Bending type crop intelligent multifunctional management platform and working method thereof

Abstract

The invention discloses a bending-over type crop intelligent multifunctional management platform and a working method thereof, and relates to the field of agricultural robots. The intelligent multifunctional crop management platform comprises an electric four-wheel-drive chassis walking system, an intelligent human body comfortable bending control system and a fruit basket auxiliary transferring system. Through an intelligent voice physical fusion control method, convenient control over the advancing speed, steering, bed bending angles and physical sudden stop in emergency situations of the electric four-wheel drive chassis is achieved; through a crop inter-row autonomous walking method, autonomous navigation of the management platform along rows is guided, and human intervention in management operation is reduced; operation jolt and bearing pressure in a complex field environment are reduced through a head bearing self-adaptive adjusting method; through an operation blind area voice regulation and control method, the working visual field of workers in prone operation is expanded. The method has a remarkable promoting effect on reducing labor force and improving economic benefits in agricultural production.

Description

Technical Field

[0001] The present invention relates to the field of agricultural robots, and particularly to a prone-type intelligent multi-functional management platform for crops and its working method. Background Art

[0002] During the growth of crops, there are production operations such as pollination, thinning of flowers and fruits, removal of stolons and old leaves, and harvesting. There are many management links and frequent operations. A large number of fine management operations completely rely on manual labor, bending down or squatting for a long time in narrow facility ridges. In developed countries, ridged crops generally have spacious fields, and their riding-assisted picking machine solutions are not applicable to the narrow ridged environment in our country; while the existing prone-type picking machines still have deficiencies such as single function, inconvenient prone operation, fatigue and discomfort during prone position, and limited prone observation operation postures. Therefore, there is an urgent need for an efficient prone-type intelligent multi-functional management platform for crops to achieve low-fatigue and high-efficiency operations in various management links during crop planting, which has important practical significance for ensuring the vigorous development of modern agriculture. Summary of the Invention

[0003] In view of this, the present invention provides a prone-type intelligent multi-functional management platform for crops and its working method, which effectively improves the mechanization level of management operations during the growth of field crops.

[0004] The present invention achieves the above technical objectives through the following technical means.

[0005] A prone-type multi-functional intelligent management operation platform for crops, comprising:

[0006] An electric four-wheel drive chassis walking system, including a vehicle body;

[0007] A human body comfortable prone intelligent control system, including a head support device, an ergonomic prone bed, a depth camera, a pressure sensor, a voice sensor, a lifting motor, a footrest, a controller, and a high-torque motor; the head support device is located directly in front of the vehicle body, and it includes a pair of support rods and a C-shaped support bracket. The ends of the pair of support rods are installed on the vehicle body, and the C-shaped support bracket is rotatably connected to the heads of the pair of support rods; the ergonomic prone bed is installed directly above the vehicle body, the depth camera is installed directly in front of the C-shaped support bracket, the pressure sensor is embedded directly above the C-shaped bracket, the voice sensor is installed on both sides of the C-shaped support bracket, the lifting motor is installed on the support rod, the footrest is installed directly behind the vehicle body, and the high-torque motor is installed in the middle of the vehicle body; the depth camera, the pressure sensor, and the voice sensor are all in communication with the controller, and the lifting motor and the high-torque motor are both controlled by the controller.

[0008] The above technical solution further includes a fruit basket auxiliary transfer system, which includes a roller conveyor belt and rollers; the roller conveyor belt is installed directly above one side of the vehicle body, and the height in front of the roller conveyor belt is greater than that at the rear; a plurality of rollers are evenly installed on the roller conveyor belt, and the fruit basket is located on the rollers; a baffle is provided at the end of the roller conveyor belt.

[0009] In the above technical solution, the electric four-wheel drive chassis walking system further includes front drive wheels, a lithium battery pack, rear drive wheels, a rotary motor, and a servo motor; the front drive wheels are installed on the left and right sides of the front part under the vehicle body through support columns, and the rear drive wheels are installed on the left and right sides of the rear part under the vehicle body through support columns; the rotary motor is used to drive the wheels to rotate 360°, the servo motor is used to drive the wheels to move forward and backward, and the lithium battery pack is used to supply power to the rotary motor, the servo motor, the depth camera, the pressure sensor, the voice sensor, the lifting motor, the controller, and the high-torque motor.

