Remote control mountain pepper harvester and method

Abstract

The application provides a remote control type mountain pepper harvester and method, which comprises the following steps: collecting scanning information of a target operation area; extracting a left side operation center line and a right side operation center line of the mountain pepper harvester during pepper picking from the scanning information based on the transverse distribution distance between each row of pepper plants in the target operation area and the density distribution characteristics of the pepper plants; determining a first operation cost and a second operation cost during pepper picking based on the left side operation center line, the right side operation center line and the motion constraint point of the mountain pepper harvester; and constraining the pose state of the mountain pepper harvester based on the constraint control quantity determined based on the first operation cost and the second operation cost. The above scheme constrains the pose state of the mountain pepper harvester based on the constraint control quantity, and can realize the automatic adjustment of the advancing pose of the mountain pepper harvester under the constraint influence of the rugged mountain, thereby improving the harvesting efficiency of the mountain pepper harvester.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of intelligent agricultural power machinery, more particularly, the present application relates to a remote control type mountain pepper harvesting machine and method. BACKGROUND

[0002] With the continuous development of the intelligent level of agricultural production, intelligent agricultural power machinery has been significantly developed. Intelligent agricultural power machinery can realize remote control, autonomous decision-making and autonomous operation. In addition, when intelligent agricultural power machinery encounters uneven terrain and uneven soil resistance, it can autonomously fine-tune engine power, gear position and suspension system to adapt to the working environment. For example, in the process of harvesting peppers by a remote control type mountain pepper harvesting machine, through the combination of fuzzy control and adaptive control algorithms, the mountain pepper harvesting machine can realize autonomous operation in a complex and variable farmland environment, ensuring accurate picking of peppers in the target working area, and the working quality far exceeds manual operation.

[0003] In the existing intelligent agricultural power machinery control, the control of intelligent agricultural power machinery is mainly based on a distributed power control system and intelligent control algorithm. Through real-time data acquisition and a multi-modal reinforcement learning framework, the precise control of mechanical movement and operation is realized to adapt to different working conditions. However, in the control of the remote control type mountain pepper harvesting machine, when the mountain pepper harvesting machine is in a mountain environment, it is affected by the constraints of rugged mountains. The working boundary of the mountain pepper harvesting machine cannot be aligned with the target harvesting boundary (i.e. the preset harvesting boundary of the mountain pepper harvesting machine), causing the advancing posture of the mountain pepper harvesting machine to deviate from the normal advancing posture, thereby reducing the harvesting efficiency of the mountain pepper harvesting machine. Therefore, how to automatically adjust the advancing posture of the mountain pepper harvesting machine under the constraints of rugged mountains and improve the harvesting efficiency of the mountain pepper harvesting machine has become a difficult problem in the industry. SUMMARY

[0004] The present application provides a remote control type mountain pepper harvesting machine and method, which can automatically adjust the advancing posture of the mountain pepper harvesting machine under the constraints of rugged mountains and improve the harvesting efficiency of the mountain pepper harvesting machine.

[0005] In a first aspect, the present application provides a pepper picking method, comprising the following steps:

[0006] When the remote control type mountain pepper harvesting machine travels to the target working area, the target working area is scanned;

[0007] The working center line of the left side and the working center line of the right side of the mountain pepper harvesting machine during pepper picking are extracted from the scanning information through the transverse distribution distance between each row of pepper plants in the target working area and the density distribution characteristics of the pepper plants;

[0008] determining a motion constraint point of the mountain pepper harvester under the constraint of the picking boundary, determining a first operation cost of the left side of the mountain pepper harvester under the constraint of the motion constraint point and the operation center line of the left side, and determining a second operation cost of the left side of the mountain pepper harvester under the constraint of the motion constraint point and the operation center line of the left side;

[0009] constraining the travel route of the mountain pepper harvester under the constraint of the first operation cost and the second operation cost to obtain a constraint control quantity of the mountain pepper harvester on the travel route;

[0010] constraining the pose state of the mountain pepper harvester based on the constraint control quantity, and controlling the mountain pepper harvester to pick peppers in the target operation area after the pose state is constrained.

[0011] In some embodiments, the operation center line of the left side and the operation center line of the right side of the mountain pepper harvester during pepper picking are extracted from the scanning information by the lateral distribution interval between the rows of pepper plants in the target operation area and the density distribution characteristics of the pepper plants, and specifically include:

[0012] extracting a current scanning image frame from the scanning information of the target operation area;

[0013] performing density clustering analysis on the scanning image frame to obtain an aggregation block of pepper plants in the target operation area, and taking the aggregation block as a pepper row in the target operation area to obtain a plurality of pepper rows in the target operation area;

[0014] determining the pepper row center line of each pepper row based on the density distribution characteristics of the pepper plants in the target operation area;

[0015] calculating the interval between the pepper row center lines to obtain the lateral distribution interval between the rows of pepper plants in the target operation area;

[0016] extracting the effective lateral distribution interval of the mountain pepper harvester during pepper picking from all the lateral distribution intervals based on the length of the picking bucket of the mountain pepper harvester;

[0017] extracting the first pepper row center line and the second pepper row center line corresponding to the effective lateral distribution interval;

[0018] taking the first pepper row center line and the second pepper row center line as the operation center line of the left side and the operation center line of the right side of the mountain pepper harvester during pepper picking, respectively.

[0019] In some embodiments, determining the pepper row centerline of each pepper row in the target working area based on the density distribution characteristics of the pepper plants in the target working area specifically comprises:

[0020] performing pixel value extraction on each pepper row in the target working area to obtain a pixel value set of each pepper row;

[0021] selecting one pepper row as a selected pepper row, and determining the density distribution characteristics of the pepper plants in the selected pepper row based on the pixel value set of the selected pepper row;

[0022] performing disturbance pixel removal on the selected pepper row based on the density distribution characteristics to obtain an anti-interference selected pepper row;

[0023] fitting all geometric center points of the anti-interference selected pepper row to obtain a pepper row centerline of the selected pepper row;

[0024] continuing to determine the pepper row centerlines of the remaining pepper rows.

