Projects & Theses
Below you find a collection of BSc and MSc projects and theses that can be carried out within the MaP community.
Development of a Heterocellular Human Bone Organoid for Precision Medicine and Treatment
Our goal is to establish a heterocellular 3D printed bone organoid model comprising all major bone cell types (osteoblasts, osteocytes, osteoclasts) to recapitulate bone remodeling units in an in vitro system. The organoids will be produced with the human cells, as they could represent human pathophysiology better than animal models, and eventually could replace them. These in vitro models could be used in the advancement of next-generation personalised treatment strategies. Our tools are different kinds of 3D bioprinting platforms, bio-ink formulations, hydrogels, mol-bioassays, and time-lapsed image processing of micro-CT scans.
Keywords
3D printing, bone organoids, co-culture, bioreactor, hydrogels, drug testing
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-19 , Earliest start: 2022-08-01 , Latest end: 2024-08-31
Organization Müller Group / Laboratory for Bone Biomechanics
Hosts Steffi Chris
Topics Engineering and Technology , Biology
Imitation Learning for Pedipulation
This project aims to enable our quadrupedal robot, ANYmal, to perform manipulation tasks with its foot. Particularly, we want to use imitation learning to learn high-level motion plans that can solve real-world tasks such as opening doors, pushing objects, and transporting payloads. This approach will allow solving simple manipulation tasks autonomously.
Keywords
legged robots, imitation learning, manipulation, pedipulation
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-19 , Earliest start: 2025-02-01 , Latest end: 2025-08-31
Organization Robotic Systems Lab
Hosts Arm Philip
Topics Information, Computing and Communication Sciences , Engineering and Technology
Navigation in 3D Environments with a Climbing Robot
In this project, you will integrate a path-planning algorithm for the magnetic climbing robot magnecko. Your work will enable the robot to navigate on 3-dimensional structures like beams, walls, and ceilings.
Keywords
Climbing robot, legged robot, navigation, path-planning
Labels
Semester Project , Bachelor Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-19 , Earliest start: 2025-02-01 , Latest end: 2025-08-31
Organization Robotic Systems Lab
Hosts Arm Philip , Roth Pascal
Topics Information, Computing and Communication Sciences , Engineering and Technology
RL-based Hybrid Control for Manipulation
Legged mobile manipulation is advancing rapidly, with applications in home assistance and robotic maintenance. While learning-based methods excel at end-effector position tracking, many real-world tasks require force control. This project develops a hybrid control policy for legged manipulators to apply force along a principal axis while controlling position tangentially. An example task is whipping a board, where the robot maintains consistent perpendicular force while moving the end-effector on the plane.
Keywords
reinforcement learning, legged robotics, hybrid control
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-19 , Earliest start: 2025-01-05
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Portela Tifanny , Arm Philip
Topics Information, Computing and Communication Sciences , Engineering and Technology
Optimization and testing on healthy participants of a smart sock with textile pressure sensors
The goal of the project is to test and optimize a smart sock prototype for plantar pressure measurements that was previously developed in the lab. The prototype will be optimized based on its ability to track pressure during everyday activities as a wearable device. Tests on healthy participants performing standard movements (e.g., walking, climbing stairs, etc) will be performed to compare the sock performance to a commercial gold standard smart insole system. This technology can be used for plantar pressure monitoring in diverse wearable applications ranging from healthcare to sports.
Keywords
wearable technology, smart sock, plantar pressure, e-textiles, gait analysis
Labels
Semester Project , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-19 , Earliest start: 2025-01-06 , Latest end: 2025-10-01
Organization Biomedical and Mobile Health Technology Lab
Hosts Galli Valeria
Topics Engineering and Technology
Generative Adversarial Networks for Autonomous Tool Wear Assessment
Tool wear has a significant impact on the productivity of machining processes. Currently, tools are replaced based on an estimate of their remaining life, often with significant safety margins that lead to inefficiency. Developments are underway to measure wear automatically by image segmentation of microscope images. Increasing the size of the training data set has a positive effect on the performance and robustness of deep learning systems. However, when it comes to tool wear, datasets are not openly available and are expensive and time consuming to create. This thesis builds on an existing segmentation pipeline and dataset of labeled war images. The goal is to leverage advanced deep learning techniques, such as Generative Adversarial Networks (GANs) to create an improved, data-efficient pipeline. You will work towards a system that can classify and quantify different types of wear for objective tool condition monitoring.
Keywords
Computer Vision, Machine Learning, Tool Wear, Manufacturing Automation
Labels
Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-18 , Earliest start: 2025-01-06 , Latest end: 2025-12-31
Organization Computational Modelling of Materials in Manufacturing
Hosts Zwicker Ruben
Topics Information, Computing and Communication Sciences , Engineering and Technology
BEV meets Semantic traversability
Enable Birds-Eye-View perception on autonomous mobile robots for human-like navigation.
Keywords
Semantic Traversability, Birds-Eye-View, Localization, SLAM, Object Detection
Labels
Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-18 , Earliest start: 2025-01-15 , Latest end: 2025-10-31
Organization Robotic Systems Lab
Hosts Gawel Abel
Topics Information, Computing and Communication Sciences , Engineering and Technology
Scene graphs for robot navigation and reasoning
Elevate semantic scene graphs to a new level and perform semantically-guided navigation and interaction with real robots at The AI Institute.
Keywords
Scene graphs, SLAM, Navigation, Spacial Reasoning, 3D reconstruction, Semantics
Labels
Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-18 , Earliest start: 2025-01-15 , Latest end: 2025-10-31
Organization Robotic Systems Lab
Hosts Gawel Abel
Topics Information, Computing and Communication Sciences , Engineering and Technology
Global robot localization in mesh models
This project includes designing and implementing a particle filter for global robot localization in the environments represented by a mesh model. Particular interest lies in the applicability of such a technique for construction sites, which are considered highly dynamic and, hence, challenging for robot localization.
Keywords
* Global robot localization * Diffusion models * Particle Filter
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-17 , Earliest start: 2024-12-18
Organization Robotic Systems Lab
Hosts Vysotska Olga , Talbot William
Topics Engineering and Technology
Development of Neuromuscular Biohybrid Robots
Biohybrid robots integrate living cells and synthetic components to achieve motion. These systems often rely on engineered skeletal muscle tissues that contract upon electrical stimulation for actuation. Neuromuscular-powered biohybrid robots take this concept further by integrating motor neurons to induce muscle contractions, mimicking natural muscle actuation. In our lab, we are developing neuromuscular actuators using advanced 3D co-culture systems and biofabrication techniques to enable functional macro-scale biohybrid robots.
Keywords
Tissue engineering, mechanical engineering, biology, neuroengineering, biomaterials, biohybrid robotics, 3D in vitro models, biofabrication, bioprinting, volumetric printing.
Labels
Semester Project , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-17 , Earliest start: 2025-01-15
Organization Soft Robotics Lab
Hosts Katzschmann Robert, Prof. Dr.
Topics Engineering and Technology , Biology
Graph-based robot localization in BIM models
This project focuses on implementing and deploying a graph-based localization framework that allows to position a robot within BIM (Building information Models).
Keywords
Graph-based robot localization BIM models
Labels
Semester Project , Bachelor Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-17 , Earliest start: 2024-12-18
Organization Robotic Systems Lab
Hosts Vysotska Olga , Talbot William
Topics Engineering and Technology
Digital Twin for Spot's Home
MOTIVATION ⇾ Creating a digital twin of the robot's environment is crucial for several reasons: 1. Simulate Different Robots: Test various robots in a virtual environment, saving time and resources. 2. Accurate Evaluation: Precisely assess robot interactions and performance. 3. Enhanced Flexibility: Easily modify scenarios to develop robust systems. 4. Cost Efficiency: Reduce costs by identifying issues in virtual simulations. 5. Scalability: Replicate multiple environments for comprehensive testing. PROPOSAL We propose to create a digital twin of our Semantic environment, designed in your preferred graphics Platform to be able to simulate Reinforcement Learning agents in the digital environment, to create a unified evaluation platform for robotic tasks.
Keywords
Digital Twin, Robotics
Labels
Semester Project , Master Thesis
Contact Details
More information
Published since: 2024-12-17 , Earliest start: 2025-01-05
Applications limited to University of Zurich , ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Computer Vision and Geometry Group
Hosts Blum Hermann , Portela Tifanny , Bauer Zuria, Dr. , Trisovic Jelena
Topics Information, Computing and Communication Sciences
KALLAX Benchmark: Evaluating Household Tasks
Motivation ⇾ There are three ways to evaluate robots for pick-and-place tasks at home: 1. Simulation setups: High reproducibility but hard to simulate real-world complexities and perception noise. 2. Competitions: Good for comparing overall systems but require significant effort and can't be done frequently. 3. Custom lab setups: Common but lead to overfitting and lack comparability between labs. Proposal ⇾ We propose using IKEA furniture to create standardized, randomized setups that researchers can easily replicate. E.g, a 4x4 KALLAX unit with varying door knobs and drawer positions, generating tasks like "move the cup from the upper right shelf into the black drawer." This prevents overfitting and allows for consistent evaluation across different labs.
Keywords
Benchmakr, Robotics, pick-and-place
Labels
Semester Project , Master Thesis
Contact Details
More information
Published since: 2024-12-17 , Earliest start: 2025-01-06
Applications limited to University of Zurich , ETH Zurich , Swiss National Science Foundation , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Computer Vision and Geometry Group
Hosts Blum Hermann , Bauer Zuria, Dr. , Zurbrügg René
Topics Information, Computing and Communication Sciences
Quantifying the impact of network conditions on telesurgery performance and reliability
In this thesis, our goal is to quantify the impact of various network conditions (e.g., latency, congestion, packet loss) on the performance of stroke treatment via telesurgery. At ETH, we have established a real testbed equipped with agent controllers and robotic instruments, and the student’s role will be to simulate the network environment connecting these devices.