[0010] In the above technical solution, shock absorption devices are respectively installed between the front drive wheels, the rear drive wheels and the vehicle body.

[0011] A working method of a prone multi-functional crop intelligent management operation platform:

[0012] Construct a depth semantic feature training set for the canopy and ridge furrows during the crop growth period. The canopy in the training set is the depth semantic upper context feature, and the ridge furrow is the depth semantic lower context feature. The features are sequentially implanted into the semantic network in the order of upper-lower context features for feature fusion training.

[0013] Overlay the canopy upper context feature and the ridge furrow lower context feature obtained from the fusion training through the feature map output by the semantic network, and then deploy them in the controller as a solidified model for extracting crop inter-row target features.

[0014] When the crop intelligent multi-functional management platform is operating between rows, the controller activates the depth camera to continuously scan the RGB video stream of the information in front of the vehicle body.

[0015] The controller inputs the two-dimensional RGB data set scanned by the depth camera into the solidified model for extracting crop inter-row target features to be segmented into a two-dimensional RGB data set containing only the canopy.

[0016] Obtain a three-dimensional depth point cloud cluster containing only the canopy through color-depth coordinate alignment, perform extraction of the longitudinal center line of the three-dimensional point cloud clustering, and the extracted longitudinal center line is the vehicle body's autonomous walking alignment line. The controller drives the crop intelligent multi-functional management platform to autonomously walk along the alignment line.

[0017] Further, when the depth camera scans, color-depth composite video stream information is acquired within a set horizontal viewing angle range [-θ, θ] and vertical viewing angle range [0, β]; wherein, the horizontal viewing angle θ and the vertical viewing angle β satisfy:

[0018]

[0019] In the formula, W1 is the width of the ridge body, W2 is the width of the furrow, H is the installation height of the depth camera, and L is the set imaging distance of the depth camera.

[0020] Further, it also includes voice control for the operation blind area: When the crop intelligent multi-functional management platform is operating between rows, the user cannot observe the left and right sides of the canopy directly below. By using the voice sensor, the user expresses the motion state voice control text {turn the vehicle body to the left, turn the vehicle body to the right} in the motion state voice control word set A1 for the vehicle body movement. After being parsed by the online voice understanding control instruction output set B1, the controller sends a signal to the high-torque motor, thereby driving the vehicle body to turn left or right, increasing the user's operation vision.

[0021] Furthermore, the motion state voice control text in the motion state voice control word set A1 for the vehicle body movement includes a specific wake-up word and {go forward, go backward, accelerate, decelerate, move left, move right, turn the vehicle body to the left, turn the vehicle body to the right}; when the user expresses {go forward, go backward} in the motion state voice control word set A1 for the vehicle body movement through the voice sensor, after being parsed by the online voice understanding control instruction output set B1, the controller sends a signal to the servo motor to drive the vehicle body to go forward or backward; when the user expresses {accelerate, decelerate} in the motion state voice control word set A1 for the vehicle body movement through the voice sensor, after being parsed by the online voice understanding control instruction output set B1, the controller sends a signal to the servo motor to drive the vehicle body to accelerate or decelerate; when the user expresses {move left, move right} in the motion state voice control word set A1 for the vehicle body movement through the voice sensor, after being parsed by the online voice understanding control instruction output set B1, the controller sends a signal to the rotary motor to drive the wheels to rotate 90°, realizing the left and right translation of the vehicle body; when the foot pedal is depressed, the controller sends a signal to the servo motor to drive the vehicle body to go forward or stop.

[0022] Further, it also includes adaptive adjustment of the head support: When the crop intelligent multi-functional management platform is operating between rows, the head pressure data z collected by the pressure sensor is compared with the pressure range P0 (z min , z max ) set by the controller in real time online; if z ∈ [z min , z max , the operator's head is in a comfortable state; if z < z min , the controller drives the lifting motor to reverse, driving the C-shaped support bracket to rise; if z > z max , the controller drives the lifting motor to rotate forward, driving the C-shaped support bracket to descend.

