Chinese cabbage row-control harvester based on machine vision and control method

Abstract

The invention provides a Chinese cabbage row-control harvester based on machine vision and a control method. The Chinese cabbage row-control harvester comprises a chassis, a row-control mechanism, a Chinese cabbage harvesting and conveying mechanism, a visual detection module, a motion control module and an edge calculation module. The Chinese cabbage harvesting and conveying mechanism is used for backwards conveying harvested Chinese cabbages; the Chinese cabbage harvesting and conveying mechanism is mounted on the chassis, and the row control mechanism is used for adjusting the Chinese cabbage harvesting and conveying mechanism for row control; the visual detection module is installed on a rack at the front end of the Chinese cabbage harvesting and conveying mechanism and used for collecting field images and transmitting the field images to the edge calculation module, and the edge calculation module is connected with the visual detection module and the motion control module; the edge calculation module processes the image acquired by the visual detection module to obtain offset data and outputs the offset data to the motion control module, and the motion control module controls the row control mechanism to adjust the Chinese cabbage harvesting and conveying mechanism to perform row control. By means of the two row control modes of active row control and chassis floating row control, the row control precision can be improved, and the row control efficiency can be improved.

Description

Technical Field

[0001] The invention belongs to the technical field of agricultural machinery, and in particular relates to a Chinese cabbage row harvester based on machine vision and a control method thereof. Background Art

[0002] Existing Chinese cabbage harvesters have technical defects in row harvesting and ground leveling. Especially when harvesting in ridge planting mode, the harvester needs to travel on the ridge. At this time, due to the incoordination between the ridge spacing and the wheelbase of the harvester, and the mismatch between the row spacing and the number of rows on the ridge, the machine body may tilt during harvesting. Excessive inclination may even cause the harvesting parts to sink into the soil, causing damage to the machine. Secondly, due to the tilt of the machine body, it is more difficult for the driver to keep driving in a straight line during single-row harvesting operations; the tilt of the harvesting parts causes the clamping part of the cabbage to change, which will also cause a decrease in the success rate of harvesting and an increase in the damage rate of the harvested cabbage. Summary of the invention

[0003] In view of the above technical problems, the present invention provides a Chinese cabbage row harvester and a control method based on machine vision.

[0004] The present invention can realize row-by-row harvesting of ridge-planted cabbages, and can improve row-by-row precision and efficiency through two row-by-row modes, namely, active row-by-row and chassis floating row-by-row.

[0005] The present invention is conducive to realizing harvesting operations of planting patterns with different ridge distances, ridge widths, number of rows, row spacings, etc., and enhances the adaptability of single-row harvesters to different planting patterns. The automatic row alignment of the device can cope with the problems of different growth cycles and uneven straightness of cabbage, which is conducive to realizing continuous and stable extraction and stable feeding.

[0006] Note that the description of these objectives does not prevent the existence of other objectives. One embodiment of the present invention does not need to achieve all of the above objectives. Objectives other than the above objectives can be extracted from the description of the specification, drawings, and claims.

[0007] The present invention achieves the above technical objectives through the following technical means.

[0008] A Chinese cabbage row harvester based on machine vision, comprising a chassis, a row mechanism, a cabbage harvesting and conveying mechanism, a visual detection module, a motion control module and an edge computing module;

[0009] The cabbage harvesting and conveying mechanism is used to harvest the cabbage and convey it backwards;

[0010] The cabbage harvesting and conveying mechanism is installed on the chassis, and the row-aligning mechanism is used to adjust the cabbage harvesting and conveying mechanism to align the rows;

[0011] The visual detection module is installed on the front end frame of the cabbage harvesting and conveying mechanism, and is used to collect field images and transmit them to the edge computing module. The edge computing module is respectively connected to the visual detection module and the motion control module, and the motion control module is connected to the row alignment mechanism; after the edge computing module processes the image collected by the visual detection module, it obtains the offset data and outputs it to the motion control module, and the motion control module controls the row alignment mechanism to adjust the cabbage harvesting and conveying mechanism for row alignment.

[0012] In the above scheme, the row-aligning mechanism includes a floating row-aligning mechanism and an active row-aligning mechanism. The floating row-aligning mechanism is installed on the chassis, and is used to make the cabbage harvesting and conveying mechanism move laterally along the chassis to align the rows; the active row-aligning mechanism is installed on the cabbage harvesting and conveying mechanism, and is used to change the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism to perform row-aligning feeding.

[0013] Furthermore, the floating alignment mechanism comprises a fixed linear bearing group, a dynamic linear bearing group, a linear drive mechanism and a connecting frame;

[0014] The chassis includes a chassis frame, a movable frame and a connecting frame; the connecting frame is installed above one side of the movable frame, and the connecting frame is connected to the cabbage harvesting and conveying mechanism;

[0015] The movable frame is installed on the chassis frame, the dynamic linear bearing group is installed on the movable frame, the fixed linear bearing group is installed on one side of the chassis frame, and the other side of the movable frame is located on the fixed linear bearing group. One end of the linear drive mechanism is installed below one side of the movable frame through a connecting frame, and one end of the linear drive mechanism is connected to the dynamic linear bearing group to drive the dynamic linear bearing group to drive the movable frame to move horizontally left and right along the chassis frame, thereby driving the cabbage harvesting and conveying mechanism to float in parallel, and the linear drive mechanism is connected to the motion control module.

