ROAIMar 6, 2024

On-device Self-supervised Learning of Visual Perception Tasks aboard Hardware-limited Nano-quadrotors

arXiv:2403.04071v16 citationsh-index: 16ICRA
AI Analysis

This addresses the problem of domain shift for nano-drones operating in unknown environments, representing an incremental advancement for hardware-limited robotics.

The paper tackles domain shift in visual perception for nano-drones by proposing on-device self-supervised fine-tuning of pre-trained CNNs, achieving up to 30% improvement in mean absolute error with only 22 seconds of fine-tuning on a low-power chip.

Sub-\SI{50}{\gram} nano-drones are gaining momentum in both academia and industry. Their most compelling applications rely on onboard deep learning models for perception despite severe hardware constraints (\ie sub-\SI{100}{\milli\watt} processor). When deployed in unknown environments not represented in the training data, these models often underperform due to domain shift. To cope with this fundamental problem, we propose, for the first time, on-device learning aboard nano-drones, where the first part of the in-field mission is dedicated to self-supervised fine-tuning of a pre-trained convolutional neural network (CNN). Leveraging a real-world vision-based regression task, we thoroughly explore performance-cost trade-offs of the fine-tuning phase along three axes: \textit{i}) dataset size (more data increases the regression performance but requires more memory and longer computation); \textit{ii}) methodologies (\eg fine-tuning all model parameters vs. only a subset); and \textit{iii}) self-supervision strategy. Our approach demonstrates an improvement in mean absolute error up to 30\% compared to the pre-trained baseline, requiring only \SI{22}{\second} fine-tuning on an ultra-low-power GWT GAP9 System-on-Chip. Addressing the domain shift problem via on-device learning aboard nano-drones not only marks a novel result for hardware-limited robots but lays the ground for more general advancements for the entire robotics community.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes