
Topics Information, Computing and Communication Sciences
Exploring upper limb impairments using explainable AI on Virtual Peg Insertion Test data
This thesis aims to apply explainable AI techniques to analyze time series data from the Virtual Peg Insertion Test (VPIT), uncovering additional metrics that describe upper limb impairments in neurological subjects, such as those with stroke, Parkinson's disease, and multiple sclerosis. By preserving the full dimensionality of the data, the project will identify new patterns and insights to aid in understanding motor dysfunctions and support rehabilitation.
Keywords
Machine learning, rehabilitation, neurology, upper limb, impairment, explainable AI, SHAP, novel technology, assessment, computer vision, artificial intelligence
Labels
Master Thesis
Description
Goal
Tasks
Your Profile
Contact Details
More information
Published since: 2025-08-21 , Earliest start: 2025-06-01
Organization Rehabilitation Engineering Lab
Hosts Domnik Nadine
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Comparing the Virtual Peg Insertion Test (VPIT) with the haptic device Inverse3 for assessing upper limb function
This thesis will compare the Virtual Peg Insertion Test (VPIT) with the Inverse3 haptic device by Haply to evaluate its effectiveness as a tool for assessing upper limb function. The focus will be on comparing both the hardware features and software capabilities to determine if the Inverse3 can serve as a valid alternative to VPIT for clinical assessments.
Keywords
Haptic device, virtual environment, rehabilitation, programming, health technology, assessment, software, hardware
Labels
Collaboration , Master Thesis
Description
Goal
Tasks
Your Profile
Contact Details
More information
Published since: 2025-08-21 , Earliest start: 2025-06-01
Organization Rehabilitation Engineering Lab
Hosts Domnik Nadine
Topics Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Markerless Pose Tracking for Remote Physical Therapy
The aim of this project is to collect a comprehensive dataset of human movements, and augmenting it with synthetic datasets, to train computer vision models for accurate human pose estimation in therapeutic settings.
Keywords
Human Pose Estimation, Physiotherapy, Movement Science, Motion Capturing
Labels
Internship
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-08-20 , Earliest start: 2025-09-01 , Latest end: 2026-09-01
Organization Sensory-Motor Systems Lab
Hosts Rode David
Topics Medical and Health Sciences , Engineering and Technology
Mechanophores for advanced wearable strain and pressure sensors
The goal of the project is to synthesize and characterize a number of small molecules capable of acting as mechanophore addition to various polymers. These polymers would then be used as wearable strain or pressure sensors.
Keywords
mechanophore, polymer, wearable, sensor, color, strain, pressure
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-19
Organization Biomedical and Mobile Health Technology Lab
Hosts Shokurov Aleksandr
Topics Engineering and Technology , Chemistry
Point-of-Care Sensor for Urinary Iodine
The goal of the project is to develop a cheap and disposable sensor capable of determination of iodine levels in human urine for early diagnostic purposes.
Keywords
electrochemistry, iodine, nutrition, health, point of care
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-19
Organization Biomedical and Mobile Health Technology Lab
Hosts Shokurov Aleksandr
Topics Medical and Health Sciences , Engineering and Technology , Chemistry
Development of a modular exosuit for upper extremity support
In this project you will advance the modularity of an upper-limb exosuit (Myoshirt, see: https://sms.hest.ethz.ch/research/current-research-projects/wearable-robots-for-assistance-and-rehabilitation/The%20Myoshirt.html) to extend the configurable support modalities.
Keywords
rehabilitation robotics, assistive robotics, therapy, assistance, exosuits, soft wearable robotics, testing, evaluation
Labels
Master Thesis
Project Background
Your Task
Your Benefits
Your Profile
Contact Details
More information
Published since: 2025-08-14 , Earliest start: 2025-09-08 , Latest end: 2026-04-30
Organization Sensory-Motor Systems Lab
Hosts Esser Adrian
Topics Medical and Health Sciences , Engineering and Technology
Kinematics and Dynamics of Human-Like Locomotion in LocoMuJoCo Simulations
This Master’s thesis at ETH Zürich focuses on benchmarking and improving humanoid locomotion policies in the LocoMuJoCo simulation environment. The project develops a complete pipeline for analyzing kinematic and dynamic metrics, enabling direct comparison of simulated humanoid locomotion with real-world human motion capture data.
