A method and system for controlling the spraying of a vineyard unmanned vehicle

Abstract

The application relates to a vineyard unmanned vehicle up-and-down pesticide spraying control method and system. The control system comprises a control unit, a moving unit, an acquisition unit and a spraying unit. The control method comprises the following steps: acquiring a regional terrain image obtained by the acquisition unit collecting a target region; recognizing regional terrain data in the regional terrain image, determining an optimal driving path according to the regional terrain data, and controlling the unmanned vehicle to drive; in the driving process, acquiring weather conditions of the target region and real-time grape plant images collected by the acquisition unit in the driving process; determining a target pesticide spraying control instruction according to the weather conditions and / or the grape plant images, wherein the target pesticide spraying control instruction comprises a leaf back spraying instruction and a leaf surface spraying instruction; and controlling the spraying unit to spray pesticide liquid according to the target pesticide spraying control instruction. The driving path is automatically planned, the pesticide spraying mode is determined by multiple factors such as weather and diseases, the pesticide spraying efficiency and precision are improved, the pesticide use effect is further improved, and pesticide pollution and spraying waste are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of unmanned vehicle pesticide spraying, and particularly relates to a method and system for controlling up-and-down pesticide spraying of an unmanned vehicle in a vineyard. BACKGROUND

[0002] There are many grape varieties, which are widely used for food and wine making and have good economic value. In different seasons of grape planting, different pesticides need to be sprayed on grape trees to prevent pests from rampant invasion. When spraying pesticides on grapes, there are generally two methods: foliage spraying and back spraying. Foliage spraying is to directly spray pesticides on grape leaves, while back spraying is to spray pesticides on the back of grape leaves. Foliage spraying can quickly act on the disease and pest area of the leaf surface, and has a rapid killing effect, but foliage spraying leaves pesticide residues on the leaf surface, which is easy to be washed away by rain. Back spraying is relatively more time-consuming and laborious than foliage spraying, but since the stomata on the back of grape leaves are more dense, the pesticide can penetrate the leaf surface more quickly and uniformly, and the effect is more significant. At the same time, back spraying does not leave pesticide stains on the leaf surface and does not affect plant growth and fruit quality. Therefore, when spraying pesticides on grapes, foliage spraying should be selected for treatment of existing diseases and pests, and back spraying should be selected for prevention of diseases and pests.

[0003] However, current grape agricultural spraying only changes the spraying method according to the size of the planting area, and uses manual spraying or mechanical spraying to directly complete pesticide spraying on grape leaves. Both of these two methods use direct spraying, and cannot distinguish between foliage and back spraying.

[0004] At the same time, in areas with small planting areas, manual back spraying with small pesticide spraying machines is used to spray pesticides on grape trees. Manual pesticide spraying equipment cannot meet the daily use of large-scale grape fields, and the process is relatively complicated. For large areas of grape planting, large pesticide spraying machines are used to spray pesticides on grape trees. The current pesticide spraying machines, including air-blast pesticide spraying machines and negative ion pesticide spraying machines, cannot be well adapted to grape planting, and may even damage grapes and cause yield reduction when turning in the field. SUMMARY

[0005] To solve or partially solve the problems in the related art, the present application provides a method and system for controlling up-and-down pesticide spraying of an unmanned vehicle in a vineyard. The image collected by the unmanned vehicle is processed by an algorithm to determine whether there are diseases and pests, and the spraying method is selected according to the weather conditions and the disease and pest situation, and the spray head is adjusted to complete pesticide spraying.

[0006] The first aspect of the present application provides a method for controlling up-and-down pesticide spraying of an unmanned vehicle in a vineyard, comprising the following steps:

[0007] First, a regional terrain image obtained by a target area acquisition unit is acquired;

[0008] Secondly, the regional terrain data in the regional terrain image is identified, the optimal driving path of the unmanned vehicle is determined according to the regional terrain data, and the unmanned vehicle is controlled to drive according to the optimal driving path;

[0009] Thirdly, during the driving of the unmanned vehicle according to the optimal driving path, the weather condition corresponding to the target region sent by the weather station and the real-time grape plant image collected by the collecting unit in the driving process are obtained;

[0010] Fourthly, the target pesticide spraying control instruction is determined according to the weather condition and / or the grape plant image, and the target pesticide spraying control instruction includes the leaf back spraying instruction and the leaf surface spraying instruction;

[0011] The target pesticide spraying control instruction is determined according to the weather condition and / or the grape plant image, and the target pesticide spraying control instruction includes:

[0012] When there is rain in the target period of the target region, the leaf back spraying instruction is output, and the target period is 24 to 48 hours in the future;

[0013] When there is no rain in the target period of the target region, the grape plant image is processed and it is judged whether there is a pest;

[0014] When it is detected that there is a pest in the target region, the leaf surface spraying instruction is output, and when it is detected that there is no pest in the target region, the leaf back spraying instruction is output.