[0025] In some embodiments, determining the first working cost of the left side of the mountain pepper harvester when working specifically comprises:

[0026] extracting a left side working point set from the working centerline of the left side, and extracting the closest left side working point and the farthest left side working point in the left side working point set to the current position of the mountain pepper harvester;

[0027] determining the orientation vector of the left side pepper row of the mountain pepper harvester based on the closest left side working point and the farthest left side working point;

[0028] determining the orientation vector of the left side of the mountain pepper harvester based on the farthest left side working point and the motion constraint point;

[0029] performing vector dot product calculation on the orientation vector of the left side pepper row of the mountain pepper harvester and the orientation vector of the left side of the mountain pepper harvester to obtain the first working cost of the left side of the mountain pepper harvester when working.

[0030] In some embodiments, determining the second working cost of the left side of the mountain pepper harvester when working specifically comprises:

[0031] extracting a right side working point set from the working centerline of the right side, and extracting the closest right side working point and the farthest right side working point in the right side working point set to the current position of the mountain pepper harvester;

[0032] determining a direction vector of the right side row of peppers of the mountainous pepper harvester through the nearest right side working point and the farthest right side working point;

[0033] determining a direction vector of the right side of the mountainous pepper harvester according to the farthest right side working point and the motion constraint point;

[0034] performing a vector dot product calculation on the direction vector of the right side row of peppers of the mountainous pepper harvester and the direction vector of the right side of the mountainous pepper harvester to obtain a second working cost of the right side of the mountainous pepper harvester when working.

[0035] In some embodiments, the remote-controlled mountainous pepper harvester is a drum-type separated pepper harvester.

[0036] In some embodiments, the scanning information of the target working area is collected through a camera.

[0037] In a second aspect, the present application provides a remote-controlled mountainous pepper harvester, which comprises a pepper picking control unit, and the pepper picking control unit comprises:

[0038] a scanning module, configured to scan a target working area when the remote-controlled mountainous pepper harvester travels to the target working area;

[0039] a processing module, configured to extract a working center line of a left side and a working center line of a right side of the mountainous pepper harvester when picking peppers from the target working area according to a transverse distribution distance between rows of pepper plants and a density distribution characteristic of the pepper plants in the target working area;

[0040] The processing module is further configured to determine a motion constraint point of the mountainous pepper harvester under the constraint of a picking boundary, determine a first working cost of the left side of the mountainous pepper harvester when working based on the motion constraint point and the working center line of the left side, and determine a second working cost of the left side of the mountainous pepper harvester when working through the motion constraint point and the working center line of the left side;

[0041] The processing module is further configured to constrain a travel route of the mountainous pepper harvester when picking peppers according to the first working cost and the second working cost to obtain a constraint control amount of the mountainous pepper harvester on the travel route;

[0042] an execution module, configured to constrain a pose state of the mountainous pepper harvester based on the constraint control amount, and control the mountainous pepper harvester after the pose state is constrained to pick peppers in the target working area.

[0043] In a third aspect, the present application provides a computer device, comprising a memory and a processor, the memory storing a code, and the processor being configured to acquire the code and execute the pepper picking method.

[0044] In a fourth aspect, the present application provides a computer readable storage medium, storing a computer program, which, when executed by a processor, implements the pepper picking method.

[0045] The technical scheme provided by the embodiments of the present application has the following beneficial effects:

[0046] In the remote mountain pepper harvester and method provided by the present application, first, when the remote mountain pepper harvester drives to the target working area, the target working area is scanned; second, the working center line of the left side and the working center line of the right side of the mountain pepper harvester during pepper picking are extracted from the scanning information through the transverse distribution distance between each row of pepper plants in the target working area and the density distribution characteristics of the pepper plants; further, the motion constraint point of the mountain pepper harvester under the picking boundary constraint is determined, the first working cost of the left side of the mountain pepper harvester during working is determined based on the motion constraint point and the working center line of the left side, the second working cost of the left side of the mountain pepper harvester during working is determined through the motion constraint point and the working center line of the left side; then, the travel route of the mountain pepper harvester during pepper picking is constrained by the minimum cost based on the first working cost and the second working cost, and the constraint control amount of the mountain pepper harvester on the travel route is obtained; finally, the pose state of the mountain pepper harvester is constrained based on the constraint control amount, and the mountain pepper harvester after the pose state constraint picks the peppers in the target working area.

[0047] It can be seen that the application can realize automatic adjustment of the advancing posture of the mountain pepper harvesting machine under the constraint of the rugged mountain, thereby improving the harvesting efficiency of the mountain pepper harvesting machine. First, the scanning information when the mountain pepper harvesting machine scans the target operation area is collected, thereby providing comprehensive and reliable basic data support for subsequent operation center line extraction, path planning and posture adjustment. Second, the operation center line on the left side and the operation center line on the right side of the mountain pepper harvesting machine during pepper picking are extracted from the scanning information through the transverse distribution distance between the rows of pepper plants in the target operation area and the density distribution characteristics of the pepper plants, thereby providing accurate working range guidance for the picking bucket, so that the harvesting machine can pick as many peppers as possible in a single pass, and avoid missing or repeated operation caused by the rugged mountain. Further, the first operation cost on the left side and the second operation cost on the right side of the mountain pepper harvesting machine during pepper picking are determined to analyze the path loss state of the left and right sides of the mountain pepper harvesting machine, and then the loss state of the mountain pepper harvesting machine is adjusted to avoid the constraint of the rugged mountain. Then, the constraint control amount of the mountain pepper harvesting machine on the advancing route is determined based on the first operation cost and the second operation cost, so as to timely adjust the operation center line of the mountain pepper harvesting machine to align with the target harvesting boundary, so as to optimize the harvesting process, ensure that the mountain pepper harvesting machine remains on the preset operation path, and reduce the instability of the harvesting process. Finally, the pose state of the mountain pepper harvesting machine is constrained based on the constraint control amount, so as to align the operation boundary of the mountain pepper harvesting machine with the target harvesting boundary, and then obtain the effective operation posture of the mountain pepper harvesting machine. In summary, the technical scheme provided by the application can realize automatic adjustment of the advancing posture of the mountain pepper harvesting machine under the constraint of the rugged mountain, thereby improving the harvesting efficiency of the mountain pepper harvesting machine. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 is an exemplary flowchart of a pepper picking method according to some embodiments of the application;

[0049] Figure 2 is an exemplary flowchart of determining a first operation cost according to some embodiments of the application;

[0050] Figure 3 is an exemplary flowchart of determining a constraint control amount according to some embodiments of the application;

[0051] Figure 4 is a structural schematic diagram of a pepper picking control unit according to some embodiments of the application;

[0052] Figure 5 is a structural schematic diagram of a computer device for implementing a pepper picking method according to some embodiments of the application. DETAILED DESCRIPTION

[0053] In order to better understand the technical solution of the present application, the technical solution of the present application will be described in detail below with reference to the accompanying drawings and specific implementation methods.

