While visual navigation has been extensively studied in agricultural robotics, most existing systems operate under the assumption of daytime conditions. In fact, deploying autonomous robots at night offers significant advantages, including 24-hour crop and soil monitoring, fruit harvesting, and nocturnal pest detection. However, modern vision-based systems heavily rely on large-scale, well-annotated image datasets, which remain challenging to obtain for nighttime operational scenarios.
To address this, we propose an unsupervised image translation framework that converts daytime plant-row RGB images into near-infrared (NIR) nighttime counterparts without requiring pixel-to-pixel supervision. This enables the direct reuse of daytime semantic labels for training nighttime perception models. In particular, by incorporating a pre-trained Contrastive Language-Image Pre-training (CLIP) model, our framework is designed to preserve semantic consistency during day-to-night translation. Additionally, a visibility mask is introduced to account for the limited effective range of NIR illumination in nighttime scenes.
We conduct comparative evaluations with state-of-the-art image translation baselines, demonstrating higher image quality and improved performance in downstream semantic segmentation for nighttime visual navigation. For evaluation, we utilize AgriNight, a novel dataset comprising 428 daytime and 549 nighttime images collected using night-vision-equipped mobile robots in agricultural fields, manually annotated with pixel-wise semantic labels, introducing it as the first benchmark for nighttime agricultural visual navigation. We also perform real-time autonomous navigation experiments with a physical robot operating at night. The data and code are available at: GitHub.
Blogger's Review: This research significantly advances nighttime agricultural robotics, particularly in unsupervised learning and image translation. By introducing the CLIP model and visibility masks, the authors effectively tackle challenges in nighttime visual navigation, providing essential technological support for future agricultural automation. The AgriNight dataset is a valuable resource for subsequent research in this domain.