[0016] Furthermore, two groups of active row alignment mechanisms are symmetrically arranged on both sides of the front end of the bracket of the cabbage harvesting and conveying mechanism, and each group of active row alignment mechanisms includes a high-torque steering gear, a steering gear mounting seat and a coupling;

[0017] The servo mounting seat is installed on the front end of the bracket of the cabbage harvesting and conveying mechanism, the high-torque servo is installed on the servo mounting seat, the output shaft of the high-torque servo is connected to the driven wheel mounting frame through a coupling, and the driven wheel mounting frame is provided with a driven wheel; the high-torque servo is used to adjust the angles of the two driven wheel mounting frames, thereby changing the angles of the two conveyor belts of the cabbage harvesting and conveying mechanism for row feeding; the high-torque servo is connected to the motion control module.

[0018] The above scheme also includes a two-stage conveying and collecting mechanism; the two-stage conveying and collecting mechanism includes a first-stage retractable horizontal conveying mechanism, a second-stage vertical conveying belt mechanism, a conveying belt fixing frame and a collecting frame;

[0019] The first-level retractable transverse conveyor belt mechanism is located at the end of the cabbage harvesting conveyor mechanism, the second-level vertical conveyor belt mechanism is located at the end of the first-level retractable transverse conveyor belt mechanism, the first-level retractable transverse conveyor belt mechanism and the second-level vertical conveyor belt mechanism are vertically arranged in space, and the collection frame is located behind the second-level vertical conveyor belt mechanism;

[0020] The first-stage retractable transverse conveyor belt mechanism comprises a conveyor belt front section mounting frame, a conveyor belt rear section mounting frame, a conveyor belt, a belt surface tensioning mechanism and a plurality of conveyor belt rollers;

[0021] The front section mounting frame of the conveyor belt and the rear section mounting frame of the conveyor belt are arranged in parallel and connected by sliding through a U-shaped slide groove; a plurality of conveyor belt rollers are mounted on the front section conveyor belt roller mounting frame and the rear section conveyor belt roller mounting frame, the conveyor belt is mounted on the front section conveyor belt mounting frame and the rear section conveyor belt mounting frame, and the belt surface tensioning mechanism is mounted on the rear section conveyor belt mounting frame and contacts the conveyor belt;

[0022] A plurality of conveyor belt fixing frames are provided below the first-stage retractable transverse conveyor mechanism and the second-stage vertical conveyor belt mechanism, wherein the bottom of the conveyor belt front section mounting frame is connected to the connecting frame through the conveyor belt fixing frame, and the conveyor belt rear section mounting frame is connected to the chassis frame through the conveyor belt fixing frame;

[0023] The connecting frame can drive the front section mounting frame of the conveyor belt to move left and right, and the belt surface tensioning mechanism is used to keep the conveyor belt in a tensioned state.

[0024] In the above scheme, the visual inspection module includes an industrial camera, a mounting rod, a camera mounting seat and an angle adjustment plate;

[0025] The industrial camera is installed on a camera mounting seat, the camera mounting seat is connected to the upper end of the mounting rod through an angle adjustment plate, and the lower end of the mounting rod is connected to the upper part of the bracket of the cabbage harvesting and conveying mechanism; the industrial camera is connected to the edge computing module.

[0026] Furthermore, one side of the angle adjustment plate is connected to the mounting rod via a rotating shaft, and the other side is provided with an arc-shaped slide groove, which is connected to the mounting rod via a bolt. Changing the position of the bolt in the arc-shaped slide groove changes the shooting angle of the industrial camera.

[0027] In the above scheme, the four corners of the chassis are respectively provided with a leveling mechanism, and each of the leveling mechanisms includes a connecting rod, a hinge, a leveling hydraulic rod, a tire connecting flange and a posture sensor;

[0028] One end of the connecting rod is connected to the tire connecting flange, and the other end is connected to the chassis frame through a hinge. One end of the leveling hydraulic rod is connected to the tire connecting flange, and the other end is connected to the chassis frame through a hinge. The attitude sensor is installed at the center of the chassis frame, the attitude sensor is connected to the motion control module, and the leveling hydraulic rod is connected to the motion control module. The attitude sensor is used to detect the attitude angle of the chassis frame and transmit it to the motion control module. The motion control module controls the leveling hydraulic rod to level the chassis frame according to the attitude angle.

[0029] A method for controlling a Chinese cabbage row harvester based on machine vision comprises the following steps:

[0030] The visual detection module collects field images and transmits them to the edge computing module; after processing the images collected by the visual detection module, the edge computing module obtains offset data and outputs it to the motion control module, and the motion control module compares the obtained offset data with a preset threshold value. When the offset data is greater than or equal to the preset displacement threshold value, the motion control module controls the floating row alignment mechanism of the row alignment mechanism so that the cabbage harvesting and conveying mechanism moves laterally along the chassis to align the rows; when the offset data is less than the preset displacement threshold value, the motion control module controls the active row alignment mechanism of the row alignment mechanism to change the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism to feed the rows.

