Intelligent all-dimensional mechanical weeding machine for orchard

Abstract

The invention relates to the technical field of agricultural machinery, in particular to an orchard intelligent all-dimensional mechanical weeding machine which comprises a frame, a controller is fixedly connected to the top of the frame, a camera is fixedly connected to one side of the frame, the controller is used for receiving image information collected by the camera, and an energy supply assembly for supplying energy to the mechanical weeding machine is arranged at the top of the frame. A driving assembly used for driving the mechanical weeding machine to move and a reversing assembly used for controlling the mechanical weeding machine to reverse are arranged at the bottom of the frame, and a monitoring assembly used for monitoring the current environment condition and an adjusting assembly used for adjusting the weeding height are arranged on one side of the frame; and a control system for controlling the mechanical weeding machine to operate is arranged in the controller. Orchard images are collected in real time through the camera and the recognition module, crops and weeds are accurately distinguished on the basis of the database established through data over the years, the weeding accuracy is improved, and meanwhile the possibility that the crops are accidentally damaged by the mechanical weeding machine is greatly reduced.

Description

Technical Field

[0001] The invention relates to the technical field of agricultural machinery, and in particular to an intelligent all-round mechanical weeder for orchards. Background Art

[0002] With the continuous expansion of orchards and the continuous increase in labor costs, traditional manual weeding methods can no longer meet the needs of orchard management. At the same time, although the use of chemical herbicides has improved the efficiency of weeding to a certain extent, long-term use will cause serious pollution to the environment and affect the growth of fruit trees and the quality of fruits. Therefore, the research and development and application of intelligent all-round mechanical weeders for orchards have become the key to solving this problem.

[0003] Existing mechanical weeders for orchards may accidentally damage fruit trees or other crops during the weeding process. When fruit trees are intertwined with weeds, they may cause damage to fruit trees or other crops in the orchard, thereby affecting the yield and quality of the orchard.

[0004] To sum up, the problem that existing orchard mechanical weeders may accidentally damage fruit trees or other crops during the weeding process, which may in turn affect the yield and quality of the orchard has become a difficult problem that needs to be solved urgently in this field. Therefore, it is necessary to propose an intelligent all-round mechanical weeder for orchards. Summary of the invention

[0005] To solve the above problems, the present invention provides an intelligent all-round mechanical weeder for orchards, which collects orchard images in real time through cameras and recognition modules, and accurately distinguishes crops and weeds based on a database established based on historical data, thereby improving the accuracy of weeding and greatly reducing the possibility of accidental damage to crops by the mechanical weeder.

[0006] In order to achieve the above-mentioned objectives, the technical solution of the present invention is as follows: an intelligent all-round mechanical weeder for an orchard, comprising a frame, a controller fixedly connected to the top of the frame, a camera fixedly connected to one side of the frame, the controller being used to receive image information collected by the camera, a power supply component for supplying power to the mechanical weeder being provided on the top of the frame, a driving component for driving the mechanical weeder to move and a reversing component for controlling the reversal of the mechanical weeder being provided at the bottom of the frame, a monitoring component for monitoring the current environmental conditions and an adjustment component for adjusting the weeding height being provided on one side of the frame.

[0007] The controller is equipped with a control system for controlling the operation of the mechanical weeder, including the following modules:

[0008] The information module is used to establish a database based on all the data of the orchard over the years, and to organize and deeply analyze the crop categories and weed categories of the database in combination with the Internet big model, and to establish different types of crop characteristic templates and weed characteristic templates.

[0009] The recognition module is used to perform image recognition and feature extraction on the current environment, compare all extracted features with all established crop feature templates and weed feature templates, distinguish crops and weeds in the current environment image based on the database, obtain recognition results, and store the recognition results in the database.

[0010] The recognition module is also used to recognize the fluctuations and complexities of the current environment, adjust the weeding height according to the fluctuations of the current environment, and adjust the weeding speed according to the complexity of the environment.

[0011] The monitoring module is used to monitor the current environmental data. Based on the database and the Internet big model, the monitoring results are analyzed and predicted. According to the historical data and monitoring results, the future weed growth trend is predicted through the decision tree algorithm. The areas with obvious weed growth trend in the future are monitored in a key manner. The environmental data include the temperature, humidity and carbon dioxide concentration of the environment.

[0012] The control module is used to formulate and execute a weed control plan based on the database, identification results and monitoring results.