[0015] Finally, the spraying unit is controlled to spray pesticide liquid according to the target pesticide spraying control instruction.

[0016] The grape plant image is processed and it is judged whether there is a pest, and the grape plant image is processed and it is judged whether there is a pest.

[0017] The grape plant image is processed and it is judged whether there is a pest, and the grape plant image is processed and it is judged whether there is a pest.

[0018] The grape plant image is processed and it is judged whether there is a pest, and the grape plant image is processed and it is judged whether there is a pest.

[0019] The grape plant image is processed and it is judged whether there is a pest, and the grape plant image is processed and it is judged whether there is a pest.

[0020] The grape plant image is processed and it is judged whether there is a pest, and the grape plant image is processed and it is judged whether there is a pest.

[0021] The optimal driving path of the unmanned vehicle is realized by a two-dimensional full coverage path planning algorithm, and is realized based on a comprehensive motion function and a steering confidence function. When trapped in a dead zone, an A-star algorithm is used to calculate the generation value of each uncovered node and the current node to determine the next node, so as to escape from the dead zone.

[0022] The spraying unit is provided with a spray head mechanism, and the spray head mechanism comprises a spray head and an adjusting structure. The spraying unit sprays pesticide liquid according to the target pesticide spraying control instruction, which comprises the following steps:

[0023] The spraying unit receives the leaf back spraying instruction, controls the adjusting structure to swing the spray head downward, and controls the spray head to spray pesticide liquid from the lower side of the leaf back after reaching the spraying position;

[0024] The spraying unit receives the leaf surface spraying instruction, controls the adjusting structure to swing the spray head upward, and controls the spray head to spray pesticide liquid from the upper side of the leaf surface after reaching the spraying position.

[0025] The second aspect of the application provides a grape vineyard unmanned vehicle up-and-down pesticide spraying control system, which is used for the grape vineyard unmanned vehicle up-and-down pesticide spraying control method provided in the first aspect, and comprises:

[0026] The moving unit comprises an unmanned vehicle, and the unmanned vehicle comprises a vehicle frame, vehicle wheels and a driving unit. The vehicle wheels are arranged on the front and rear sides of the vehicle frame, and the driving unit drives the vehicle wheels to rotate and controls the steering of the vehicle wheels.

[0027] The collecting unit comprises a visual image collecting structure and a vehicle speed collecting structure. The visual image collecting structure is used for collecting grape plant images and regional terrain images of a target region, and the vehicle speed collecting structure is used for detecting the driving speed of the unmanned vehicle.

[0028] The control unit comprises:

[0029] The central processing unit is used for acquiring the regional terrain images of the target region collected by the collecting unit, identifying regional terrain data in the regional terrain images, determining an optimal driving path of the unmanned vehicle according to the regional terrain data, acquiring weather conditions of the target region sent by a weather station and real-time grape plant images collected by the collecting unit in the driving process of the unmanned vehicle according to the optimal driving path, and determining a target pesticide spraying control instruction according to the weather conditions and / or the grape plant images.

[0030] The controller is used for controlling the unmanned vehicle to drive according to the optimal driving path and controlling the spraying unit to spray pesticide liquid according to the target pesticide spraying control instruction.

[0031] The spraying unit comprises a pesticide liquid tank installed on the unmanned vehicle, and at least two groups of spray head mechanisms are installed on the pesticide liquid tank. The spray head mechanism comprises a spray head and an adjusting structure, the adjusting structure comprises a support seat and a rocker arm, and the spray head is installed at the top end of the rocker arm.