[0054] refer to Figure 1 , which is an exemplary flow chart of a pepper picking method according to some embodiments of the present application. The pepper picking method 100 mainly includes the following steps:

[0055] In step 101, when the remote-controlled mountain pepper harvester travels to a target operating area, the target operating area is scanned.

[0056] It should be noted that the remote-controlled mountain pepper harvester in this embodiment can be a drum-type separation pepper harvester. The remote-controlled mountain pepper harvester is an agricultural machinery specially designed for harvesting peppers in complex terrains such as mountains. The remote-controlled mountain pepper harvester usually has good adaptability and can adapt to the mechanized harvesting of string peppers and flat peppers in different regions and varieties. In specific implementation, it can also be other mountain pepper harvesters with similar functions, which will not be repeated here.

[0057] In specific implementation, the mountain pepper harvester is remotely controlled by the remote control terminal to travel to the target operating area, and the target operating area is scanned by the camera to obtain scanning information of the target operating area. The scanning information includes multiple scanning image frames, and the scanning image frames represent images obtained by scanning the target picking area. In addition, in other embodiments, other acquisition sensors can also be used to collect scanning information of the target operating area, such as laser radar, which is not limited here.

[0058] It should be noted that the target operation area in this application represents the area where peppers need to be picked; the scanning information in this embodiment represents the regional feature information obtained by scanning the target operation area. By determining the scanning information, it can effectively provide comprehensive and reliable basic data support for subsequent operation centerline extraction, path planning, posture adjustment and picking execution.

[0059] In step 102, the operation center line on the left side and the operation center line on the right side of the mountain pepper harvester when picking peppers are extracted from the scanning information through the lateral distribution spacing between the rows of pepper plants in the target operation area and the density distribution characteristics of the pepper plants.

[0060] In some embodiments, the following steps can be used to extract the left and right operation center lines of the mountain pepper harvester during pepper picking by using the lateral distribution spacing between rows of pepper plants and the density distribution characteristics of the pepper plants in the target operation area from the scanning information, namely:

[0061] extracting a current scanning image frame from the scanning information of the target working area;

[0062] performing density clustering analysis on the scanning image frame to obtain a cluster block of the pepper plants in the target working area, and taking the cluster block as a pepper row in the target working area to obtain a plurality of pepper rows in the target working area;

[0063] determining a pepper row center line of each pepper row based on a density distribution feature of the pepper plants in the target working area;

[0064] calculating a spacing between the pepper row center lines to obtain a transverse distribution spacing between the rows of pepper plants in the target working area;

[0065] extracting an effective transverse distribution spacing of the mountainous pepper harvester during pepper picking from all the transverse distribution spacings based on a bucket length of a picking bucket of the mountainous pepper harvester;

[0066] extracting a first pepper row center line and a second pepper row center line corresponding to the effective transverse distribution spacing;

[0067] taking the first pepper row center line and the second pepper row center line as a working center line of a left side and a working center line of a right side of the mountainous pepper harvester during pepper picking, respectively.

[0068] In a specific implementation, the density clustering analysis is performed on the scanning image frame to obtain the cluster block of the pepper plants in the target working area, and the cluster block is taken as the pepper row in the target working area to obtain the plurality of pepper rows in the target working area, that is, a density-based clustering algorithm (such as DBSCAN) is used to analyze the pixel data in the scanning image frame to identify the cluster block of the pepper plants, and the cluster block is taken as the pepper row in the target working area to obtain the plurality of pepper rows in the target working area.

[0069] It should be noted that the cluster block in the embodiment represents a concentrated distribution region formed by the pepper plants in space, and the pepper plants are usually planted in the form of rows in pepper planting. Therefore, each cluster block usually corresponds to a row of pepper plants, and the cluster block reflects the distribution characteristics of the pepper plants in the target working area, including the spatial position, morphological features, and relative relationship with adjacent plants. The pepper row in the embodiment represents the arrangement structure of the pepper plants in the target working area. By determining the pepper row, the distribution state of the pepper in the target working area can be effectively identified, thereby providing conditions for subsequent picking of the mountainous pepper harvester.

[0070] In some embodiments, the determination of the pepper row centerline of each pepper row in the target working area based on the density distribution characteristics of the pepper plants in the target working area can be achieved by the following steps, that is:

[0071] Pixel value extraction is performed on each pepper row in the target working area to obtain a pixel value set of each pepper row.

[0072] A pepper row is selected as a selected pepper row, and the density distribution characteristics of the pepper plants in the selected pepper row in the target working area are determined based on the pixel value set of the selected pepper row.

[0073] The selected pepper row is disturbed and the disturbed pixels are removed based on the density distribution characteristics to obtain an anti-interference selected pepper row.

[0074] All geometric center points of the anti-interference selected pepper row are fitted to obtain a pepper row centerline of the selected pepper row.

[0075] The pepper row centerlines of the remaining pepper rows are continuously determined.

[0076] In specific implementation, first, pixel value extraction is performed on each pepper row in the target working area to obtain a pixel value set of each pepper row; second, a pepper row is selected as a selected pepper row, mean value calculation is performed on all pixel values in the pixel value set of the selected pepper row, and the mean value calculation result is taken as the density distribution characteristics of the pepper plants in the selected pepper row in the target working area; further, the selected pepper row is disturbed and the disturbed pixels are removed based on the density distribution characteristics to obtain an anti-interference selected pepper row, that is, pixel points with a difference greater than a preset difference value between the pixel value and the density distribution characteristics in the selected pepper row are removed to obtain the anti-interference selected pepper row; then, all geometric center points of the anti-interference selected pepper row are fitted by an image processing tool OpenCV to obtain a pepper row centerline of the selected pepper row; finally, the pepper row centerlines of the remaining pepper rows are continuously determined by the method of "fitting all geometric center points of the anti-interference selected pepper row to obtain a pepper row centerline of the selected pepper row".