Keywords
Surgical robotics, network emulation, test setup design, user study
Labels
Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-17 , Earliest start: 2025-02-01 , Latest end: 2025-12-31
Organization Multiscale Robotics Lab
Hosts Heemeyer Florian
Topics Information, Computing and Communication Sciences , Engineering and Technology
Activity and fatigue detection using machine learning based on real-world data from smart clothing
The aim of this project is to use machine learning methods to extract useful information such as activity type and fatigue level from real-world data acquired from our textile-based wearable technology during sport activities.
Keywords
smart clothing, wearable technology, textile sensor, fitness tracking, sports medicine, fatigue, machine learning, artificial intelligence, computer science
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-16 , Earliest start: 2023-09-15 , Latest end: 2024-05-31
Organization Biomedical and Mobile Health Technology Lab
Hosts Ahmadizadeh Chakaveh
Topics Information, Computing and Communication Sciences , Engineering and Technology
Develop software for wearable technologies
The aim of this project is to develop mobile software to communicate with our already developed textile-based wearable technology and process sensor data for movement monitoring.
Keywords
smart clothing, wearable technology, software development, fitness tracking, sports medicine, mobile application, computer science
Labels
Semester Project , Bachelor Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-16 , Earliest start: 2024-07-22 , Latest end: 2025-08-31
Organization Biomedical and Mobile Health Technology Lab
Hosts Ahmadizadeh Chakaveh
Topics Information, Computing and Communication Sciences , Engineering and Technology
Language-Conditioned Interaction Trajectories Using Diffusion Algorithms for Robotic Manipulation
Train a diffusion model to predict 3D interaction trajectories from ego-centric images and task descriptions. The final model will be evaluated on a real robotic arm with five fingered hand.
Keywords
egocentric vision, diffusion models, representation learning, multimodal models, robotic manipulation
Labels
Master Thesis , ETH Zurich (ETHZ)
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-16
Organization Robotic Systems Lab
Hosts Zurbrügg René , Zurbrügg René
Topics Information, Computing and Communication Sciences
Ultrasound-driven dynamics of confined microbubbles in capillary-scale hydrogel microchannel networks
Coated microbubbles are commercially available and clinically approved ultrasound contrast agents used in ultrasound imaging to enhance imaging contrast when injected systemically, thereby improving diagnostic capabilities. Recently, microbubbles have been proposed as therapeutic agents for targeted drug delivery. In order to reliably characterise these agents, suitable in vitro phantoms that mimic the mechanical properties of vascular networks are required to study the dynamics of microbubbles in a confined environment.
Keywords
Microbubble, bubble dynamics, microchannels, hydrogels
Labels
Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-16 , Earliest start: 2025-03-02 , Latest end: 2025-09-01
Applications limited to ETH Zurich
Organization Group Supponen
Hosts Guerriero Giulia
Topics Engineering and Technology
Developing Multi-Functional Microrobots Using Microfluidic Chips (3M project)
We are looking for a motivated Master’s student to join an exciting interdisciplinary thesis project, collaborating between the Multi-Scale Robotics Lab (D-MAVT) and the deMello group (D-CHAB) at ETH Zurich. This project focuses on creating a novel microfluidic-based bottom-up method to fabricate multifunctional microrobots. This innovative approach seeks to revolutionize microrobot fabrication, opening the door to diverse new applications.
Keywords
Microfluidics, Self-assembly, Microrobots
Labels
Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-16 , Earliest start: 2025-02-17
Organization Multiscale Robotics Lab
Hosts Hu Minghan
Topics Engineering and Technology , Chemistry
Development of a ray tracing algorithm for light distortion correction
When optical imaging is used for multiphase fluid dynamics experiments, the presence of multiple interfaces can deflect light rays passing through the test volume, thus creating a distorted image on the camera sensor. To correctly extract quantitative data, a numerical simulation of the test case of interest can be performed and then validated by synthetically recreating the distortion effects and comparing the result with the experimental image. Ray tracing algorithms, widely used in the videogame industry for they computational efficiency, are one of the common choices to perform this task.
Keywords
experimental fluid dynamics, shock wave, optical distortion, ray tracing
Labels
Semester Project , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-16 , Earliest start: 2025-02-01 , Latest end: 2025-06-30
Organization Group Supponen
Hosts Fiorini Samuele
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Physics
Design and development of a diffusion cell for sonophoresis treatment
Sonophoresis is the physical method of using low frequency ultrasound to increase skin permeability for transdermal drug delivery. Although acoustic cavitation in the coupling medium placed on the skin for ultrasound transmission has been identified as the primary mechanical effect responsible for the increase in skin permeability after low frequency ultrasound treatment, the complex interplay between physical and biological processes is still unclear. Franz diffusion cells are a commercially available platform for performing in vitro skin permeation and drug release studies. They consist of two compartments, the donor and the receptor. The receptor compartment contains the dissolution medium, and the donor part contains the drug formulation. The skin model is placed between the receptor and donor compartment and acts as a barrier. The application of low frequency ultrasound in the donor compartment of the Franz diffusion cell is limited by the size of the donor compartment itself, but more importantly its geometry does not allow lateral optical access to the skin surface to study cavitation bubble dynamics using high speed imaging, limiting the study of cavitation bubble-tissue interactions.
Keywords
Sonophoresis, diffusion cell, drug delivery, cavitation
Labels
Bachelor Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-16 , Earliest start: 2025-03-02 , Latest end: 2025-06-30
Applications limited to ETH Zurich
Organization Group Supponen
Hosts Guerriero Giulia
Topics Engineering and Technology
Theoretical and numerical modeling of High-Intensity Focused Ultrasound (HIFU) transducers
High-Intensity Focused Ultrasound (HIFU) transducers are widely used in ultrasound-based therapies such as targeted drug delivery, lithotripsy and histotripsy. Predicting the pressure field generated by their interaction with obstacles is of paramount importance both for research and applications. A numerical approach can be adopted to create a model that can be used as an acoustic source in subsequent calculations.
Keywords
Acoustics, HIFU transducers, numerical simulation, pressure measurements
Labels
Semester Project , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-16 , Earliest start: 2025-02-01 , Latest end: 2025-06-30
Organization Group Supponen
Hosts Fiorini Samuele
Topics Mathematical Sciences , Engineering and Technology , Physics
Modeling cavitation bubble dynamics inside a droplet influenced by acoustic driving
Cavitation bubble dynamics confined within droplets is a crucial aspect to evaluate their applicability in several biomedical applications. The goal of this project involves adapting an existing boundary integral method (BIM) interface to simulate the dynamics of an acoustically-activated bubble confined in a droplet.
Keywords
Bubble dynamics, Boundary Integral Method, Acoustic activation
Labels
Semester Project , Bachelor Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-13 , Earliest start: 2025-02-03 , Latest end: 2025-06-30
Applications limited to ETH Zurich
Organization Group Supponen
Hosts Anunay Anunay
Topics Engineering and Technology
Clinical investigation of the pathophysiology and reciprocal relationships between progressive disc degeneration and spinal (mal-) alignment
Lumbar intervertebral disc degeneration (LDD) is estimated to affect 400M individuals worldwide annually, which causes pain and disability for the patients affected by it [1]. Continuous localized, and structural degeneration on multiple intervertebral disc (IVD) levels can progress to accumulated macroscopic deformities, manifesting in abnormal spinal curvatures, also known as Adult Spinal Deformities (ASD). LDD and spinal alignment are hypothesized to be closely interrelated, with specific alignment parameters showing mutual correlations with the degree of LDD [2]. Through analyzing longitudinal data - including X-ray-derived biplanar alignment parameters and MRI-based Pfirrmann grading - this study aims to identify extensive patterns between alignment characteristics and the individual degenerative state of lumbar IVDs. Such insights could improve our understanding of degenerative cascades and influence decision making in clinical treatment approaches by identifying alignment profiles at higher risk for degeneration. The proposed study includes a clinical cohort of degenerative lumbar back pain patients who were treated conservatively, and followed up over a period of 3-5 years, before eventually undergoing surgery at Schulthess Clinic Zurich. This project is jointly supervised by an interdisciplinary group of researchers and clinicians of ETH Zurich and Schulthess clinic Zurich, which offers insights into fundamental spinal research, as well as daily clinical practice. Your tasks will include data annotation, model development, and collaborative discussions of results, providing a comprehensive experience in interdisciplinary research. This project offers an excellent opportunity to engage in impactful research at the intersection of biomechanics, clinical practice, and data-driven modeling.
Keywords
Spinal Alignment, Intervertebral Disc Degeneration, Adult Spinal Deformity, Clinical Research, Longitudinal Study
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-13 , Earliest start: 2025-02-01
Organization Musculoskeletal Biomechanics
Hosts Rieger Florian
Topics Medical and Health Sciences , Engineering and Technology
Improvements of the Detection Block of a Rapid PCR diagnostic device in a medtech startup
diaxxo, a start-up from ETH Zürich, is transforming molecular diagnostics with an innovative Point-of-Care Polymerase Chain Reaction (PCR) device. This project aims to enhance the detection module of Diaxxo's Point-of-Care Polymerase Chain Reaction (PCR) system by improving light uniformity to ensure precise and reliable optical detection. The redesigned detection block will be compatible with Diaxxo's existing cartridges and support multiplexing with up to three distinct light channels. High light uniformity across the cartridge will be a key focus to enhance detection accuracy and reliability. Additionally, the project emphasizes cost-effective design, leveraging affordable materials and methods to maintain performance without increasing production costs.
Keywords
Optical Detection Fluorescence-Based Methods Light Uniformity Fiber Optics Multiplexing PCR Diagnostics Thermocycling Cost-Effective Design Point-of-Care Systems Molecular Diagnostics
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-13 , Earliest start: 2025-01-01 , Latest end: 2025-09-01
Organization Functional Materials Laboratory
Hosts Gregorini Michele
Topics Medical and Health Sciences , Engineering and Technology , Physics
Dynamic Feed Rate Adjustment for Recoating Overflow Based on CLI File Data
In powder bed-based additive manufacturing, the feed rate during recoating determines the amount of powder distributed to maintain a uniform layer across the build platform. An optimal feed rate ensures a consistent powder bed while minimizing material waste and ensuring process stability. Traditional approaches often use a static feed rate, which may lead to inconsistencies in layer thickness or material overflow, especially for geometries with complex scanning patterns or varying powder requirements.
