Unmanned Aerial Vehicles in Smart Agriculture

Chili recognition is one of the critical technologies for robots to pick chilies. The robots need locate the fruit. Furthermore, chilies are always planted intensively and their fruits are always clustered. It is a challenge to recognize and locate the chilies that are blocked by branches and leaves, or other chilies. However, little is known about the recognition algorithms considering this situation. Failure to solve this problem will mean that the robot cannot accurately locate and collect chilies, which may even damage the picking robot’s mechanical arm and end effector. Additionally, most of the existing ground target recognition algorithms are relatively complex, and there are many problems, such as numerous parameters and calculations. Many of the existing models have high requirements for hardware and poor portability. It is very difficult to perform these algorithms if the picking robots have limited computing and battery power. In view of these practical issues, we propose a target recognition-location scheme GNPD-YOLOv5s based on improved YOLOv5s in order to automatically identify the occluded and non-occluded chilies. Firstly, the lightweight optimization for Ghost module is introduced into our scheme. Secondly, pruning and distilling the model is designed to further reduce the number of parameters. Finally, the experimental data show that compared with the YOLOv5s model, the floating point operation number of the GNPD-YOLOv5s scheme is reduced by 40.9%, the model size is reduced by 46.6%, and the reasoning speed is accelerated from 29 ms/frame to 14 ms/frame. At the same time, the Mean Accuracy Precision (MAP) is reduced by 1.3%. Our model implements a lightweight network model and target recognition in the dense environment at a small cost. In our locating experiments, the maximum depth locating chili error is 1.84 mm, which meets the needs of a chili picking robot for chili recognition. Full article

Agricultural droughts cause a great reduction in winter wheat productivity; therefore, timely and precise irrigation recommendations are needed to alleviate the impact. This study aims to assess drought stress in winter wheat with the use of an unmanned aerial system (UAS) with multispectral and thermal sensors. High-resolution Water Deficit Index (WDI) maps were derived to assess crop drought stress and evaluate winter wheat actual evapotranspiration rate (ET_a). However, the estimation of WDI needs to be improved by using more appropriate vegetation indices as a proximate of the fraction of vegetation cover. The experiments involved six irrigation levels of winter wheat in the harvest years 2019 and 2020 at Luancheng, North China Plain on seasonal and diurnal timescales. Additionally, WDI derived from several vegetation indices (VIs) were compared: near-infrared-, red edge-, and RGB-based. The WDIs derived from different VIs were highly correlated with each other and had similar performances. The WDI had a consistently high correlation to stomatal conductance during the whole season (R² between 0.63–0.99) and the correlation was the highest in the middle of the growing season. On the contrary, the correlation between WDI and leaf water potential increased as the season progressed with R² up to 0.99. Additionally, WDI and ET_a had a strong connection to soil water status with R² up to 0.93 to the fraction of transpirable soil water and 0.94 to the soil water change at 2 m depth at the hourly rate. The results indicated that WDI derived from multispectral and thermal sensors was a reliable factor in assessing the water status of the crop for irrigation scheduling. Full article

Recent advances in unmanned aerial vehicles (UAV), mini and mobile sensors, and GeoAI (a blend of geospatial and artificial intelligence (AI) research) are the main highlights among agricultural innovations to improve crop productivity and thus secure vulnerable food systems. This study investigated the versatility of UAV-borne multisensory data fusion within a framework of multi-task deep learning for high-throughput phenotyping in maize. UAVs equipped with a set of miniaturized sensors including hyperspectral, thermal, and LiDAR were collected in an experimental corn field in Urbana, IL, USA during the growing season. A full suite of eight phenotypes was in situ measured at the end of the season for ground truth data, specifically, dry stalk biomass, cob biomass, dry grain yield, harvest index, grain nitrogen utilization efficiency (Grain NutE), grain nitrogen content, total plant nitrogen content, and grain density. After being funneled through a series of radiometric calibrations and geo-corrections, the aerial data were analytically processed in three primary approaches. First, an extended version normalized difference spectral index (NDSI) served as a simple arithmetic combination of different data modalities to explore the correlation degree with maize phenotypes. The extended NDSI analysis revealed the NIR spectra (750–1000 nm) alone in a strong relation with all of eight maize traits. Second, a fusion of vegetation indices, structural indices, and thermal index selectively handcrafted from each data modality was fed to classical machine learning regressors, Support Vector Machine (SVM) and Random Forest (RF). The prediction performance varied from phenotype to phenotype, ranging from R² = 0.34 for grain density up to R² = 0.85 for both grain nitrogen content and total plant nitrogen content. Further, a fusion of hyperspectral and LiDAR data completely exceeded limitations of single data modality, especially addressing the vegetation saturation effect occurring in optical remote sensing. Third, a multi-task deep convolutional neural network (CNN) was customized to take a raw imagery data fusion of hyperspectral, thermal, and LiDAR for multi-predictions of maize traits at a time. The multi-task deep learning performed predictions comparably, if not better in some traits, with the mono-task deep learning and machine learning regressors. Data augmentation used for the deep learning models boosted the prediction accuracy, which helps to alleviate the intrinsic limitation of a small sample size and unbalanced sample classes in remote sensing research. Theoretical and practical implications to plant breeders and crop growers were also made explicit during discussions in the studies. Full article

The aim of this work is to establish a method for real-time calculating droplet deposition distribution of a six-rotor plant protection unmanned aerial vehicle (UAV). The numerical simulation of the airflow field was carried out using computational fluid dynamics (CFD). The airflow field distribution was obtained under seven flight speeds, six flight heights, and seven crosswind speeds. The relative error verified the accuracy of the numerical model within 12% between the spatial point wind speed test and the simulated value. The numerical simulation results showed that with the improvement of the UAV flight speed and the crosswind, the relative airflow produces a vortex in the downwash wind field below the UAV and reduces the stability of the downwash wind field. The discrete droplet phase was introduced in the flow field. The ground regions were divided using a small grid of 0.5 m × 0.5 m, and statistical calculations of droplet deposition rates within each grid yielded the distribution of droplets under 294 different parameter combinations. The statistical results show that the relative airflow and crosswind caused droplet convolution, and droplet drift was increased. In the actual operation of the UAV, the flight speed should be well controlled under the condition of low environmental wind to reduce the droplet drift rate and improve the utilization rate of pesticides. Based on the distribution under 294 different parameter combinations, one droplet deposition prediction method was established using inverse distance weighting (IDW). The proposed method lays a foundation for the cumulative calculation of droplet deposition distribution during continuous operation of plant protection UAV. It provides a basis for objectively evaluating the operational quality of plant protection UAVs and optimizing the setting of operation parameters. Full article

Road detection is a crucial part of the autonomous driving system, and semantic segmentation is used as the default method for this kind of task. However, the descriptive categories of agroforestry are not directly definable and constrain the semantic segmentation-based method for road detection. This paper proposes a novel road detection approach to overcome the problem mentioned above. Specifically, a novel two-stage method for road detection in an agroforestry environment, namely ARDformer. First, a transformer-based hierarchical feature aggregation network is used for semantic segmentation. After the segmentation network generates the scene mask, the edge extraction algorithm extracts the trail’s edge. It then calculates the periphery of the trail to surround the area where the trail and grass are located. The proposed method is tested on the public agroforestry dataset, and experimental results show that the intersection over union is approximately 0.82, which significantly outperforms the baseline. Moreover, ARDformer is also effective in a real agroforestry environment. Full article