[0077] It should be noted that the density distribution characteristics in the present application represent the density of the pepper plants in the pepper row, that is, the greater the density distribution characteristics, the greater the density of the pepper plants in the pepper row, and vice versa; and the pepper row centerline in the present embodiment represents a fitting line along the center of the pepper row.

[0078] In a specific implementation, the distance between the center lines of each pepper row can be calculated by a vector method to obtain the lateral distribution distance between each pepper row in the target working area, that is, for each two pepper rows, a common normal vector of the center lines of the two pepper rows is first calculated, then the projection length of the connecting line segment of any two points on the center lines of the two pepper rows on the common normal vector is calculated, and the projection length is taken as the lateral distribution distance between the two pepper rows, and then the lateral distribution distance between each pepper row in the target working area is obtained. In this application, the lateral distribution distance between each pepper row is calculated in sequence with the first pepper row as the reference row, and the distance between the reference row and the remaining pepper rows is calculated.

[0079] It should be noted that the lateral distribution distance in this application represents the lateral distance between two pepper rows in the target working area, and the lateral distribution distance reflects the arrangement state of the pepper rows in space. The determination of the lateral distribution distance is an important parameter for planning the picking path of the mountain pepper harvester. The extraction of the effective lateral distribution distance further considers the physical size of the picking bucket and the working requirements to ensure the efficiency and accuracy of the harvester.

[0080] In a specific implementation, the effective lateral distribution distance of the mountain pepper harvester when picking peppers is extracted from all lateral distribution distances based on the bucket length of the picking bucket of the mountain pepper harvester, that is, the difference value calculation between the bucket length and each lateral distribution distance is calculated, and the lateral distribution distance corresponding to the minimum difference value calculation result is extracted as the effective lateral distribution distance of the mountain pepper harvester when picking peppers. In the mountain pepper harvester harvesting, the picking efficiency of the pepper needs to be improved, and the picking bucket needs to cover more pepper plants to the greatest extent. Therefore, the lateral distribution distance corresponding to the minimum difference value calculation result can be taken as the effective lateral distribution distance of the mountain pepper harvester when picking peppers. In this embodiment, the effective lateral distribution distance represents the effective picking distance range of the mountain pepper harvester.

[0081] In a specific implementation, the first pepper row center line and the second pepper row center line corresponding to the effective lateral distribution distance are extracted, that is, the left pepper row center line corresponding to the effective lateral distribution distance is taken as the first pepper row center line, and the right pepper row center line corresponding to the effective lateral distribution distance is taken as the first pepper row center line. The first pepper row center line represents the pepper row center line close to the left side of the mountain pepper harvester, and the second pepper row center line represents the pepper row center line close to the right side of the mountain pepper harvester.

[0082] It should be noted that the left side working center line in the present application represents the reference center line when the left side of the mountain pepper harvester is working; the right side working center line in the present application represents the reference center line when the right side of the mountain pepper harvester is working. In the complex terrain and uneven plant distribution of mountainous environment, through the determination of the left side working center line and the right side working center line, the mountain pepper harvester can provide the basis for harvesting, so as to adapt to the operation requirements of different areas flexibly. The working center line provides accurate working range guidance for the picking shovel, so that the harvester can pick as many peppers as possible in a single pass, while reducing missed or repeated work.

[0083] In step 103, the movement constraint point of the mountain pepper harvester under the picking boundary constraint is determined, the first operation cost of the left side of the mountain pepper harvester is determined based on the movement constraint point and the left side working center line, and the second operation cost of the left side of the mountain pepper harvester is determined through the movement constraint point and the left side working center line.

[0084] In some embodiments, the determination of the movement constraint point of the mountain pepper harvester under the picking boundary constraint can be achieved by the following steps, that is:

[0085] The left picking boundary extension line and the right picking boundary extension line of the picking shovel of the mountain pepper harvester are calibrated;

[0086] According to the boundary margin parameter preset by the mountain pepper harvester, the left picking boundary extension line and the right picking boundary extension line, the effective working path line of the mountain pepper harvester is determined;

[0087] The points on the effective working path line where there are obstacles are marked as movement constraint points.

[0088] In a specific implementation, first, the left picking boundary extension line and the right picking boundary extension line of the picking shovel of the mountain pepper harvester are calibrated using a spatial projection technology, for example, the left picking boundary extension line and the right picking boundary extension line of the picking shovel of the mountain pepper harvester are calibrated using a point cloud processing spatial projection technology, the point cloud processing spatial projection technology processes the point cloud data of the picking shovel of the mountain pepper harvester in a three-dimensional space, extracts boundary points, and projects them onto a two-dimensional plane to determine the picking boundary, and then extends the picking boundary points according to a predetermined delay scale to obtain the left picking boundary extension line and the right picking boundary extension line of the picking shovel of the mountain pepper harvester; then, the effective working path line of the mountain pepper harvester is determined according to the boundary margin parameter preset for the mountain pepper harvester, the left picking boundary extension line, and the right picking boundary extension line, that is, the left picking boundary extension line and the right picking boundary extension line are offset corrected according to the boundary margin parameter (such as the working safety distance) preset for the mountain pepper harvester, and the center line between the corrected left picking boundary extension line and the right picking boundary extension line is taken as the effective working path line of the mountain pepper harvester, for example, the left picking boundary extension line and the right picking boundary extension line are inwardly corrected according to the boundary margin parameter, and the center line between the corrected left picking boundary extension line and the right picking boundary extension line is taken as the effective working path line of the mountain pepper harvester; finally, obstacles on the effective working path line are identified by an image segmentation algorithm, and their position coordinates are marked as motion constraint points, for example, the Canny edge detection algorithm in the image segmentation algorithm is used.

