MetaCropFollow: Few-Shot Adaptation with Meta-Learning for Under-Canopy Navigation

TL;DR

This paper introduces MetaCropFollow, a meta-learning approach enabling autonomous under-canopy robots to adapt quickly to new agricultural environments with minimal data, improving navigation robustness.

Contribution

It proposes a meta-learning framework for rapid adaptation of visual navigation models to diverse crop conditions, addressing domain shift challenges in agriculture.

Findings

Meta-learning enables quick adaptation to new environments.

Improved navigation robustness in low-data scenarios.

Effective handling of domain shifts in agricultural settings.

Abstract

Autonomous under-canopy navigation faces additional challenges compared to over-canopy settings - for example the tight spacing between the crop rows, degraded GPS accuracy and excessive clutter. Keypoint-based visual navigation has been shown to perform well in these conditions, however the differences between agricultural environments in terms of lighting, season, soil and crop type mean that a domain shift will likely be encountered at some point of the robot deployment. In this paper, we explore the use of Meta-Learning to overcome this domain shift using a minimal amount of data. We train a base-learner that can quickly adapt to new conditions, enabling more robust navigation in low-data regimes.