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-13 , Earliest start: 2025-01-01 , Latest end: 2025-12-01
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Information, Computing and Communication Sciences , Engineering and Technology , Physics
Prediction of the acoustic field generated by a single oscillating bubble
Ultrasonically driven microbubbles can produce cyclic jets driven by interfacial instabilities that can result in cellular drug uptake in targeted drug delivery. In this project, the student will develop a model to predict the acoustic signals produced by these jets.
Keywords
Acoustics, bubbles, finite element methods
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-13 , Earliest start: 2025-02-01 , Latest end: 2025-12-31
Organization Group Supponen
Hosts Supponen Outi
Topics Engineering and Technology
Advanced Topological Design for Vibration Dampening in Cryogenic TEM Systems Using Lattice Structures and Additive Manufacturing
Cryogenic transmission electron microscopy (Cryo- TEM) requires precise control of vibration dampening to maintain imaging quality, particularly in the 20-50 Hz frequency range. Conventional vibration dampening approaches often struggle with achieving the required precision and adaptability for such specialized equipment. This project focuses on the topological optimization of vibration dampening components using lattice structures, advanced CAD tools (e.g., Fusion360), and AI-based design methodologies, validated through Laser Powder Bed Fusion (L-PBF) manufacturing.
Keywords
Cryo-TEM, Topological optimisation, AM, L-PBF, Design, Dampening, Vibration
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-13 , Earliest start: 2025-01-01 , Latest end: 2026-02-01
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Medical and Health Sciences , Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Education , Earth Sciences , Chemistry , Biology , Architecture, Urban Environment and Building , Physics
Implementation of a Quality Management System in a diagnostic startup
This project focuses on design and implementation of a Quality Management System (QMS) for the production of PCR diagnostic devices and cartridges. The primary goals include developing and implementing high-level Standard Operating Procedures (SOPs), establishing robust Quality Control (QC) processes in manufacturing, and designing comprehensive training programs for staff. The project aims to ensure consistent product quality, regulatory compliance, and operational efficiency, contributing to the delivery of reliable and high-performing diagnostic solutions.
Keywords
Quality Management System (QMS) Standard Operating Procedures (SOPs) Quality Control (QC) Regulatory Compliance Manufacturing Processes Risk Management Continuous Improvement
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-13 , Earliest start: 2025-01-01 , Latest end: 2025-09-01
Organization Functional Materials Laboratory
Hosts Gregorini Michele
Topics Engineering and Technology , Commerce, Management, Tourism and Services , Chemistry
MICRO-MULTI-PHYSICS AGENT-BASED MODELLING OF THE TRABECULAR BONE RESPONSE TO ESTROGEN DEPLETION
The proposed project will investigate the trabecular bone response to estrogen depletion and will be used to investigate the probability of the chosen mechanism of action for estrogen. The chosen mechanism of action will be validated using available experimental reference data.
Keywords
Osteoporosis, trabecular bone, python programming, simulation
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-12 , Earliest start: 2024-09-01 , Latest end: 2025-08-31
Organization Müller Group / Laboratory for Bone Biomechanics
Hosts Schulte Friederike
Topics Medical and Health Sciences , Information, Computing and Communication Sciences
Adapting to Injuries for Dexterous In-Hand Manipulation
Develop a reinforcement learning-based method for training adaptive policies for dexterous in-hand manipulation systems to deal with actuator failure on the fly.
Keywords
Dexterous Manipulation, Reinforcement Learning, Adaptive Learning
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-12 , Earliest start: 2024-12-16 , Latest end: 2025-06-01
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Robotic Systems Lab
Hosts Bhardwaj Arjun , Ma Yuntao
Topics Information, Computing and Communication Sciences
Extending Functional Scene Graphs to Include Articulated Object States
While traditional [1] and functional [2] scene graphs are capable of capturing the spatial relationships and functional interactions between objects and spaces, they encode each object as static, with fixed geometry. In this project, we aim to enable the estimation of the state of articulated objects and include it in the functional scene graph.
Keywords
scene understanding, scene graph, exploration
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-11
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Computer Vision and Geometry Group
Hosts Bauer Zuria, Dr. , Trisovic Jelena , Zurbrügg René
Topics Information, Computing and Communication Sciences , Engineering and Technology
Offline LLM-Based Planning for Mobile Manipulation
Large Language Models have enormous potential to increase the generalization capability of robots. Specifically, they can allow mobile platforms to navigate successfully in unknown environments towards open-vocabulary goals. Current applications to mobile platforms are limited due to online models being both slow to query and requiring stable internet connection, which can not always be guaranteed. This project aims to explore the possibility of using a miniature LLM locally on the ALMA robot for navigation though unmapped space, with the goal of moving towards specifying objects using an open-vocabulary and interacting with them.
Keywords
Large Language Models, Mobile Manipulation, Robotics, Machine Learning, Planning
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-11 , Earliest start: 2025-02-01 , Latest end: 2025-12-31
Organization Robotic Systems Lab
Hosts Elanjimattathil Aravind , Scheidemann Carmen
Topics Engineering and Technology
Active Object Localization with Touch
Develop active exploration strategies for object identification and localization with tactile feedback.
Keywords
Dexterous Manipulation, Object Retrieval, Active Localization, Tactile Sensing
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-11 , Earliest start: 2024-12-15 , Latest end: 2025-06-01
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Robotic Systems Lab
Hosts Bhardwaj Arjun , Zurbrügg René
Topics Information, Computing and Communication Sciences
In vitro liver tissue models for studying liver regeneration and drug delivery systems
Many organ grafts are not suitable for transplantation due to excessive ischemic injury. In an effort to save these discarded grafts, ex vivo perfusion systems have been developed to extend the time window for organ repair. The liver, in particular, has a remarkable regenerative capacity and its ex vivo perfusion provides a unique opportunity to trigger regeneration pathways. Thus far, advanced perfusion technologies have enabled the preservation of the human liver outside of the body for up to two weeks using normothermic machine perfusion. Until now, this liver perfusion machine has only been employed to treat bacterial infections, determine tumour malignancy and assess liver function, yet how to stimulate growth and repair of liver grafts ex vivo remains unexplored. In order to effectively develop regeneration strategies, in vitro liver models are necessary since ex vivo human liver experiments are low-throughput, confounded by patient to patient variability and costly. Liver tissue slices, which are directly obtained from native liver tissue, preserve the intact hepatocellular architecture and microenvironment of the liver unlike 2D cell culture and organoid models. Thus, we aim to use liver tissue slices as a screening platform to identify pro-regenerative biomolecules and drugs. In addition, we will explore mRNA lipid nanoparticles to improve the delivery and therapeutic effect of candidate biomolecules and drugs for ex vivo liver perfusion.
Keywords
in vitro liver models, regeneration, drug screening, cell culture, molecular biology, biomedical engineering
Labels
Internship , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-10 , Earliest start: 2025-01-12 , Latest end: 2026-05-31
Organization Macromolecular Engineering Laboratory
Hosts Cunningham Leslie
Topics Engineering and Technology , Biology
Precipitation-Based Processing of Lactide Copolymers for Shape Memory Polymer Development
Shape memory polymers (SMPs) are advanced materials capable of returning to their original shape when exposed to external stimuli such as heat or stress. Lactide copolymers, derived from renewable resources, are particularly attractive for SMP applications due to their tunable mechanical properties, thermal responsiveness, and biocompatibility. Optimizing precipitation-based processing techniques is critical to unlocking the full potential of lactide copolymers for advanced applications.
Labels
Semester Project , Collaboration , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-10 , Earliest start: 2025-01-01 , Latest end: 2026-02-01
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Agricultural, Veterinary and Environmental Sciences , Medical and Health Sciences , Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Education , Earth Sciences , Chemistry , Biology , Architecture, Urban Environment and Building
Selective Laser Sintering Parameter Development for Polyhydroxyalkanoates (PHA)
Selective Laser Sintering (SLS) is a versatile additive manufacturing process that creates complex components by fusing powdered materials layer by layer. This project focuses on optimizing SLS parameters for polyhydroxyalkanoates (PHA), a biodegradable thermoplastic derived from renewable resources. PHA is of great interest due to its biodegradability and potential applications in sustainable manufacturing.
Labels
Semester Project , Collaboration , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-10 , Earliest start: 2025-01-01 , Latest end: 2026-03-01
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Agricultural, Veterinary and Environmental Sciences , Medical and Health Sciences , Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Earth Sciences , Chemistry , Biology , Architecture, Urban Environment and Building
Selective Laser Sintering of Cerium Oxide for High-Performance CO₂ Capture
Selective Laser Sintering (SLS) is a versatile additive manufacturing technique used to produce components by fusing powder materials layer by layer. This project focuses on the SLS processing of cerium oxide (ceria) for applications in carbon dioxide (CO2) capture. Ceria is a promising material in this domain due to its redox properties and potential for use in thermochemical cycles for CO₂ conversion and storage. However, optimizing its SLS processing parameters to ensure high-quality functional parts remains a challenge.
Labels
Semester Project , Collaboration , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-10 , Earliest start: 2025-01-01 , Latest end: 2026-01-01
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Agricultural, Veterinary and Environmental Sciences , Medical and Health Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Education , Economics , Earth Sciences , Chemistry , Biology , Behavioural and Cognitive Sciences , Architecture, Urban Environment and Building
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: 2024-12-10 , 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
Single-cell Encapsulation for RNA Sequencing
This project focuses on understanding the metabolism of bacteria exposed to an enzymatic inhibitor. By employing microfluidic systems, we aim to perform a single-based encapsulation in hydrogel beads to be exposed to inhibitors. Using FACS-based sorting and bulk sequencing, bacteria in hydrogel beads will be sorted and the RNA from pool or individual beads will be extracted for sequencing. We aim to create an RNA sequencing protocol for sorted hydrogel beads containing bacteria to analyze the total gene expression profiles of our candidates.