[0032] Further, the support seat is fixed on the upper end of the liquid medicine tank, and comprises two groups of support plates fixed with each other, and a group of baffles is symmetrically arranged on the inner side of the support plate and is fixed through the baffle support arranged on the support seat, an actuator is installed on the baffle support and connected with the baffle, and is used for controlling the angle of the baffle; a rocker arm is arranged above the baffle and is rotatably connected with the support seat through a connecting shaft.

[0033] Further, the unmanned vehicle is provided with two front wheels and two rear wheels arranged horizontally, and a side wheel is arranged on the outer side of the rear wheel and is higher than the rear wheel.

[0034] Further, the swing angle of the rocker arm is between 45°-135°, when the liquid medicine is sprayed on the leaf face, the rocker arm swings upward, the swing angle is between 45°-90°, when the liquid medicine is sprayed on the back of the leaf, the rocker arm swings downward, and the swing angle is between 90°-135°, so that the liquid medicine is sprayed to the target area.

[0035] The technical scheme provided in the application can have the following beneficial effects:

[0036] The application provides a grape vine unmanned vehicle up-and-down spraying control method and system, which automatically plans a driving path and completes grape pesticide spraying, improves labor productivity, reduces labor cost, determines the pesticide spraying mode through multiple factors such as weather and disease while realizing automatic pesticide spraying, improves pesticide spraying efficiency and accuracy, further improves the use effect of the pesticide, and reduces pesticide pollution and spraying waste.

[0037] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the application. BRIEF DESCRIPTION OF DRAWINGS

[0038] The above and other objects, features and advantages of the present application will become more apparent from the following detailed description when taken in conjunction with the accompanying drawings in which like reference characters refer to like parts throughout the several views, and in which the exemplary embodiments of the present application are shown.

[0039] Figure 1 is a flowchart of the grape vine unmanned vehicle up-and-down spraying control method shown in the first embodiment of the application;

[0040] Figure 2 is a structural schematic diagram of the grape vine unmanned vehicle up-and-down spraying control system device shown in the second embodiment of the application;

[0041] Figure 3 is a partial enlarged schematic diagram of the grape vine unmanned vehicle up-and-down spraying control system shown in the second embodiment of the application;

[0042] REFERENCE SIGNS:

[0043] In the figure, 1 - front wheel, 2 - environment vision collector, 3 - frame, 4 - spray head mechanism, 4-1 - support seat, 4-2 - baffle, 4-3 - rocker arm, 4-4 - spray head, 4-5 - support plate, 4-6 - connecting shaft, 5 - liquid medicine tank, 6 - control unit, 7 - rotating door, 8 - rear wheel, 9 - side wheel. DETAILED DESCRIPTION

[0044] Embodiments of the present application will be described in more detail with reference to the drawings. Although embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present application is more thoroughly and completely conveyed to those skilled in the art, and the scope of the present application is fully conveyed to those skilled in the art.

[0045] It should be understood that although the terms "first", "second", "third" and the like can be used herein to describe various information, these information should not be limited by these terms. These terms are only used to distinguish the same type of information from each other. For example, the first information can also be referred to as the second information, and similarly, the second information can also be referred to as the first information, without departing from the scope of the present application. Therefore, the features defined as "first", "second" can explicitly or implicitly include one or more of the features. In the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0046] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application.

[0047] Unless otherwise explicitly specified and limited, the terms "mounting", "connection", "connection", "fixing" and the like should be understood broadly, for example, it can be fixed connection, or detachable connection, or integral; it can be mechanical connection, or electrical connection; it can be directly connected, or indirectly connected through intermediate medium; it can be the internal communication of two elements or the interaction relationship between two elements. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0048] The technical solutions of the embodiments of the present application will be described in detail below with reference to the drawings.

[0049] Embodiment one,

[0050] As Figure 1 shown in a kind of vineyard unmanned vehicle up and down spray control method, applied to the vineyard unmanned vehicle up and down spray control system shown in example two, vineyard unmanned vehicle up and down spray control method includes the following steps:

[0051] S101, the region topographic image obtained by the acquisition unit acquisition target area is acquired.

[0052] S102, the region topographic image in region topographic data is identified, and the optimal driving path of unmanned vehicle is determined according to region topographic data, and the unmanned vehicle is controlled to travel according to the optimal driving path.