[0023] Further, when the crop intelligent multi-functional management platform is about to operate, the user takes the fruit basket at the very end of the roller conveyor belt to place fruits or leaves. The other fruit baskets move towards the very end of the roller conveyor belt due to the height difference. When the fruit basket in the user's hand is full of fruits or leaves, it is placed at the very front end of the roller conveyor belt. During subsequent operations, the user continues to take the fruit basket at the very end of the roller conveyor belt, and this cycle repeats.

[0024] The beneficial effects of the present invention are as follows: Aiming at the deficiencies of existing prone harvesters, such as single function, inconvenient prone operation, fatigue and discomfort during prone operation, and limited prone observation of the operation position, etc., the present invention realizes the convenient control of the traveling speed, steering, angle of the prone bed, and physical emergency stop in case of emergencies of the electric four-wheel drive chassis through the intelligent voice physical fusion control method; through the method of autonomous walking between crop rows, it guides the management platform to perform autonomous navigation along the row to reduce human intervention in management operations; through the head support operation adaptive adjustment method and the operation blind area voice control method, it reduces the operation bumps and support pressure in the complex field environment and expands the working vision of the laborer during prone operation. The present invention has a significant promoting effect on reducing labor and improving economic benefits in agricultural production. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 It is a schematic structural diagram of the multi-functional management platform of the present invention;

[0026] Figure 2 It is a schematic diagram of a comfortable prone position of the human body on the multi-functional management platform of the present invention;

[0027] Figure 3 It is a schematic diagram of the ergonomic prone bed of the present invention;

[0028] Figure 4 It is a schematic structural diagram of the comfortable prone intelligent control system of the human body of the present invention;

[0029] Figure 5(a) is a side view of the fruit basket auxiliary transfer system of the present invention;

[0030] Figure 5(b) is a top view of the fruit basket auxiliary transfer system of the present invention;

[0031] Figure 6 It is a schematic diagram of the multi-functional management platform of the present invention during field operation;

[0032] Figure 7 It is a schematic diagram of the vertical view imaging of the depth camera of the present invention;

[0033] Figure 8 It is a schematic diagram of the horizontal view imaging of the depth camera of the present invention;

[0034] In the figure, 1. vehicle body, 2. support column, 3. front drive wheel, 4. shock absorber, 5. lithium battery pack, 6. rear drive wheel, 7. rotary motor, 8. servo motor, 9. head support device, 10. ergonomic prone bed, 11. depth camera, 12. pressure sensor, 13. voice sensor, 14. lifting motor, 15. footrest I, 16. footrest II, 17. controller, 18. high-torque motor, 19. roller conveyor line, 20. roller, 21. baffle, 22. fruit basket, 23. canopy, 24. furrow, 25. ridge body. Detailed implementation mode

[0035] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but the protection scope of the present invention is not limited thereto.

[0036] A prone crop intelligent multi-functional management platform and its working method of the present invention include a prone crop intelligent multi-functional management platform and a working method of the prone crop intelligent multi-functional management platform.

[0037] The present invention is applicable to field management operations such as pollination, flower thinning, fruit thinning, removal of stolons and old leaf harvesting of ridge crops. The management operation of facility ridge strawberries is taken as an example for description below.

[0038] The prone crop intelligent multi-functional management platform includes an electric four-wheel drive chassis walking system, a human body comfortable prone intelligent control system and a fruit basket auxiliary transfer system. The front in this embodiment is the forward direction of the multi-functional management platform.

