Matej Dobrevski

Papers

Total Citations

137

H-Index

About

Matej Dobrevski is a leading researcher in mobile robotics, specializing in local navigation and autonomous systems. His work focuses on two critical challenges: robust obstacle avoidance in dynamic environments and map-less goal-driven navigation. Dobrevski’s most significant contributions center on the Dynamic Window Approach (DWA), a widely used local navigation method. In his highly cited 2020 paper, “Adaptive Dynamic Window Approach for Local Navigation” (42 citations), he pioneered a method to automatically tune DWA’s cost function parameters, solving a long-standing problem of manual, environment-specific configuration. He extended this work in 2024 with “Dynamic Adaptive Dynamic Window Approach” (55 citations), introducing a robust variant that excels in human-populated, unstructured spaces. Complementing these advances, his 2021 paper “Deep reinforcement learning for map-less goal-driven robot navigation” (40 citations) demonstrates a novel deep reinforcement learning framework that enables robots to navigate without pre-built maps, a breakthrough for dynamic or unknown environments. With over 137 total citations across his key works, Dobrevski’s research has directly improved the safety and adaptability of mobile robots, making him a notable figure in the field of autonomous navigation.