[0031] In the above scheme, the displacement threshold p is calculated according to the following formula:

[0032]

[0033] The mathematical model of the floating row alignment mechanism of the row alignment mechanism that makes the cabbage harvesting and conveying mechanism move laterally along the chassis to match the row speed is as follows:

[0034]

[0035] Among them, d1 and d3 are the outputs of the edge computing module after processing the acquired images, d1 is the longitudinal distance between the two cabbages to be harvested, d3 is the lateral offset distance between the two cabbages, d2 is the distance between the position of the visual detection module and the cabbages to be harvested in the driving direction, v1 is the operating speed of the harvester, t1 is the arrival time from the position of the visual detection module to the cabbages to be harvested, t2 is the time for the edge computing module to receive the image, process the image and obtain the output, and v2 is the optimal speed to reach the optimal harvesting position in theory, that is, the driving speed of the linear drive mechanism;

[0036] The speed matching calculation method of the active row-directing mechanism changing the angle of the two conveyor belts of the cabbage harvesting conveying mechanism to feed the cabbage in rows is as follows:

[0037]

[0038] Among them, d4 is the lateral distance between the visual inspection module and the cabbage to be harvested, the output obtained by the edge computing module, d5 is the center distance between the two large torque servos on the left and right of the front end of the bracket of the cabbage harvesting and conveying mechanism, L1 is the center distance from the output shaft of the large torque servo to the driven wheel axle of the conveyor belt, w is the average angular velocity of the large torque servo during the operation, and θ is the angle between the connecting line between the driving shaft of the large torque servo and the driven wheel axle and the conveyor belt frame.

[0039] Compared with the prior art, the present invention has the following beneficial effects:

[0040] 1. The present invention can realize row-by-row harvesting of ridge-planted cabbages, and can improve row-by-row accuracy and efficiency through two row-by-row modes: active row-by-row and chassis floating row-by-row.

[0041] 2. The automatic leveling chassis of the present invention adopts four hydraulic rods for adjustment, and can be ridden on the ridge for harvesting. It is also suitable for planting patterns with large ridge spacing and multiple rows, and has more practicality. The leveling mechanism keeps the harvesting platform stable in real time, reducing damage during harvesting.

[0042] 3. The visual detection device of the present invention can realize the rapid measurement of the lateral offset of the cabbage row by combining a monocular camera with an optimized target detection algorithm. The visual detection method has stronger robustness than the sensor.

[0043] 4. The conveying and collecting device of the present invention adopts a two-level conveying structure, and designs a retractable conveyor belt structure for the floating alignment of the chassis, and designs a pulley tensioning mechanism to ensure smooth conveying and collecting operations during the alignment process.

[0044] Note that the description of these effects does not prevent the existence of other effects. One embodiment of the present invention does not necessarily have all of the above effects. Effects other than the above can be clearly seen and extracted from the description of the specification, drawings, claims, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a schematic structural diagram of a Chinese cabbage row harvester based on machine vision according to an embodiment of the present invention.

[0046] Figure 2 It is a structural schematic diagram of a chassis, a row-aligning mechanism, a cabbage harvesting and conveying mechanism, and a visual inspection module according to one embodiment of the present invention.

[0047] Figure 3 It is a structural schematic diagram of a floating alignment mechanism according to an embodiment of the present invention.

[0048] Figure 4 It is a schematic structural diagram of a two-stage conveying and collecting mechanism according to one embodiment of the present invention.

[0049] Figure 5 It is a schematic structural diagram of a conveyor belt fixing frame according to an embodiment of the present invention.

[0050] Figure 6 It is a structural schematic diagram of a leveling mechanism according to an embodiment of the present invention.

[0051] Figure 7 It is a structural schematic diagram of an active alignment mechanism and a visual detection module according to an embodiment of the present invention.

[0052] Figure 8 It is a flow chart of a control method for a Chinese cabbage row harvester based on machine vision according to an embodiment of the present invention.

[0053] Fig. 9 It is a schematic diagram of row operation according to one embodiment of the present invention.

[0054] Fig.10 Schematic diagram of angle θ according to one embodiment of the present invention.

[0055] In the figure, 1. tractor; 2. two-stage conveying and collecting mechanism; 201. first-stage retractable horizontal conveying mechanism; 202. second-stage vertical conveyor belt mechanism; 203. collecting frame; 204. conveyor belt front section mounting frame; 205. conveyor belt rear section mounting frame; 206. conveyor belt; 207. conveyor belt fixing frame; 208. belt surface tensioning mechanism; 209. conveyor belt roller; 3. chassis; 301. chassis frame; 302. three-point suspension connecting frame; 303. connecting rod; 304. wheel; 305. hinge; 306. leveling hydraulic rod; 307. leveling hydraulic rod; 308. attitude sensor; 4. line-up mechanism ; 401. Floating alignment mechanism; 402. Fixed linear bearing group; 403. Fixed linear bearing group; 404. Linear drive mechanism; 405. Active alignment mechanism; 406. High-torque servo; 407. Servo mounting seat; 408. Coupling; 409. Connecting frame; 5. Cabbage harvesting and conveying mechanism; 501 bracket; 502. Drive motor; 503. Conveyor belt; 504. Driven wheel; 505. Driven wheel mounting frame; 6. Visual inspection module; 601. Industrial camera; 602. Mounting rod; 603. Camera mounting seat; 604. Angle adjustment plate; 7. Motion control module; 8. Edge computing module. DETAILED DESCRIPTION

[0056] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present invention, and should not be construed as limiting the present invention.

[0057] In the description of the present invention, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "front", "back", "left", "right", "up", "down", "axial", "radial", "vertical", "horizontal", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the present invention, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.

