A method, system and equipment for controlling autonomous walking operation of open-field vegetables

By using the cameras on the main support frame and the side support frame in the open-field vegetable fields for image processing and identification, navigation path planning and spray control are realized, and the problems of poor target performance and low effective utilization of operating machinery in the prior art are solved. Improves the autonomy and accuracy of the work.

CN117707131BActive Publication Date: 2024-11-15NANJING AGRI MECHANIZATION INST MIN OF AGRI
View PDF 10 Cites 0 Cited by

Patent Information

Application Number
CN202311390915.3
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2023-10-25
Publication Date
2024-11-15
Estimated Expiration
2043-10-25

AI Technical Summary

Technical Problem

The field operation machinery of open-field vegetable vegetables has poor target performance, low effective utilization rate, serious drift loss, and lacks intelligence. It depends on the experience of operators and cannot identify the operating environment and plant status in real time, resulting in the inability to accurately apply fertilizer spray.

Method used

The cameras on the main support frame and the side support frame are used to obtain the field images in real time. Through image processing and recognition technology, crop areas and non-crop areas are determined, navigation paths are planned, spray assembly is controlled for precise spraying, and spraying is adjusted according to crop row information. mode to realize intelligent ridge-to-ridge operation.

Benefits of technology

It improves the autonomy, accuracy and efficiency of operations, ensures the precise spraying of fertilizers, reduces the dependence on the experience of operators, and improves the coverage and effective utilization of operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN117707131B_ABST
    Figure CN117707131B_ABST
Patent Text Reader

Abstract

The method, system and equipment for controlling the autonomous walking operation of open-field vegetables disclosed in the present invention obtain the field ridge images, including the center ridge and the side ridges, in real time through the cameras on the main support frame and the side support frame; identify the crop area and the non-crop area after image processing; determine the centerline position, ridge width and crop row information of the center ridge by analyzing the center ridge image, so as to plan the navigation path and make the frame adapt to the ridge width of the center ridge. At the same time, by processing the side ridge image, determine the centerline position, ridge width and crop row information of the side ridge, so as to adjust the air-supplied spray bar spray assembly on the side frame to ensure that it is aligned with the side ridge in front; based on the crop row information of the center ridge and the side ridge, intelligently control the corresponding spray assembly to perform the spray operation. The present invention realizes the efficient operation of autonomous walking and precise fertilizer and pesticide spraying in the open-field vegetable field through key steps such as intelligent perception, working environment recognition, navigation path planning and intelligent control.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of vegetable plant protection machinery, and in particular to a method, system and equipment for controlling autonomous walking operations of open-field vegetables. Background Art

[0002] There is no targeted operation machinery for open-field vegetables. The use of field crop sprayers for spraying has problems such as poor targeting, low effective utilization, and serious drift loss. Traditional operation equipment generally has a low degree of intelligence, and the operation quality is heavily dependent on the experience of the operator. Although image processing and recognition technology has been widely used in the agricultural field in recent years, it has the problem of insufficient adaptability in the field environment of open-field vegetables, lacks perception of the operating environment and real-time recognition of the plant status, and the fertilizer and pesticide spraying during the operation cannot be adjusted according to actual needs, resulting in low effective utilization of fertilizers and pesticides. Summary of the invention

[0003] Purpose of the invention: The purpose of the present invention is to address the deficiencies in the prior art and to provide a method, system and equipment for controlling autonomous walking operations in open-field vegetable fields. Through key steps such as intelligent perception, working environment recognition, navigation path planning and intelligent control, efficient operations of autonomous walking and precise fertilizer and pesticide spraying in open-field vegetable fields are achieved.

[0004] Technical solution: In the first aspect, the method for controlling the autonomous walking operation of open-field vegetables in the present invention comprises the following steps:

[0005] Acquire the field ridge image of the working area in real time, the field ridge image includes the center ridge image collected by the main camera fixed on the main support frame and the side ridge image collected by the side camera fixed on the side support frame, wherein the main support frame is fixed in front of the frame, the side support frames are arranged on both sides of the main support frame, and the main support frame and the side support frame are connected with the air-conveying spray boom spray assembly through the transverse movement mechanism;