[0039] As Figure 1 shown, the electric four-wheel drive chassis walking system includes a vehicle body 1, support columns 2, front drive wheels 3, shock absorbers 4, lithium battery packs 5, rear drive wheels 6, rotary motors 7 and servo motors 8. Four support columns 2 are respectively installed at the four corners directly below the vehicle body 1. Two front drive wheels 3 and two rear drive wheels 6 are respectively installed directly below the support columns 2. Shock absorbers 4 are respectively installed between the front drive wheels 3 and the vehicle body 1 and between the rear drive wheels 6 and the vehicle body 1. The lithium battery packs 5 are installed directly below the middle of the left and right sides of the vehicle body 1. The rotary motors 7 are installed directly above the shock absorbers 4. The servo motors 8 are installed directly above the rotary motors 7. The rotary motors 7 are used to drive the wheels to rotate 360°, and the servo motors 8 are used to drive the wheels to move forward and backward.

[0040] As Figure 2 、 3, as shown in FIGS. 4, the human body comfortable prone intelligent control system includes a head support device 9, an ergonomic prone bed 10, a depth camera 11, a pressure sensor 12, a voice sensor 13, a lifting motor 14, a footrest I 15, a footrest II 16, a controller 17 and a high-torque motor 18. The ergonomic prone bed 10 is installed directly above the vehicle body 1, and the head support device 9 is installed directly in front of the vehicle body 1. The head support device 9 includes a pair of support rods and a C-shaped support bracket. The ends of the pair of support rods are installed on the vehicle body 1, and the C-shaped support bracket is rotatably connected to the heads of the pair of support rods; a plurality of pressure sensors 12 are embedded directly above the head support device 9, the depth camera 11 is installed directly in front of the head support device 9, and two voice sensors 13 are respectively installed on the left and right sides of the head support device 9. Two lifting motors 14 are respectively installed on the support rods for driving the C-shaped support bracket to rotate around the support rods. The footrest I 15 and the footrest II 16 are respectively installed on the left and right sides at the rear of the vehicle body 1, the controller 17 is installed directly behind the footrest I 15 and the footrest II 16, and the high-torque motor 18 is installed in the middle of the vehicle body 1. In this embodiment, S = 10 cm.

[0041] As Figure 5(a) , 5(b) shown, the fruit basket auxiliary transfer system includes a roller conveyor belt 19, rollers 20, a baffle 21 and a fruit basket 22. The front and rear ends of the roller conveyor belt 19 are installed directly above the left side of the vehicle body 1 with a height difference of Δh. A plurality of rollers 20 are installed on the roller conveyor belt 19 in an array at an interval of ΔC. The baffle 21 is installed at the end of the roller conveyor belt 19 to prevent the fruit basket 22 from falling. A plurality of fruit baskets 22 are placed on the roller conveyor belt 19 and automatically move to the end through the height difference of Δh of the roller conveyor belt 19. In this embodiment, Δh = 5 cm and ΔC = 6 cm.

[0042] Among them, the rotary motor 7, the servo motor 8, the depth camera 11, the pressure sensor 12, the voice sensor 13, the lifting motor 14, the controller 17 and the high-torque motor 18 are all powered by the lithium battery pack 5.

[0043] The working method of the prone multi-functional crop intelligent management operation platform includes an intelligent voice and physical fusion control method, a method for autonomous walking between crop rows, a head support adaptive adjustment method and an operation blind area voice control method.

[0044] The intelligent voice physical fusion control method includes a vehicle body movement voice control entry set A1, an online voice understanding control instruction output set B1, and an auxiliary foot control instruction output set C1; the vehicle body movement voice control entry set A1, which includes specific wake-up words and movement state voice control texts of {forward, backward, acceleration, deceleration, left translation, right translation, vehicle body left turn, vehicle body right turn}; the online voice understanding control instruction output set B1, which includes the vehicle body movement voice control information encoding and the pulse output mode of the corresponding servo motor 8 or high-torque motor 18; the auxiliary foot control instruction output set C1, which includes the physical control information encoding of the vehicle body movement corresponding to the foot pedal I 15 and the foot pedal II 16 for vehicle body forward and stop respectively, and the pulse output mode of the corresponding servo motor 8; by expressing the movement state voice control texts {forward, backward} in the vehicle body movement voice control entry set A1 through the voice sensor 13, and after being parsed by the online voice understanding control instruction output set B1, the controller 17 sends a positive pulse or negative pulse output signal to the servo motor 8 to drive the vehicle body 1 to move forward or backward; by expressing the movement state voice control texts {acceleration, deceleration} in the vehicle body movement voice control entry set A1 through the voice sensor 13, and after being parsed by the online voice understanding control instruction output set B1, the controller 17 sends a corresponding pulse frequency output signal to the servo motor 8 to drive the vehicle body 1 to accelerate or decelerate; by expressing the movement state voice control texts {left translation, right translation} in the vehicle body movement voice control entry set A1 through the voice sensor 13, and after being parsed by the online voice understanding control instruction output set B1, the controller 17 sends a corresponding pulse frequency output signal to the rotary motor 7 to drive the wheel to rotate 90°, realizing the left translation and right translation of the vehicle body 1; when the foot pedal I 15 or the foot pedal II 16 is stepped on, the controller 17 sends a positive pulse or negative pulse output signal to the servo motor 8 to drive the vehicle body 1 to move forward or stop.

