Research

My research lies at the intersection of Robotics and Machine Learning, with a focus on enabling autonomous systems to learn complex behaviours in unstructured environments.

  • Space Robotics: Autonomous manipulation for in-orbit servicing and planetary exploration, including contact-rich tasks such as assembly and maintenance
  • Reinforcement Learning: Model-based and model-free methods for dexterous control, with emphasis on sample efficiency and reward shaping
  • Modeling and Simulation: Procedural scene generation, domain randomization, and high-fidelity physics for sim-to-real transfer

Publications

2025

Robot Learning Beyond Earth: Enabling Adaptive Autonomy in Space

Andrej Orsula

PhD Thesis, University of Luxembourg

RL-AVIST: Reinforcement Learning for Autonomous Visual Inspection of Space Targets

Matteo El-Hariry, Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez

preprint

Space Robotics Bench: Robot Learning Beyond Earth

Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez

preprint

Learning Tool-Aware Adaptive Compliant Control for Autonomous Regolith Excavation

Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez

ASTRA 2025

Sim2Dust: Mastering Dynamic Waypoint Tracking on Granular Media

Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez

iSpaRo 2025

Learning Compliant Manipulation in Space

Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez

ICRA 2025 — Workshop on Enhancing Dexterity in Space Environments

Advancing Adaptive Autonomy Through Procedural Space Environments

Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez

IAC 2025

2024

Towards Benchmarking Robotic Manipulation in Space

Towards Benchmarking Robotic Manipulation in Space

Andrej Orsula, Antoine Richard, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez

CoRL 2024 — Workshop on Mastering Robot Manipulation in a World of Abundant Data

GraspLDM: Generative 6-DoF Grasp Synthesis using Latent Diffusion Models

GraspLDM: Generative 6-DoF Grasp Synthesis using Latent Diffusion Models

Kuldeep R. Barad, Andrej Orsula, Antoine Richard, Jan Dentler, Miguel Olivares-Mendez, Carol Martinez

IEEE Access

Leveraging Procedural Generation for Learning Autonomous Peg-in-Hole Assembly in Space

Leveraging Procedural Generation for Learning Autonomous Peg-in-Hole Assembly in Space

Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez

iSpaRo 2024

A Retrospective on the Robot Air Hockey Challenge: Benchmarking Robust, Reliable, and Safe Learning Techniques for Real-world Robotics

Puze Liu, Jonas Günster, Niklas Funk, Simon Gröger, Dong Chen, Haitham Bou-Ammar, Julius Jankowski, Ante Marić, Sylvain Calinon, Andrej Orsula, et al.

NeurIPS 2024

Immersive Rover Control and Obstacle Detection based on Extended Reality and Artificial Intelligence

Sofía Coloma, Alexandre Frantz, Dave van der Meer, Ernest Skrzypczyk, Andrej Orsula, Miguel Olivares-Mendez

VR 2024

Evaluating Image-Based Visual Servoing Techniques for Robotic Manipulation in Space

Lina María Amaya-Mejía, Andrej Orsula, Mohamed Ghita, Miguel Olivares-Mendez & Carol Martinez

ERF 2024

Grasp-O: A Generative System for Object-Centric 6-DoF Grasping of Unknown Objects

Kuldeep R. Barad, Andrej Orsula, Antoine Richard, Jan Dentler, Miguel Olivares-Mendez, Carol Martinez

ERF 2024

2023

Learning to Play Air Hockey with Model-Based Deep Reinforcement Learning

Learning to Play Air Hockey with Model-Based Deep Reinforcement Learning

Andrej Orsula

NeurIPS 2023 — Robot Air Hockey Challenge

2022

Learning to Grasp on the Moon from 3D Octree Observations with Deep Reinforcement Learning

Learning to Grasp on the Moon from 3D Octree Observations with Deep Reinforcement Learning

Andrej Orsula, Simon Bøgh, Miguel Olivares-Mendez, Carol Martinez

IROS 2022

2021

Deep Reinforcement Learning for Robotic Grasping from Octrees

Deep Reinforcement Learning for Robotic Grasping from Octrees

Andrej Orsula

Master's Thesis, Aalborg University