Robotics and Perception

Open Opportunities

Deep object pose estimation University of Zurich Robotics and Perception Train a neural network to detect poses of arbitrary objects Computer Vision Master Thesis

Pushing hard cases in tag detection with a CNN University of Zurich Robotics and Perception Try to handle failures in april/aruco tag detection using deep learning. Computer Vision Bachelor Thesis, Master Thesis, Semester Project

Learning Robust Visual Place Recognition Combining Appearance, Sequence and Structure University of Zurich Robotics and Perception Build a Neural Network that takes as input an image, a sequence of previous images, and the geometry extracted from that, and produces as output a descriptor which can be used for place recognition. Computer Vision Master Thesis

Teach and Aggressive Repeat University of Zurich Robotics and Perception Teach and repeat, but try to be as fast as possible on repeat. Goal is to deploy this on a quad. Computer Vision, Intelligent Robotics Master Thesis

Learn Depth from Multiple Input Modalities University of Zurich Robotics and Perception Learn depth from RGB frames and sparse depth information. Computer Vision, Neural Networks, Genetic Alogrithms and Fuzzy Logic Master Thesis, Semester Project

Analyzing The Role of Perception Latency in High Speed Sense and Avoid University of Zurich Robotics and Perception To prevent a collision with an obstacle or an incoming object, a robot needs to detect them as fast as possible and execute a safe maneuver to avoid them. The higher the relative speed between the robot and the object, the more critical the role of perception latency becomes. The goal of this project is to extend our previous work (http://rpg.ifi.uzh.ch/docs/RAL19_Falanga.pdf) on the analysis of the role of perception latency in high speed sense and avoid, in order to consider different robot dynamics and multiple scenarios, such as cases including dynamic obstacles. Engineering and Technology Master Thesis

Performance analysis and characterization of a morphing quadrotor University of Zurich Robotics and Perception The goal of this project is to study the impact of the morphology assumed by a quadrotor able to change its shape in-flight on the performance of its propulsion system. Engineering and Technology Semester Project

Learning minimal representations of places University of Zurich Robotics and Perception Use machine learning to condense visual observations of places into minimal representations that allow for 6dof relative pose estimation. Computer Vision, Intelligent Robotics Master Thesis

Safe Reinforcement Learning for Robotics University of Zurich Robotics and Perception During this project, we will develop machine learning based techniques to let a (real) drone learn to fly nimbly through gaps and gates, while minimizing the risk of critical failures and collisions. Intelligent Robotics Master Thesis, Semester Project

Unsupervised Obstacle Detection Learning University of Zurich Robotics and Perception In this project, we aim to build a self-supervised depth estimation and segmentation algorithm by embedding classic computer vision principles (e.g. brightness constancy) into a neural network. Computer Vision, Intelligent Robotics, Knowledge Representation and Machine Learning Master Thesis, Semester Project