[0053] The region topographic data in the region topographic image can be realized using a variety of existing software, such as the image processing algorithm in OpenCV, such as edge detection, feature extraction, image segmentation, etc., to identify and extract the terrain data in the terrain image. Remote sensing image analysis software can also be used: some remote sensing image analysis software, such as ENVI, ERDASIMAGINE, provides the tools for feature classification and analysis. These software usually have the functions of image classification and segmentation, which can be used to identify and extract terrain data.

[0054] The optimal driving path of unmanned vehicle is realized by two-dimensional full coverage path planning algorithm, based on comprehensive motion function and steering confidence function. The comprehensive motion function includes position function and steering confidence function, and the position function is used to distinguish the uncovered grid, the covered grid and the obstacle; while the steering confidence function is used to guide the travel to the uncovered area, while controlling the steering angle, so that the path tends to be flat.

[0055] Position function: the purpose of distinguishing different grids can be achieved by assigning values to the grid.

[0056] Let the position function be X,

[0057] Where: i, j represents the grid in the i-th row, the j-th column of grid map.

[0058] Steering confidence function: to control steering, make it tend to uncovered area.

[0059] Set the steering confidence function C,

[0060] Where: The steering angle.

[0061] Comprehensive consideration of position function and steering confidence function, redefine a comprehensive motion function as the basis for selecting the next motion node, which is defined as: Y K =X K +aC KK = 1,2, …, i

[0062] wherein: a∈[0,1], is a weighted coefficient; the value of i depends on the number of uncovered points connected to the current node. Select the value Y K The largest node is used as the next moving direction.

[0063] When the current node is adjacent to the node without uncovered nodes, all of which are covered nodes or obstacles, or the boundary, it is called into a dead zone. When falling into a dead zone, the A-star algorithm is used to calculate the generation value of each uncovered node and the current node to determine the next node.

[0064] wherein, the algorithm steps for escaping the dead zone are:

[0065] ① Search for uncovered nodes according to the shortest distance principle.

[0066] ② Calculate the distance from the current node to the uncovered node, and obtain the minimum distance.

[0067] ③ Whether the minimum distance is equal to the shortest distance. If so, select the corresponding node as the nearest uncovered node, and the unmanned vehicle drives to this node to escape the dead zone. Otherwise, execute the next step.

[0068] ④ The shortest distance is increased by one grid, and the above steps are executed again.

[0069] In the above steps, the minimum distance refers to the minimum value in all ring searches; the shortest distance refers to the value in each ring search, which will increase according to the specific situation of the search.

[0070] S103, in the process of driving the unmanned vehicle according to the optimal driving path, obtaining the weather conditions corresponding to the target area sent by the weather station and the real-time grape plant image collected by the collecting unit in the driving process.

[0071] The absorption time of grape spraying is affected by many factors, including drug type, drug concentration, leaf surface conditions, etc. The absorption speed of the drug on the leaf surface is fast, usually within a few hours; while it takes longer time to enter the plant and conduct to the target tissue. For the case of leaf spraying, choose to spray in the weather without rain, and there is no rain in the next few days. If it rains immediately after spraying, the drug may be washed away, resulting in reduced efficacy.

[0072] Back spraying can reduce the risk of pesticide being washed away by rain to some extent. Because when back spraying, the pesticide is directly applied to the back of the leaf, it has less contact with the external environment, including rain. But back spraying may cause slower absorption of the pesticide, because the pesticide needs to penetrate the leaf surface to reach the inside of the leaf. Therefore, for back spraying, it is still recommended to avoid spraying when there is a rain forecast to ensure effective absorption and utilization of the pesticide.

[0073] Therefore, it is necessary to avoid rain within a few hours after foliar or back spraying, to give the pesticide enough time to interact with plant tissues and improve the absorption and utilization of the pesticide.

[0074] The real-time grape plant image is used to identify whether there is a pest or disease in the current target area. Grape pests and diseases include various diseases such as grape anthracnose, which can cause a large number of fruits to rot. The initial disease appears as a circular, slightly concave, and light brown spot on the surface of the diseased fruit. The surface of the disease spot is densely arranged with black small dot particles (conidia discs). When the weather is humid, a bright red sticky substance (spore mass) can be discharged from the conidia discs. The diseased fruit gradually dries up and finally becomes a mummy. Grape black scab disease causes light purple spots on the surface of young fruits. The spots are nearly circular and gradually expand to 3-7 mm in diameter. The edge of the disease spot is purple-brown, and the center gradually changes to gray-white, with a slight depression and black spots. The diseased young leaves show sparse and uneven brown circular spots, with a diameter of 1-4 mm. The edge of the disease spot is dark brown, and the center is light brown. The diseased leaves show brown circular spots, with a gray-white center, and later the disease spot is perforated and cracked in a star shape, with a purple-brown halo around the edge. After the diseased young shoots are infected, long oval or irregular brown stripes appear. The disease spot is gray-brown, with a dark brown or purple edge. The new shoots are severely distorted and dried up.