[0045] As Figure 6 、 7 、8 shown, the method for autonomous walking between crop rows, its operation process is as follows:

[0046] Step 1, construct a deep semantic feature training set of the canopy 23 and ridge groove 24 information during the crop growth period. The canopy 23 in the training set is the deep semantic upper context feature, and the ridge groove 24 is the deep semantic lower context feature. The features are implanted into the semantic network in the order of upper-lower context features for feature fusion training;

[0047] Step 2, superimpose the canopy 23 upper context feature and the ridge groove 24 lower context feature obtained by the fusion training through the feature map output by the semantic network, and then deploy them in the controller 17 as a fixed model for extracting the features of the target object between crop rows;

[0048] Step 3: The management platform is about to start operation. The user takes the fruit basket 22 at the end of the roller conveyor belt 19 to place the fruit or leaves. The remaining fruit baskets 22 move toward the end of the roller conveyor belt 19 due to the height difference Δh. When the fruit basket 22 in the user's hand is full of fruit or leaves, it is placed at the front end of the roller conveyor belt 19. In subsequent operations, the user continues to take the fruit basket 22 at the end of the roller conveyor belt 19, and the cycle continues;

[0049] Step 4: When the management platform is operating, the controller 17 automatically activates the depth camera 11 to scan the continuous RGB video stream signal of the information in front of the vehicle body 1, and the controller 17 automatically calls the inter-row target feature extraction solidification model; when scanning, the depth camera 11 acquires color-depth composite video stream information at the horizontal viewing angle [-θ, θ] and vertical viewing angle [0, β] set by the controller 17;

[0050] Wherein, the horizontal viewing angle θ and the vertical viewing angle β of the depth camera 11 satisfy:

[0051]

[0052] In formula (1), W1 is the ridge width, W2 is the ridge ditch width 21, H is the installation height of the depth camera 11, and L is the imaging distance set by the depth camera 11;

[0053] Step 5, the controller 17 inputs the two-dimensional RGB data set B0 (x, y) scanned by the depth camera 11 into the crop inter-row target feature extraction solidification model to segment it into a two-dimensional RGB data set B (x, y) containing only the canopy 23;

[0054] Step six, obtain the three-dimensional depth point cloud cluster D(x, y, d(x, y)) containing only the canopy 23 through color-depth coordinate alignment, and extract the clustered longitudinal centerline of the three-dimensional point cloud. The extracted longitudinal centerline is the autonomous walking alignment line of the vehicle body 1, and the controller 17 will drive the management platform to walk autonomously along the alignment line.

[0055] In this embodiment, W1=60 cm, W2=30 cm, H=100 cm, and L=600 cm.

[0056] The head support adaptive adjustment method is used to adjust the head pressure data z collected by the pressure sensor 12 and the pressure interval P0 (z min , z max ) to perform real-time online comparison; if z∈[z min , z max ], the operator’s head is in a comfortable state; if z<z min , the controller 17 drives the lifting motor 14 to reverse, driving the C-shaped support frame to rise; if z>z max, the controller 17 drives the lifting motor 14 to rotate forward, driving the C-shaped support bracket to descend.