[0058] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be an indirect connection through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0059] Figure 1 The figure shows a preferred embodiment of the Chinese cabbage row harvester based on machine vision, which includes a chassis 3, a row mechanism 4, a cabbage harvesting and conveying mechanism 5, a visual detection module 6, a motion control module 7 and an edge computing module 8;

[0060] The cabbage harvesting and conveying mechanism 5 is used to harvest the cabbage and lift and convey it backwards;

[0061] The cabbage harvesting and conveying mechanism 5 is installed on the chassis 3, and the row-aligning mechanism 4 is used to adjust the cabbage harvesting and conveying mechanism 5 to align the rows;

[0062] The visual detection module 6 is installed on the front end frame of the cabbage harvesting and conveying mechanism 5, and is used to collect field images and transmit them to the edge computing module 8. The edge computing module 8 is respectively connected to the visual detection module 6 and the motion control module 7, and the motion control module 7 is connected to the row alignment mechanism 4; after the edge computing module 8 processes the image collected by the visual detection module 6, it obtains the offset data and outputs it to the motion control module 7. The motion control module 7 controls the row alignment mechanism 4 to adjust the cabbage harvesting and conveying mechanism 5 for row alignment.

[0063] like Figure 2 As shown, the row-aligning mechanism 4 includes a floating row-aligning mechanism 401 and an active row-aligning mechanism 405. The floating row-aligning mechanism 401 is installed on the chassis 3, and is used to make the cabbage harvesting and conveying mechanism 5 move horizontally along the chassis 3 to align the rows; the active row-aligning mechanism 405 is installed on the cabbage harvesting and conveying mechanism 5, and is used to change the angle of the two conveyor belts 503 of the cabbage harvesting and conveying mechanism 5 to perform row-aligning feeding, and the conveyor belts 503 are driven by a driving motor 502.

[0064] like Figure 3 As shown, the floating alignment mechanism 401 includes a fixed linear bearing group 402, a dynamic linear bearing group 403, a linear drive mechanism 404 and a connecting frame 409;

[0065] The chassis 3 includes a chassis frame 301, a movable frame 309 and a connecting frame 310; the connecting frame 310 is installed above one side of the movable frame 309, and the connecting frame 310 is connected to the cabbage harvesting and conveying mechanism 5;

[0066] The movable frame 309 is installed on the chassis frame 301, the dynamic linear bearing group 403 is installed on the movable frame 309, the fixed linear bearing group 402 is installed on one side of the chassis frame 301, and the other side of the movable frame 309 is located on the fixed linear bearing group 402. One end of the linear drive mechanism 404 is installed below one side of the movable frame 309 through the connecting frame 409, and one end of the linear drive mechanism 404 is connected to the dynamic linear bearing group 403, which is used to drive the dynamic linear bearing group 403 to drive the movable frame 309 to move horizontally left and right along the chassis frame 301, thereby driving the cabbage harvesting and conveying mechanism 5 to float in parallel. The linear drive mechanism 404 is connected to the motion control module 7.

[0067] Preferably, a three-point suspension connecting frame 302 is provided at the front of the chassis frame 301, and the three-point suspension connecting frame 302 is connected to the three-point suspension pin of the tractor 1 to achieve rapid disassembly and assembly for use in traction and harvesting operations.

[0068] like Figure 7 As shown, the two groups of active alignment mechanisms 405 are symmetrically arranged on both sides of the front end of the bracket 501 of the cabbage harvesting and conveying mechanism 5, and each group of active alignment mechanisms 405 includes a large torque steering gear 406, a steering gear mounting seat 407 and a coupling 408;

[0069] The servo mounting seat 407 is installed on the front end of the bracket 501 of the cabbage harvesting and conveying mechanism 5, and the high-torque servo 406 is installed on the servo mounting seat 407. The output shaft of the high-torque servo 406 is connected to the driven wheel mounting frame 505 through a coupling 408, and the driven wheel mounting frame 505 is provided with a driven wheel 504; the high-torque servo 406 is used to adjust the angles of the two driven wheel mounting frames 505, thereby changing the angles of the two conveyor belts of the cabbage harvesting and conveying mechanism 5 for row feeding; the high-torque servo 406 is connected to the motion control module 7.

[0070] like Figure 2 , Figure 4 and Figure 5 As shown, it also includes a two-stage conveying and collecting mechanism 2; the two-stage conveying and collecting mechanism 2 includes a first-stage telescopic transverse conveying mechanism 201, a second-stage vertical conveying belt mechanism 202, a conveying belt fixing frame 207 and a collecting frame 203;

[0071] The first-level retractable transverse conveyor belt mechanism 201 is located at the end of the cabbage harvesting conveyor mechanism 5, and the second-level vertical conveyor belt mechanism 202 is located at the end of the first-level retractable transverse conveyor belt mechanism 201. The first-level retractable transverse conveyor belt mechanism 201 and the second-level vertical conveyor belt mechanism 202 are vertically arranged in space, and the collection frame 203 is located behind the second-level vertical conveyor belt mechanism 202;

[0072] The first-stage retractable transverse conveyor belt mechanism 201 includes a conveyor belt front section mounting frame 204, a conveyor belt rear section mounting frame 205, a conveyor belt 206, a belt surface tensioning mechanism 208 and a plurality of conveyor belt rollers 209;

[0073] The conveyor belt front section mounting frame 204 and the conveyor belt rear section mounting frame 205 are arranged in parallel and connected by sliding through a U-shaped slideway; a plurality of conveyor belt rollers 209 are mounted on the conveyor belt front section roller mounting frame 204 and the conveyor belt rear section roller mounting frame 205; a conveyor belt 206 is mounted on the conveyor belt front section mounting frame 204 and the conveyor belt rear section mounting frame 205; a belt tensioning mechanism 208 is mounted on the conveyor belt rear section mounting frame 205 and contacts the conveyor belt 206;

[0074] A plurality of conveyor belt fixing frames 207 are provided below the first-stage telescopic transverse conveyor mechanism 201 and the second-stage vertical conveyor belt mechanism 202, wherein the bottom of the conveyor belt front section mounting frame 204 is connected to the connecting frame 310 through the conveyor belt fixing frame 207, and the conveyor belt rear section mounting frame 205 is connected to the chassis frame 301 through the conveyor belt fixing frame 207;

[0075] The connecting frame 310 can drive the conveyor belt front section mounting frame 204 to move left and right, and the belt surface tensioning mechanism 208 is used to keep the conveyor belt 206 in a tensioned state.