[0075] S104、According to the weather condition and / or the grape plant image, determine the target spraying control instruction, the target spraying control instruction including a back spraying instruction and a foliar spraying instruction;

[0076] According to the weather condition and / or the grape plant image, determining the target spraying control instruction includes:

[0077] S1041, obtaining the weather condition, when there is rain in the target period of the target area, outputting the back spraying instruction, the target period being 24 to 48 hours in the future;

[0078] It should be noted that when there is rain within 24 hours in the future, both back spraying and foliar spraying will be affected, and no spraying operation will be performed to avoid wasting pesticides.

[0079] S1042, when there is no rain in the target period of the target area, processing the grape plant image and determining whether there is a pest or disease;

[0080] S10421, process the grape plant image based on the ASCAE feature extraction algorithm, extract representative color features and generate a feature image;

[0081] Representative color features are extracted from the original RGB image using color attenuation technology. This process helps to reduce redundant information and highlight the color differences of pest and disease features.

[0082] ASCAE feature extraction algorithm flow:

[0083] Input: dataset X = {x1, x2, …, xN};

[0084] Output: F layer features B = {b1, b2, …, bN};

[0085] Initialization: initialize the grid weight with random numbers following the Gaussian distribution G(0, 0.12), and set the number of neurons in the fully connected layer to k;

[0086] Iteration process: minimize the target loss function J(θ; X; λ) to train the asymmetric convolution autoencoder;

[0087] Iteratively update the network parameter values through the backpropagation algorithm;

[0088] Stopping condition: target loss no longer decreases or reaches maximum iteration number;

[0089] Output features:

[0090] S10422, convert the feature image to an HSV image, and cluster the HSV image based on the K-means method to obtain the saliency image of the grape plant image;

[0091] Convert the feature image to an HSV color space and further extract color features similar to human visual perception. Use the K-means method to cluster the HSV image. K-means algorithm is an unsupervised machine learning algorithm that clusters data points into K different groups. Use K-means algorithm to cluster pixels in HSV image into several color approximation regions.

[0092] Clustering steps:

[0093] Step 1: Calculate the distance and area between regions;

[0094] Step 2: Calculate the RGB color space distance;

[0095] Step 3: Calculate the saliency value S,

[0096] Where: r k is the current region; ri for any other region; Ds is the spatial distance between the centers of two regions; A(r i ) is the area of any other region; D r is the RGB color distance between two regions; and ε is in the range of 0.3-0.5.

[0097] Through the above steps, the saliency map of the image can be clustered, and whether there is a pest and disease can be judged by comparing the collected images.

[0098] S10423, segment the saliency image based on the binary image, and extract the pest and disease target in the saliency image;

[0099] The method for segmenting the pest and disease target from the binary image mainly includes the following steps:

[0100] First step: preprocessing and thresholding.

[0101] Second step: extracting initial regions.

[0102] Third step: GrabCut algorithm iterative segmentation.

[0103] Fourth step: iteration number control. The GrabCut algorithm is iterated for 4 times to segment the initial region. In each iteration, the algorithm updates the segmentation boundary according to the characteristics and prior knowledge of the image until the predetermined number of iterations is reached or the convergence condition is met.

[0104] Fifth step: post-processing and extraction. The segmented pest and disease target is post-processed, such as removing noise, smoothing the boundary, etc., to improve the accuracy and integrity of the target. According to the processed results, the pests are extracted, and necessary feature extraction and classification recognition are performed.

[0105] S10424, compare the pest and disease target with the preset pest and disease expert library to determine whether there is a pest and disease.

[0106] Store various grape pest and disease characteristics in the preset pest and disease expert library. Suppose that the preset pest and disease expert library stores n characteristics of grape pest and disease. If there are m grape pest and disease characteristics in the pest and disease target.