Keywords
Humanoid Locomotion, MuJoCo, Motion Analysis, Deep Reinforcement Learning, Simulation
Labels
Internship , Master Thesis , ETH Zurich (ETHZ)
Project Background
Your Task
Your Benefits
Your Profile
Contact Details
More information
Published since: 2025-08-12 , Earliest start: 2025-09-01
Organization Sensory-Motor Systems Lab
Hosts Paez Diego, Dr.
Topics Medical and Health Sciences , Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Foundation models for generalizable construction machine automation
Autonomous operation of heavy construction machinery promises major gains in safety, efficiency, and scalability for the multi-trillion-dollar construction industry. Excavators present a unique challenge, combining the navigation demands of autonomous driving with the precision of robotic manipulation in unstructured, dynamic environments. This project explores Vision-Language-Action (VLA) and diffusion-based policy learning to develop generalizable excavation policies. Leveraging large-scale demonstration data naturally collected during human operation, we aim to train multimodal control models capable of adapting across sites, lighting conditions, and machine types. The approach will be evaluated through simulation benchmarks and, if possible, real-world trials, advancing the path toward scalable, generalist autonomy for robotic construction.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-08-12 , Earliest start: 2025-08-31 , Latest end: 2026-03-31
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Robotic Systems Lab
Hosts Nan Fang , Zhang Weixuan
Topics Engineering and Technology
Collision Avoidance - Master Thesis at Avientus
Avientus is a startup that specializes in developing cutting-edge, heavy-duty automated drone transportation systems designed to revolutionize logistics and industrial applications. To further enhance the safety and reliability of their drones, we are offering a Master Thesis opportunity in the field of collision avoidance for drones.
Keywords
Collision avoidance, Computer vision, Drones
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-08-12 , Earliest start: 2025-08-01 , Latest end: 2026-02-28
Applications limited to ETH Zurich
Organization Autonomous Systems Lab
Hosts Mascaro Rubén , Chli Margarita
Topics Information, Computing and Communication Sciences
From Human Videos to Robot Control: Correcting the Embodiment Gap for Dexterous Manipulation
Several strategies tackle the embodiment gap between human demonstrations and robot execution. Some remove human or robot hands from videos via segmentation and inpainting to produce agent‑agnostic inputs or uniform end-effector cues, while others replace human hands with rendered robot manipulators to synthesize robot views. Latent space alignment methods enforce a shared embedding for human and robot observations, and recent cross‑embodiment techniques inpaint source robot embodiments into target frames during testing to maintain visual consistency across domains. In this project, we aim to systematically compare these embodiment gap-bridging methods under a unified evaluation protocol. Unlike prior work that typically focuses on one technique in isolation, we implement and benchmark multiple approaches on a real robotic platform equipped with a dexterous five-fingered hand.
Keywords
egocentric vision, representation learning, robotic manipulation
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-11 , Earliest start: 2025-10-01 , Latest end: 2026-05-01
Organization Computer Vision and Geometry Group
Hosts Gavryushin Alexey , Portela Tifanny , Zurbrügg René
Topics Information, Computing and Communication Sciences
Wearable device for non-invasive assessment of fatigue
The goal of the project is to develop a wearable device capable of non-invasive measurement of human biomarkers related to performance and fatigue during exercise.
Keywords
wearable, fatigue, exercise, sensor, athlete, non-invasive, sport
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-11
Organization Biomedical and Mobile Health Technology Lab
Hosts Shokurov Aleksandr
Topics Medical and Health Sciences , Engineering and Technology , Chemistry
ADHD-Biomarker Study: Software development of Data Collection and Data Management Structures
The Rehabilitation Engineering Laboratory (ReLab), ETH Zurich is running a paediatric study that integrates fNIRS, EEG, eye-tracking and motion sensing to establish objective biomarkers for Attention-Deficit/Hyperactivity Disorder (ADHD). You will create the technical backbone of a multimodal data-collection setup – coding computerized attention tests, synchronizing hardware, and delivering clean, well-documented datasets for downstream research.