[0076] The two-stage conveying and collecting mechanism 2 is beneficial to maintaining stable conveying and framing of the harvested cabbages during the rowing process.

[0077] like Figure 7 As shown, the visual inspection module 6 includes an industrial camera 601, a mounting rod 602, a camera mounting seat 603 and an angle adjustment plate 604;

[0078] The industrial camera 601 is installed on a camera mounting base 603, and the camera mounting base 603 is connected to the upper end of the mounting rod 602 through an angle adjustment plate 604, and the lower end of the mounting rod 602 is connected to the upper part of the bracket 501 of the cabbage harvesting and conveying mechanism 5; the industrial camera 601 is connected to the edge computing module 8.

[0079] Preferably, the visual inspection module 6 is located at the center of the front end bracket 501 of the cabbage harvesting and conveying mechanism 5 .

[0080] like Figure 7 As shown, one side of the angle adjustment plate 604 is connected to the mounting rod 602 via a rotating shaft, and the other side is provided with an arc-shaped slide groove, which is connected to the mounting rod 602 via bolts. Changing the position of the bolts in the arc-shaped slide groove changes the shooting angle of the industrial camera 601.

[0081] like Figure 6 As shown, the four corners of the chassis 3 are respectively provided with a leveling mechanism, and each of the leveling mechanisms includes a connecting rod 303, a hinge 305, a leveling hydraulic rod 306, a tire connecting flange 307 and a posture sensor 308;

[0082] One end of the connecting rod 303 is connected to the tire connection flange 307, and the other end is connected to the chassis frame 301 through a hinge 305. One end of the leveling hydraulic rod 306 is connected to the tire connection flange 307, and the other end is connected to the chassis frame 301 through a hinge 305. The posture sensor 308 is installed at the center of the chassis frame 301, the posture sensor 308 is connected to the motion control module 7, and the leveling hydraulic rod 306 is connected to the motion control module 7. The posture sensor 308 is used to detect the posture angle of the chassis frame 301 and transmit it to the motion control module 7. The motion control module 7 controls the leveling hydraulic rod 306 to level the chassis frame 301 according to the posture angle to ensure that the cabbage harvesting and conveying mechanism 5 is level with the ground. Wheels 304 are provided at the bottom of the chassis frame 301.

[0083] Preferably, the linear drive mechanism 404 is equipped with a lateral displacement sensor, which is connected to the motion control module 7 and can measure the extension amount of the linear drive mechanism 404 in real time and feed back to the motion control module 7 .

[0084] Preferably, the four leveling hydraulic rods 306 are equipped with displacement sensors, which are connected to the motion control module 7 , and provide real-time feedback of the extension amount of the leveling hydraulic rods 306 to the motion control module 7 during the leveling process.

[0085] like Figure 8 As shown, a method for controlling a Chinese cabbage row harvester based on machine vision comprises the following steps:

[0086] The visual detection module 6 collects field images and transmits them to the edge computing module 8; after the edge computing module 8 processes the images collected by the visual detection module 6, it obtains the offset data and outputs it to the motion control module 7. The motion control module 7 compares the obtained offset data with the preset threshold value. When the offset data is greater than or equal to the preset displacement threshold value, the motion control module 7 controls the floating row alignment mechanism 401 of the row alignment mechanism 4 to make the cabbage harvesting and conveying mechanism 5 move horizontally along the chassis 3 in rows; when the offset data is less than the preset displacement threshold value, the motion control module 7 controls the active row alignment mechanism 405 of the row alignment mechanism 4 to change the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism 5 to feed the rows, align the front end of the cabbage harvesting and conveying mechanism 5 with the cabbage to be harvested, and the cabbage is pulled up and lifted to the two-stage conveying and collecting mechanism 2, and is transported by the first-stage retractable horizontal conveying mechanism 201 and the second-stage vertical conveyor belt mechanism 202, and the cabbage reaches the collection frame 203.

[0087] like Fig. 9 and 10 As shown, the displacement threshold p is calculated according to the following formula:

[0088]

[0089] The mathematical model of the floating row alignment mechanism 401 of the row alignment mechanism 4 making the cabbage harvesting and conveying mechanism 5 move laterally along the chassis 3 to match the speed of the row alignment is as follows:

[0090]

[0091] Wherein, d1 and d3 are the outputs after the edge computing module 8 processes the acquired images, d1 is the longitudinal spacing between the two cabbages to be harvested, d3 is the lateral offset distance between the two cabbages, d2 is the distance between the position of the visual detection module 6 and the cabbages to be harvested in the driving direction, v1 is the operating speed of the harvester, t1 is the arrival time from the position of the visual detection module 6 to the cabbages to be harvested, t2 is the time taken by the edge computing module 8 to receive the image, process the image and obtain the output, and v2 is the optimal speed for theoretically reaching the optimal harvesting position, i.e., the driving speed of the linear drive mechanism 404;