[0107] If m is greater than or equal to a certain proportion of n (such as more than 50%), it can be preliminarily judged that there is a pest and disease. In this case, the grape pest and disease characteristics existing in the target match the characteristics in the expert library, and it can be considered that there may be a pest and disease.

[0108] If m is less than a certain proportion of n (such as less than 50%), it can be preliminarily judged that there is no pest and disease. In this case, the grape pest and disease characteristics existing in the target match the characteristics in the expert library less, and it can be considered that there may be no pest and disease.

[0109] The actual judgment criteria are determined based on specific circumstances, including factors such as the selection of pest and disease characteristics, the updating and accuracy of the expert database.

[0110] S1043. When pests and diseases are detected in the target area, output a foliar spraying command; when pests and diseases are not detected in the target area, output a leaf underside spraying command.

[0111] S105. The spraying unit controls the spraying of pesticide solution according to the target spraying control command, including:

[0112] S1051. The spraying unit receives the spraying command on the back of the leaf, controls the adjustment structure to swing the nozzle 4-4 downward, and controls the nozzle 4-4 to spray the liquid from the back of the leaf after reaching the spraying position.

[0113] S1052. The spraying unit receives the leaf spraying command, controls the adjustment structure to make the nozzle 4-4 swing upward, and controls the nozzle 4-4 to spray the liquid from above the leaf after reaching the spraying position.

[0114] Example 2

[0115] like Figure 2 The control system for unmanned vineyard vehicles spraying pesticides vertically and horizontally, as shown in Embodiment 1, is used in conjunction with the control method for unmanned vineyard vehicles spraying pesticides vertically and horizontally as described in Embodiment 1, and includes:

[0116] The mobile unit includes an unmanned vehicle, which includes a frame 3, wheels, and a drive unit. The wheels are located on the front and rear sides of the frame 3, and the drive unit drives the wheels and controls their steering. The unmanned vehicle has six wheels, including two horizontally arranged front wheels 1 and two rear wheels 8, and side wheels 9 located on the outer sides of the two rear wheels 8.

[0117] The front wheel 1 and the two rear wheels 8 are on the same plane to adapt to relatively flat terrain. The two side wheels 9 are 40cm higher than the two middle rear wheels 8 (the height of two rows of grapes is approximately 40cm). When the road surface is uneven, the vehicle is balanced by the two front wheels 1 and the two rear side wheels 9. The power drive consists of two independent and identical power drive units. Each power drive unit consists of a DC brushless motor, a worm gear reducer, and a synchronous belt drive connected to the wheels, enabling the wheels to rotate at a specified speed. The two power drive units drive the left and right wheels respectively, and the forward direction of the vehicle is adjusted in real time through the speed difference. The steering power consists of a DC brushless motor, a brake, and a chain drive system. All four body support rods are equipped with sprockets of the same size, connected by a chain, which can achieve synchronous rotation of the wheels within a certain range to achieve steering.

[0118] The data acquisition unit includes a visual image acquisition structure and a vehicle speed acquisition structure. The visual image acquisition structure is used to acquire images of grapevines and terrain in the target area; in this embodiment, it is specifically an environmental visual acquisition device 2. The vehicle speed acquisition structure is used to detect the driving speed of the unmanned vehicle, constantly monitoring its forward speed and controlling its speed accordingly. The acquired data is then transmitted to the central processing unit.

[0119] Control unit 6, mounted on the frame 3, prevents liquid from dripping and causing corrosion by closing the rotating door 7, including:

[0120] The central processing unit is used to acquire regional terrain images of the target area obtained by the acquisition unit, identify regional terrain data in the regional terrain images, determine the optimal driving path of the unmanned vehicle based on the regional terrain data, acquire the weather conditions corresponding to the target area sent by the weather station and the real-time grape vine images acquired by the acquisition unit during the driving process, and determine the target spraying control command based on the weather conditions and / or grape vine images.

[0121] The controller is used to control the unmanned vehicle to travel along the optimal driving path and to control the spraying unit to spray the pesticide according to the target spraying control command;

[0122] The spraying unit includes a liquid tank 5 installed on the unmanned vehicle. The liquid tank 5 is equipped with a pump and a flow meter. Two sets of nozzle mechanisms 4 are installed on both sides of the liquid tank 5. The nozzle mechanisms 4 are connected to the pump through a liquid pipe.

