William Fu
Papers
4
Total Citations
120
H-Index
4
About
William Fu is a leading researcher at the intersection of deep reinforcement learning (RL) and autonomous aerial robotics, with a particular focus on resource-constrained systems. His major contributions center on enabling sophisticated AI capabilities—such as visual navigation and source seeking—to run directly onboard tiny, low-power drones, a domain where computational resources are extremely limited. Fu is the creator of the Air Learning platform, an open-source simulator and gym environment that serves as a benchmark for algorithm-hardware co-design in aerial robots. This work, which has garnered over 70 citations across its key papers, provides essential tools for the community to train and test deep-RL policies under realistic, domain-randomized conditions. Notably, his research on tinyRL demonstrated the first fully autonomous source-seeking mission executed entirely onboard a nano quadcopter’s microcontroller, a breakthrough that pushes the boundaries of edge AI. By bridging the gap between high-performance algorithms and severe hardware constraints, Fu’s work is paving the way for a new generation of intelligent, agile, and ultra-compact flying robots.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3Tiny Robot Learning (tinyRL) for Source Seeking on a Nano Quadcopter26 citations · 2021
- 4