Keywords
ADHD, Biomarker, fNIRS, EEG, Eye-tracking, Motion sensing, Data collection, Data management, Software engineering
Labels
Semester Project , Course Project , Internship , Lab Practice , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Tasks
Your Profile
Contact Details
More information
Published since: 2025-08-08 , Earliest start: 2025-08-10 , Latest end: 2026-03-31
Organization Rehabilitation Engineering Lab
Hosts Schönhammer Josef
Topics Medical and Health Sciences , Engineering and Technology , Behavioural and Cognitive Sciences
ADHD-Biomarker Study: Neuropsychological Task Development and Data Analytics
The Rehabilitation Engineering Laboratory (ReLab), ETH Zurich is running a pediatric study that integrates fNIRS, EEG, eye-tracking and motion sensing to establish objective biomarkers for Attention-Deficit/Hyperactivity Disorder (ADHD). You will drive the scientific backbone of the project – optimizing child-specific task parameters, leading participant interaction, and conducting and presenting statistical analyses.
Keywords
ADHD, Biomarkers, Neuropsychology, fNIRS, EEG, Eye-tracking, Motion Capture, Data Analytics, Statistical Analysis
Labels
Semester Project , Course Project , Internship , Lab Practice , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Tasks
Your Profile
Contact Details
More information
Published since: 2025-08-08 , Earliest start: 2025-08-10 , Latest end: 2026-03-31
Organization Rehabilitation Engineering Lab
Hosts Schönhammer Josef
Topics Medical and Health Sciences , Engineering and Technology , Behavioural and Cognitive Sciences
Learning-based Control for Magnetic Levitation
This project explores the control of freely levitating objects in space using electromagnetic navigation, as illustrated in the image above. By generating controlled magnetic fields, we can wirelessly manipulate the position and orientation of the levitating magnetic object. A fully functional experimental platform is already available, providing a rich testbed for exploring advanced control strategies.
Keywords
magnetic levitation control, learning-based control, Model Predictive Control
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-06 , Earliest start: 2025-08-07 , Latest end: 2026-12-31
Organization Multiscale Robotics Lab
Hosts Zughaibi Jasan
Topics Engineering and Technology
Neural Architecture Knowledge Transfer for Event-based Vision
Perform knowledge distillation from Transformers to more energy-efficient neural network architectures for Event-based Vision.
Keywords
deep neural networks, knowledge transfer, knowledge distillation, event cameras, event-based vision, sequence modeling, transformers, low-energy vision
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-08-04
Applications limited to ETH Zurich , University of Zurich
Organization Robotics and Perception
Hosts Zubic Nikola
Topics Information, Computing and Communication Sciences , Engineering and Technology
Mechanistic Evaluation of taVNS in Motor Adaptation for Stroke Rehabilitation
This study aims to evaluate the mechanisms by which transauricular vagus nerve stimulation (taVNS) may facilitate motor learning and adaptation, focusing on reticulospinal tract (RST) activation. A set of assessments will be used to build a comprehensive neurophysiological profile, providing insights relevant to developing taVNS-based clinical interventions. In this role, you will engage in study design, data collection, and analysis, gaining hands-on experience in mechanistic neurophysiological research with direct clinical applications.
Keywords
Transauricular vagus nerve stimulation (taVNS) Reticulospinal tract (RST) Motor learning StartReact paradigm Stroke rehabilitation Neurophysiology Multisite EMG Motor pathway activation Clinical neurostimulation
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-08-04 , Earliest start: 2024-11-24 , Latest end: 2025-10-26
Organization Rehabilitation Engineering Lab
Hosts Viskaitis Paulius
Topics Medical and Health Sciences , Engineering and Technology
Reiforcement Learning of Pretrained Trasformer Models
We want to train RL agents on our new particle simulator, accelerated on the GPU via warp in Isaac sim.
Keywords
Keywords: particle simulation, omniverse, warp, reinforcement learning
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-29 , Earliest start: 2025-09-01 , Latest end: 2026-03-31
Organization Robotic Systems Lab
Hosts Terenzi Lorenzo
Topics Information, Computing and Communication Sciences
Multiagent Reinforcement Learning in Terra
We want to train multiple agents in the Terra environment, a fully end-to-end GPU-accelerated environment for RL training.
Keywords
multiagent reinforcement learning, jax, deep learning, planning
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-29 , Earliest start: 2025-09-01 , Latest end: 2026-03-31
Organization Robotic Systems Lab
Hosts Terenzi Lorenzo
Topics Information, Computing and Communication Sciences
Bimanual Dexterity: Coordinated Dual-Arm Policies for Dexterous Object Manipulation
Dexterous bimanual manipulation remains a difficult challenge, requiring tight coordination between both arms and hands—especially when handling large objects. While humans perform such tasks effortlessly, robotic systems often lack the precision and adaptability needed. This project aims to tackle this problem by learning coordinated dual-arm and hand policies from a few human demonstrations, followed by reinforcement learning in simulation for real-world deployment.
