Radu Grosu
Papers
17
Total Citations
270
H-Index
9
About
Radu Grosu is a prominent researcher whose work spans cyber-physical systems, robotics, autonomous vehicles, and biologically inspired neural networks. His contributions have significantly advanced how machines perceive, learn, and navigate complex real-world environments. Among his most influential achievements is the development of OpenUAV, a UAV testbed that democratized drone research and education by reducing the costly burden of physical testing — a work that has garnered over 60 citations. Grosu has also pioneered the use of biologically inspired neural architectures, drawing from the nervous system of the nematode *C. elegans* to design interpretable liquid time-constant recurrent neural networks for robotic control, earning widespread recognition within the machine learning and robotics communities. His work on model-based deep reinforcement learning for autonomous racing cars and world-model-driven zero-shot transfer further demonstrates his commitment to sample-efficient, generalizable AI systems. Grosu has also tackled critical safety challenges, examining the limitations of adversarial training in robot learning environments. Spanning foundational infrastructure like ROS sensor fusion to cutting-edge reinforcement learning, his research consistently bridges theoretical innovation with practical robotic deployment, establishing him as a leading voice in intelligent autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1OpenUAV: A UAV Testbed for the CPS and Robotics Community62 citations · 2018
- 2Designing Worm-inspired Neural Networks for Interpretable Robotic Control43 citations · 2019
- 3Latent Imagination Facilitates Zero-Shot Transfer in Autonomous Racing35 citations · 2022
- 4
- 5
- 6
- 7Adversarial Training is Not Ready for Robot Learning12 citations · 2021
- 8Generic sensor fusion package for ROS12 citations · 2015
- 9
- 10