[0089] It should be noted that the left picking boundary extension line in the present embodiment represents the extension line of the left side boundary line of the picking shovel of the mountain pepper harvester; the right picking boundary extension line in the present embodiment represents the extension line of the right side boundary line of the picking shovel of the mountain pepper harvester; the boundary margin parameter in the present embodiment represents a predefined parameter for adjusting the picking boundary to ensure the safety and effectiveness of the mountain pepper harvester in complex and irregular terrains; the effective working path line in the present embodiment represents the effective working trajectory of the mountain pepper harvester; the motion constraint point in the present application represents a position point with a motion path restriction, that is, a restrictive position point for the motion trajectory and path planning of the mountain pepper harvester generated by boundary conditions, terrain features, or obstacle distribution, which can dynamically adjust the pose and travel route of the harvester, and the determination of the motion constraint point can effectively ensure the continuity and stability of the working path, while reducing unnecessary path deviation or repeated work.

[0090] In some embodiments, reference is made to Figure 2As shown, the figure is an exemplary flow chart for determining the first work cost according to some embodiments of the present application, and the first work cost of the left side of the mountain pepper harvester can be determined based on the work center line of the left side and the motion constraint point according to the following steps:

[0091] Firstly, in step 1031, a set of left side work points is extracted from the work center line of the left side, and the closest left side work point and the farthest left side work point in the set of left side work points to the current position of the mountain pepper harvester are extracted;

[0092] Secondly, in step 1032, the orientation vector of the pepper row on the left side of the mountain pepper harvester is determined by the closest left side work point and the farthest left side work point;

[0093] Then, in step 1033, the orientation vector of the left side of the mountain pepper harvester is determined according to the farthest left side work point and the motion constraint point;

[0094] Finally, in step 1034, the first work cost of the left side of the mountain pepper harvester is calculated by vector dot product of the orientation vector of the pepper row on the left side of the mountain pepper harvester and the orientation vector of the left side of the mountain pepper harvester.

[0095] In specific implementation, firstly, a set of left side work points is extracted from the work center line of the left side, and the closest left side work point and the farthest left side work point in the set of left side work points to the current position of the mountain pepper harvester are extracted, that is, all left side work points are extracted from the work center line of the left side according to a preset sampling step to obtain a set of left side work points, the distances between each left side work point in the set of left side work points and the current position of the mountain pepper harvester are calculated by Euclidean distance, and the closest left side work point and the farthest left side work point in the set of left side work points to the current position of the mountain pepper harvester are extracted; secondly, the closest left side work point and the farthest left side work point are calculated by vector subtraction to obtain the orientation vector of the pepper row on the left side of the mountain pepper harvester; then, the farthest left side work point and the motion constraint point are calculated by vector subtraction to obtain the orientation vector of the left side of the mountain pepper harvester; finally, the first work cost of the left side of the mountain pepper harvester is calculated by vector dot product of the orientation vector of the pepper row on the left side of the mountain pepper harvester and the orientation vector of the left side of the mountain pepper harvester.

[0096] It should be noted that the vector subtraction in the embodiment refers to the subtraction between two coordinate vectors, which will not be described herein; the left side edge working point set in the embodiment represents a combination containing multiple left side edge working points, and the left side edge working point represents a position point on the working center line when the mountain pepper harvester works on the left side; the orientation vector of the left side pepper row of the mountain pepper harvester in the embodiment represents the extension direction of the pepper row of the left side of the mountain pepper harvester in space, which is usually expressed in the form of a vector, and the orientation vector is used as a direction guide for the left side pepper row of the mountain pepper harvester, helping the harvester to determine the travel route of the left side working range, and providing a reference for path planning and working orientation of the mountain pepper harvester; the orientation vector of the left side of the mountain pepper harvester in the embodiment represents the direction of the mountain pepper harvester extending to the working center line in the left side space; the first working cost in the application represents the working loss of the left side of the mountain pepper harvester, which is characterized by the included angle between the orientation vector of the working center line of the left side of the mountain pepper harvester and the orientation vector of the left side of the mountain pepper harvester, and the included angle describes the deviation between the left side pepper row direction and the picking direction of the left side of the mountain pepper harvester; the greater the deviation between the left side pepper row direction and the picking direction of the left side of the mountain pepper harvester, the greater the working loss of the left side of the mountain pepper harvester, and vice versa.

[0097] In some embodiments, the second working cost of the mountain pepper harvester when working on the left side can be determined by the motion constraint point and the working center line of the left side, which can be achieved by the following steps:

[0098] The right side working point set is extracted from the working center line of the right side, and the nearest right side working point and the farthest right side working point in the right side working point set from the current position of the mountain pepper harvester are extracted;

[0099] The orientation vector of the right side pepper row of the mountain pepper harvester is determined by the nearest right side working point and the farthest right side working point;

[0100] The orientation vector of the right side of the mountain pepper harvester is determined according to the farthest right side working point and the motion constraint point;

[0101] The vector dot product calculation is performed on the orientation vector of the right side pepper row of the mountain pepper harvester and the orientation vector of the right side of the mountain pepper harvester, and the second working cost of the mountain pepper harvester when working on the right side is obtained.

[0102] In specific implementation, first, a right side operation point set is extracted from the operation center line of the right side, and the right side operation point closest to the current position of the mountain pepper harvester and the farthest right side operation point in the right side operation point set are extracted, that is: all right side operation points are extracted from the operation center line of the right side according to the preset acquisition step length to obtain the right side operation point set, and the distance between each right side operation point in the right side operation point set and the current position of the mountain pepper harvester is calculated by Euclidean distance, and the right side operation point set closest to the current position of the mountain pepper harvester is extracted. The right side working point and the farthest right side working point; secondly, the nearest right side working point and the farthest right side working point are calculated by vector subtraction to obtain the orientation vector of the pepper row on the right side of the mountain pepper harvester; then, the farthest right side working point and the center position are calculated according to vector subtraction to obtain the orientation vector of the right side of the mountain pepper harvester; finally, the orientation vector of the pepper row on the right side of the mountain pepper harvester is calculated by vector dot product with the orientation vector of the right side of the mountain pepper harvester to obtain the second operation cost when the mountain pepper harvester is operating on the right side.