Keywords
Keywords: Bacteria, RNA, droplets, microfluidic, high-throughput, hydrogel, gene expression, sequencing
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-09 , Earliest start: 2024-08-01 , Latest end: 2025-07-31
Organization Complex Materials
Hosts Enrriquez Nadia
Topics Medical and Health Sciences , Engineering and Technology , Chemistry , Biology
pH Quantification in Emulsified Droplets for High-Throughput Screening
This project focuses on finding a suitable emulsification system for encapsulating bacteria in water-in-oil droplets. By employing microfluidic systems and versatile pH dyes, we aim to create emulsified droplets with reduced crosstalk rates to screen bacteria based on the pH changes as a result of their metabolic activity.
Keywords
Keywords: Bacteria, pH, droplets, microfluidic, high-throughput.
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-09 , Earliest start: 2024-08-01 , Latest end: 2025-07-31
Organization Complex Materials
Hosts Enrriquez Nadia
Topics Medical and Health Sciences , Engineering and Technology , Biology
Evolve to Grasp: Learning Optimal Finger Configuration for a Dexterous Multifingered Hand
Use evolutionary algorithms with analytical force closure metrics to learn the optimal morphology of a dexterous hand.
Keywords
Evolutionary Algorithm, Machine Learning, Grasping, Robotics
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-09 , Earliest start: 2024-12-09 , Latest end: 2025-10-31
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Church Joseph , Zurbrügg René
Topics Information, Computing and Communication Sciences
Learn to Reach: Collision Aware End-Effector Path Planning & Tracking using Reinforcement Learning
Develop a method for collision aware reaching tasks using reinforcement learning and shape encodings of the environment
Keywords
Reinforcement Learning, Robotics, Perception, Machine Learning
Labels
Semester Project , Master Thesis , ETH Zurich (ETHZ)
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-09 , Earliest start: 2024-12-09 , Latest end: 2025-12-31
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Zurbrügg René , Zurbrügg René , Portela Tifanny
Topics Information, Computing and Communication Sciences
Differential Particle Simulation for Robotics
This project focuses on applying differential particle-based simulation to address challenges in simulating real-world robotic tasks involving interactions with fluids, granular materials, and soft objects. Leveraging the differentiability of simulations, the project aims to enhance simulation accuracy with limited real-world data and explore learning robotic control using first-order gradient information.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-09 , Earliest start: 2025-01-01 , Latest end: 2025-12-31
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Robotic Systems Lab
Hosts Nan Fang , Ma Hao
Topics Engineering and Technology
Development of an acoustic trapping device for microbubbles and droplets
This project focuses on developing an acoustic trapping device to control micrometric bubbles, droplets and particles in suspension and subsequently on using their precise positioning to investigate bubble dynamics in ultrasound fields.
Keywords
bubbles, droplets, particles, acoustic trapping
Labels
Semester Project , Bachelor Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-09 , Earliest start: 2025-02-01 , Latest end: 2025-06-30
Applications limited to ETH Zurich
Organization Group Supponen
Hosts Bauer Tobias
Topics Engineering and Technology
Numerical and experimental investigations of powder stream behaviors in high-speed laser cladding (HSLC).
Laser cladding (LC) and high-speed laser cladding (HSLC) are direct metal deposition (DMD) techniques where metal powder is delivered to a substrate using a carrier gas, and a laser melts the powder and substrate to create a coating. The primary difference between LC and HSLC lies in the powder-laser interaction, as shown in Figure 1. In LC, the powder is injected into a molten pool on the substrate, while in HSLC, the powder is melted in flight before reaching the substrate. This distinction allows HSLC to achieve deposition speeds up to two orders of magnitude higher than LC while reducing the heat input to the substrate. Achieving these benefits, however, depends on the efficient and predictable interaction between the powder and the laser beam. This project investigates the behavior of powder streams in HSLC using a dual approach: advanced numerical simulations and experimental validations. It explores the influence of key input parameters, such as gas flow settings, nozzle geometry, and material properties, on powder stream dynamics. By combining numerical modeling and experimental analysis, the study aims to uncover new insights into powder stream behaviors, optimize the process, and refine the robustness of the model under diverse conditions.
Keywords
Additive manufacturing, 3D printing, Multi-physics, Numerical modeling
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-09 , Earliest start: 2024-12-16
Organization Advanced Manufacturing Laboratory
Hosts Zhang Zhilang , Rippa Francesco
Topics Engineering and Technology
Conformal Prediction for Distribution Shift Detection in Online Learning
This project investigates the use of conformal prediction for detecting distribution shifts in online learning scenarios, with a focus on robotics applications. Distribution shifts, arising from deviations in task distributions or changes in robot dynamics, pose significant challenges to online learning systems by impacting learning efficiency and model performance. The project aims to develop a robust detection algorithm to address these shifts, classifying task distribution shifts as outliers while dynamically retraining models for characteristic shifts.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-09 , Earliest start: 2025-01-01 , Latest end: 2025-12-31
Organization Robotic Systems Lab
Hosts Ma Hao , Nan Fang
Topics Information, Computing and Communication Sciences , Engineering and Technology
Biomineralization of Hydrogels-Based Structures
Currently, the mineralization capacity of S. pasteurii is being exploited in developing construction materials in the form of bio-bricks and bio-cement. These materials are mostly compact structures with different degrees of porosity to increase the diffusion of nutrients through the material. Nevertheless, one recurrent challenge in biomineralized structures is the limited precipitation across the structure.
Keywords
Biomineralization, Hydrogel Scaffolds, Bacteria, 3D Printing
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-09 , Earliest start: 2024-08-01 , Latest end: 2025-04-01
Organization Complex Materials
Hosts Enrriquez Nadia
Topics Engineering and Technology , Chemistry , Biology
Conductive polymer pattern deposition for smart textile applications
The goal of the project is to develop a simple and versatile method for production of robust conductive patterns on textile via deposition of conductive polymers. This technology will allow further development of wearable electronics for biomedical applications.
Keywords
wearable, smart textile, conducting polymer, polymerization, capacitance, conductivity
Labels
Bachelor Thesis , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-06 , Earliest start: 2024-08-01 , Latest end: 2025-08-01
Organization Biomedical and Mobile Health Technology Lab
Hosts Shokurov Aleksandr
Topics Medical and Health Sciences , Engineering and Technology , Chemistry
Visual Language Models for Long-Term Planning in Construction
This project uses Visual Language Models (VLMs) for high-level planning and supervision in construction tasks, enabling task prioritization, dynamic adaptation, and multi-robot collaboration for excavation and site management. prioritization, dynamic adaptation, and multi-robot collaboration for excavation and site management
Keywords
Visual Language Models, Long-term planning, Robotics
Labels
Semester Project , Master Thesis
Description
Work Packages
Contact Details
More information
Published since: 2024-12-06 , Earliest start: 2025-01-24 , Latest end: 2025-10-29
Organization Robotic Systems Lab
Hosts Terenzi Lorenzo
Topics Information, Computing and Communication Sciences
Learning Acrobatic Excavator Maneuvers
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on the automation of heavy machinery (https://gravisrobotics.com/). In this project you will be working with the Gravis team to develop an algorithm that allows a 25-ton excavator to perform an acrobatics maneuver, the jump turn.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2025-12-01
Organization Robotic Systems Lab
Hosts Egli Pascal Arturo , Zhang Weixuan
Topics Engineering and Technology
RL-Based Autonomous Excavation
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on the automation of heavy machinery (https://gravisrobotics.com/). In this project, you will be working with the Gravis team to develop a vision-based system to detect anomalies during the autonomous operation of the machine. You will conduct your project at Gravis under joint supervision from RSL.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2025-12-01
Organization Robotic Systems Lab
Hosts Zhang Weixuan , Egli Pascal Arturo
Topics Engineering and Technology
Data-Driven Joint Control
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on the automation of heavy machinery (https://gravisrobotics.com/). In this project, you will be working with the Gravis team to develop a perceptive navigation system for an autonomous CAT323 excavator. You will conduct your project at Gravis with joint supervision with RSL.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2026-01-01
Organization Robotic Systems Lab
Hosts Jelavic Edo , Egli Pascal Arturo
Topics Engineering and Technology
Full Coverage Navigation for Construction Sites
Monitoring the development of a construction site is essential to ensure the project stays on schedule, within budget, and meets quality and safety standards. This can be achieved through regular 3D scans of the full construction site. However, the current approach demands substantial human time and effort, limiting efficiency and scalability. To address this, we aim to automate the process using a legged robot, capable of navigating complex terrains and performing scans autonomously.
Keywords
Global Planning, Navigation, Autonomy, Construction Side
Labels
Semester Project , Bachelor Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06
Organization Robotic Systems Lab
Hosts Richter Julia
Topics Information, Computing and Communication Sciences
Gas Distribution Mapping with ANYmal
Gas source localisation has many applications, such as responding to chemical accidents, detecting explosives, and identifying methane leaks from landfill sites. These tasks often take place in hazardous environments, posing significant risks to human operators. To mitigate these dangers, robotic systems are increasingly being developed to undertake these challenging missions. However, real-world gas localization presents a significant challenge: gas plumes are highly dynamic, exhibiting both temporal and spatial inconsistencies due to environmental factors such as wind, turbulence, and diffusion. Overcoming these complexities is essential for reliable robotic gas detection and localization.
Keywords
Navigation, Gas Sensing, Neural Network
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06
Organization Robotic Systems Lab
Hosts Richter Julia
Topics Information, Computing and Communication Sciences
Sniffing ANYmal: Detecting Explosives Through Smell
The most effective method for locating explosives is through vapor detection, traditionally performed by trained sniffing dogs. While highly effective, these animals are costly to train, have limited operational endurance, and require the constant presence of a handler. Most critically, the significant risks of detonation endanger both the dogs and their handlers, highlighting the need for autonomous solutions in explosive detection.
