CLEVER: Stream-Based Active Learning for Robust Semantic Perception From Human Instructions
Published in IEEE RA-L, 2025
TL;DR: World-wide first implementation of stream-based active learning on a real robot.
Publications
A complete list of my publications is on my Google Scholar profile. Feel free to click the links below. You can find additional materials like slides, posters, links to the websites, codes, and more. I am only listing my lead author papers here because I can share more materials without asking permissions from my collaborators.
Selected Papers
Introspective Perception for Long-term Aerial Telemanipulation with Virtual Reality
Published in T-FR, 2024
TL;DR: A probabilistic approach to building a perception system for a long-term deployment in the field.
Learning Expressive Priors for Generalization and Uncertainty Estimation in Neural Networks
Published in ICML, 2023
TL;DR: Dreaming about foundational priors for approximate Bayesian inference.
Learning Fluid Flow Visualizations from In-Flight Images with Tufts
Published in IEEE RA-L, 2023
TL;DR: A probabilistic approach to a real-world semantic segmentation problem.
Trust Your Robots! Predictive Uncertainty Estimation of Neural Networks with Sparse Gaussian Processes
Published in CoRL, 2021
TL;DR: On how Gaussian Processes can be combined with deep learning for reliable and fast uncertainty estimation.
Estimating Model Uncertainty of Neural Networks in Sparse Information Form
Published in ICML, 2020
TL;DR: On how Bayesian Neural Networks can scale to ImageNet size data and architecture, or even beyond.
Visual-Inertial Telepresence for Aerial Manipulation
Published in IEEE ICRA, 2020
TL;DR: First roll-out of the DLR cable-Suspended Aerial Manipulator (SAM) in outdoor environments.
Towards Autonomous Stratospheric Flight: A Generic Global System Identification Framework for Fixed-Wing Platforms
Published in IEEE IROS, 2018
TL;DR: Envisioning to build mathematical models of autonomous stratospheric flight.