Keywords
Dexterous Manipulation, Bimanual Manipulation, Learning from Demonstrations, Reinforcement Learning, Humanoid Robotics
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-25 , Earliest start: 2025-07-28 , Latest end: 2026-05-31
Organization Robotic Systems Lab
Hosts He Junzhe , Mittal Mayank , Bhardwaj Arjun
Topics Information, Computing and Communication Sciences
From Grasp to Twist: Composing Dexterous Grasping and In-Hand Manipulation under Motion
Dexterous in-hand manipulation has advanced rapidly, yet most approaches assume the object is already grasped or the hand is fixed in place—conditions rarely met in real-world scenarios. This project addresses the challenge of enabling robots to both grasp and manipulate objects dynamically by learning modular skills that can be coordinated for effective manipulation under motion.
Keywords
Dexterous Manipulation, Reinforcement Learning, Skill Learning
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-25 , Earliest start: 2025-07-28 , Latest end: 2026-05-31
Organization Robotic Systems Lab
Hosts Bhardwaj Arjun , Zurbrügg René , Mittal Mayank
Topics Information, Computing and Communication Sciences
3D Pose Estimation to Understand Orthosis Effects in Cerebral Palsy
This Master’s thesis at ETH Zürich develops a 3D pose estimation pipeline to evaluate the effects of orthotic devices on gait patterns in children with cerebral palsy (CP). Using a CP-specific dataset, the project benchmarks multiple 2D and 3D pose estimation models, fine-tunes the best-performing model, and applies time-series analysis to assess joint-level kinematic differences across healthy, CP without orthosis, and CP with orthosis conditions. The goal is to support clinical assessment through non-invasive motion analysis.
Keywords
3D Pose Estimation, Cerebral Palsy, Gait Analysis, Orthosis, Deep Learning
Labels
Internship , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2025-07-23 , Earliest start: 2025-07-23 , Latest end: 2025-09-30
Organization Spinal Cord Injury & Artificial Intelligence Lab
Hosts Paez Diego, Dr. , Paez Diego, Dr.
Topics Medical and Health Sciences , Information, Computing and Communication Sciences
Internship / Research assistant / Semester project in Biosensing for Robotics Care and Body Simulation (~12 months)
Join a team of scientists improving the long-term prognosis and treatment of Spinal Cord Injury (SCI) through mobile and wearable systems and personalized health monitoring. Joining the SCAI Lab part of the Sensory-Motor Systems Lab at ETH, you will have the unique opportunity of working at one of the largest and most prestigious health providers in Switzerland: Swiss Paraplegic Center (SPZ) in Nottwil (LU).
Keywords
App development, Machine Learning, Data bases, Data engineering, Systems Engineering, Data Modelling
Labels
Internship , Lab Practice , Student Assistant / HiWi , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2025-07-23 , Earliest start: 2025-09-01 , Latest end: 2026-12-31
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne , IBM Research Zurich Lab , Institute for Research in Biomedicine , Hochschulmedizin Zürich , Swiss Institute of Bioinformatics , University of Lucerne , University of Zurich , Zurich University of Applied Sciences , Zurich University of the Arts , Lucerne University of Applied Sciences and Arts , Berner Fachhochschule
Organization Spinal Cord Injury & Artificial Intelligence Lab
Hosts Paez Diego, Dr. , Paez Diego, Dr. , Paez Diego, Dr.
Topics Information, Computing and Communication Sciences
Development of flexible, biocompatible electrodes for vagus nerve stimulation therapy
Stroke is the most common cause of disability and as many as 14 million people suffer a stroke every year. Typical therapy today is focused on physical exercises and rarely combats the core of the problem - the interactions between the motor actions and the activity patterns of the injured brain. Therefore, we are developing a cutting-edge medical device that helps combine physical therapy with a non-invasive brain stimulation. This is expected to boost recovery after stroke and make this advanced neuro-therapy accessible to millions of people who need it.