Keywords
Navigation, Autonomy, Gas Detection
Labels
Semester Project , Bachelor Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06
Organization Robotic Systems Lab
Hosts Richter Julia
Topics Information, Computing and Communication Sciences
RL Digging Policy Interpretability
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on the automation of heavy machinery (https://gravisrobotics.com/). In this project, you will be working with the Gravis team to evaluate interpretability methods for neural networks and use them to analyze undesirable behavior of RL digging policies deployed on real excavators. You will conduct your project at Gravis under joint supervision from RSL.
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2025-12-01
Organization Robotic Systems Lab
Hosts Egli Pascal Arturo
Topics 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: 2024-12-06 , Earliest start: 2024-09-01 , Latest end: 2025-09-01
Organization Biomedical and Mobile Health Technology Lab
Hosts Shokurov Aleksandr
Topics Engineering and Technology , Chemistry
Establishing a novel high throughput Drug Screening in vitro
The development of advanced drug formulations is a cornerstone of pharmaceutical innovation, directly influencing therapeutic efficacy, patient outcomes, and market success. Achieving optimal drug absorption and bioavailability remains one of the most significant challenges in formulation design, particularly for oral and parenteral delivery systems. Addressing this challenge is critical for advancing scientific understanding and also for accelerating drug discovery and reducing time-to-market for new therapies. This Master’s thesis project aims to develop an advanced cell culture assay to model drug absorption, providing a scientifically robust and commercially valuable platform for drug screening and optimizing novel drug formulations. By bridging gaps in current drug screening methodologies, this project will contribute to innovation in drug delivery technologies and enhance competitive positioning in the growing global market for pharmaceutical solutions.
Keywords
cell culture, drug screening, drug formulation, polymer,
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-06 , Earliest start: 2024-12-09 , Latest end: 2025-12-31
Applications limited to EPFL - Ecole Polytechnique Fédérale de Lausanne , ETH Zurich , Hochschulmedizin Zürich , IBM Research Zurich Lab , Institute for Research in Biomedicine , Zurich University of Applied Sciences , Wyss Translational Center Zurich , University of Zurich , University of Berne , University of Geneva , University of Basel , University of Fribourg , Swiss National Science Foundation , Empa , CSEM - Centre Suisse d'Electronique et Microtechnique , Department of Quantitative Biomedicine , Balgrist Campus , [nothing]
Organization Macromolecular Engineering Laboratory
Hosts Guzzi Elia
Topics Medical and Health Sciences , Engineering and Technology , Chemistry , Biology
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: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2025-10-01
Organization Biomedical and Mobile Health Technology Lab
Hosts Shokurov Aleksandr
Topics Medical and Health Sciences , Engineering and Technology , Chemistry
Perceptive Navigation
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on automation of heavy machinery (https://gravisrobotics.com/). In this project you will be working with the Gravis team to develop a perceptive navigation system for autonomous CAT323 excavator. You will conduct your project at Gravis with joint supervision with RSL.
Keywords
Perceptive Navigation, Autonomous Excavator
Labels
Internship , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2026-01-01
Organization Robotic Systems Lab
Hosts Jelavic Edo , Egli Pascal Arturo , Cizek Petr
Topics Engineering and Technology
Conductive thread modification for wearable strain sensors
The goal of the project is to modify commercially available conductive yarns to improve their operational properties for potential employment in novel garment-embedded sensors for human motion detection.
Keywords
wearable, smart textile, conductive, e-textile, sensor
Labels
Semester Project
Description
Goal
Contact Details
More information
Published since: 2024-12-06 , Earliest start: 2024-10-01
Organization Biomedical and Mobile Health Technology Lab
Hosts Shokurov Aleksandr
Topics Medical and Health Sciences , Engineering and Technology , Chemistry
Autonomous Wear Evaluation of 3D Scanned Cutting Tool Geometry
Today, setting up new tools for machining is a very time-consuming process. Only a few parameters, such as length and nose radius, are known to the machine. In addition, wear must be assessed manu-ally by looking at the tool under a microscope. Even then, only a visual inspection is possible. The actual changes in tool geometry are not detected. In this thesis you will develop a system to evaluate tool wear on 3D scans of the geometry of the tools. The focus will be on two different technologies: Photogrammetry, where the 3D model is re-constructed from 2D images, and structured light scanning, where 3D positions can be measured directly. There is an existing system for photogrammetry that was trained on synthetic data to which a structured light sensor will be added
Keywords
Machine Vision, Structured Light Scanner, Tool Wear
Labels
Semester Project , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
Description
Goal
Contact Details
More information
Published since: 2024-12-06 , Earliest start: 2025-01-13 , Latest end: 2026-01-31
Organization ETH Zurich
Hosts Zwicker Ruben
Topics Information, Computing and Communication Sciences , Engineering and Technology
RL-Based Stockpile Management for Autonomous Excavators
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on the automation of heavy machinery (https://gravisrobotics.com/). In this project, you will be working with the Gravis team to develop an RL-based perceptive planning and control system for stockpile management for an autonomous excavator. You will conduct your project at Gravis under joint supervision from RSL.
Keywords
Reinforcement Learning, Autonomous Excavation
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2026-01-01
Organization Robotic Systems Lab
Hosts Egli Pascal Arturo
Topics Engineering and Technology
Volumetric Bucket-Fill Estimation
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on the automation of heavy machinery (https://gravisrobotics.com/). In this project, you will be working with the Gravis team to develop a perceptive bucket-fill estimation system. You will conduct your project at Gravis under joint supervision from RSL.
Keywords
Autonomous Excavation
Labels
Internship
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2026-01-01
Organization Robotic Systems Lab
Hosts Egli Pascal Arturo
Topics Engineering and Technology
Vision-Based Anomaly Detection for Autonomous Excavators
Gravis Robotics is an ETH spinoff from the Robotic Systems Lab (RSL) working on the automation of heavy machinery (https://gravisrobotics.com/). In this project, you will be working with the Gravis team to develop a vision-based system to detect anomalies during the autonomous operation of the machine. You will conduct your project at Gravis under joint supervision from RSL.
Keywords
Vision, Autonomous Excavator
Labels
Semester Project , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-06 , Earliest start: 2025-01-01 , Latest end: 2026-01-01
Organization Robotic Systems Lab
Hosts Zhang Weixuan , Egli Pascal Arturo
Topics Engineering and Technology
Boosting Reinforcement Learning with High-Speed Gaussian Splatting
This project addresses the computational bottlenecks in model-free reinforcement learning (RL) with high-dimensional image inputs by optimizing Gaussian Splatting—a GPU-accelerated technique for photorealistic image generation from point clouds—for RL applications. By integrating pre-sorting methods, the project aims to enhance rendering speeds, enabling broader RL applications beyond geometric constraints or abstraction layers. Building on previous work involving risk annotations in Gaussian splats, the project seeks to develop generalizable RL policies that leverage real-world knowledge.
Keywords
Gaussian Splatting, Navigation, Reinforcement Learning
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-05
Organization Robotic Systems Lab
Hosts Wilder-Smith Max , Roth Pascal
Topics Information, Computing and Communication Sciences
Semantic Graph-based Large Scale Navigation
This project aims to enhance RL-based planners by addressing their limited memory and short goal-reaching range (15–20m). Inspired by human navigation aided by positional memory, it proposes constructing a graph of past positions and providing it as input to the RL policy, enabling better memory utilization and extended goal distances. By leveraging RL's exploratory behavior, the approach seeks to resolve local minima and achieve human-like goal-reaching capabilities. Current advancements include a tenfold reduction in memory footprint and faster inference for far-away goals. The project will further integrate geometric and semantic environmental data to improve understanding and real-world applicability.
Keywords
Navigation, Graph, Reinforcement Learning
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-05
Organization Robotic Systems Lab
Hosts Richter Julia , Roth Pascal
Topics Information, Computing and Communication Sciences
Exploring interfacial dynamics in microgravity
This project will review different experiments involving bubble dynamics in microgravity and explore new topics where microgravity experiments can teach about the physics of dynamic gas-liquid or liquid-liquid interfaces.
Keywords
microgravity, interfaces, bubbles
Labels
Semester Project , Bachelor Thesis
Description
Goal
Contact Details
More information
Published since: 2024-12-05 , Earliest start: 2025-02-01 , Latest end: 2025-06-30
Applications limited to ETH Zurich
Organization Group Supponen
Hosts Supponen Outi
Topics Engineering and Technology
From Spatial to Functional: Functional Scene Graph for Enhanced Robotic Decision Making
This project explores the concept of Functional Scene Graphs, an extension of traditional scene graphs that capture not just spatial relationships but also the functional interactions between objects and spaces. For example, a light switch enables illumination of a room, or a key provides access to a locked door. Such relationships, while intuitive for humans, are often overlooked in robotics systems, limiting a robot’s ability to reason about and interact with its environment effectively. The core challenge lies in understanding these functional relationships. While a robot might attempt to explore and infer such connections autonomously, humans could assist by demonstrating interactions, offering a means for robots to learn more efficiently. This project will focus on integrating functional understanding into scene graphs, enabling robots to infer high-level semantic interactions and make better decisions during tasks like navigation and manipulation.
Keywords
scene understanding, scene graph, exploration
Labels
Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-03
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Robotic Systems Lab
Hosts Bauer Zuria, Dr. , Qu Kaixian , Zurbrügg René , Blum Hermann
Topics Information, Computing and Communication Sciences , Engineering and Technology
Securing the Future: Enhance Human Safety in Robot Decision Making
Robots have become increasingly advanced recently, capable of performing challenging tasks such as taking elevators and cooking shrimp. Moreover, their ability to accomplish long-horizon tasks given simple natural language instructions is also made possible by large language models. However, with this increased functionality comes the risk that intelligent robots might unintentionally or intentionally harm people based on instructions from an operator. On the other hand, significant efforts have been made to restrain large language models from generating harmful content. Can these efforts be applied to robotics to ensure safe interactions between robots and humans, even as robots become more capable? This project aims to answer this question.