[0013] The technical principles of the above scheme are as follows:

[0014] The controller first collects all the data of the orchard over the years through the information module and establishes a database. Then the recognition module collects the current environment image through the camera to obtain image information, analyzes the image information, and identifies and distinguishes the crops and weeds in the image information based on the database to obtain the recognition result. At the same time, the monitoring module will monitor the environmental data of the current environment in real time through the monitoring component and obtain the monitoring results. The control module will formulate and execute a targeted weeding plan based on the database, recognition results and monitoring results. When the mechanical weeder executes the weeding plan, the controller starts the mechanical weeder and drives the mechanical weeder to move through the driving component. When the weeding height needs to be adjusted, the adjustment component will adjust the weeding height to the appropriate height. When the mechanical weeder operates in a sunny environment, the energy supply component will supply it with energy.

[0015] The above scheme has the following beneficial effects:

[0016] 1. The present invention uses a camera and a recognition module to enable the mechanical weeder to collect orchard images in real time and accurately distinguish between crops and weeds based on a database built with historical data. This not only improves the accuracy of weeding, but also greatly reduces the possibility of the mechanical weeder accidentally damaging crops.

[0017] 2. The present invention, through the built-in control system of the controller, formulates a targeted weeding plan according to various indicators of the current environment, and realizes the self-operation of the mechanical weeder. At the same time, the control system can also be seamlessly connected with the orchard management system to realize real-time transmission and sharing of data. This enables the staff to understand the weeding situation of the orchard anytime and anywhere, adjust the management strategy in time, and improve the overall level and efficiency of orchard management.

[0018] 3. The present invention, through precise weeding operations and real-time monitoring, enables the staff to formulate production plans more scientifically and optimize resource allocation, thereby improving the overall economic benefits of the orchard.

[0019] Furthermore, when the recognition module performs image recognition and feature extraction, it first performs grayscale processing on the image. The grayscale processing formula is as follows:

[0020]

[0021] Among them, Gray represents the grayscale value, R represents the red component of the pixel in the image, G represents the green component of the pixel in the image, and B represents the blue component of the pixel in the image.

[0022] After the grayscale processing is completed, the recognition module uses the energy gradient formula to clarify the image. The energy gradient formula is as follows:

[0023]

[0024] The pixel at the lower left corner of the image is taken as the origin, x is the horizontal distance from the pixel to the origin, and y is the vertical distance from the pixel to the origin.

[0025] Beneficial effects: Grayscale processing of images can reduce the complexity of subsequent processing, and by using the energy gradient formula to calculate and process the image, the structure in the image can be made more obvious. This enhancement can make the features more prominent, which is beneficial to the subsequent feature extraction and recognition process.

[0026] Furthermore, the calculation formula of the decision tree algorithm used by the monitoring module in predicting the weed growth trend is as follows:

[0027] Information entropy calculation formula:

[0028] Info(D)=-∑i=1kpilog2(pi)Info(D) (3).

[0029] Among them, D represents the data set, k represents the number of categories in the data set, and pi represents the ratio of the i-th category in the data set.

[0030] Information gain calculation formula:

[0031] Gain(A)=Info(D)-InfoA(D) Gain(A)=Info(D)-InfoA(D) (4).

[0032] Among them, Gain(A) represents the information gain that can be obtained by using feature A for partitioning, Info(D) represents the information entropy of data set D, and Info_A(D) represents the weighted sum of the information entropies of each subset after data set D is partitioned using feature A.

[0033] Information gain ratio calculation formula:

[0034] gr(D,A)=Gain(D,A)HA(D)g_r(D,A)=HA(D)Gain(D,A) (5).

[0035] Among them, Gain(D,A) represents the information gain that can be brought by using feature A for partitioning on dataset D, and H_A(D) represents the inherent information entropy of feature A.

[0036] Beneficial effects: By accurately predicting the growth trend and distribution of weeds through the decision tree algorithm, workers can formulate weeding plans more accurately. This means that weeding operations can be more concentrated on areas with dense weeds, avoiding blind weeding and excessive weeding, which not only improves weeding efficiency, but also reduces the waste of manpower, material resources and time.

[0037] Furthermore, the energy supply component includes a rechargeable battery fixedly connected to the top of the frame, a solar panel fixedly connected to the top of the rechargeable battery, and the solar panel is electrically connected to the rechargeable battery.

[0038] Beneficial effects: Solar panels can capture and utilize solar energy, convert it into electrical energy and store it in rechargeable batteries. The design of fixing the solar panel to the top of the rechargeable battery realizes independent energy acquisition, reduces dependence on traditional energy, and improves the sustainability of energy utilization.

[0039] Furthermore, the driving assembly includes a first driving box fixedly connected to the bottom of the frame, a first driving member is fixedly connected to the top wall of the first driving box, a controller is used to control the operation of the first driving member, a worm is fixedly connected to the output shaft of the first driving member, the worm is meshed with a worm wheel, the worm wheel is coaxially fixedly connected to a driving rod, both ends of the driving rod pass through the first driving box and are rotatably coordinated with the frame, and both ends of the driving rod are symmetrically fixedly connected to driving wheels.