[0057] Operation blind area voice control method: When the management platform is performing row operations, the user cannot observe the left and right sides of the canopy 23 when looking straight down. The user expresses the motion state voice control texts {vehicle body turns left, vehicle body turns right} in the motion state voice control lexicon set A1 through the voice sensor 13. After being parsed by the online voice understanding control instruction output set B1, the controller 17 sends a positive pulse or negative pulse output signal to the high-torque motor 18 to drive the vehicle body 1 to turn left or right, increasing the user's operation vision.

[0058] The described embodiments are the preferred embodiments of the present invention, but the present invention is not limited to the above embodiments. Without departing from the essential content of the present invention, any obvious improvements, substitutions or modifications that those skilled in the art can make all fall within the protection scope of the present invention.

Claims

1. A prone multi-functional intelligent crop management operation platform, characterized in that, Including: An electric four-wheel drive chassis walking system, including a vehicle body (1); A human body comfortable prone intelligent control system, including a head support device (9), a human body engineering prone bed (10), a depth camera (11), a pressure sensor (12), a voice sensor (13), a lifting motor (14), a footrest, a controller (17) and a high-torque motor (18); the head support device (9) is located directly in front of the vehicle body (1), and it includes a pair of support rods and a C-shaped support bracket. The ends of the pair of support rods are installed on the vehicle body (1), and the C-shaped support bracket is rotatably connected to the heads of the pair of support rods; the human body engineering prone bed (10) is installed directly above the vehicle body (1), the depth camera (11) is installed directly in front of the C-shaped support bracket, the pressure sensor (12) is embedded directly above the C-shaped bracket, the voice sensor (13) is installed on both the left and right sides of the C-shaped support bracket, the lifting motor (14) is installed on the support rod, the footrest is installed directly behind the vehicle body (1), and the high-torque motor (18) is installed directly in the middle of the vehicle body (1); the depth camera (11), the pressure sensor (12), and the voice sensor (13) are all in communication with the controller (17), and the lifting motor (14) and the high-torque motor (18) are both controlled by the controller (17).

2. The crop intelligent management operation platform according to claim 1, wherein, It further includes a fruit basket auxiliary transfer system, and the fruit basket auxiliary transfer system includes a roller conveyor belt (19) and rollers (20); the roller conveyor belt (19) is installed directly above one side of the vehicle body (1), and the height in front of the roller conveyor belt (19) is greater than the height at the rear; a plurality of rollers (20) are evenly installed on the roller conveyor belt (19), and the fruit basket (22) is located on the rollers (20); a baffle (21) is provided at the very end of the roller conveyor belt (19).

3. The crop intelligent management operation platform according to claim 2, characterized in that, The electric four-wheel drive chassis walking system further includes front drive wheels (3), a lithium battery pack (5), rear drive wheels (6), a rotary motor (7) and a servo motor (8); the front drive wheels (3) are installed on the left and right sides of the front part below the vehicle body (1) through support columns, and the rear drive wheels (6) are installed on the left and right sides of the rear part below the vehicle body (1) through support columns; the rotary motor (7) is used to drive the wheels to rotate 360°, and the servo motor (8) is used to drive the wheels to move back and forth. The lithium battery pack (5) is used to supply power to the rotary motor (7), the servo motor (8), the depth camera (11), the pressure sensor (12), the voice sensor (13), the lifting motor (14), the controller (17) and the high-torque motor (18).

4. The crop intelligent management operation platform according to claim 3, characterized in that, Shock-absorbing devices (4) are respectively installed between the front drive wheels (3), the rear drive wheels (6) and the vehicle body (1).

