Automatic navigation (AN) is an essential component to ensure the safety of pesticide application in orchards, whereas precision variable-rate spraying (PVS) serves as an indispensable technology for reducing the application of pesticides and protecting the environment. At present, AN and PVS are not closely combined. In this case, a single three-dimension (3D) light detection and ranging (LIDAR) sensor is hereby adopted to sense the information of fruit trees around the robot and determine the region of interest (ROI). Moreover, two-dimensional (2D) processing is conducted over the point clouds within the ROI to obtain the center-of-mass coordinates of fruit trees, and determine the vertical distance of the robot to the center line of the fruit tree row (FTR) based on the FTR on both sides using the Random Sample Consensus (RANSAC) algorithm. Then, the robot is controlled to drive along the center line of the FTR. At the same time, the speed and position of the robot are determined by the encoder and inertial measurement unit (IMU), and the IMU corrects the information collected from the zoned canopy of the fruit trees. The results present a lateral deviation (LD) of less than 22 cm and a course deviation (CD) of less than 4.02° during AN. Compared with the traditional spraying (TS), the PVS applies 32.46%, 44.34% and 58.14% less pesticide application, air drift and ground loss, respectively. With the spraying effect guaranteed, the single 3D LIDAR, the encoder and IMU realize the AN and PVS of the robot, reduce the volume of pesticide application, ground loss and air drift, and effectively control the pollution caused by pesticides to the environment.

With the aim of solving the problems of high labor intensity, low operational efficiency, and poor deposition distribution uniformity in the mango canopy associated with traditional plant protection devices, an air-ground co-operation stereoscopic plant protection system consisting of an orchard caterpillar mist sprayer and a six-rotor plant protection UAV was designed to jointly undertake plant protection operations in mango orchards. We tested the spraying performance of the system on mango trees, compared with the single-machine operation, the air–ground co-operation system could significantly increase the droplet coverage on the upperside of mango leaves in each part of the canopy (a 14.7% increase for the mist sprayer and 12.9% for the UAV). This increased the active component deposition distribution uniformity in the mango canopy but could not significantly improve the deposition and coverage of droplets on the underside of leaves compared with the mist sprayer and plant protection UAV. Due to the characteristics of the mango canopy such as large leaf length and thickness and complex leaf inclination distribution, this led to poor deposition distribution uniformity of the two spray units, and the overall CV was over 150%. The pesticide active ingredients were almost uniformly distributed in the vertical direction when the application ratios (ground implements/plant protection drones) were 8/2 and 7/3, offering a promising protocol for reduced pesticide application in mango orchards. This study presents promising data that support the innovative integration of drones into crop protection programs for large canopy crops (e.g., mango) and provides guidance for the ACSPPS system in reduction and precision application research.

As an essential element, the effect of Phosphorus (P) on plant growth is very significant. In the early growth stage of maize, it has a high sensitivity to the deficiency of phosphorus. The main purpose of this paper is to monitor the maize status under two phosphorus levels in soil by a nondestructive testing method and identify different phosphorus treatments by spectral data. Here, the Analytical Spectral Devices (ASD) spectrometer was used to obtain canopy spectral data of 30 maize inbred lines in two P-level fields, whose reflectance differences were compared and the sensitive bands of P were discovered. Leaf Area Index (LAI) and yield under two P levels were quantitatively analyzed, and the responses of different varieties to P content in soil were observed. In addition, the correlations between 13 vegetation indexes and eight phenotypic parameters were compared under two P levels so as to find out the best vegetation index for maize characteristics estimation. A Back Propagation (BP) neural network was used to evaluate leaf area index and yield, and the corresponding prediction model was established. In order to classify different P levels of soil, the method of support vector machine (SVM) was applied. The results showed that the sensitive bands of P for maize canopy included 763 nm, 815 nm, and 900–1000 nm. P-stress had a significant effect on LAI and yield of most varieties, whose reduction rate reached 41% as a whole. In addition, it was found that the correlations between vegetation indexes and phenotypic parameters were weakened under low-P level. The regression coefficients of 0.75 and 0.5 for the prediction models of LAI and yield were found by combining the spectral data under two P levels. For the P-level identification in soil, the classification accuracy could reach above 86%. These abilities potentially allow for phenotypic parameters prediction of maize plants by spectral data and different phosphorus contents identification with unknown phosphorus fertilizer status.

Liquid spray pollination (LSP) is widely used in fruit tree pollination. However, the LSP parameters that affect the pollen activity are still unclear. In this study, three LSP parameters that mainly affect the pollen activity were studied: storage time of pollen suspensions, sprayer parameters and unmanned aerial vehicle sprayer (UAVS) downwash airflow. In addition, sprayer parameters include the recirculation device, pump type, spraying pressure, nozzle size and revolutions per minute (rpm) of the rotary atomizer (RA). The results showed that, with the exception of nozzle size and UAVS downwash airflow, the pollen activity was significantly influenced by LSP parameters. The mean pollen activity decreased by 20.20% when the pollen suspension was stored for 30 min compared to 0 min. The activity of pollen in the tank was dramatically reduced using the recirculation device. The mean pollen activity decreases as the pump production maximum pressure increases. The mean pollen activity decreased from 40.7% to 29.02% when the spraying pressure increased from 0.3 MPa to 2.5 MPa. Additionally, the mean pollen activity decreased from 44.25% to 14.14% as the rpm of RA increased from 3000 rpm to 14,000 rpm. Our study demonstrated that pollen activity would be ensured by appropriate LSP parameters. This study provides a reliable theoretical basis for optimizing and advancing pear LSP technology.

In turning, tool wear and cutting vibration are inevitable, which have great influence on surface quality. Analyzing the influence mechanism of tool wear and cutting vibration on surface quality is important to achieve the accurate prediction of surface roughness before machining and improve machining quality. In this paper, a turning vibration experiment is conducted to reveal that the diameter of shafts is an important factor affecting the vibration amplitude and frequency. In addition, based on machining parameters, tool wear and workpiece diameter, this empirical model, the response surface method and a support vector machine are used to model and predict surface roughness. The fitting accuracy, prediction accuracy and generalization performance of the proposed methods are compared in detail. The results show that the response surface modeling method has the highest fitting accuracy, but the exponential empirical modeling method has the highest prediction accuracy and best generalization performance. Additionally, the prediction results indicate that the surface roughness increases with the increase in tool wear and decreases with the increase in workpiece diameter.

To demonstrate the feasibility and improve the implementation of laser weeding, a static movable lift-adjustable closed fiber laser weeding device and system have been developed, followed by experiments and performance evaluations. Physical experiments were conducted on the energy required for laser cutting of weed stems, targeting four common larger weeds (Chenopodium album, Amaranthus spinosus, Setaria viridis, and Eleusine indica) in farmland and orchards. At the same irradiation distances, the energy required to cut the same type of large weed generally increases with increasing distances and stem diameters but decreases with increasing irradiation time. The variance of stems’ power density after irradiation was larger and the values were more dispersed for Chenopodium album and Setaria viridis weeds, and the values were relatively scattered, while the power density values of Amaranthus spinosus and Eleusine indica weeds were more concentrated. When the irradiation time was 10 s, the 3.892 W/mm² laser was sufficient to eliminate weeds and plants with the irradiation distances of 2 m. The 2.47 W/mm² laser was more effective, as it killed weeds within a distance of 1 m in less than 1 s. This work demonstrates the feasibility of the laser weeding device and system that can completely cut off the stems of large weeds, and this technology has the potential to promote sustainable agriculture.