[0092] The active row-aligning mechanism 405 of the row-aligning mechanism 4 changes the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism 5 to perform row-aligning feeding speed matching calculation method as follows:

[0093]

[0094] Among them, d4 is the lateral distance between the visual inspection module 6 and the cabbage to be harvested, the output obtained by the edge computing module 8, d5 is the center distance between the two large torque servos 406 on the left and right front ends of the bracket 501 of the cabbage harvesting and conveying mechanism 5, L1 is the center distance from the output shaft of the large torque servo 406 to the axle of the conveyor belt driven wheel 504, w is the average angular velocity of the large torque servo 406 during the row process, and θ is the angle between the connecting line between the driving shaft of the large torque servo 406 and the axle of the driven wheel 504 and the conveyor belt frame.

[0095] In a specific embodiment of the present invention, preferably, the edge computing module 8 adopts a target detection algorithm based on deep learning, takes the field image collected by the visual detection module 6 as input, performs a target detection task on the cabbage in the image through the pre-trained cabbage recognition model weights, outputs the anchor point and anchor frame size of the cabbage position in a regression manner, and uses the center point of the anchor frame as the characterization feature of the crop row, and then fits the crop row line through clustering and straight line fitting algorithms, and takes the slope and intercept of the crop row line, the longitudinal spacing d1 between the two cabbages to be harvested, the lateral offset distance d3 between the two cabbages, and the lateral distance d4 between the visual detection module 6 and the cabbage to be harvested as output quantities, and realizes the calculation of the lateral offset with the vertical reference line of the camera screen through the slope of the crop row line and the tangent calculation of the intercept.

[0096] In a specific embodiment of the present invention, preferably, the cabbage row harvester based on machine vision keeps moving straight at the operating speed, and keeps the right wheel set of the chassis 3 on the ridge and preliminarily aligns with the cabbage row to be harvested;

[0097] The visual detection module 6 collects field images and transmits them to the edge computing module 8; the edge computing module 8 processes the images collected by the visual detection module 6 to obtain offset data and outputs it to the motion control module 7;

[0098] The posture sensor 308 detects the posture angle of the chassis frame 301 and transmits it to the motion control module 7. The motion control module 7 controls the leveling hydraulic rod 306 to level the chassis frame 301 according to the posture angle. The initial position is that the right wheel group is on the ridge and the chassis 3 is tilted to the left. At this time, the two leveling hydraulic rods 306 on the left are extended and the two leveling hydraulic rods 306 on the right are contracted until leveling is achieved. A stop command is received to complete the leveling action of the chassis 3.

[0099] The motion control module 7 compares the obtained offset data with a preset threshold value. When the offset data is greater than or equal to the preset displacement threshold value, the motion control module 7 controls the floating row alignment mechanism 401 of the row alignment mechanism 4 to make the cabbage harvesting and conveying mechanism 5 move horizontally along the chassis 3 in rows; when the offset data is less than the preset displacement threshold value, the motion control module 7 controls the active row alignment mechanism 405 of the row alignment mechanism 4 to change the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism 5 to feed the rows, complete the separation of the cabbage roots from the soil, complete the pulling action, and lift the pulled cabbage. The cabbage is pulled up and lifted to the two-stage conveying and collecting mechanism 2, and is transported by the first-stage retractable horizontal conveying mechanism 201 and the second-stage vertical conveyor belt mechanism 202, and the cabbage reaches the collection frame 203.

[0100] The present invention solves the problem of high cabbage damage rate during harvesting caused by inaccurate row alignment and difficulty in leveling in the prior art. The row alignment mechanism 4 has two row alignment modes: a floating row alignment mechanism 401 and an active row alignment mechanism 405, which can coordinate the two modes to achieve a more precise and rapid row alignment action and row alignment range.

[0101] The present invention measures the posture by the posture sensor 308 installed at the center of the chassis 3, and measures the speed by the GPS positioning system; and inputs the obtained measurement value and offset into the motion control module 7, and outputs the command to control the motion speed and motion direction of the row linear drive mechanism 404 after planning. When the offset data is less than the preset displacement threshold, the motion control module 7 controls the high torque steering engine 406 of the active row mechanism 405 of the row mechanism 4 to rotate synchronously, and changes the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism 5 to assist the row adjustment. When the offset data is greater than or equal to the preset displacement threshold, the motion control module 7 controls the linear drive mechanism 404 of the floating row mechanism 401 of the row mechanism 4 to make the cabbage harvesting and conveying mechanism 5 move laterally along the chassis 3 to perform active row adjustment.

[0102] The present invention is conducive to keeping the cabbage harvesting and conveying mechanism 5 in the best extraction position and feeding position, and finally realizing stable and continuous cabbage row-by-row profiling harvesting. The present invention has the advantages of automatic leveling function, efficient and accurate row-by-row harvesting and strong adaptability, is suitable for the harvesting needs of different varieties and types of cabbages, can further improve the harvesting automation and efficiency, and effectively reduce harvesting losses.

[0103] The present invention uses machine vision combined with deep learning to complete the recognition of Chinese cabbage and crop row fitting, has the characteristics of strong robustness and high accuracy, and has the advantages of automatic leveling function, efficient and accurate row harvesting and strong adaptability.

[0104] It should be understood that although this specification is described according to various embodiments, not every embodiment contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art. The series of detailed descriptions listed above are only specific descriptions of the feasible embodiments of the present invention. They are not intended to limit the scope of protection of the present invention. All equivalent embodiments or changes that do not deviate from the technical spirit of the present invention should be included in the scope of protection of the present invention.