[0006] Processing the center ridge image and the side ridge image to obtain crop areas and non-crop areas; identifying the processed center ridge image to determine the center line position, ridge width and crop row information of the center ridge, and identifying the side ridge image to determine the center line position, ridge width and crop row information of the side ridge;

[0007] Based on the width of the center ridge, the wheelbase of the frame is determined; based on the centerline position of the center ridge, the relative position of the frame and the center ridge is determined; a navigation path is planned; and the frame is controlled to travel along the navigation path with a wheelbase that matches the width of the center ridge;

[0008] Based on the centerline position of the side ridge, determine the offset distance between the air-conveying spray bar spray assembly on the side frame and the side ridge, and control the air-conveying spray bar spray assembly to move laterally on the side frame according to the offset distance so that the air-conveying spray bar spray assembly is aligned with the side ridge in front;

[0009] Based on the crop row information of the central ridge and the side ridge, the spraying operation of the corresponding air-conveying spray boom spray assembly is controlled.

[0010] To further improve the above technical solution, the central ridge image and the side ridge image are processed to obtain the crop area and the non-crop area, including:

[0011] Image preprocessing to ensure a high-quality image set;

[0012] Convert the image to HSV color space and select H component, S component and V component as feature components;

[0013] Use fuzzy clustering algorithm to segment the image and create sample space , set the number of clusters c=2;

[0014] Initialize the parameters of the fuzzy clustering algorithm, and then perform iterative optimization. In each iteration, calculate the membership of each sample in the cluster, calculate the membership matrix U and the cluster center matrix V according to the update rule, and iterate until the stop condition is met to obtain the cluster centers corresponding to the two clusters. ;

[0015] Assumptions It is a cluster related to crops. H and S components are selected as the main basis for judging crops and thresholds are set. Through logical comparison, it is determined whether the specified feature component is greater than the threshold to identify the crop part in the image;

[0016] The recognized image is binarized to obtain a binary image, in which white represents the crop area and black represents the non-crop area.

[0017] Furthermore, the processed central ridge image and side ridge image are identified to determine the center line position, ridge width and crop row information of the central ridge and side ridge, including the following steps:

[0018] Using morphological operations on the binary image, connecting adjacent crop regions, and identifying and marking connected regions of the crop regions;

[0019] For each connected area, calculate its bounding box and extract the coordinate information of the bounding box as crop row information;

[0020] The center position of each connected area is calculated using the geometric center coordinates of the bounding box as the centerline position;

[0021] Calculate the width of the connected region and use the maximum width of the region as the ridge width.

[0022] Furthermore, for each crop row information, the area of ​​the connected region is calculated And the number of crop characteristic components belonging to the connected area , calculate the ratio of the number of characteristic components to the area , set the threshold ,when , mark the connected area as sparse, otherwise mark it as dense;

[0023] For crop row information marked as sparse, the working mode of the air-assisted spray boom spray assembly adopts a spot spray mode, and for crop row information marked as dense, the working mode of the air-assisted spray boom spray assembly adopts a direct spray mode.

[0024] Further, the determining the relative position of the frame and the center ridge based on the centerline position of the center ridge and planning the navigation path includes:

[0025] A centerline scatter plot is generated according to the centerline position of the center ridge, and a fitting equation of the centerline is obtained by performing least square regression on the centerline scatter plot;

[0026] Get the current position, direction and speed of the frame, and calculate the offset d and angle of the frame relative to the centerline of the center ridge. , according to the offset and offset angle, determine the path curve , calculate the path curve The curvature k of each point on the image is used to adjust the steering angle of the frame. and velocity information v;

[0027] Turn the angle The speed information v is sent to the frame control module to control the steering and driving of the frame.

[0028] A control system for implementing the above method comprises:

[0029] An image acquisition module acquires the center ridge image in front of the frame and the side ridge images on both sides of the frame in real time;

[0030] An image recognition module is used to process the central ridge image and the side ridge image to obtain a crop area and a non-crop area;

[0031] A ridge recognition module recognizes the processed central ridge image and side ridge image, and determines the centerline position and ridge width of the central ridge and the centerline position and ridge width of the side ridge;

[0032] A crop row recognition module recognizes the processed central ridge image and determines the crop row information and crop row features, wherein the crop row features are classified into sparse and dense.