[0103] It should be noted that, in this embodiment, vector subtraction refers to the subtraction between two coordinate vectors, which will not be repeated here; in this embodiment, the right side operation point set represents a combination of multiple right side operation points, and the right side operation point represents the position point on the operation center line when the mountain pepper harvester is operating on the right side; in this embodiment, the orientation vector of the pepper row on the right side of the mountain pepper harvester represents the extension direction of the pepper row on the right side of the mountain pepper harvester in space, which is often expressed in the form of a vector. The orientation vector is used as a direction guide for the pepper row on the right side of the mountain pepper harvester, helping the harvester to determine the route of the right operating range, and is for path planning and operation orientation of the mountain pepper harvester. Provide reference; in this embodiment, the orientation vector of the right side of the mountain pepper harvester represents the direction in which the right side of the mountain pepper harvester extends toward the operation center line in space; in this application, the second operation cost represents the operation loss amount of the right side of the mountain pepper harvester, that is, it is characterized by the angle between the orientation vector of the operation center line of the right side of the mountain pepper harvester and the orientation vector of the right side of the mountain pepper harvester. The angle describes the degree of deviation between the right side pepper row direction and the picking direction of the right side of the mountain pepper harvester. The greater the degree of deviation between the right side pepper row direction and the picking direction of the right side of the mountain pepper harvester, the greater the operation loss amount of the right side of the mountain pepper harvester, and vice versa.

[0104] In step 104, the travel route of the mountain pepper harvester when picking peppers is subjected to minimum cost constraint by the first job cost and the second job cost, to obtain the constraint control amount of the mountain pepper harvester on the travel route.

[0105] In some embodiments, referring to Figure 3 As shown in the figure, which is an exemplary flow chart for determining the constraint control amount according to some embodiments of the present application, the constraint control amount of the mountain pepper harvester on the travel route can be obtained by subjecting the travel route of the mountain pepper harvester when picking peppers to minimum cost constraint by the first job cost and the second job cost, which can be achieved by the following steps:

[0106] First, in step 1041, the reference job center line of the left side and the reference job center line of the right side of the mountain pepper harvester when picking peppers are obtained;

[0107] Secondly, in step 1042, the travel route reference line of the mountain pepper harvester is generated according to the reference job center line of the left side and the reference job center line of the right side;

[0108] Then, in step 1043, the first travel constraint route and the second travel constraint route of the mountain pepper harvester are determined based on the first job cost and the first job cost, respectively;

[0109] Finally, in step 1044, the constraint control amount of the mountain pepper harvester on the travel route is determined by the first travel constraint route, the second travel constraint route and the travel route reference line.

[0110] In the specific implementation, first, the reference work center line of the left side edge and the reference work center line of the right side edge of the pepper picking machine are acquired from the pepper harvesting monitoring database when the peppers are picked, and the reference work center line of the left side edge and the reference work center line of the right side edge are set in advance according to the picking range of the picking bucket of the mountain pepper harvesting machine, which is not limited here; second, the reference work center line of the left side edge and the reference work center line of the right side edge are processed by the bilateral least square fitting method in the geometric fitting technology to obtain the reference line of the travel route of the mountain pepper harvesting machine, that is, the average distance between the reference work center line of the left side edge and the reference work center line of the right side edge is calculated, and the points corresponding to the average distance are connected to obtain the reference line of the travel route of the mountain pepper harvesting machine; then, the first travel constraint route and the second travel constraint route of the mountain pepper harvesting machine are determined based on the first work cost and the first work cost, that is, the motion constraint point of the mountain pepper harvesting machine and the preset route step length are acquired, the first constraint unit vector is constructed according to the first work cost, all path points are generated in the direction of the first constraint unit vector from the motion constraint point with the preset route step length, all path points are sequentially connected to obtain the first travel constraint route of the mountain pepper harvesting machine, the second constraint unit vector is constructed according to the second work cost, all path points are generated in the direction of the second constraint unit vector from the motion constraint point with the preset route step length, and all path points are sequentially connected to obtain the second travel constraint route of the mountain pepper harvesting machine; finally, the constraint control quantity of the mountain pepper harvesting machine on the travel route is determined by the first travel constraint route, the second travel constraint route and the reference line of the travel route, that is, the first travel constraint route and the reference line of the travel route are compared according to the Euclidean distance to obtain the left travel deviation of the mountain pepper harvesting machine, the second travel constraint route and the reference line of the travel route are compared according to the Euclidean distance to obtain the right travel deviation of the mountain pepper harvesting machine, the left travel deviation and the right travel deviation are averaged, and the average value calculation result is taken as the constraint control quantity of the mountain pepper harvesting machine on the travel route.

[0111] It should be noted that the minimum cost constraint in the present application represents the process of minimizing the cost of the left side and right side travel routes of the mountain pepper harvester, wherein the first operation cost and the second operation cost constrain the travel route of the mountain pepper harvester during pepper picking, that is, the reference operation center lines of the left side and the right side of the mountain pepper harvester during pepper picking are obtained, the travel route reference line of the mountain pepper harvester is generated according to the reference operation center lines of the left side and the right side, the first travel constraint route and the second travel constraint route of the mountain pepper harvester are respectively determined based on the first operation cost and the first operation cost, and the constraint control amount of the mountain pepper harvester on the travel route is determined through the first travel constraint route, the second travel constraint route and the travel route reference line, that is, the minimum cost constraint of the travel route of the mountain pepper harvester during pepper picking is completed.