Claims

1. A Chinese cabbage row harvester based on machine vision, characterized in that: It comprises a chassis (3), a row alignment mechanism (4), a cabbage harvesting and conveying mechanism (5), a visual detection module (6), a motion control module (7) and an edge computing module (8); The cabbage harvesting and conveying mechanism (5) is used to harvest the cabbage and convey it backwards; The cabbage harvesting and conveying mechanism (5) is installed on the chassis (3), and the row alignment mechanism (4) is used to adjust the cabbage harvesting and conveying mechanism (5) to align the rows; The visual detection module (6) is installed on the front frame of the cabbage harvesting and conveying mechanism (5) and is used to collect field images and transmit them to the edge computing module (8). The edge computing module (8) is respectively connected to the visual detection module (6) and the motion control module (7), and the motion control module (7) is connected to the row alignment mechanism (4). After the edge computing module (8) processes the image collected by the visual detection module (6), it obtains offset data and outputs it to the motion control module (7). The motion control module (7) controls the row alignment mechanism (4) to adjust the cabbage harvesting and conveying mechanism (5) to perform row alignment.

2. The Chinese cabbage row harvester based on machine vision according to claim 1, characterized in that: The row-aligning mechanism (4) comprises a floating row-aligning mechanism (401) and an active row-aligning mechanism (405); the floating row-aligning mechanism (401) is mounted on the chassis (3) and is used to enable the cabbage harvesting and conveying mechanism (5) to move laterally along the chassis (3) to align the rows; the active row-aligning mechanism (405) is mounted on the cabbage harvesting and conveying mechanism (5) and is used to change the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism (5) to perform row-aligning feeding.

3. The Chinese cabbage row harvester based on machine vision according to claim 2, characterized in that: The floating alignment mechanism (401) comprises a fixed linear bearing group (402), a dynamic linear bearing group (403), a linear drive mechanism (404) and a connecting frame (409); The chassis (3) comprises a chassis frame (301), a movable frame (309) and a connecting frame (310); the connecting frame (310) is installed above one side of the movable frame (309), and the connecting frame (310) is connected to the cabbage harvesting and conveying mechanism (5); The movable frame (309) is mounted on the chassis frame (301), the dynamic linear bearing group (403) is mounted on the movable frame (309), the fixed linear bearing group (402) is mounted on one side of the chassis frame (301), and the other side of the movable frame (309) is located on the fixed linear bearing group (402). One end of the linear drive mechanism (404) is mounted below one side of the movable frame (309) through a connecting frame (409), and one end of the linear drive mechanism (404) is connected to the dynamic linear bearing group (403) and is used to drive the dynamic linear bearing group (403) to drive the movable frame (309) to move horizontally left and right along the chassis frame (301), thereby driving the cabbage harvesting and conveying mechanism (5) to float and align, and the linear drive mechanism (404) is connected to a motion control module (7).

4. The Chinese cabbage row harvester based on machine vision according to claim 2, characterized in that: The two groups of active alignment mechanisms (405) are symmetrically arranged on both sides of the front end of the bracket (501) of the cabbage harvesting and conveying mechanism (5), and each group of active alignment mechanisms (405) includes a large torque steering gear (406), a steering gear mounting seat (407) and a coupling (408); The steering gear mounting seat (407) is mounted on the front end of the bracket (501) of the cabbage harvesting and conveying mechanism (5); the high-torque steering gear (406) is mounted on the steering gear mounting seat (407); the output shaft of the high-torque steering gear (406) is connected to the driven wheel mounting frame (505) through a coupling (408); the driven wheel mounting frame (505) is provided with a driven wheel (504); the high-torque steering gear (406) is used to adjust the angles of the two driven wheel mounting frames (505), thereby changing the angles of the two conveyor belts (503) of the cabbage harvesting and conveying mechanism (5) for row feeding; the high-torque steering gear (406) is connected to the motion control module (7).

5. The Chinese cabbage row harvester based on machine vision according to claim 3, characterized in that: It also includes a two-stage conveying and collecting mechanism (2); the two-stage conveying and collecting mechanism (2) includes a first-stage telescopic transverse conveying mechanism (201), a second-stage vertical conveying belt mechanism (202), a conveying belt fixing frame (207) and a collecting frame (203); The first-level telescopic transverse conveyor belt mechanism (201) is located at the end of the cabbage harvesting conveyor mechanism (5), the second-level vertical conveyor belt mechanism (202) is located at the end of the first-level telescopic transverse conveyor belt mechanism (201), the first-level telescopic transverse conveyor belt mechanism (201) and the second-level vertical conveyor belt mechanism (202) are vertically arranged in space, and the collection frame (203) is located behind the second-level vertical conveyor belt mechanism (202); The first-stage telescopic transverse conveyor belt mechanism (201) comprises a conveyor belt front section mounting frame (204), a conveyor belt rear section mounting frame (205), a conveyor belt (206), a belt surface tensioning mechanism (208) and a plurality of conveyor belt rollers (209); The conveyor belt front section mounting frame (204) and the conveyor belt rear section mounting frame (205) are arranged in parallel and slidably connected via a U-shaped slide groove; a plurality of conveyor belt rollers (209) are mounted on the conveyor belt front section roller mounting frame (204) and the conveyor belt rear section roller mounting frame (205); the conveyor belt (206) is mounted on the conveyor belt front section mounting frame (204) and the conveyor belt rear section mounting frame (205); and the belt surface tensioning mechanism (208) is mounted on the conveyor belt rear section mounting frame (205) and is in contact with the conveyor belt (206); A plurality of conveyor belt fixing frames (207) are provided below the first-stage telescopic transverse conveying mechanism (201) and the second-stage vertical conveyor belt mechanism (202), wherein the bottom of the conveyor belt front section mounting frame (204) is connected to the connecting frame (310) via the conveyor belt fixing frame (207), and the conveyor belt rear section mounting frame (205) is connected to the chassis frame (301) via the conveyor belt fixing frame (207); The connecting frame (310) can drive the conveyor belt front section mounting frame (204) to move left and right, and the belt surface tensioning mechanism (208) is used to keep the conveyor belt (206) in a tensioned state.

