This academic article presents a vision-based crop row-switching algorithm for efficient navigation in arable fields. The algorithm uses a single front-mounted camera on a mobile robot to navigate an entire field. It employs deep learning-based RGB image segmentation and depth data to detect the end of a crop row and the re-entry point to the next row. The navigation system was tested in a real sugar beet field and showed promising results. This approach aims to reduce costs associated with agricultural robotics and increase the adoption of such technologies.
Publication date: 22 Sep 2023
Project Page: Not provided
Paper: https://arxiv.org/pdf/2309.11989
