Papers
5
Total Citations
113
H-Index
4
About
Sascha Hornauer is a researcher working at the intersection of sensory perception, autonomous systems, and machine learning, with particular expertise in bio-inspired perception and autonomous driving. His most recognized contribution is the **BatVision** project, a pioneering line of work that draws inspiration from echolocation in bats and dolphins to teach artificial systems how to infer 3D spatial layouts from sound alone. By training neural networks on binaural audio echoes, Hornauer demonstrated that low-cost acoustic sensing could generate meaningful depth maps and scene layouts — a compelling alternative to expensive LiDAR and radar setups. This work, which has accumulated over 55 citations, spans multiple publications including a dedicated dataset release to support the broader research community. Beyond perception, Hornauer has made notable strides in autonomous driving through his work on General Reinforced Imitation (GRI), which cleverly combines deep reinforcement learning with expert demonstrations to overcome the notorious instability and sample inefficiency of standard DRL approaches, earning 44 citations since 2023. Together, his contributions reflect a creative research vision that bridges biological inspiration with practical robotics challenges, making his work highly relevant for students exploring multimodal perception and intelligent autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1BatVision: Learning to See 3D Spatial Layout with Two Ears55 citations · 2020
- 2
- 3BatVision: Learning to See 3D Spatial Layout with Two Ears6 citations · 2019
- 4BatVision with GCC-PHAT Features for Better Sound to Vision Predictions5 citations · 2020
- 5The Audio-Visual BatVision Dataset for Research on Sight and Sound3 citations · 2023