PeAR WPI

EdgeFlowNet : 100FPS@1W Dense Optical Flow For Tiny Mobile Robots

Sai Ramana Kiran¹ Rishabh Singh¹ Manoj Velmurugan¹ Nitin J. Sanket¹

¹ Perception and Autonomous Robotics Group (PeAR)
at
Worcester Polytechnic Institute

Abstract

Optical flow estimation is a critical task for tiny mobile robotics to enable safe and accurate navigation, obstacle avoidance, and other functionalities. However, optical flow estimation on tiny robots is challenging due to limited onboard sensing and computation capabilities. In this paper, we propose EdgeFlowNet, a high-speed, low-latency dense optical flow approach for tiny autonomous mobile robots by harnessing the power of edge computing. We demonstrate the efficacy of our approach by deploying EdgeFlowNet on a tiny quadrotor to perform static obstacle avoidance and flight through unknown gaps and dynamic obstacle dodging. EdgeFlowNet is about 20X faster than the previous state-of-the-art approaches while improving accuracy by over 20% and using only 1.08W of power enabling advanced autonomy on palm-sized tiny mobile robots.

Figure: Left to right: Different applications of our EdgeFlowNet network estimating high-speed accurate optical flow: static obstacle avoidance, flight through unknown gaps and dodging dynamic obstacles.

Resources

Arxiv Github IEEE RA-L Paper

Cite

@ARTICLE{Raju2024EdgeFlowNet,
author={Raju, Sai Ramana Kiran Pinnama and Singh, Rishabh and Velmurugan, Manoj and Sanket, Nitin J.},
journal={IEEE Robotics and Automation Letters},
title={EdgeFlowNet: 100FPS@1W Dense Optical Flow For Tiny Mobile Robots},
year={2024},
volume={},
number={},
pages={1-8},
doi={10.1109/LRA.2024.3496336}}