[0033] The navigation control module adjusts the wheelbase of the vehicle frame based on the width of the center ridge, determines the relative position of the vehicle frame and the center ridge based on the centerline position of the center ridge, plans the navigation path, and controls the vehicle frame to travel along the navigation path with a wheelbase that matches the width of the center ridge;

[0034] The spray control module controls the spray operation of the corresponding air-assisted spray boom spray assembly based on the crop row information of the central ridge and the side ridge, wherein for the crop row information marked as sparse, the working mode of the air-assisted spray boom spray assembly adopts the spot spray mode, and for the crop row information marked as dense, the working mode of the air-assisted spray boom spray assembly adopts the direct spray mode.

[0035] Equipment used to perform autonomous walking operations in open-field vegetable fields, including:

[0036] A vehicle frame, a four-wheel drive hub assembly is installed at the bottom of the frame, a main support frame is connected to the front side of the frame, side support frames are connected to the two sides of the main support frame, and a pneumatic spray bar spray assembly is connected to the bottom of the main support frame and the side support frame through a transverse movement mechanism, the four-wheel drive hub assembly controls the steering and movement of the frame through a frame control module, and the transverse movement mechanism drives the pneumatic spray bar spray assembly to move laterally through a linear motor;

[0037] A camera unit, the camera unit comprising a main camera above the main support frame and a side camera above the side support frame, the main camera is used to capture the image of the center ridge in front of the frame, and the side camera is used to capture the image of the side ridge in front of the side support frame;

[0038] The control system is used to implement the above-mentioned open-field vegetable field autonomous walking operation control method.

[0039] Beneficial effects: Compared with the prior art, the advantages of the present invention are: the present invention obtains the field ridge image in real time through the main camera on the main support frame and the side camera on the side support frame, ensuring all-round monitoring of the field environment. Image processing and recognition technology, including image preprocessing, HSV color space conversion and fuzzy clustering algorithm, is used to process the center ridge and side ridge images, realize accurate division of crop area and non-crop area, and determine the center line position, ridge width and crop row information of the center ridge and side ridge.

[0040] Based on the width of the center ridge, the wheelbase of the frame is determined, and the navigation path is planned according to the centerline position of the center ridge, which helps to ensure that the frame adapts to the width of the center ridge, thereby ensuring that the driving track covers the target operation area. Through real-time identification of the centerline position of the center ridge and the side ridge, the offset distance of the wind-supplied spray boom spray assembly on the side frame is controlled to achieve lateral movement of the wind-supplied spray boom spray assembly, ensuring that the spray is aimed at the center ridge and side ridge in front, realizing intelligent ridge operation, and improving the precise spraying effect of fertilizers and pesticides. According to the crop row information of the center ridge and the side ridge, the working mode of the spot spray mode or the direct spray mode is adopted according to the dense or sparse markings, thereby realizing intelligent control of the corresponding wind-supplied spray boom spray assembly.

[0041] By generating a centerline scatter plot based on the centerline position of the center ridge and performing least squares regression, high-precision navigation path planning is achieved. Combined with the current position and direction of travel of the frame, the offset and offset angle are calculated, the steering angle of the frame is adjusted, and the speed curve is designed according to the curvature information to enable the frame to travel along the navigation path.

[0042] Based on the field ridge images collected by the main camera and the side camera, the present invention realizes the efficient operation of autonomous walking and precise fertilizer and pesticide spraying in the open field of vegetables through key steps such as intelligent perception, working environment recognition, navigation path planning and intelligent control, thereby improving the autonomy, accuracy and efficiency of the operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is the overall flow chart of the method of the present invention;

[0044] Figure 2 is a flow chart of image processing in the present invention;

[0045] Figure 3 It is a flow chart of image recognition in the present invention;

[0046] Figure 4 It is the navigation control process in the present invention;

[0047] Figure 5 It is a simulation schematic diagram of the equipment of the present invention in the intelligent ridge alignment process;

[0048] Figure 6 It is a simulation schematic diagram of the equipment of the present invention in the spray control process;

[0049] Figure 7 It is a schematic diagram of the structure of the equipment side support frame provided by the present invention after being stored;

[0050] Figure 8 It is a schematic structural diagram of the equipment side support frame provided by the present invention after being unfolded. DETAILED DESCRIPTION

[0051] The technical solution of the present invention is described in detail below with reference to the accompanying drawings, but the protection scope of the present invention is not limited to the embodiments.

