ETH Zurich

Open Opportunities

Tree species identification using deep learning ETH Zurich Forest Resources Management Other organizations: Photogrammetry and Remote Sensing (Prof. Schindler) Tree species maps are crucial for effective forest management, biomass assessment, and biodiversity monitoring. Remote sensing products offer flexible and cost-effective ways to assess forest characteristics, while deep learning methods promise high predictive accuracy and transformative applications in forestry. This study aims to apply novel deep learning approaches to detect and identify individual trees and tree species in mixed forests. By addressing the challenges of tree species identification, this research will enhance biodiversity assessment, forest resilience understanding, and management strategies. Artificial Intelligence and Signal and Image Processing, Forestry Sciences, Geomatic Engineering ETH Zurich (ETHZ), Master Thesis

Agile Flight of Flexible Drones in Confined Spaces University of Zurich Robotics and Perception The project aims to create a controller for an interesting and challenging type of quadrotor, where the rotors are connected via flexible joints. Control Engineering, Flight Control Systems, Intelligent Robotics, Systems Theory and Control Master Thesis, Semester Project

Automated Interpretation of Core Fracture Surfaces Using Deep Learning Swiss Federal Institute for Forest, Snow and Landscape Research WSL Institute for Snow and Avalanche Research SLF Fracture surfaces in rock cores contain valuable structural information crucial for geological interpretation, engineering design, and are commonly mapped and analyzed by geologists. With advancements in camera technologies and computational techniques, it is now possible to digitize these surfaces in high resolution and apply automated methods for fracture analysis. Computer Vision, Geology, Image Processing, Photogrammetry and Remote Sensing Bachelor Thesis, Master Thesis, Semester Project

Vision-Based World Models for Real-Time Robot Control University of Zurich Robotics and Perception This project aims to use vision-based world models as a basis for model-based reinforcement learning, aiming to achieve a generalizable approach for drone navigation. Computer Vision, Intelligent Robotics, Simulation and Modelling Master Thesis, Semester Project

Vision-Based Reinforcement Learning in the Real World University of Zurich Robotics and Perception We aim to learn vision-based policies in the real world using state-of-the-art model-based reinforcement learning. Computer Vision, Flight Control Systems, Intelligent Robotics Master Thesis, Semester Project

Fast Quantized Speech-Synthesis on Mobile Device ETH Zurich Center for Project-Based Learning D-ITET The proliferation of mobile and embedded devices has spurred the demand for efficient, high-quality speech synthesis systems that operate entirely on-device. This project aims to develop a fast, quantized speech synthesis pipeline optimized for mobile platforms (i.e. Samsung Galaxy, Google Pixel Pro 8), focusing on reducing computational load and memory usage without compromising audio quality.​ Engineering and Technology, Information, Computing and Communication Sciences Bachelor Thesis, ETH Zurich (ETHZ), Master Thesis, Semester Project

Fully Electric Railway Supply Chains: Rail-City Portal as Key Infrastructure for Sustainable E-Mobility ETH Zurich Chair of Logistics Management Switzerland is committed to achieving net-zero greenhouse gas emissions by 2050. Innovative and sustainable freight solutions are essential as the transport sector accounts for a significant share of CO2 emissions. Currently, road transport dominates the modal split (62%), with rail transport contributing only 38%. Traditional urban logistics rely heavily on road-based transport, contributing to congestion, emissions, and inefficiencies in last-mile delivery. While rail offers a sustainable alternative for long-distance freight, its integration into city logistics remains limited. Therefore, this thesis investigates how a fully electric railway supply chain centered around a rail-city portal (e.g., intermodal urban connectors between rail and last-mail logistics) can reduce emissions and improve the efficiency of urban freight distribution. A rail-city portal can be a transshipment node, bridging rail and e-mobility for last-mile logistics. Transport Economics, Transport Engineering, Transportation Master Thesis

Evaluating the Potential of Any-Aged Forestry Compared to Regular and Close-to-Nature Management: A Forest Modeling Study ETH Zurich Forest Resources Management This Master’s thesis explores whether Any-Aged Forestry (AAF), could offer a more optimal management strategy compared to Regular Forest Management (RFM) and Close-to-Nature management in a Swiss forest context. Using Forest Studio, a forest modeling platform under development, the project will implement and test the AAF approach alongside conventional management systems in a case study area in Canton Zurich. The goal is to evaluate performance across ecological and economic indicators. Forestry Sciences not elsewhere classified ETH Zurich (ETHZ), Master Thesis

Making aviation's climate claims credible ETH Zurich Climate Policy Group Can we keep flying while still meeting climate goals? This Master thesis dives into a bold idea: balancing fossil jet fuel emissions by locking away an equivalent amount of CO₂ underground. But how can we be sure it actually works — and that the public trusts it?With this work, you will explore the rules, standards, and safeguards needed to make this vision credible. You’ll help shape the future of climate accountability in aviation as part of a broader project on geologically-balanced fuels. The thesis will be supervised by researchers from ETH Zurich and the University of Oxford, with the possibility to conduct the research at either locations. 👉 Curious? Find out more in the full project description in the leaflet attached. Accounting, Auditing and Accountability, Business and Management, Environmental Sciences, Justice and Legal Studies, Policy and Administration, Tourism, Transportation ETH Zurich (ETHZ), Master Thesis, Other specific labels

Does Management Unit Definition Matter? Impacts on Ecosystem Services in Contrasting Landscapes ETH Zurich Forest Resources Management Management units (MUs) are the core spatial entities where forest management strategies are applied. Their delineation is typically based on stand characteristics (e.g., tree species, age structure), ownership boundaries, or legacy units established decades ago. Before a new forest management plan is developed—typically every ten years—foresters revise these MU boundaries using aerial imagery and ground-based assessments. However, the implications of how these boundaries are defined are rarely questioned. This Master’s thesis investigates how different approaches to MU definition, and how frequently they are revised, may alter the expected provisioning of ecosystem services (ES) such as timber and biodiversity. The study compares these impacts across two contrasting landscapes: one topographically complex and one relatively homogeneous. Forestry Sciences not elsewhere classified ETH Zurich (ETHZ), Master Thesis