Editorial: Vision, learning, and robotics: AI for plants in the 2020s

摘要

IntroductionWith the growth of the global population and increasing demand for food, agricultural production is under significant pressure. At the same time, climate change and resource constraints exacerbate these challenges, further heightening the need for sustainable agricultural practices. To address these complex issues, the field of plant science is undergoing a technological revolution. The rapid advancement of artificial intelligence (AI), computer vision, and robotics is redefining how plants are studied and agricultural practices are managed. From high-throughput phenotyping to precision agriculture and real-time monitoring, these technologies are dramatically improving efficiency and accuracy, laying a foundation for more resilient and sustainable agricultural systems. This research topic brings together pioneering studies to demonstrate how AI is advancing plant science and providing innovative solutions for modern agriculture.Research ContributionsThe articles in this research topic showcase innovations across multiple fields. These contributions can be summarized into five key areas, each highlighting significant advancements in the study and application of plant science.High-Throughput Phenotyping and Crop MonitoringHigh-throughput phenotyping is a critical component of precision agriculture. By incorporating advanced deep learning models, researchers have significantly enhanced the efficiency and accuracy of crop phenotyping. For instance, Li et al. (https://doi.org/10.3389/fpls.2024.1376915) proposed a residual network approach based on hyperspectral imaging, enabling rapid identification of corn varieties while adapting to varying growth conditions. This method not only improves prediction accuracy but also demonstrates the potential of hyperspectral data in agriculture. Additionally, the integration of RGB imaging with environmental variables broadens the scope of crop monitoring, driving the adoption of multimodal data fusion in agricultural applications.Applications of Robotics and Automation in AgricultureAgricultural automation is transforming traditional farming practices. Guo et al. (https://doi.org/10.3389/fpls.2024.1377269) developed an autonomous navigation system for a greenhouse electric crawler tractor based on LiDAR, demonstrating its ability to navigate complex environments accurately. This system combines high-precision sensors with AI algorithms, reducing dependence on manual operation and significantly improving operational efficiency. Furthermore, solutions that integrate ground-based robots with drones have been applied to canopy imaging, weed detection, and disease monitoring, opening new avenues for smart farming.Plant Disease Detection and ManagementPlant disease detection remains a critical area of agricultural research. Zhou et al. (https://doi.org/10.3389/fpls.2024.1342123) developed an improved ShuffleNetV2 model for rapid identification of field crop leaf diseases. This lightweight model maintains high accuracy while reducing computational requirements, making it well-suited for deployment in resource-constrained agricultural environments. Additionally, Ye et al. (https://doi.org/ 10.3389/fpls.2024.1373104) proposed enhancements to the YOLOv7 model for large-scale tea leaf disease detection in complex environments. The dual-level routing dynamic sparse attention mechanism employed significantly improves detection accuracy, offering robust support for precision agriculture.Predicting Plant Growth and Pruning BehaviorUsing machine learning to predict plant growth and pruning behavior provides new tools for agricultural decision-making. Shu et al. (https://doi.org/10.3389/fpls.2024.1297390) employed machine learning algorithms to predict the resprouting of Platanus × hispanica after pruning. This study not only reveals the relationship between pruning and plant growth but also offers practical guidance for forestry and horticulture. Moreover, multimodal modeling that integrates