6. The Chinese cabbage row harvester based on machine vision according to claim 1, characterized in that: The visual inspection module (6) comprises an industrial camera (601), a mounting rod (602), a camera mounting seat (603) and an angle adjustment plate (604); The industrial camera (601) is mounted on a camera mounting seat (603), the camera mounting seat (603) is connected to the upper end of a mounting rod (602) via an angle adjustment plate (604), and the lower end of the mounting rod (602) is connected to the upper part of a bracket (501) of a cabbage harvesting and conveying mechanism (5); the industrial camera (601) is connected to an edge computing module (8).

7. The Chinese cabbage row harvester based on machine vision according to claim 6, characterized in that: One side of the angle adjustment plate (604) is connected to the mounting rod (602) via a rotating shaft, and the other side is provided with an arc-shaped slide groove, which is connected to the mounting rod (602) via a bolt. The shooting angle of the industrial camera (601) can be changed by changing the position of the bolt in the arc-shaped slide groove.

8. The Chinese cabbage row harvester based on machine vision according to claim 3, characterized in that: The four corners of the chassis (3) are respectively provided with a leveling mechanism, each of which comprises a connecting rod (303), a hinge (305), a leveling hydraulic rod (306), a tire connecting flange (307) and a posture sensor (308); One end of the connecting rod (303) is connected to the tire connecting flange (307), and the other end is connected to the chassis frame (301) through a hinge (305); one end of the leveling hydraulic rod (306) is connected to the tire connecting flange (307), and the other end is connected to the chassis frame (301) through a hinge (305); a posture sensor (308) is installed at the center of the chassis frame (301), the posture sensor (308) is connected to the motion control module (7), and the leveling hydraulic rod (306) is connected to the motion control module (7); the posture sensor (308) is used to detect the posture angle of the chassis frame (301) and transmit it to the motion control module (7); the motion control module (7) controls the leveling hydraulic rod (306) according to the posture angle to level the chassis frame (301).

9. A method for controlling a Chinese cabbage row harvester based on machine vision according to any one of claims 1 to 8, characterized in that: The following steps are involved: The visual detection module (6) collects field images and transmits them to the edge computing module (8); after the edge computing module (8) processes the images collected by the visual detection module (6), it obtains offset data and outputs it to the motion control module (7); the motion control module (7) compares the obtained offset data with a preset threshold value; when the offset data is greater than or equal to the preset displacement threshold value, the motion control module (7) controls the floating row alignment mechanism (401) of the row alignment mechanism (4) so ​​that the cabbage harvesting and conveying mechanism (5) moves horizontally along the chassis (3) to align the rows; when the offset data is less than the preset displacement threshold value, the motion control module (7) controls the active row alignment mechanism (405) of the row alignment mechanism (4) to change the angle of the opening of the two conveyor belts of the cabbage harvesting and conveying mechanism (5) to perform row feeding.

10. The method for controlling a Chinese cabbage row harvester based on machine vision according to claim 9, characterized in that: The displacement threshold p is calculated according to the following formula: The mathematical model of the floating row alignment mechanism (401) of the row alignment mechanism (4) making the cabbage harvesting and conveying mechanism (5) move laterally along the chassis (3) to match the speed of the rows is as follows: Wherein, d1 and d3 are the outputs after the edge computing module (8) processes the acquired image, d1 is the longitudinal distance between the two cabbages to be harvested, d3 is the lateral offset distance between the two cabbages, d2 is the distance between the position of the visual detection module (6) and the cabbages to be harvested in the driving direction, v1 is the operating speed of the harvester, t1 is the arrival time from the position of the visual detection module (6) to the cabbages to be harvested, t2 is the time taken by the edge computing module (8) to receive the image, process the image and obtain the output, and v2 is the optimal speed for theoretically reaching the optimal harvesting position, that is, the driving speed of the linear drive mechanism (404); The active row-aligning mechanism (405) of the row-aligning mechanism (4) changes the angle of the two conveyor belts of the cabbage harvesting and conveying mechanism (5) to perform row-aligning feeding speed matching calculation method as follows: Wherein, d4 is the lateral distance between the visual inspection module (6) and the cabbage to be harvested, the output obtained by the edge computing module (8), d5 is the center distance between the two large torque servos (406) on the left and right front ends of the bracket (501) of the cabbage harvesting and conveying mechanism (5), L1 is the center distance between the output shaft of the large torque servo (406) and the axle of the conveyor belt driven wheel (504), w is the average angular velocity of the large torque servo (406) during the travel process, and θ is the angle between the line between the driving shaft of the large torque servo (406) and the axle of the driven wheel (504) and the conveyor belt frame.

Images