Jonas Frey
Papers
11
Total Citations
147
H-Index
8
About
Jonas Frey is a robotics researcher whose work sits at the intersection of autonomous navigation, legged locomotion, and machine learning for real-world robot deployment. His research addresses some of the most demanding challenges in field robotics: enabling robots to move safely and intelligently through unstructured, hazardous, and visually complex outdoor environments. Frey has made notable contributions to traversability estimation, developing systems like RoadRunner (31 citations) and Wild Visual Navigation (13 citations) that allow robots to learn which terrain is safe to traverse using onboard sensing and self-supervised learning. His work on risk-aware quadrupedal locomotion using distributional reinforcement learning (19 citations) broke new ground by explicitly modeling uncertainty and danger in legged robot movement — a critical step toward deploying robots in genuinely hazardous settings. He has also advanced resilient navigation under compromised perception (17 citations) and versatile skill learning through adversarial imitation (19 citations), broadening the behavioral repertoire of autonomous systems. Applying these capabilities to real-world missions, Frey has contributed to autonomous forest inventory systems and safe multi-goal planning frameworks. With over 140 cumulative citations across recent publications, his research is rapidly shaping how robots perceive, reason about, and navigate the physical world beyond controlled laboratory conditions.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10