[0052] Example 1: Figure 1 The open-field vegetable field autonomous walking operation control method shown includes:

[0053] Acquire the field ridge image of the working area in real time, the field ridge image includes the center ridge image collected by the main camera fixed on the main support frame and the side ridge image collected by the side camera fixed on the side support frame, wherein the main support frame is fixed in front of the frame, the side support frames are arranged on both sides of the main support frame, and the main support frame and the side support frame are connected with the air-conveying spray boom spray assembly through the transverse movement mechanism;

[0054] Processing the center ridge image and the side ridge image to obtain crop areas and non-crop areas; identifying the processed center ridge image to determine the center line position, ridge width and crop row information of the center ridge, and identifying the side ridge image to determine the center line position, ridge width and crop row information of the side ridge;

[0055] Based on the width of the center ridge, the wheelbase of the frame is determined; based on the centerline position of the center ridge, the relative position of the frame and the center ridge is determined; a navigation path is planned; and the frame is controlled to travel along the navigation path with a wheelbase that matches the width of the center ridge;

[0056] Based on the centerline position of the side ridge, determine the offset distance between the air-conveying spray bar spray assembly on the side frame and the side ridge, and control the air-conveying spray bar spray assembly to move laterally on the side frame according to the offset distance so that the air-conveying spray bar spray assembly is aligned with the side ridge in front;

[0057] Based on the crop row information of the central ridge and the side ridge, the spraying operation of the corresponding air-conveying spray boom spray assembly is controlled.

[0058] Step 1: Image acquisition

[0059] A main support frame is installed in the front of the frame, and a main camera is fixed on it to collect the image of the center ridge; side support frames are respectively arranged on both sides of the main support frame, wherein there are four groups of side support frames distributed on both sides of the main support frame, and each side support frame is fixed with a side camera to collect the image of the side ridge; the main support frame and the side support frame are both equipped with a transverse movement mechanism to ensure that the spray assemblies on the main support frame and the side support frame can be horizontally moved according to the needs of the working area.

[0060] Step 2: Image processing, such as Figure 2 As shown, the following steps are included:

[0061] Preprocessing: First, the center ridge image and side ridge images collected from the main camera and side camera are preprocessed, including color correction, denoising, histogram equalization and contrast enhancement to ensure high-quality images.

[0062] Data preparation: Convert the image to HSV color space for better separation of crop and other parts.

[0063] Use FCM or other fuzzy clustering algorithms to segment the image. In FCM, each sample vector A membership value is assigned to each sample, indicating the possibility of belonging to each category. U represents the membership of each sample to each cluster, V represents the cluster center of each cluster, and the number of clusters c=2 is set to divide the samples into two categories, representing crops and others respectively.

[0064] Creating a sample space , each sample vector , whose components are the three HSV feature components at the original image position, d=3.

[0065] Initialization: Linear transformation of samples , so that its value range is between;

[0066] Calculate the mean vector of the transformed samples and Transform Range ;

[0067] Initialize the cluster center matrix ,in, Is the size of , R is a matrix of size A matrix whose elements are A random number between is element-by-element multiplication.

[0068] The initialization step is performed before each iteration, which makes the algorithm more robust by linearly transforming the training samples and randomly generating initial cluster centers.

[0069] Iterative update: In each iteration, for each sample And each cluster ,calculate ,express To cluster center The square of the Euclidean distance;

[0070] Calculate the membership matrix U and cluster center matrix V according to the update rule:

[0071] ;

[0072] ;

[0073] Wherein, b is the fuzziness parameter, b>1.

[0074] Iterate until the stopping condition is met. For all holds, or the number of iterations t reaches the maximum number of iterations .

[0075] After the iteration, two clusters are obtained. The corresponding cluster center , assuming are the clusters associated with crops, , the amount of which Respectively represent the average values ​​of the corresponding HSV components, select H and S as the main basis for judging crops, and set the threshold , for each HSV component in the image, compare its HSV component with the threshold , if the H and S components are greater than the threshold , it is considered to belong to the crop, thus identifying the crop part in the image. Crop area ,in, Represents the coordinates in the image.

[0076] The processed image is binarized, where the crop part is white and the other parts are black, for further processing and analysis.