Keywords
human safety, large language models, human-robot interaction
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-03
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization Robotic Systems Lab
Hosts Shi Fan , Qu Kaixian
Topics Information, Computing and Communication Sciences , Engineering and Technology
LLM-Driven Skill Acquisition for the ANYmal robot
The goal of this project is to apply LLMs to teach the ANYmal robot new low-level skills via Reinforcement Learning (RL) that the task planner identifies to be missing.
Keywords
Large Language Models, Reinforcement Learning, Robotics
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-03
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Roth Pascal , Portela Tifanny
Topics Information, Computing and Communication Sciences
Diffusion-based Shared Autonomy System for Telemanipulation
Robots may not be able to complete tasks fully autonomously in unstructured or unseen environments, however direct teleoperation from human operators may also be challenging due to the difficulty of providing full situational awareness to the operator as well as degradation in communication leading to the loss of control authority. This motivates the use of shared autonomy for assisting the operator thereby enhancing the performance during the task. In this project, we aim to develop a shared autonomy framework for teleoperation of manipulator arms, to assist non-expert users or in the presence of degraded communication. Imitation learning, such as diffusion models, have emerged as a popular and scalable approach for learning manipulation tasks [1, 2]. Additionally, recent works have combined this with partial diffusion to enable shared autonomy [3]. However, the tasks were restricted to simple 2D domains. In this project, we wish to extend previous work in the lab using diffusion-based imitation learning, to enable shared autonomy for non-expert users to complete unseen tasks or in degraded communication environments.
Keywords
Imitation learning, Robotics, Manipulation, Teleoperation
Labels
Semester Project , ETH Zurich (ETHZ)
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-12-02 , Earliest start: 2024-11-01 , Latest end: 2025-11-01
Applications limited to ETH Zurich , University of Zurich
Organization Robotic Systems Lab
Hosts Elanjimattathil Aravind
Topics Information, Computing and Communication Sciences , Engineering and Technology
Image-based analysis of liver tissue viability
The goal of this project is to establish a python code to analyse microscopy images of liver tissue to detect changes in phenotype in response to pro-regenerative biomolecules and drugs. The tissue will be stained with hematoxylin and eosin to visualize nuclei and cytoplasmic structures respectively. Phenotypic changes in nuclei number and size as well as sinusoidal lumen dilation will be computationally analyzed. The automated analysis of microscopy images will enable the quantitative assessment of phenotypic changes in liver tissue in a more robust and high throughput manner. The tasks will include - Literature search, what tools are already in use for automatic processing of microscopy images of liver tissue? How is the pipeline set up? - Understanding the provided microscopy images and which features are of interest - Extracting image features using computer vision (CV) algorithms - Verifying the generated algorithm on a large set of data We are looking for a motivated student for a bachelor thesis or semester project. The student should have some experience using python. As the project is centered around CV, strong interest in that topic is necessary. Prior knowledge in CV is favored but not necessary and can be learned during the project.
Keywords
microscopy image analysis python biomedical engineering computational liver
Labels
Semester Project , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-12-02 , Earliest start: 2024-12-15 , Latest end: 2025-12-31
Organization Macromolecular Engineering Laboratory
Hosts Cunningham Leslie , Binz Jonas
Topics Information, Computing and Communication Sciences , Biology
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: 2024-12-01 , Earliest start: 2024-08-01 , Latest end: 2025-06-30
Organization Soft Robotics Lab
Hosts Michelis Mike , Katzschmann Robert, Prof. Dr.
Topics Information, Computing and Communication Sciences , Engineering and Technology
Feedback Loop Control for L-PBF Process Using in situ Sensor Data
This project proposal aims to develop JavaScript algorithms that facilitate seamless interaction with the webpage controlling a Laser Powder Bed Fusion (L-PBF) machine. Currently, the operation of L-PBF machines involves repetitive manual steps in preparing new build jobs, limiting efficiency and real-time adaptability. Additionally, the project seeks to integrate in situ sensors like pyrometers, cameras, and eddy current sensors to enable feedback loop control. These sensors will provide real-time data to the JavaScript script, allowing it to autonomously pause, adjust parameters for different parts, and resume printing at the next layer. Furthermore, the project aims to integrate Python/Netfabb slicing to re-slice parts with varying parameters, enhancing the adaptability and precision of the L-PBF process.This project aims to enhance the efficiency and precision of L-PBF machine operations, reduce manual labor, and enable real-time adaptability through the use of JavaScript algorithms.
Labels
Semester Project , Course Project , Collaboration , Internship , Bachelor Thesis , Master Thesis , Other specific labels , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-11-28 , Earliest start: 2024-01-01 , Latest end: 2025-06-20
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Medical and Health Sciences , Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Education , Physics
Powder Bed Fusion of Shape Memory Alloys as actuators for lightweight and complex soft robots
Soft robots are a rapidly growing field that has a variety of applications in areas such as healthcare, automation, and exploration. One of the major challenges in designing soft robots is the development of actuators that can mimic the movement and flexibility of natural muscles. Shape memory alloys (SMAs) are materials that can change shape when heated, making them ideal for use as actuators. Powder bed fusion (PBF) is an additive manufacturing process that can be used to create complex shapes with high accuracy, making it an ideal technique for manufacturing SMA actuators.
Keywords
Soft robots, shape memory alloys, additive manufacturing, PBF
Labels
Semester Project , Collaboration , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-11-28 , Earliest start: 2023-05-11 , Latest end: 2024-07-05
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Engineering and Technology
Computer Vision and Artificial Intelligence for 3D Powder Bed Reconstruction
Laser powder bed fusion (L-PBF) is an additive manufacturing (AM) technique capable of producing 3D parts from metallic powders in a layer-by-layer fashion starting from a 3D Computer-Aided Design (CAD) model. The L-PBF process also presents some new challenges such as part porosity, rough surface finish, cracking, etc., which may compromise the integrity of the produced parts. The aim of this project is to investigate the robustness of artificial intelligence and machine learning algorithms for the visual inspection of layer-wise powder bed images.
Keywords
Machine learning, Additive Manufacturing, L-PBF, PBF, Computer Vision, Process Monitoring
Labels
Semester Project , Internship , Bachelor Thesis , Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-11-28 , Earliest start: 2024-01-01 , Latest end: 2025-04-11
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Engineering and Technology
Understanding and Mitigating Spatter in Laser Powder Bed Fusion (L-PBF)
This master's thesis proposal aims to tackle the challenge of spatter formation in Laser Powder Bed Fusion (L-PBF). Spatter, the unintended ejection of molten material droplets, can compromise component quality and increase production costs. The study will involve hardware and process parameter analysis to prevent or capture spatter, extending its focus to assess spatter's impact on aluminum and steel alloys commonly used in L-PBF.
Keywords
Laser Powder Bed Fusion, 3D Printing, Alloys, Metal, Simulations, COMSOL, Gas flow, Physics, Electronics, Mechanics, Mechanical Engineering, Electrical Engineering, Computer Science, Design, CAD modelling
Labels
IDEA League Student Grant (IDL) , Master Thesis , ETH Zurich (ETHZ)
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-11-28 , Earliest start: 2023-10-12 , Latest end: 2025-04-30
Organization Advanced Manufacturing Laboratory
Hosts Tosoratti Enrico
Topics Mathematical Sciences , Information, Computing and Communication Sciences , Engineering and Technology , Education , Physics
DynaGELLO: A Low Cost "Puppet" Teleoperation System for the DynaArm
The GELLO system proposed in [1] is a low-cost “puppet” robot arm that is used to teleoperate a larger, main robot arm. This project aims to adapt this open source design to enable teleoperation of the DynaArm, which is a robot manipulator arm custom designed by the Robotic Systems Lab to be mounted on the ANYmal quadruped platform. Such a system provides a simplification over the existing DynaArm teleoperation interface consisting of a second identical DynaArm used purely as a human interface device [2], which may be an unnecessarily expensive and cumbersome solution. The system developed may have applications in remote teleoperation for industrial inspection or disaster response scenarios, as well as providing an interface for training imitation learning models, which may optionally be explored as time permits. [1] Wu, Philip et al. "GELLO: A General, Low-Cost, and Intuitive Teleoperation Framework for Robot Manipulators". arXiv preprint (2024) [2] Fuchioka, Yuni et al. AIRA Challenge: Teleoperated Mobile Manipulation for Industrial Inspection. Youtube Video. (2024)
Keywords
Robotics, Teleoperation, Manipulation, Imitation Learning
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-11-27 , Earliest start: 2025-01-06 , Latest end: 2025-07-31
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Fuchioka Yuni
Topics Information, Computing and Communication Sciences , Engineering and Technology
Responsive ultrasound contrast agents for molecular imaging
Ultrasound imaging with contrast agents such as microbubbles and nanodroplets is a promising tool to diagnose and monitor diseases at the molecular level. In our lab, we are interested in detecting soluble molecular targets such as proteases in vivo. To achieve this, we aim to modify the acoustic properties of microbubbles in response to protease activity by altering their shell properties using tight peptide-crosslinked networks.
Keywords
synthesis, organic chemistry, ultrasound, microbubbles, liposomes, mechanical properties, biomedical engineering, polymers
Labels
Semester Project , Bachelor Thesis , Master Thesis
PLEASE LOG IN TO SEE DESCRIPTION
More information
Published since: 2024-11-27 , Earliest start: 2024-07-01
Applications limited to ETH Zurich , CERN , Eawag , Empa , EPFL - Ecole Polytechnique Fédérale de Lausanne , IBM Research Zurich Lab , Institute for Research in Biomedicine , Paul Scherrer Institute , University of Zurich , Wyss Translational Center Zurich
Organization Responsive Biomedical Systems - Prof. Simone Schürle
Hosts Oberhuber Ines
Topics Medical and Health Sciences , Engineering and Technology , Chemistry
Pretraining for RL
This project addresses sampling inefficiency in classical reinforcement learning by exploring smart weight initialization. Inspired by computer vision, we aim to enhance learning across different hardware (cross embodiment) and skills (cross skills) using pre-trained representations, reducing training times and potentially improving the overall effectiveness of reinforcement learning policies.