Keywords
Biocompatibility Conductive Materials Vagus Nerve Stimulation (VNS) Transcutaneous VNS (tVNS) Stroke Rehabilitation Therapy Medical Device Engineering Electrode Design and Prototyping Materials Science Biomedical Engineering Neurorehabilitation Clinical Usability Patient-Centred Design
Labels
Semester Project , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-07-22 , Earliest start: 2024-11-24 , Latest end: 2025-10-31
Organization Rehabilitation Engineering Lab
Hosts Viskaitis Paulius
Topics Medical and Health Sciences , Engineering and Technology
Better View Learning Using RL
Selecting the right viewpoint is crucial for effective perception and task execution in robotics and autonomous systems. Traditional rule-based methods often struggle in complex or dynamic settings. This project explores reinforcement learning (RL) to enable agents to autonomously learn adaptive viewpoint selection strategies. By maximizing task performance and information gain, the resulting policies aim to enhance system efficiency in applications such as active perception, autonomous exploration, and intelligent robotic control.
Keywords
Active Perception, Optimal Viewpoint Learning, Reinforcement Learning, Robotic Manipulation
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-07-21 , Earliest start: 2025-07-28 , Latest end: 2026-04-30
Organization Robotic Systems Lab
Hosts Zurbrügg René , Bhardwaj Arjun
Topics Information, Computing and Communication Sciences , Engineering and Technology
Embedded algorithms of IMUs in a neurorehabilitation device
The goal of this project is to help develop embedded firmware for a imu based rehabilitation device. This project is part of the SmartVNS project which utilizes movement-gated control of vagus nerve stimulation for stroke rehabilitation.
Keywords
electrical engineering PCB Embedded systems neurorehabilitation
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-07-17 , Earliest start: 2024-01-06 , Latest end: 2024-12-31
Organization Rehabilitation Engineering Lab
Hosts Donegan Dane , Viskaitis Paulius
Topics Medical and Health Sciences , Engineering and Technology
Development and Testing of Electrical Systems for a SmartVNS Docking Station with Focus on Wireless Data Management
We are looking for an enthusiastic electrical/firmware engineer to design and implement the electrical and firmware aspects of a docking station for the SmartVNS device. The station will charge the device components (pulse generator and wrist motion tracker) and pull data from the pulse generator and motion tracker, uploading it to an online server via Wi-Fi. This project will also involve testing the reliability of data transfer and power systems under real-world conditions, providing valuable insights into the practical application of this technology.
Keywords
Electrical, embedded, electronic, engineering, biomedical
Labels
Internship , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2025-07-17 , Earliest start: 2024-08-18 , Latest end: 2025-10-01
Organization Rehabilitation Engineering Lab
Hosts Viskaitis Paulius
Topics Information, Computing and Communication Sciences , Engineering and Technology
Reinforcement Learning for Loco-manipulation
This project aims to develop a reinforcement learning (RL) framework to enable quadruped robots to perform dynamic locomotion and manipulation simultaneously by leveraging advanced RL algorithms . We will develop control policies that can predict future states and rewards, enabling the robot to adapt its behavior on-the-fly. The primary focus will be on achieving stable and adaptive walking patterns while reaching and grasping objects. The outcome will provide insights into the integration of complex behaviors in robotic systems, with potential applications in service robotics and automated object handling.
Keywords
Robotics, quadrupeds, RL, AI,
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-11 , Earliest start: 2025-07-31 , Latest end: 2026-01-31
Organization Robotic Systems Lab
Hosts Mirrazavi Sina
Topics Information, Computing and Communication Sciences , Engineering and Technology
Computational Modeling of Muscle Dynamics for Biohybrid Robots
This research aims to advance biohybrid robotics by integrating living biological components with artificial materials. The focus is on developing computational models for artificial muscle cells, a critical element in creating biohybrid robots. Challenges include modeling the complex and nonlinear nature of biological muscles, considering factors like elasticity and muscle fatigue, as well as accounting for fluid-structure interaction in the artificial muscle's environment. The research combines first principle soft body simulation methods and machine learning to improve understanding and control of biohybrid systems.
Keywords
Biohybrid Robotics, Computational Models, Soft Body Simulation, Finite Element Method (FEM), Muscle Dynamics, Soft Robotics
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-11 , Earliest start: 2025-02-01 , Latest end: 2026-02-01
Organization Soft Robotics Lab
Hosts Mekkattu Manuel , Katzschmann Robert, Prof. Dr.
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Biology , Physics
GPU Acceleration of Soft Robot Modeling: Enhancing Performance with CUDA
We are enhancing soft robot modeling by developing a GPU-accelerated version of our FEM-based framework using CUDA. This research focuses on optimizing parallel computations to significantly speed up simulations, enabling larger problem sizes and real-time control. By improving computational efficiency, we aim to advance soft robotics research and facilitate more detailed, dynamic simulations.