[0077] Step 3: Identification of centerline position, ridge width, and crop row information, such as Figure 3 As shown,

[0078] For the binary image, morphological operations are used to connect adjacent crop regions, identify and mark the connected regions of the crop regions, and exclude connected regions with smaller areas, which may be noise or obstacles.

[0079] For each connected region retained, calculate its bounding box and extract the coordinate information of the bounding box as crop row information;

[0080] The center position of each connected area is calculated using the geometric center coordinates of the bounding box as the centerline position;

[0081] Calculate the width of the connected region and use the maximum width of the region as the ridge width.

[0082] For each crop row information, calculate the area of ​​the connected region And the number of crop characteristic components belonging to the connected area , calculate the ratio of the number of characteristic components to the area , set the threshold ,when , mark the connected area as sparse, otherwise it is marked as dense.

[0083] Step 4: Frame travel control, such as Figure 4 As shown,

[0084] Based on the width of the center ridge, determine the wheelbase of the frame;

[0085] Based on the centerline position of the center ridge, the relative position between the frame and the center ridge is determined, and the navigation path planning includes:

[0086] A centerline scatter plot is generated according to the centerline position of the center ridge, and a fitting equation of the centerline is obtained by performing least square regression on the centerline scatter plot;

[0087] The main cameras above the main support frame are used to shoot the center ridge in front. Assume that the initial position of the frame is , the direction of travel is , the current driving speed is , the position information of the center line is , in The index of the center line scatter point is calculated as follows:

[0088] Calculate the relative position information of the frame and the center line, including the offset and offset angle :

[0089] ;

[0090] ;

[0091] According to the offset of the frame relative to the center line of the center ridge and offset angle , determine the path curve , calculate the curvature k of each point on the path curve;

[0092] According to the calculated curvature information k, the steering angle of the frame is adjusted through the following control strategy: :

[0093] ;

[0094] in, are the proportional and derivative gains of the controller.

[0095] The speed curve is designed according to the curvature information k. The frame needs to slow down where the curvature is large, and the frame can speed up where the curvature is small. The speed v adjustment formula is as follows:

[0096] ;

[0097] in, is the adjustment factor.

[0098] Through the above calculation, the steering angle calculated is The speed information v is sent to the frame control module to control the steering and driving of the frame, so that the frame can move along the navigation path, thereby realizing intelligent autonomous walking of the frame.

[0099] Step 5: Spray Control

[0100] For crop row information marked as sparse, the working mode of the wind-driven spray boom spray assembly adopts the spot spray mode, and for crop row information marked as dense, the working mode of the wind-driven spray boom spray assembly adopts the direct spray mode. The operation simulation diagram of intelligent recognition and spraying of ridges and spraying for sparse crop rows is shown in the figure. Figure 5 , Figure 6 shown.

[0101] Embodiment 2: This embodiment provides a control system for implementing the method described in Embodiment 1, including:

[0102] An image acquisition module acquires the center ridge image in front of the frame and the side ridge images on both sides of the frame in real time;

[0103] An image recognition module is used to process the central ridge image and the side ridge image to obtain a crop area and a non-crop area;

[0104] A ridge recognition module recognizes the processed central ridge image and side ridge image, and determines the centerline position and ridge width of the central ridge and the centerline position and ridge width of the side ridge;

[0105] A crop row recognition module recognizes the processed central ridge image and determines the crop row information and crop row features, wherein the crop row features are classified into sparse and dense.

[0106] The navigation control module adjusts the wheelbase of the vehicle frame based on the width of the center ridge, determines the relative position of the vehicle frame and the center ridge based on the centerline position of the center ridge, plans the navigation path, and controls the vehicle frame to travel along the navigation path with a wheelbase that matches the width of the center ridge;

[0107] The spray control module controls the spray operation of the corresponding air-assisted spray boom spray assembly based on the crop row information of the central ridge and the side ridge, wherein for the crop row information marked as sparse, the working mode of the air-assisted spray boom spray assembly adopts the spot spray mode, and for the crop row information marked as dense, the working mode of the air-assisted spray boom spray assembly adopts the direct spray mode.

