A lightweight optical flow network combined with image chunking achieves 93.6 FPS on a 1-watt Edge TPU and enables a tiny quadrotor to avoid static and dynamic obstacles in real time.
Active Vision Based Embodied-AI Design for Nano-UAV Autonomy
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.RO 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
EdgeFlowNet: 100FPS@1W Dense Optical Flow For Tiny Mobile Robots
A lightweight optical flow network combined with image chunking achieves 93.6 FPS on a 1-watt Edge TPU and enables a tiny quadrotor to avoid static and dynamic obstacles in real time.