• People
  • Research
  • Projects
  • Publications
  • Resources
ViCoS Lab

Matej Dobrevski

Researcher, Teaching Assistant
matej.dobrevski@fri.uni-lj.si

About

I recently finished my PhD dissertation on using deep reinforcement learning methods in robot navigation, under the supervision of Prof. Danijel Skočaj, PhD. I have worked on numerous projects involving deep learning in computer vision and robotics. I am currenly working on the SABUVIS II project, financed by the European Defense Agency.

My research interests are deep learning methods in computer vision and robot navigation and localization.

I am a teaching assistant for the courses Introduction to Computer Science and Development of Intelligent Systems.

Current projects

Sabuvis II - Swarm of biomimetic underwater vehicles for underwater intelligence, surveillance and reconnaissance

June 2023 - December 2025
This project is a continuation of the successful collaboration enjoyed between Poland, Germany and Portugal in the first SABUVIS project. In this first iteration, three different Biomimetic Underwater Vehicles (BUVs) were constructed with the resulting conclusion that BUVs can be designed with varying degrees of similarity with living organisms. Some of the identified benefits are that undulating propulsion consumes less electrical energy than conventional propulsion systems utilising screw based propellers, and different hydroacoustic signatures are produced with a corresponding lower noise level.

Research

Deep reinforcement learning for navigation

This research investigates the development of autonomous mobile robot navigation methods using deep reinforcement learning. Our methods aim to produce navigation policies which are learned completely in simulation and deployed on real robots.

Publications

Faculty of Computer and Information Science

Visual Cognitive Systems Laboratory

University of Ljubljana

Faculty of Computer and Information Science

Večna pot 113
SI-1000 Ljubljana
Slovenia
Tel.: +386 1 479 8245