[0108] Embodiment 3: The equipment for precise application of fertilizer and pesticide for open-field vegetables provided in this embodiment is as follows: Figure 7 , Figure 8 As shown, it includes a frame 20, a four-wheel drive hub assembly 10 is installed at the bottom of the frame 20, a clean water tank, a medicine box, a fertilizer box, a diaphragm pump, and a compression motor 30 are fixed on the frame 20, and the front side of the frame 20 is connected to a main support frame 301 through a four-bar linkage, and side support frames 401 are connected to both sides of the main support frame 301. A main camera 401 is arranged above the main support frame 301, and a side camera 402 is arranged above the side support frame 302. The main support frame 301 and the side support frame 302 are both connected to the air-supplied spray rod spray assembly 70 through a transverse mechanism 60.

[0109] The lateral movement mechanism 60 includes a linear motor and a guide rod group. The top of the linear motor is fixedly connected to the main support frame and the side support frame. The guide rod group includes a bottom lateral movement guide rod and two top lateral movement guide rods. The top lateral movement guide rod is connected to the linear motor and can move in the linear motor to drive the air-conveying spray rod spray assembly connected to the bottom lateral movement guide rod to move linearly. The lateral movement mechanism can realize the lateral movement of the air-conveying spray rod spray assembly, so as to achieve precise application of fertilizers and pesticides to open-field vegetables such as cabbage and broccoli.

[0110] The structure of the four-wheel drive hub assembly 10 is shown in the publication number CN114771538A. It has a variable characteristic steering system and a wheelbase adjustment mechanism. It adopts a four-wheel independent electric drive variable wheelbase working chassis, which can realize chassis wheelbase control, stepless control of walking speed, and arbitrary radius turning and U-turn, providing a high-passability and high-adaptability working chassis for the precise spraying of fertilizers and pesticides on open-field vegetables, and realizing high-precision positioning, operation path planning, and intelligent autonomous walking on ridges during the precise spraying of fertilizers and pesticides on open-field vegetables.

[0111] As described above, although the present invention has been shown and described with reference to specific preferred embodiments, it should not be construed as limiting the present invention itself. Various changes in form and details may be made without departing from the spirit and scope of the present invention as defined in the appended claims.

Claims

1. A method for controlling autonomous walking operations of open-field vegetables, characterized in that: The steps include: Acquire the field ridge image of the working area in real time, the field ridge image includes the center ridge image collected by the main camera fixed on the main support frame and the side ridge image collected by the side camera fixed on the side support frame, wherein the main support frame is fixed in front of the frame, the side support frames are arranged on both sides of the main support frame, and the main support frame and the side support frame are connected with the air-conveying spray boom spray assembly through the transverse movement mechanism; The center ridge image and the side ridge image are processed to obtain the crop area and the non-crop area, including: image preprocessing to obtain a high-quality image set; converting the image into the HSV color space, selecting the H component, the S component, and the V component as feature components; using the fuzzy clustering algorithm to segment the image and create a sample space , set the number of clusters c=2; initialize the parameters of the fuzzy clustering algorithm, and then perform iterative optimization. In each iteration, calculate the membership of each sample in the cluster, calculate the membership matrix U and the cluster center matrix V according to the update rule, and iterate until the stop condition is met to obtain the cluster centers corresponding to the two clusters. Assumptions It is a cluster related to crops. The H and S components are selected as the main basis for judging crops and the threshold is set. The specified feature component is judged by logical comparison whether it is greater than the threshold to identify the crop part in the image. The identified image is binarized to obtain a binary image, in which white is the crop area and black is the non-crop area. Identify the processed center ridge image to determine the centerline position, ridge width and crop row information of the center ridge, and identify the side ridge image to determine the centerline position, ridge width and crop row information of the side ridge; Based on the width of the center ridge, the wheelbase of the frame is determined; based on the centerline position of the center ridge, the relative position of the frame and the center ridge is determined; a navigation path is planned; and the frame is controlled to travel along the navigation path with a wheelbase that matches the width of the center ridge; Based on the centerline position of the side ridge, determine the offset distance between the air-conveying spray bar spray assembly on the side frame and the side ridge, and control the air-conveying spray bar spray assembly to move laterally on the side frame according to the offset distance so that the air-conveying spray bar spray assembly is aligned with the side ridge in front; Based on the crop row information of the central ridge and the side ridge, the spraying operation of the corresponding air-conveying spray boom spray assembly is controlled.

