Student Projects

Main content

ETH Zurich uses SiROP to publish and search scientific projects. For more information visit

Spatio-temporal-Spectral Environment Modeling using a UAV

The goal of this thesis is to use a Micro Air Vehicle equipped with inertial sensors, a high resolution colour camera and lower resolution multispectral camera to collect and measure field parameters which will enable an analysis of crop health, determination of variations within the field in space Show details 

Human-Robot Cooperative Interaction for a Manipulation Task

The goal of this project is to develop a system for human-robot cooperation for a manipulation task on a tabletop-sized workspace. Show details 

Line Descriptors for Place Recognition

The goal of this project is to implement the current state-of-the-art line descriptors and evaluate their performance on some of the publicly available datasets for place recognition. Show details 

Semantic 3D SLAM

Robots map their environment with their onboard sensors. However, the representations robots use are often abstract and can be ambiguous. For facilitating scene understanding and increased robustness, we aim to include machine learning and semantic relations into the mapping process. Show details 

Multi-robot Mapping, Navigation and Exploration (MA/SA)

Master/Semester thesis at the Autonomous Systems Lab at ETH Zurich. The mapping and localization team is looking for motivated students for projects in the field of mapping, localization, machine learning, exploration and 3D reconstruction. Show details 

Deep Reinforcement Learning for Motion Planning

Tuning a motion planner can be a cumbersome endeavor. In our recent work we showed how neural networks can be used for imitation learning. This work builds on top of supervised imitation learning, such that reinforcement learning techniques should be applied to robot navigation. Show details 

Evaluation of Sparse, Semi-Dense and Dense Visual-Odometry Methods

The goal of this project is to implement an evaluation framework that determines the best suited visual odometry (VO) concept for aerial imagery. Show details 

Real-Time Point-Cloud Summarization

The goal of this project is to develop a method for real-time summarization of point-clouds using geometric primitives or repetitive structures, such as planes, cylinders, and more complex manifolds. This project has as input point-clouds generated either by ground robots or Unmanned Aerial Vehicles Show details 

Augmented Reality based Assembling Guidance Using Microsoft HoloLens

The goal of this project is to develop a system to support assembly workers by means of visual assembly guidance using augmented reality. Show details 

Detection, Pose Estimation and Manipulation of Household Objects

The goal of this project is to evaluate current state-of-the-art object detectors. Furthermore, the most suitable technique should be implemented and tested on a small dataset. Show details 

Aerial Benchmark: Evaluation of Publicly Available Visual-Inertial Odometry and SLAM Frameworks based on Aerial Groundtruth Datasets

The goal of this project is to evaluate publicly available visual-inertial odometry (VIO) and simultaneous localization and mapping (SLAM) frameworks applied to fixed-wing unmanned aerial vehicles (UAVs) based on ground-truth datasets. Show details 

Evaluation of Visual Inertial Map Quality

The first goal of this project is to implement a GUI-based visualization of a visual-inertial sparse feature map using QT. In a second step, the map quality is to be evaluated and directly visualized in QT. Show details 

Autonomous Airshow Using Variable Pitch Quadrotor

Variable pitch quadrotors have very fast dynamics and the capability to achieve negative thrust. Exploiting these properties, we can achieve very aggressive aerobatic trajectories. The aim of this project is to design and implement a fully autonomous airshow based on variable pitch quadrotors. Show details 

Collaborative Visual-Inertial based Sensing using Multiple UAVs

This project is aimed at the development of a collaborative visual-inertial state estimation for multiple UAV's. The goal is to enable teams of robots with the sensing capability to perform tasks which require collaboration. Show details 

Fall Recovery for a Quadrupedal Robot

The ASL ( develops quadrupedal robots that can statically walk and dynamically run on different types of terrain. In contrast to wheeled or tracked vehicles, legged robots have the ability to recover from falls. So far, Show details 

Trajectory Optimization for Quadrupedal Locomotion

A challenging problem in legged robotics is to generate motion plans that are feasible with respect to the robot’s skills (range of motion, actuator limits) and the underlying physics (dynamics, contact constraints). State-of-the-art approaches tackl Show details 

Note on plagiarism

We would like to suggest every student, irrespective of the type of project (Bachelor, Semester, Master, ...), to make himself/herself familiar with ETH rules regarding plagiarism

Page URL:
© 2017 Eidgenössische Technische Hochschule Zürich