I am a PhD student at TU Wien, working in the Cyber-Physical Systems Group under the supervision of Radu Grosu.

My research interests lie in the study of learning-based approaches able to cope with safety-critical environments, under strict operational constraints. The goal of my work is the enforcement of formal requirements on the behavior emergent by automated learning approaches, specifically reinforcement learning. The application area is autonomous driving and, more in general, robotics.

My main interests:

  • Safe and robust learning in real-world systems
  • Reinforcement Learning in robotics
  • Assessment and validation of autonomous controllers

News


  • [Paper Accepted - Jan 2024] The paper "Learning Adaptive Safety for Multi-Agent Systems" has been accepted at ICRA 2024!
  • [Seminar - Nov 2023] I presented the work on adaptive safety for multi-agent systems at the DCRES seminar series of TU Wien.
  • [Hackaton - Oct 2023] Winner team (EUR 1250) at the AI Safety Hackaton for Agency-Preserving Reinforcement Learning.
  • [Preprint - Sep 2023] The paper "Learning Adaptive Safety for Multi-Agent Systems" is out.
  • [Research & Networking - Sep 2023] I had the pleasure to participate at the first High-Performance Trip organized by the Marshall Plan Foundation.
  • [Paper Accepted - May 2023] The paper "Enhancing Robot Learning through Learned Human-Attention Feature Maps" has been accepted at ICRA 2023's workshop on Effective Representations, Abstractions, and Priors for Robot Learning (RAP4Robots).
  • [Research Exchange - Feb 2023] I am starting a research exchange at the xLab of the University of Pennsylvania, Philadelphia, USA.
  • [Grant - Dec 2022] I have been nominated for the Marshall Plan Scholarship (EUR 6500) to support my research on safe learning for intelligent autonomous driving systems.
  • [Science Booth - Sep 2022] Our most-recent research activity on autonomous driving has been presented at the Forschungsfest (Science Festival) in Vienna.
  • [Paper Accepted - Aug 2022] The paper "Safe Policy Improvement in Constrained Markov Decision Processes" has been accepted to ISoLA 2022!
  • [Seminar - Jun 2022] I presented the work on safe policy improvement from formal requirements for RL.
  • [F1/10th Grand Prix at ICRA - May 2022] Our team *"Scuderia Segfault"* placed 3rd at the 10th F1TENTH Autonomous Grand Prix at ICRA2022! Video.
  • [Paper Accepted - Jan 2022] The paper "Latent Imagination Facilitates Zero-Shot Transfer in Autonomous Racing" has been accepted to ICRA 2022!
  • [Seminar - Nov 2021] I presented the work on automatic reward shaping from multiple formal requirements for RL control tasks.
  • [F1/10th Grand Prix at IROS - Oct 2021] Our team "Scuderia Segfault" won the 9th F1TENTH Autonomous Grand Prix at IROS2021! Video.
  • [Poster Session - Sep 2021] I presented the project ADEX on Reinforcement Learning for autonomous-driving scenario generation at AVL Research Networking Day 2021.