Leandro Di Bella

Leandro Di Bella

Perception & robotics engineer. Physical AI for autonomous systems: drone localization & navigation, world models, tracking, VLA.

Leandro Di Bella

PhD researcher · ETRO, Vrije Universiteit Brussel (VUB) & imec · Technical lead, ORIGAMI (European Defence Fund)

Perception & physical AI

Robotics, computer vision & generative AI for autonomous systems.

I build deployable autonomy: technical lead on drone localization & navigation in a European Defence Fund project, with a research focus on perception that stays robust in the real world.

  • World models: perception & navigation for autonomous systems.
  • Localization: sensor fusion & state estimation in 3D environments.
  • Tracking & forecasting: 3D MOT and generative motion prediction.
  • VLA: vision-language-action models for robotics.
CV Publications Projects

Featured publications

All →
FlowS: One-Step Motion Prediction via Local Transport Conditioning IEEE Robotics and Automation Letters (RA-L), accepted HybridTrack: A Hybrid Approach for Robust Multi-Object Tracking IEEE Robotics and Automation Letters (RA-L/ICRA) Spectral-Aware Multi-Object Tracking in Harsh Aerial Perception Domains IEEE ICRA 2026 Workshop S2S ReferGPT: Towards Zero-Shot Referring Multi-Object Tracking IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

What I work on

Drone localization & navigation World models Sensor fusion SLAM Multi-object tracking Generative forecasting Vision-language-action ROS 2

Current focus: world-model-based navigation, localization in 3D environments, and perception that holds up under distribution shift.

Personal project

mappx.app →

Mappx

Map-centric social travel app (1000+ users).

Flutter FastAPI PostgreSQL Azure Mapbox

Contact

Email →

Open to full-time perception & robotics roles for autonomous systems.

leandro.dibella@gmail.com GitHub LinkedIn Scholar