[0123] like Figure 3 As shown, the nozzle mechanism 4 includes a nozzle 4-4 and an adjustment structure. The adjustment structure includes a support base 4-1 and a rocker arm 4-3. The nozzle 4-4 is mounted on the top of the rocker arm 4-3. The support base 4-1 is fixed to the upper end of the medicine tank 5 and includes two sets of mutually fixed support plates 4-5. A set of baffles 4-2 are symmetrically arranged on the inner side of the support plates 4-5. The angle between the baffles 4-2 and the support base 4-1 is 45°. The baffles 4-2 are fixed by baffle 4-2 brackets set on the support base 4-1. An actuator is installed on the baffle 4-2 bracket and connected to the baffle 4-2 to control the angle of the baffle 4-2. The rocker arm 4-3 is arranged above the baffle 4-2, and the bottom of the rocker arm 4-3 is rotatably connected to the support base 4-1 through a connecting shaft 4-6.

[0124] Because the rocker arm 4-3 can swing up and down, and the swing angle range can be flexibly adjusted, fixing the nozzle 4-4 to the rocker arm 4-3 allows for switching between two spraying modes. Therefore, the swing angle of the rocker arm 4-3 is set between 45° and 135°. By controlling the swing angle of the rocker arm 4-3, the spraying direction of the nozzle 4-4 can be switched.

[0125] When spraying from top to bottom: the controller controls the actuator to adjust the angle of the baffle 4-2, the baffle 4-2 rotates to make the rocker arm 4-3 swing downward around the connecting shaft 4-6, and the nozzle 4-4 sprays the liquid medicine from the bottom. In this mode, the swing angle of the rocker arm 4-3 is between 90°-135° to ensure that the liquid medicine is sprayed to the target area.

[0126] When spraying from top to bottom: the controller controls the actuator to adjust the angle of the baffle 4-2, the baffle 4-2 rotates to make the rocker arm 4-3 swing downward around the connecting shaft 4-6, and the nozzle 4-4 sprays the liquid medicine from the bottom. In this mode, the swing angle of the rocker arm 4-3 is between 90°-135° to ensure that the liquid medicine is sprayed to the target area.

[0127] Finally, it should be noted that relationships such as first and second, and the like, are merely used to distinguish one entity or action from another, without necessarily requiring or implying any such actual relationship or order between or among the entities or actions, and the like. Moreover, the terms "comprises", "comprising", or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus.

[0128] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e. may be located in one place, or may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

[0129] The embodiments of the present application have been described above, and the above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes are obvious to those skilled in the art without departing from the scope and spirit of the described embodiments. The choice of terms used herein is intended to best explain the principles, practical applications or improvements to the technology in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A method for controlling the up and down spraying of unmanned vehicles in vineyards, characterized in that, The application is applied to the up-and-down pesticide spraying control system of the vineyard unmanned vehicle, and the control system comprises a control unit and a moving unit, a collecting unit and a spraying unit connected with the control unit. The vineyard unmanned vehicle up-and-down pesticide spraying control method comprises the following steps: Obtain the regional terrain image obtained by the collecting unit collecting the target area; Identify the regional terrain data in the regional terrain image, determine the optimal driving path of the unmanned vehicle according to the regional terrain data, and control the unmanned vehicle to drive according to the optimal driving path; In the process of driving the unmanned vehicle according to the optimal driving path, obtain the weather condition of the target area sent by the weather station and the real-time grape plant image collected by the collecting unit in the driving process; According to the weather condition and / or the grape plant image, determine the target pesticide spraying control instruction, and the target pesticide spraying control instruction comprises a leaf back spraying instruction and a leaf surface spraying instruction; The target pesticide spraying control instruction is determined according to the weather condition and / or the grape plant image, which comprises: When there is rain in the target period of the target area, the leaf back spraying instruction is output, and the target period is 24 to 48 hours in the future; When there is no rain in the target period of the target area, process the grape plant image and judge whether there is disease and pest; When it is detected that there is disease and pest in the target area, the leaf surface spraying instruction is output, and when it is detected that there is no disease and pest in the target area, the leaf back spraying instruction is output; Control the spraying unit to spray pesticide liquid according to the target pesticide spraying control instruction, and a spray head mechanism is arranged on the spraying unit, the spray head mechanism comprises a spray head and an adjusting structure, and the spraying unit sprays pesticide liquid according to the target pesticide spraying control instruction, which comprises: The spraying unit receives the leaf back spraying instruction, controls the adjusting structure to swing the spray head downward, and controls the spray head to spray pesticide liquid from the lower side of the leaf back after reaching the spraying position; 2. The vineyard unmanned vehicle up-and-down spraying control method according to claim 1, characterized in that, The spraying unit receives the leaf surface spraying instruction, controls the adjusting structure to swing the spray head upward, and controls the spray head to spray pesticide liquid from the upper side of the leaf surface after reaching the spraying position. The processing of the grape plant image and the judgment of whether there is disease and pest comprise: Based on the ASCAE feature extraction algorithm, the grape plant image is processed, the representative color features are extracted, and a feature image is generated; The feature image is converted into an HSV image, and the HSV image is clustered based on the K-means method to obtain a saliency image of the grape plant image; The saliency image is segmented based on a binary image, and a disease and pest target in the saliency image is extracted; 3. The vineyard unmanned vehicle up-and-down spraying control method according to claim 1, characterized in that, The disease and pest target is compared with a preset disease and pest expert library to determine whether there is disease and pest.