5. A working method of the crop intelligent management operation platform according to any one of claims 1-4, characterized in that: Construct a depth semantic feature training set for the canopy and the ridge ditch during the crop growth period. The canopy in the training set is the depth semantic upper context feature, and the ridge ditch is the depth semantic lower context feature. The features are sequentially implanted into the semantic network in the order of upper-lower context features for feature fusion training; Overlay the canopy upper - context features and the furrow lower - context features obtained from the fusion training through the feature map output by the semantic network, and then deploy them in the controller (17) as the solidified model for crop - row target feature extraction. When the crop intelligent multi - functional management platform is operating between rows, the controller (17) activates the depth camera (11) to scan the continuous RGB video stream of the information in front of the vehicle body (1). The controller (17) inputs the two - dimensional RGB data set scanned by the depth camera (11) into the solidified model for crop - row target feature extraction to segment it into a two - dimensional RGB data set containing only the canopy. Obtain a three - dimensional depth point cloud cluster containing only the canopy through color - depth coordinate alignment, extract the longitudinal center line of the three - dimensional point cloud clustering. The extracted longitudinal center line is the reference line for the autonomous walking of the vehicle body (1), and the controller (17) drives the crop intelligent multi - functional management platform to walk autonomously along the reference line.

6. The working method according to claim 5, characterized in that, When the depth camera (11) scans, it acquires color - depth composite video stream information within the set horizontal viewing angle range [-θ, θ] and vertical viewing angle range [0, β]; where the horizontal viewing angle θ and the vertical viewing angle β satisfy: In the formula, W1 is the width of the ridge body, W2 is the width of the furrow, H is the installation height of the depth camera, and L is the set imaging distance of the depth camera.

7. The working method according to claim 5, characterized in that, It also includes voice control for the operation blind area: When the crop intelligent multi - functional management platform is operating between rows, the user cannot observe the left and right sides of the canopy directly below. By expressing the motion - state voice control text {vehicle body turns left, vehicle body turns right} in the motion - state voice control term set A1 of the vehicle body through the voice sensor (13), after being parsed by the online voice understanding control instruction output set B1, the controller (17) sends a signal to the high - torque motor (18), and then drives the vehicle body (1) to turn left or right, increasing the user's operation vision.

8. The working method according to claim 7, characterized in that, The motion - state voice control text in the motion - state voice control term set A1 of the vehicle body includes a specific wake - up word and {forward, backward, accelerate, decelerate, left translation, right translation, vehicle body turns left, vehicle body turns right}; when the user expresses {forward, backward} in the motion - state voice control term set A1 of the vehicle body through the voice sensor (13), after being parsed by the online voice understanding control instruction output set B1, the controller (17) sends a signal to the servo motor (8) to drive the vehicle body (1) to move forward or backward; when the user expresses {accelerate, decelerate} in the motion - state voice control term set A1 of the vehicle body through the voice sensor (13), after being parsed by the online voice understanding control instruction output set B1, the controller (17) sends a signal to the servo motor (8) to drive the vehicle body (1) to accelerate or decelerate; when the user expresses {left translation, right translation} in the motion - state voice control term set A1 of the vehicle body through the voice sensor (13), after being parsed by the online voice understanding control instruction output set B1, the controller (17) sends a signal to the rotary motor (7) to drive the wheel to rotate 90°, realizing the left and right translation of the vehicle body (1); when the foot pedal is pressed, the controller (7) sends a signal to the servo motor (8) to drive the vehicle body (1) to move forward or stop.

9. The working method according to claim 5, characterized in that, It also includes head support adaptive adjustment: When the crop intelligent multi-functional management platform is operating between rows, the head pressure data z collected by the pressure sensor (12) is compared online in real time with the pressure range P0 (z min , z max ) set by the controller (17); If z ∈ [z min , z max , the operator's head is in a comfortable state; if z < z min , the controller (17) drives the lifting motor (14) to reverse, driving the C-shaped support bracket to rise; if z > z max , the controller (17) drives the lifting motor (14) to rotate forward, driving the C-shaped support bracket to descend.

10. The working method according to claim 5, characterized in that, When the crop intelligent multi-functional management platform is about to operate, the user takes the fruit basket (22) at the very end of the roller conveyor belt (19) to place fruits or leaves. The other fruit baskets (22) move towards the very end of the roller conveyor belt (19) due to the height difference. When the fruit basket (22) in the user's hand is full of fruits or leaves, it is placed at the very front of the roller conveyor belt (19). During subsequent operations, the user continues to take the fruit basket (22) at the very end of the roller conveyor belt (19), and this process repeats in sequence.

Images