Keywords
reinforcement learning weight initialization cross embodiment cross skills
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-11-26 , Earliest start: 2024-09-01 , Latest end: 2025-03-31
Organization Robotic Systems Lab
Hosts Portela Tifanny , Cramariuc Andrei
Topics Information, Computing and Communication Sciences , Engineering and Technology
Periodic Motion Priors for General Quadruped Locomotion Learning
In recent years, advancements in reinforcement learning have achieved remarkable success in quadruped locomotion tasks. Despite their similar structural designs, quadruped robots often require uniquely tailored reward functions for effective motion pattern development, limiting the transferability of learned behaviors across different models. This project proposes to bridge this gap by developing a unified, continuous latent representation of quadruped motions applicable across various robotic platforms. By mapping these motions onto a shared latent space, the project aims to create a versatile foundation that can be adapted to downstream tasks for specific robot configurations.
Keywords
representation learning, periodic autoencoders, policy modulating trajectory generators
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Miki Takahiro
Topics Information, Computing and Communication Sciences , Engineering and Technology
Lifelike Agility on ANYmal by Learning from Animals
The remarkable agility of animals, characterized by their rapid, fluid movements and precise interaction with their environment, serves as an inspiration for advancements in legged robotics. Recent progress in the field has underscored the potential of learning-based methods for robot control. These methods streamline the development process by optimizing control mechanisms directly from sensory inputs to actuator outputs, often employing deep reinforcement learning (RL) algorithms. By training in simulated environments, these algorithms can develop locomotion skills that are subsequently transferred to physical robots. Although this approach has led to significant achievements in achieving robust locomotion, mimicking the wide range of agile capabilities observed in animals remains a significant challenge. Traditionally, manually crafted controllers have succeeded in replicating complex behaviors, but their development is labor-intensive and demands a high level of expertise in each specific skill. Reinforcement learning offers a promising alternative by potentially reducing the manual labor involved in controller development. However, crafting learning objectives that lead to the desired behaviors in robots also requires considerable expertise, specific to each skill.
Keywords
learning from demonstrations, imitation learning, reinforcement learning
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Li Chenhao , Klemm Victor
Topics Information, Computing and Communication Sciences
Pushing the Limit of Quadruped Running Speed with Autonomous Curriculum Learning
The project aims to explore curriculum learning techniques to push the limits of quadruped running speed using reinforcement learning. By systematically designing and implementing curricula that guide the learning process, the project seeks to develop a quadruped controller capable of achieving the fastest possible forward locomotion. This involves not only optimizing the learning process but also ensuring the robustness and adaptability of the learned policies across various running conditions.
Keywords
curriculum learning, fast locomotion
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization Robotic Systems Lab
Hosts Li Chenhao , Bagatella Marco , Li Chenhao , Li Chenhao , Li Chenhao
Topics Engineering and Technology
Leveraging Human Motion Data from Videos for Humanoid Robot Motion Learning
The advancement in humanoid robotics has reached a stage where mimicking complex human motions with high accuracy is crucial for tasks ranging from entertainment to human-robot interaction in dynamic environments. Traditional approaches in motion learning, particularly for humanoid robots, rely heavily on motion capture (MoCap) data. However, acquiring large amounts of high-quality MoCap data is both expensive and logistically challenging. In contrast, video footage of human activities, such as sports events or dance performances, is widely available and offers an abundant source of motion data. Building on recent advancements in extracting and utilizing human motion from videos, such as the method proposed in WHAM (refer to the paper "Learning Physically Simulated Tennis Skills from Broadcast Videos"), this project aims to develop a system that extracts human motion from videos and applies it to teach a humanoid robot how to perform similar actions. The primary focus will be on extracting dynamic and expressive motions from videos, such as soccer player celebrations, and using these extracted motions as reference data for reinforcement learning (RL) and imitation learning on a humanoid robot.
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-11-26
Applications limited to ETH Zurich , EPFL - Ecole Polytechnique Fédérale de Lausanne
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Kaufmann Manuel , Li Chenhao , Li Chenhao , Kaufmann Manuel , Li Chenhao
Topics Engineering and Technology
Learning Real-time Human Motion Tracking on a Humanoid Robot
Humanoid robots, designed to mimic the structure and behavior of humans, have seen significant advancements in kinematics, dynamics, and control systems. Teleoperation of humanoid robots involves complex control strategies to manage bipedal locomotion, balance, and interaction with environments. Research in this area has focused on developing robots that can perform tasks in environments designed for humans, from simple object manipulation to navigating complex terrains. Reinforcement learning has emerged as a powerful method for enabling robots to learn from interactions with their environment, improving their performance over time without explicit programming for every possible scenario. In the context of humanoid robotics and teleoperation, RL can be used to optimize control policies, adapt to new tasks, and improve the efficiency and safety of human-robot interactions. Key challenges include the high dimensionality of the action space, the need for safe exploration, and the transfer of learned skills across different tasks and environments. Integrating human motion tracking with reinforcement learning on humanoid robots represents a cutting-edge area of research. This approach involves using human motion data as input to train RL models, enabling the robot to learn more natural and human-like movements. The goal is to develop systems that can not only replicate human actions in real-time but also adapt and improve their responses over time through learning. Challenges in this area include ensuring real-time performance, dealing with the variability of human motion, and maintaining stability and safety of the humanoid robot.
Keywords
real-time, humanoid, reinforcement learning, representation learning
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts He Junzhe , Li Chenhao , Li Chenhao
Topics Information, Computing and Communication Sciences
Continuous Skill Learning with Fourier Latent Dynamics
In recent years, advancements in reinforcement learning have achieved remarkable success in teaching robots discrete motor skills. However, this process often involves intricate reward structuring and extensive hyperparameter adjustments for each new skill, making it a time-consuming and complex endeavor. This project proposes the development of a skill generator operating within a continuous latent space. This innovative approach contrasts with the discrete skill learning methods currently prevalent in the field. By leveraging a continuous latent space, the skill generator aims to produce a diverse range of skills without the need for individualized reward designs and hyperparameter configurations for each skill. This method not only simplifies the skill generation process but also promises to enhance the adaptability and efficiency of skill learning in robotics.
Keywords
representation learning, periodic autoencoders, learning from demonstrations, policy modulating trajectory generators
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Rudin Nikita
Topics Information, Computing and Communication Sciences , Engineering and Technology
Handstands with a quadruped robot
Doing handstands (one arm and two legs) with a quadrupedal robot equipped with an arm using reinforcement learning. After getting into a stable upright position the next step will also be locomoting in this tripod configuration.
Keywords
reinforcement learning agile control quadrupedal robots
Labels
Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-11-26 , Earliest start: 2024-12-01
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Cramariuc Andrei
Topics Information, Computing and Communication Sciences , Engineering and Technology
DynaHead++: Audio-visual telepresence system for ANYmal
The DynaHead system created at the Robotic Systems Lab enables intuitive telepresence, giving operators depth perception from a stereo camera whose pan-and-tilt motion is controlled through a Virtual Reality (VR) goggle that is worn by the operator. The goal of the project is to further advance this system to enable increased telepresence capabilities. In particular, the system proposed in [1] represents a more advanced version of the concept, providing audio feedback through microphones and the VR goggle’s built-in speaker system, as well as full 6 Degree of Freedom (DOF) motions of the camera compared to the DynaArm’s pan/tilt. We would like to similarly achieve immersive remote telepresence for ANYmal through the addition of directional audio feedback, as well as through additional optional extensions to the system. [1] Lenz, Christian et al. "NimbRo wins ANA Avatar XPRIZE Immersive Telepresence Competition: Human-Centric Evaluation and Lessons Learned". SORO, pp.1-25, (2023)
Keywords
Teleoperation, Telepresence, Virtual Reality, Robotics
Labels
Semester Project , Bachelor Thesis , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-11-26 , Earliest start: 2025-01-06 , Latest end: 2025-07-31
Applications limited to ETH Zurich
Organization Robotic Systems Lab
Hosts Fuchioka Yuni , Wilder-Smith Max
Topics Information, Computing and Communication Sciences , Engineering and Technology
Simulation Environment for Imitation Learning with Dexterous Robotic Hands
This thesis focuses on creating a robust simulation environment using Isaac Lab for imitation learning and policy development in dexterous robotic manipulation
Keywords
Imitation Learning, Dexterous Manipulation, Simulation
Labels
Semester Project , Master Thesis
Description
Work Packages
Requirements
Contact Details
More information
Published since: 2024-11-26 , Earliest start: 2025-01-01 , Latest end: 2026-01-01
Applications limited to ETH Zurich
Organization Soft Robotics Lab
Hosts Yang Chenyu , Liconti Davide
Topics Information, Computing and Communication Sciences , Engineering and Technology
Universal Humanoid Motion Representations for Expressive Learning-based Control
Recent advances in physically simulated humanoids have broadened their application spectrum, including animation, gaming, augmented and virtual reality (AR/VR), and robotics, showcasing significant enhancements in both performance and practicality. With the advent of motion capture (MoCap) technology and reinforcement learning (RL) techniques, these simulated humanoids are capable of replicating extensive human motion datasets, executing complex animations, and following intricate motion patterns using minimal sensor input. Nevertheless, generating such detailed and naturalistic motions requires meticulous motion data curation and the development of new physics-based policies from the ground up—a process that is not only labor-intensive but also fraught with challenges related to reward system design, dataset curation, and the learning algorithm, which can result in unnatural motions. To circumvent these challenges, researchers have explored the use of latent spaces or skill embeddings derived from pre-trained motion controllers, facilitating their application in hierarchical RL frameworks. This method involves training a low-level policy to generate a representation space from tasks like motion imitation or adversarial learning, which a high-level policy can then navigate to produce latent codes that represent specific motor actions. This approach promotes the reuse of learned motor skills and efficient action space sampling. However, the effectiveness of this strategy is often limited by the scope of the latent space, which is traditionally based on specialized and relatively narrow motion datasets, thus limiting the range of achievable behaviors. An alternative strategy involves employing a low-level controller as a motion imitator, using full-body kinematic motions as high-level control signals. This method is particularly prevalent in motion tracking applications, where supervised learning techniques are applied to paired input data, such as video and kinematic data. For generative tasks without paired data, RL becomes necessary, although kinematic motion presents challenges as a sampling space due to its high dimensionality and the absence of physical constraints. This necessitates the use of kinematic motion latent spaces for generative tasks and highlights the limitations of using purely kinematic signals for tasks requiring interaction with the environment or other agents, where understanding of interaction dynamics is crucial. We would like to extend the idea of creating a low-level controller as a motion imitator to full-body motions from real-time expressive kinematic targets.