[0112] It should also be noted that the travel route reference line in the present embodiment represents the reference travel route of the mountain pepper harvester, the first constraint unit vector in the present embodiment represents the unit vector constructed from the first operation cost, for example, when the first operation cost is θ1, the first constraint unit vector constructed from the first operation cost is (cos(θ1), sin(θ1)), and θ1 represents the first operation cost, the second constraint unit vector in the present embodiment represents the unit vector constructed from the second operation cost, for example, when the first operation cost is θ2, the second constraint unit vector constructed from the second operation cost is (cos(θ2), sin(θ2)), and θ2 represents the second operation cost, the first travel constraint route in the present embodiment represents the constrained travel route of the left side of the mountain pepper harvester, the second travel constraint route in the present embodiment represents the constrained travel route of the right side of the mountain pepper harvester, the preset route step in the present embodiment is a pre-set route step, which can be set according to actual needs, and is not limited here, the left travel deviation amount in the present embodiment represents the deviation degree between the actual constrained travel route of the left side of the mountain pepper harvester and the travel route reference line, the right travel deviation amount in the present embodiment represents the deviation degree between the actual constrained travel route of the right side of the mountain pepper harvester and the travel route reference line, and the constraint control amount in the present application represents the control amount of the motion constraint of the mountain pepper harvester, when the constraint control amount is larger, the control amount of the motion constraint of the mountain pepper harvester is larger, when the constraint control amount is smaller, the control amount of the motion constraint of the mountain pepper harvester is smaller, by calculating the constraint control amount, the path deviation of the mountain pepper harvester can be adjusted in time, so as to optimize the harvesting process, ensure that the mountain pepper harvester remains on the preset operation path, and reduce the instability of the harvesting process.

[0113] In step 105, the pose state of the mountainous chili harvester is constrained based on the constraint control quantity, and the mountainous chili harvester after the pose state constraint is controlled to pick chilies in the target working area.

[0114] In some embodiments, the constraint of the pose state of the mountainous chili harvester based on the constraint control quantity can be achieved by the following steps, that is:

[0115] The pose constraint signal of the mountainous chili harvester is generated from the constraint control quantity, and the mountainous chili harvester is controlled in response to the pose constraint signal to constrain the pose state.

[0116] In specific implementation, the constraint control quantity is input into a control algorithm, such as a proportional-integral-derivative (PID) controller or a fuzzy controller, to generate a corresponding pose constraint signal, which includes a translation instruction for adjusting the lateral position of the mountainous chili harvester and an angle correction instruction for adjusting the direction of travel, to ensure that the actual path of the mountainous chili harvester gradually approaches the reference path. The pose constraint signal is used to correct the pose state of the mountainous chili harvester through the control system driving the mountainous chili harvester to act on the walking mechanism and direction control system of the mountainous chili harvester, until the constraint control quantity is reduced to an acceptable range, so as to align the working center line of the mountainous chili harvester with the target harvesting boundary, thereby achieving precise path alignment and stable working state.

[0117] In specific implementation, the mountainous chili harvester after the pose state constraint is controlled to pick chilies in the target working area, that is, the picking bucket of the mountainous chili harvester after the pose state constraint is controlled by the remote control terminal to vibrate and pick chilies in the target working area.

[0118] In addition, another aspect of the present application, in some embodiments, the present application provides a remote control type mountainous chili harvester, which includes a chili picking control unit, which is described with reference to Figure 4 The figure is a structural schematic diagram of the chili picking control unit according to some embodiments of the present application. The chili picking control unit 200 includes a scanning module 201, a processing module 202, and an execution module 203, which are described as follows:

[0119] The scanning module 201 is mainly used in the present application to scan the target working area when the remote control mountainous chili harvester travels to the target working area;

[0120] The processing module 202 is mainly used in the present application to extract the working center line of the left side and the working center line of the right side of the mountainous chili harvester during chili picking from the scanning information through the lateral distribution distance between each row of chili plants in the target working area and the density distribution characteristics of the chili plants.

[0121] The processing module 202 is further configured to determine a motion constraint point of the mountain pepper harvesting machine under the constraint of the picking boundary, determine a first operation cost of the left side of the mountain pepper harvesting machine based on the motion constraint point and the operation center line of the left side, and determine a second operation cost of the left side of the mountain pepper harvesting machine based on the motion constraint point and the operation center line of the left side.

[0122] In addition, the processing module 202 is further configured to constrain the travel route of the mountain pepper harvesting machine when picking peppers by the first operation cost and the second operation cost, and obtain a constraint control amount of the mountain pepper harvesting machine on the travel route.

[0123] The execution module 203 is mainly configured to constrain the pose state of the mountain pepper harvesting machine based on the constraint control amount, and control the mountain pepper harvesting machine to pick peppers in the target operation area after the pose state is constrained.

[0124] In addition, the present application also provides a computer device, which comprises a memory and a processor, the memory stores code, and the processor is configured to acquire the code and execute the above-mentioned pepper picking method.

[0125] In some embodiments, with reference to Figure 5 The figure is a structural schematic diagram of a computer device for implementing the pepper picking method according to some embodiments of the present application. The pepper picking method in the above-mentioned embodiments can be implemented by the computer device shown in the figure, which comprises at least one processor 301, a communication bus 302, a memory 303 and at least one communication interface 304. Figure 5 The processor 301 can be a general central processing unit (CPU), an application-specific integrated circuit (ASIC) or one or more circuits for controlling the execution of the pepper picking method in the present application.

[0126] The communication bus 302 can be used to transmit information between the above-mentioned components.

[0127]

[0128] ​The memory 303 can be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM), or other type of dynamic storage device that can store information and instructions, and can be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disk storage, a magnetic disk storage or other magnetic storage devices, or any other medium capable of storing desired program code in the form of instructions or data structures and that can be accessed by a computer, but is not limited to this. The memory 303 can exist independently and be connected to the processor 301 through the communication bus 302. The memory 303 can also be integrated with the processor 301.

[0129] The memory 303 is configured to store program codes for implementing the solutions of the present application, and the processor 301 is configured to control the execution of the program codes. The processor 301 is configured to execute the program codes stored in the memory 303. The program codes can include one or more software modules. The determination of the pepper picking method in the above embodiments can be implemented by one or more software modules in the program codes of the processor 301 and the memory 303.

[0130] The communication interface 304 is configured to communicate with other devices or communication networks, such as an Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc., using any transceiver-like device.

[0131] In specific implementations, as an example, the computer device can include multiple processors, each of which can be a single-CPU processor or a multi-CPU processor. The processor herein can refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0132] The computer device described above can be a general-purpose computer device or a special-purpose computer device. In a specific implementation, the computer device can be a desktop computer, a laptop computer, a network server, a personal digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. The embodiments of the present application do not limit the type of the computer device.

[0133] In addition, the present application also provides a computer readable storage medium, which stores a computer program. The computer program is executed by a processor to implement the pepper picking method.