4. A vineyard unmanned vehicle up-and-down pesticide spraying control system for the vineyard unmanned vehicle up-and-down pesticide spraying control method according to any one of claims 1 to 3, characterized in that, The optimal driving path of the unmanned vehicle is realized by a two-dimensional full coverage path planning algorithm, and is realized based on a comprehensive motion function and a steering confidence function. When it is trapped in a dead zone, an A-star algorithm is used to calculate the generation value of each uncovered node and the current node to determine the next node, so as to escape from the dead zone. It comprises: The mobile unit comprises an unmanned vehicle, which comprises a vehicle frame, vehicle wheels arranged on the front and rear sides of the vehicle frame, and a driving unit for driving the vehicle wheels to run and controlling the steering of the vehicle wheels. The acquisition unit comprises a visual image acquisition structure for acquiring grape plant images and regional terrain images of a target region, and a vehicle speed acquisition structure for detecting the running speed of the unmanned vehicle. The control unit comprises: A central processing unit for acquiring regional terrain images of a target region obtained by the acquisition unit, identifying regional terrain data in the regional terrain images, determining an optimal running path of the unmanned vehicle according to the regional terrain data, acquiring weather conditions of the target region sent by a weather station and real-time grape plant images acquired by the acquisition unit during the running of the unmanned vehicle, and determining a target pesticide spraying control instruction according to the weather conditions and / or the grape plant images. A controller for controlling the unmanned vehicle to run along the optimal running path and controlling the spraying unit to spray pesticide liquid according to the target pesticide spraying control instruction. The spraying unit comprises a pesticide liquid tank mounted on the unmanned vehicle, at least two groups of spraying head mechanisms mounted on the pesticide liquid tank, the spraying head mechanisms comprising spraying heads and adjusting structures, and the adjusting structures comprising support seats and rocker arms, wherein the spraying heads are mounted at the top ends of the rocker arms.

5. The vineyard unmanned vehicle up-and-down spraying control system of claim 4, wherein, The support seat is fixed to the upper end of the pesticide liquid tank and comprises two groups of support plates fixed to each other, a group of baffles symmetrically arranged on the inner sides of the support plates, and baffle supports arranged on the support seat and used for fixing the baffles, wherein actuators are mounted on the baffle supports and connected with the baffles to control the angles of the baffles, and the rocker arms are arranged above the baffles and rotatably connected with the support seat through connecting shafts.

6. The vineyard unmanned vehicle up-and-down spraying control system of claim 4, wherein, The unmanned vehicle is provided with two front wheels and two rear wheels arranged horizontally, and side wheels arranged on the outer sides of the rear wheels and higher than the rear wheels.

7. The vineyard unmanned vehicle up-and-down spraying control system of claim 4, wherein, The swinging angle of the rocker arm is between 45° and 135°, the rocker arm swings upward when spraying the leaf surface, the swinging angle is between 45° and 90°, the rocker arm swings downward when spraying the leaf back, and the swinging angle is between 90° and 135°, so that the pesticide liquid is sprayed to the target region.

Images