Keywords
Soft Body Simulation, high-performance computing, GPU programming, Parallel Computing, Finite Element Method (FEM), Multiphysics Simulation
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-11 , Earliest start: 2025-02-01 , Latest end: 2026-02-01
Organization Soft Robotics Lab
Hosts Katzschmann Robert, Prof. Dr. , Mekkattu Manuel
Topics Information, Computing and Communication Sciences , Engineering and Technology
Advancing Soft Robot Modeling: Integrating Physics, Optimization, and Control
We are advancing soft robot simulation with FEM and energy-based methods to model complex, adaptive behaviors. This research entails developing the framework to support diverse designs, integrate new physics models, and optimize performance, enabling enhanced control and real-world applications of soft robots.
Keywords
Soft Robotics, Finite Element Method (FEM), Physical Modeling, Benchmarking, Optimization, Multiphysics Simulation, Sim-to-Real
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-11 , Earliest start: 2025-02-01 , Latest end: 2026-02-01
Organization Soft Robotics Lab
Hosts Mekkattu Manuel , Katzschmann Robert, Prof. Dr.
Topics Information, Computing and Communication Sciences , Engineering and Technology
Perceptive Bipedal‑Wheeled Locomotion Using Direct Depth‑Camera Inputs
Recent work [1] trains perceptive quadrupedal locomotion policies directly from depth images using teacher‑student distillation and RL fine‑tuning, achieving robust traversal over challenging terrain. The egocentric vision problem usually requires RL fine-tuning or more complex training schemes to enable active perception due to the information gap between teacher and student observations. This project aims to bring depth camera-based perceptive locomotion policies to a wheeled-bipedal robot. We believe that this might be more adequate than elevation map-based perception, due to the highly dynamic nature of its locomotion. We put high emphasis on novel training schemes to reduce the number of training phases. Initial inspiration could be drawn from [2]. References: [1]: https://arxiv.org/abs/2505.11164 [2]: https://arxiv.org/abs/2412.09149
Keywords
Perceptive locomotion · Depth camera · Bipedal‑wheeled Robot · End‑to‑end RL · Sim‑to‑real
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-09
Organization Robotic Systems Lab
Hosts Schwarke Clemens , Klemm Victor
Topics Information, Computing and Communication Sciences , Engineering and Technology
Learning Continuous-Time Feedback Laws for Hierarchical Reinforcement Learning
Hierarchical reinforcement learning (HRL) typically operates by decomposing tasks into high-level planners and low-level controllers. While high-level policies often evolve at discrete timescales, the low-level controller must generate fast, robust behaviors. This is usually framed as feedback laws or motion trajectories. In this project, we propose to learn continuous-time dynamical systems in the form of ODEs as low-level feedback controllers. The idea is to learn an ordinary differential equation (ODE) whose integral curve realizes the low-level trajectory, or whose vector field provides feedback behavior. These models offer several advantages: they are interpretable, temporally coherent, and can generalize better under time perturbations. The student will implement and evaluate ODE-based controllers within a hierarchical RL setup and compare their performance against standard low-level policy architectures (e.g., feedforward MLPs, recurrent policies). We are also open to shape the projects in the context of operator learning. References as inspiration: https://arxiv.org/pdf/1806.07366 https://arxiv.org/pdf/1909.12077 https://arxiv.org/pdf/2006.04439 https://arxiv.org/pdf/2402.15715
Keywords
Reinforcement Learning, Neural ODEs, Feedback Control, Hierarchical Policies, Continuous-Time Policies, Operator Learning
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-07
Organization Robotic Systems Lab
Hosts Klemm Victor
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Automatic Failure Detection for Drones
Automatic failure detection is an essential topic for aerial robots as small failures can already lead to catastrophic crashes. Classical methods in fault detection typically use a system model as a reference and check that the observed system dynamics are within a certain error margin. In this project, we want to explore sequence modeling as an alternative approach that feeds all available sensor data into a neural network. The network will be pre-trained on simulation data and finetuned on real-world flight data. Such a machine learning-based approach has significant potential because neural networks are very good at picking up patterns in the data that are hidden/invisible to hand-crafted detection algorithms.