[0134] Although the preferred embodiments of the present application have been described, those skilled in the art who are familiar with the basic inventive concept can make additional changes and modifications to the embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0135] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A pepper picking method for picking peppers using a remote-controlled mountain pepper harvester, characterized in that: The steps include: When the remote-controlled mountain pepper harvester drives to the target operation area, it scans the target operation area; Extracting the left and right operating center lines of the mountain pepper harvester during pepper picking from the scanning information based on the lateral distribution spacing between rows of pepper plants and the density distribution characteristics of the pepper plants in the target operating area; Determine a motion constraint point of the mountain pepper harvester under the picking boundary constraint, determine a first operation cost of the mountain pepper harvester when operating on the left side based on the motion constraint point and the operation centerline of the left side, and determine a second operation cost of the mountain pepper harvester when operating on the right side based on the motion constraint point and the operation centerline of the right side; Using the first operation cost and the second operation cost, a minimum cost constraint is imposed on the route of the mountain pepper harvester when picking peppers, thereby obtaining a constraint control amount of the mountain pepper harvester on the route; The posture state of the mountain pepper harvester is constrained based on the constraint control amount, and the mountain pepper harvester with the constrained posture state is controlled to pick peppers in the target operation area.

2. The method according to claim 1, wherein The operation center line on the left side and the operation center line on the right side of the mountain pepper harvester when picking peppers are extracted from the scanning information by the lateral distribution spacing between the rows of pepper plants in the target operation area and the density distribution characteristics of the pepper plants. Specifically, the following are included: extracting a current scan image frame from the scan information of the target operation area; Performing density cluster analysis on the scanned image frames to obtain clustered blocks of pepper plants in the target operation area, and using the clustered blocks as pepper rows in the target operation area to obtain multiple pepper rows in the target operation area; Determining the pepper row centerline of each pepper row based on the density distribution characteristics of pepper plants in the target operation area; Calculate the spacing between the center lines of each pepper row to obtain the lateral distribution spacing between pepper plants in each row in the target operation area; Extracting the effective lateral distribution spacing when the mountain pepper harvester is picking peppers from all lateral distribution spacings based on the bucket length of the mountain pepper harvester picking bucket; Extracting the center line of the first pepper row and the center line of the second pepper row corresponding to the effective transverse distribution spacing; The first pepper row center line and the second pepper row center line are used as the left side operation center line and the right side operation center line of the mountain pepper harvester when picking peppers.

3. The method according to claim 2, wherein Determining the center line of each pepper row based on the density distribution characteristics of pepper plants in the target operation area specifically includes: Extracting pixel values ​​of each pepper row in the target operation area to obtain a pixel value set of each pepper row; Selecting a pepper row as a selected pepper row, and determining a density distribution feature of pepper plants in the selected pepper row in a target operation area according to a set of pixel values ​​of the selected pepper row; Remove disturbed pixels from the selected pepper row using the density distribution feature to obtain an interference-resistant selected pepper row; Fitting all geometric center points of the selected anti-interference pepper row to obtain a pepper row center line of the selected pepper row; Continue determining the pepper row center lines for the remaining pepper rows.

4. The method according to claim 1, wherein Determining the first operation cost of the mountain pepper harvester when operating on the left side based on the motion constraint point and the operation center line of the left side specifically includes: Extracting a left side operation point set from the operation center line of the left side, and extracting the left side operation point closest to the current position of the mountain pepper harvester and the left side operation point farthest from the left side operation point set; Determining the orientation vector of the left side pepper row of the mountain pepper harvester by the nearest left side operation point and the farthest left side operation point; Determining a direction vector of the left side of the mountain pepper harvester according to the farthest left side operation point and the motion constraint point; A vector dot product calculation is performed on the orientation vector of the pepper row on the left side of the mountain pepper harvester and the orientation vector of the left side of the mountain pepper harvester to obtain the first operation cost when the mountain pepper harvester operates on the left side.

5. The method according to claim 1, wherein Determining the second operation cost of the mountain pepper harvester when operating on the right side through the motion constraint point and the operation center line of the right side specifically includes: Extracting a right side operation point set from the operation center line of the right side, and extracting the right side operation point closest to the current position of the mountain pepper harvester and the farthest right side operation point from the right side operation point set; Determining the orientation vector of the pepper row on the right side of the mountain pepper harvester through the nearest right side operation point and the farthest right side operation point; Determining a direction vector of the right side of the mountain pepper harvester according to the farthest right side operation point and the motion constraint point; A vector dot product calculation is performed on the orientation vector of the pepper row on the right side of the mountain pepper harvester and the orientation vector of the right side of the mountain pepper harvester to obtain a second operation cost when the mountain pepper harvester operates on the right side.

6. The method according to claim 1, wherein The remote-controlled mountain pepper harvester is a drum-type separation pepper harvester.

7. The method according to claim 1, wherein Scan the target work area through the camera.

8. A remote-controlled mountain pepper harvester, comprising a pepper picking control unit, characterized in that: The pepper picking control unit includes: A scanning module is used to scan the target operating area when the remote-controlled mountain pepper harvester drives to the target operating area; a processing module for extracting from the scanning information the left side operation center line and the right side operation center line of the mountain pepper harvester when picking peppers based on the lateral distribution spacing between the rows of pepper plants and the density distribution characteristics of the pepper plants in the target operation area; The processing module is further configured to determine a motion constraint point of the mountain pepper harvester under the picking boundary constraint, determine a first operation cost of the mountain pepper harvester when operating on the left side based on the motion constraint point and the operation centerline of the left side, and determine a second operation cost of the mountain pepper harvester when operating on the right side based on the motion constraint point and the operation centerline of the right side; The processing module is further configured to perform a minimum cost constraint on the route of the mountain pepper harvester when picking peppers based on the first operation cost and the second operation cost, thereby obtaining a constraint control amount of the mountain pepper harvester on the route; An execution module is used to constrain the posture state of the mountain pepper harvester based on the constraint control amount, and control the mountain pepper harvester with constrained posture state to pick peppers in the target working area.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores codes, and the processor is configured to obtain the codes and execute the pepper picking method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the pepper picking method according to any one of claims 1 to 7 is implemented.

Images