Plant protection unmanned aerial vehicles (UAVs) play a crucial role in agricultural aviation services. In recent years, plant protection UAVs, which improve the accuracy and eco-friendliness of agricultural techniques, have been used to overcome the shortcomings of traditional agricultural operations. First, this paper introduces the development scale, main types, and operation scenarios of China’s plant protection UAVs. Subsequently, the key technologies of plant protection UAVs, such as precision autonomous flight control, pesticide spraying, drift control, and spraying quality measurement technologies, are reviewed. Next, the emergent technologies of plant protection UAVs are studied and analyzed with a focus on better spray effects, calculation models of droplet drift, controllable droplet size atomization technology, droplet drift detection technology, and droplet deposition quality detection technology in the application of plant protection UAVs. Moreover, the technologies of plant protection UAV application are summarized and future research prospects are presented, offering ideas for follow-up research on the key technologies of plant protection UAVs and encouraging agricultural production management to move toward better efficiency, eco-friendliness, and accuracy.

Pine wilt disease (PWD) is a serious threat to pine forests. Combining unmanned aerial vehicle (UAV) images and deep learning (DL) techniques to identify infected pines is the most efficient method to determine the potential spread of PWD over a large area. In particular, image segmentation using DL obtains the detailed shape and size of infected pines to assess the disease’s degree of damage. However, the performance of such segmentation models has not been thoroughly studied. We used a fixed-wing UAV to collect images from a pine forest in Laoshan, Qingdao, China, and conducted a ground survey to collect samples of infected pines and construct prior knowledge to interpret the images. Then, training and test sets were annotated on selected images, and we obtained 2352 samples of infected pines annotated over different backgrounds. Finally, high-performance DL models (e.g., fully convolutional networks for semantic segmentation, DeepLabv3+, and PSPNet) were trained and evaluated. The results demonstrated that focal loss provided a higher accuracy and a finer boundary than Dice loss, with the average intersection over union (IoU) for all models increasing from 0.656 to 0.701. From the evaluated models, DeepLLabv3+ achieved the highest IoU and an F1 score of 0.720 and 0.832, respectively. Also, an atrous spatial pyramid pooling module encoded multiscale context information, and the encoder–decoder architecture recovered location/spatial information, being the best architecture for segmenting trees infected by the PWD. Furthermore, segmentation accuracy did not improve as the depth of the backbone network increased, and neither ResNet34 nor ResNet50 was the appropriate backbone for most segmentation models.

Extensive research has been conducted on plant protection unmanned aerial vehicle (UAV) chemical application technology in recent years owing to its importance as a means of pest and disease control. UAV spraying in orchards faces the drawback of drift risk and can be hazardous to non-targeted crops, humans, and the environment. A detailed and systematic analysis must be performed to determine the uniformity and drift risk of plant UAV sprays. In this study, a peach orchard is sprayed with a plant-protection UAV at three different flight velocities and we evaluate the combined pesticide deposition performance of the canopy, ground loss, downwind ground drift, and airborne drift. Additionally, the droplet size and coverage rate in the canopy are calculated by using water-sensitive paper. The results demonstrate that there is significant difference in the droplet size at flight velocities of 1–3 m/s. The droplet size in the lower canopy is slightly smaller than those in the middle and upper parts. Increasing the flight velocity helps the pesticide droplets to spread and penetrate the canopy. However, it also causes a non-uniform pesticide deposition, reduced effective coverage ratio and effective density ratio. Among the three pesticides used in the experiment, imidacloprid exhibits the best deposition efficiency. The deposition amount and normalized deposition amount in the canopy were the highest at a flight velocity of 2 m/s, accompanied by a lower ground loss under the canopy. The highest near-field ground drift is observed at a velocity of 1 m/s, and the far-field airborne drift is highest at 3 m/s. Lastly, this study provides a reference for the commercial application of plant-protection UAVs.