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-07-07 , Earliest start: 2025-02-01 , Latest end: 2026-01-31
Organization Robotics and Perception
Hosts Bauersfeld Leonard
Topics Engineering and Technology
Rigid Body Dynamics as Sum of Squares: Reformulation and Investigation with Optimality Certificates
Rigid-body dynamics are foundational to robotics and mechanical systems. Surprisingly, although their equations often contain trigonometric or rational terms, they can typically be reformulated as (rational) polynomial systems. This opens the door to a powerful toolset: Sum-of-Squares (SOS) optimization. In this project, the student will explore how to reformulate the equations of rigid-body systems into polynomial forms suitable for SOS-based analysis (or its dual, the moment hierarchy). Through this lens, many classic robotics problems, such as verifying the global stability of visuomotor controller [1] or certifying optimality of solutions [2], can be recast as (convex) optimization problems. This project will investigate how far these methods scale: Can we extend them to complex robots? Can we prove properties like exponential stability or generate globally valid Lyapunov functions for general mechanical systems? Can we extract optimality certificates for complex dynamic trajectories? The student will implement tools for converting rigid-body models (e.g., simple pendulums, planar robots) into their polynomial equivalents, and then apply SOS programming and S-procedure techniques to verify Lyapunov stability or derive optimal controllers. References: [1] G. Chou and R. Tedrake, “Synthesizing Stable Reduced-Order Visuomotor Policies for Nonlinear Systems via Sums-of-Squares Optimization,” in IEEE Conference on Decision and Control (CDC), 2023. doi: 10.48550/arXiv.2304.12405. [2] S. Teng, A. Jasour, R. Vasudevan, and M. Jadidi, “Convex Geometric Motion Planning on Lie Groups via Moment Relaxation,” in Robotics: Science and Systems XIX, 2023. doi: 10.15607/RSS.2023.XIX.058.
Keywords
Rigid-Body Dynamics, Polynomial Systems, SOS Optimization, Lyapunov Stability, Convex Programming
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
Student(s) Name(s)
Project Report Abstract
More information
Published since: 2025-07-07
Organization Robotic Systems Lab
Hosts Klemm Victor
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology
Learning Robust Agile Flight via Adaptive Curriculum
This project focuses on developing robust reinforcement learning controllers for agile drone navigation using adaptive curricula. Commonly, these controllers are trained with a static, pre-defined curriculum. The goal is to develop a dynamic, adaptive curriculum that evolves online based on the agents' performance to increase the robustness of the controllers.
Keywords
Reinforcement Learning, Drones
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-07-07 , Earliest start: 2023-11-01 , Latest end: 2024-12-31
Organization Robotics and Perception
Hosts Xing Jiaxu
Topics Engineering and Technology
Vision-based Navigation in Dynamic Environment via Reinforcement Learning
In this project, we are going to develop a vision-based reinforcement learning policy for drone navigation in dynamic environments. The policy should adapt to two potentially conflicting navigation objectives: maximizing the visibility of a visual object as a perceptual constraint and obstacle avoidance to ensure safe flight.
Keywords
Reinforcement Learning, Computer Vision, Drones
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-07-07 , Earliest start: 2023-11-01 , Latest end: 2024-12-31
Organization Robotics and Perception
Hosts Xing Jiaxu
Topics Engineering and Technology
Learning Rapid UAV Exploration with Foundation Models
Recent research has demonstrated significant success in integrating foundational models with robotic systems. In this project, we aim to investigate how these foundational models can enhance the vision-based navigation of UAVs. The drone will utilize learned semantic relationships from extensive world-scale data to actively explore and navigate through unfamiliar environments. While previous research primarily focused on ground-based robots, our project seeks to explore the potential of integrating foundational models with aerial robots to enhance agility and flexibility.
Keywords
Visual Navigation, Foundation Models, Drones
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-07-07 , Earliest start: 2024-01-25 , Latest end: 2024-12-31
Organization Robotics and Perception
Hosts Xing Jiaxu
Topics Engineering and Technology
Energy-Efficient Path Planning for Autonomous Quadrotors in Inspection Tasks
Autonomous quadrotors are increasingly used in inspection tasks, where flight time is often limited by battery capacity. his project aims to explore and evaluate state-of-the-art path planning approaches that incorporate energy efficiency into trajectory optimization.
Keywords
Path Planning, Quadrotors
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2025-07-07 , Earliest start: 2024-11-11 , Latest end: 2025-08-13
Organization Robotics and Perception
Hosts Bauersfeld Leonard
Topics Engineering and Technology