Keywords
representation learning, periodic autoencoders
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Li Chenhao , Li Chenhao
Topics Information, Computing and Communication Sciences , Engineering and Technology
Humanoid Locomotion Learning and Finetuning from Human Feedback
In the burgeoning field of deep reinforcement learning (RL), agents autonomously develop complex behaviors through a process of trial and error. Yet, the application of RL across various domains faces notable hurdles, particularly in devising appropriate reward functions. Traditional approaches often resort to sparse rewards for simplicity, though these prove inadequate for training efficient agents. Consequently, real-world applications may necessitate elaborate setups, such as employing accelerometers for door interaction detection, thermal imaging for action recognition, or motion capture systems for precise object tracking. Despite these advanced solutions, crafting an ideal reward function remains challenging due to the propensity of RL algorithms to exploit the reward system in unforeseen ways. Agents might fulfill objectives in unexpected manners, highlighting the complexity of encoding desired behaviors, like adherence to social norms, into a reward function. An alternative strategy, imitation learning, circumvents the intricacies of reward engineering by having the agent learn through the emulation of expert behavior. However, acquiring a sufficient number of high-quality demonstrations for this purpose is often impractically costly. Humans, in contrast, learn with remarkable autonomy, benefiting from intermittent guidance from educators who provide tailored feedback based on the learner's progress. This interactive learning model holds promise for artificial agents, offering a customized learning trajectory that mitigates reward exploitation without extensive reward function engineering. The challenge lies in ensuring the feedback process is both manageable for humans and rich enough to be effective. Despite its potential, the implementation of human-in-the-loop (HiL) RL remains limited in practice. Our research endeavors to significantly lessen the human labor involved in HiL learning, leveraging both unsupervised pre-training and preference-based learning to enhance agent development with minimal human intervention.
Keywords
reinforcement learning from human feedback, preference learning
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Li Chenhao , Chen Xin , Li Chenhao
Topics Information, Computing and Communication Sciences , Engineering and Technology
Online Safe Locomotion Learning in the Wild
Reinforcement learning (RL) can potentially solve complex problems in a purely data-driven manner. Still, the state-of-the-art in applying RL in robotics, relies heavily on high-fidelity simulators. While learning in simulation allows to circumvent sample complexity challenges that are common in model-free RL, even slight distribution shift ("sim-to-real gap") between simulation and the real system can cause these algorithms to easily fail. Recent advances in model-based reinforcement learning have led to superior sample efficiency, enabling online learning without a simulator. Nonetheless, learning online cannot cause any damage and should adhere to safety requirements (for obvious reasons). The proposed project aims to demonstrate how existing safe model-based RL methods can be used to solve the foregoing challenges.
Keywords
safe mode-base RL, online learning, legged robotics
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Li Chenhao , Li Chenhao , Li Chenhao
Topics Engineering and Technology
Autonomous Curriculum Learning for Increasingly Challenging Tasks
While the history of machine learning so far largely encompasses a series of problems posed by researchers and algorithms that learn their solutions, an important question is whether the problems themselves can be generated by the algorithm at the same time as they are being solved. Such a process would in effect build its own diverse and expanding curricula, and the solutions to problems at various stages would become stepping stones towards solving even more challenging problems later in the process. Consider the realm of legged locomotion: Training a robot via reinforcement learning to track a velocity command illustrates this concept. Initially, tracking a low velocity is simpler due to algorithm initialization and environmental setup. By manually crafting a curriculum, we can start with low-velocity targets and incrementally increase them as the robot demonstrates competence. This method works well when the difficulty correlates clearly with the target, as with higher velocities or more challenging terrains. However, challenges arise when the relationship between task difficulty and control parameters is unclear. For instance, if a parameter dictates various human dance styles for the robot to mimic, it's not obvious whether jazz is easier than hip-hop. In such scenarios, the difficulty distribution does not align with the control parameter. How, then, can we devise an effective curriculum? In the conventional RSL training setting for locomotion over challenging terrains, there is also a handcrafted learning schedule dictating increasingly hard terrain levels but unified with multiple different types. With a smart autonomous curriculum learning algorithm, are we able to overcome separate terrain types asynchronously and thus achieve overall better performance or higher data efficiency?
Keywords
curriculum learning, open-ended learning, self-evolution, progressive task solving
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization Robotic Systems Lab
Hosts Li Chenhao , Li Chenhao , Li Chenhao , Bagatella Marco , Li Chenhao
Topics Engineering and Technology
Humanoid Locomotion Learning with Human Motion Priors
Humanoid robots, designed to replicate human structure and behavior, have made significant strides in kinematics, dynamics, and control systems. Research aims to develop robots capable of performing tasks in human-centric settings, from simple object manipulation to navigating complex terrains. Reinforcement learning (RL) has proven to be a powerful method for enabling robots to learn from their environment, enhancing their performance over time without explicit programming for every possible scenario. In the realm of humanoid robotics, RL is used to optimize control policies, adapt to new tasks, and improve the efficiency and safety of human-robot interactions. However, one of the primary challenges is the high dimensionality of the action space, where handcrafted reward functions fall short of generating natural, lifelike motions. Incorporating motion priors into the learning process of humanoid robots addresses these challenges effectively. Motion priors can significantly reduce the exploration space in RL, leading to faster convergence and reduced training time. They ensure that learned policies prioritize stability and safety, reducing the risk of unpredictable or hazardous actions. Additionally, motion priors guide the learning process towards more natural, human-like movements, improving the robot's ability to perform tasks intuitively and seamlessly in human environments. Therefore, motion priors are crucial for efficient, stable, and realistic humanoid locomotion learning, enabling robots to better navigate and interact with the world around them.
Keywords
motion priors, humanoid, reinforcement learning, representation learning
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization ETH Competence Center - ETH AI Center
Hosts Li Chenhao , Li Chenhao , Li Chenhao , Li Chenhao
Topics Information, Computing and Communication Sciences
Learning World Models for Legged Locomotion
Model-based reinforcement learning learns a world model from which an optimal control policy can be extracted. Understanding and predicting the forward dynamics of legged systems is crucial for effective control and planning. Forward dynamics involve predicting the next state of the robot given its current state and the applied actions. While traditional physics-based models can provide a baseline understanding, they often struggle with the complexities and non-linearities inherent in real-world scenarios, particularly due to the varying contact patterns of the robot's feet with the ground. The project aims to develop and evaluate neural network-based models for predicting the dynamics of legged environments, focusing on accounting for varying contact patterns and non-linearities. This involves collecting and preprocessing data from various simulation environment experiments, designing neural network architectures that incorporate necessary structures, and exploring hybrid models that combine physics-based predictions with neural network corrections. The models will be trained and evaluated on prediction autoregressive accuracy, with an emphasis on robustness and generalization capabilities across different noise perturbations. By the end of the project, the goal is to achieve an accurate, robust, and generalizable predictive model for the forward dynamics of legged systems.
Keywords
forward dynamics, non-smooth dynamics, neural networks, model-based reinforcement learning
Labels
Master Thesis
Description
Contact Details
More information
Published since: 2024-11-26
Organization Robotic Systems Lab
Hosts Li Chenhao , Li Chenhao , Li Chenhao , Klemm Victor , Li Chenhao
Topics Engineering and Technology
Next-Generation of Catalytic Plasma Coatings for Green Water Purification
Water contamination is a pressing issue for society, with Emerging Contaminants posing growing risks to human health and the environment. These substances include Persistent Organic Pollutants (POPs), often referred to as 'forever chemicals' due to their tendency to accumulate in water, soil, and living organisms. This project focuses on developing a new generation of catalytic nanomaterials designed to degrade POPs in water, by using plasma technology, with an emphasis on improving their stability and long-term performance.
Keywords
water purification; plasma technology; catalytic materials; thin films; photocatalysis
Labels
Internship , Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-11-25 , Earliest start: 2025-02-02 , Latest end: 2025-12-31
Organization Advanced Fibers
Hosts Denavascues Paula
Topics Engineering and Technology , Chemistry , Physics
Master thesis project: Deploying a new technology to find ghost nets for ocean cleanup
Master thesis project: Deploying a new technology to find ghost nets for ocean cleanup
Labels
Master Thesis
Description
Goal
Contact Details
More information
Published since: 2024-11-25 , Earliest start: 2025-01-01 , Latest end: 2025-10-31
Organization Biomimetic Membranes and Textiles
Hosts Defraeye Thijs
Topics Engineering and Technology
GPU Acceleration of Soft Body Modeling: Enhancing Performance with CUDA
The Soft Robotics Lab is developing a GPU-accelerated soft body modeling framework using the Finite Element Method (FEM). This enhancement aims to improve computational efficiency and enable more complex, real-time simulations. By leveraging GPUs' parallel processing power, simulations will be significantly faster. The project seeks to advance soft robotics research and enable innovative applications.
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
More information
Published since: 2024-11-25 , Earliest start: 2024-08-01
Organization Soft Robotics Lab
Hosts Mekkattu Manuel , Katzschmann Robert, Prof. Dr.
Topics Information, Computing and Communication Sciences , Engineering and Technology , Physics