2. The method for controlling the autonomous walking operation of open-field vegetables according to claim 1, characterized in that: The processed center ridge image and side ridge image are identified to determine the center line position, ridge width and crop row information of the center ridge and side ridge, including the following steps: Using morphological operations on the binary image, connecting adjacent crop regions, and identifying and marking connected regions of the crop regions; For each connected area, calculate its bounding box and extract the coordinate information of the bounding box as crop row information; The center position of each connected area is calculated using the geometric center coordinates of the bounding box as the centerline position; Calculate the width of the connected region and use the maximum width of the region as the ridge width.

3. The method for controlling the autonomous walking operation of open-field vegetables according to claim 2, characterized in that: For each crop row information, calculate the area of ​​the connected region And the number of crop characteristic components belonging to the connected area , calculate the ratio of the number of characteristic components to the area , set the threshold ,when , mark the connected area as sparse, otherwise mark it as dense; For crop row information marked as sparse, the working mode of the air-assisted spray boom spray assembly adopts a spot spray mode, and for crop row information marked as dense, the working mode of the air-assisted spray boom spray assembly adopts a direct spray mode.

4. The method for controlling the autonomous walking operation of open-field vegetables according to claim 2, characterized in that: The determining of the relative position between the vehicle frame and the center ridge based on the centerline position of the center ridge and planning the navigation path includes: A centerline scatter plot is generated according to the centerline position of the center ridge, and a fitting equation of the centerline is obtained by performing least square regression on the centerline scatter plot; Get the current position, direction and speed of the frame, and calculate the offset d and angle of the frame relative to the centerline of the center ridge. , according to the offset and offset angle, determine the path curve , calculate the path curve The curvature k of each point on the image is used to adjust the steering angle of the frame. and velocity information v; Turn the angle The speed information v is sent to the frame control module to control the steering and driving of the frame.

5. A control system for implementing the method described in any one of claims 1 to 4, characterized in that: include: An image acquisition module acquires the center ridge image in front of the frame and the side ridge images on both sides of the frame in real time; An image recognition module is used to process the central ridge image and the side ridge image to obtain a crop area and a non-crop area; A ridge recognition module recognizes the processed central ridge image and side ridge image, and determines the centerline position and ridge width of the central ridge and the centerline position and ridge width of the side ridge; A crop row recognition module recognizes the processed central ridge image and determines the crop row information and crop row features, wherein the crop row features are classified into sparse and dense. The navigation control module adjusts the wheelbase of the vehicle frame based on the width of the center ridge, determines the relative position of the vehicle frame and the center ridge based on the centerline position of the center ridge, plans the navigation path, and controls the vehicle frame to travel along the navigation path with a wheelbase that matches the width of the center ridge; The spray control module controls the spray operation of the corresponding air-assisted spray boom spray assembly based on the crop row information of the central ridge and the side ridge, wherein for the crop row information marked as sparse, the working mode of the air-assisted spray boom spray assembly adopts the spot spray mode, and for the crop row information marked as dense, the working mode of the air-assisted spray boom spray assembly adopts the direct spray mode.

6. An autonomous walking equipment for open-field vegetable operation, characterized in that: include: A vehicle frame, a four-wheel drive hub assembly is installed at the bottom of the frame, a main support frame is connected to the front side of the frame, side support frames are connected to the two sides of the main support frame, and a pneumatic spray bar spray assembly is connected to the bottom of the main support frame and the side support frame through a transverse movement mechanism, the four-wheel drive hub assembly controls the steering and movement of the frame through a frame control module, and the transverse movement mechanism drives the pneumatic spray bar spray assembly to move laterally through a linear motor; A camera unit, the camera unit comprising a main camera above the main support frame and a side camera above the side support frame, the main camera is used to capture the image of the center ridge in front of the frame, and the side camera is used to capture the image of the side ridge in front of the side support frame; A control system for implementing the method according to any one of claims 1 to 4.

Citation Information

Patent Citations

  • Agricultural four-wheel drive chassis and control method thereof

    CN114771538A

  • Greenhouse suspended self-propelled target-oriented sprayer system and operation method thereof

    CN102017938A

  • Real-time field robot visual navigation method based on FPGA and real-time field robot visual navigation system based on FPGA

    CN110196053A

  • Method for detecting field navigation line after ridge sealing of crops

    CN112146646A

  • Pesticide application control method and device, spraying machine and storage medium

    CN115336498A