[0040] Beneficial effects: The worm gear transmission system has a higher transmission ratio and greater torque transmission capacity, and can adapt to the complex and changeable terrain and operating conditions in the orchard. At the same time, the structure of the worm gear transmission method is relatively simple, easy to maintain and maintain, and reduces the operating cost and difficulty of the mechanical weeder.

[0041] Furthermore, the commutation assembly includes a second drive box fixedly connected to the bottom of the vehicle frame. A second drive member is fixedly connected to the side wall of the second drive box. The controller is used to control the operation of the second drive member. The output shaft of the second drive member penetrates through the bottom wall of the second drive box and extends outside the second drive box and is rotationally matched with the second drive box. A commutation gear is coaxially fixedly connected to the output shaft of the second drive member. The commutation gear meshes with a commutation rack. A slider is fixedly connected to the side of the commutation rack away from the commutation gear.

[0042] A fixed block is fixedly connected to the bottom of the vehicle frame. Steering wheels are symmetrically hinged to both sides of the fixed block. A slide rail is also fixedly connected to one side of the fixed block. The slider is slidably matched with the slide rail.

[0043] Hinge rods are symmetrically hinged to both ends of the commutation rack. The ends of the hinge rods away from the commutation rack are both hinged to the steering wheels.

[0044] Beneficial effects: Through the precise control of the second drive member, the meshing of the commutation gear and the commutation rack can achieve precise commutation actions, ensuring that the mechanical weeding machine can maintain stability and accuracy when turning. At the same time, the stable steering performance also ensures that the mechanical weeding machine can maintain a uniform weeding effect during operation, avoiding problems such as missed weeding or repeated weeding caused by unstable turning.

[0045] Furthermore, the monitoring assembly includes a temperature and humidity sensor and a carbon dioxide sensor fixedly connected to one side of the vehicle frame. The controller is used to receive the temperature and humidity signals sent by the temperature and humidity sensor and the carbon dioxide concentration signal sent by the carbon dioxide sensor.

[0046] Beneficial effects: The temperature and humidity sensor can real-time monitor the temperature and humidity changes in the orchard, and the carbon dioxide sensor is used to real-time monitor the carbon dioxide concentration in the orchard. Both can provide key environmental data for the management of the orchard, so as to provide data support for the subsequent optimization of the orchard.

[0047] Furthermore, the adjustment assembly includes a fixing plate fixedly connected to the top of the vehicle frame. The shape of the fixing plate is "C"-shaped. A third drive box is fixedly connected to the top of the fixing plate. A third drive member is fixedly connected to the inner side wall of the third drive box. The controller is used to control the operation of the third drive member. The output shaft of the third drive member penetrates through the bottom wall of the third drive box and the fixing plate and is rotationally matched with the bottom wall of the third drive box and the fixing plate. A screw rod is coaxially fixedly connected to the output shaft of the third drive member. The screw rod is in threaded cooperation with an inverted L-shaped adjustment plate. The adjustment plate is vertically slidably matched with the inner side wall of the fixing plate; Limiting rods are symmetrically fixedly connected to the fixing plate. The limiting rods all penetrate through the adjustment plate and are slidably matched with the adjustment plate. A weeding assembly for weeding is provided on one side of the adjustment plate.

[0048] Beneficial effects: The screw rod and the adjusting plate are threadedly matched, and the adjusting plate and the inner side wall of the fixed plate are vertically slidably matched, so that the height of the weeding component can be flexibly adjusted according to the requirements of different terrains and crop heights in the orchard. This highly adaptable feature enables the mechanical weeder to be widely used in various orchard environments, improving its application range and practicality. By accurately adjusting the height of the weeding component, it can be ensured that the distance between the weeding blade and the ground or crops is kept appropriate, thereby avoiding damage to the crops and damage to the ground.

[0049] Furthermore, the weeding assembly includes a weeding plate fixedly connected to the end of the adjusting plate away from the limiting rod, the top of the weeding plate is fixedly connected to a fourth drive box, the bottom of the weeding plate is symmetrically provided with weeding disks, the tops of the weeding disks are coaxially fixedly connected with rotating shafts, the rotating shafts penetrate the weeding plate and the bottom wall of the fourth drive box and extend into the fourth drive box and are rotatably coordinated with the weeding plate and the fourth drive box, and one end of the rotating shaft extending into the fourth drive box is coaxially fixedly connected with a first bevel gear.

[0050] A fourth driving member is fixedly connected to the inner bottom wall of the fourth driving box. The controller is used to control the operation of the fourth driving member. The output shafts of the fourth driving member are coaxially fixedly connected to the second bevel gears. The second bevel gears are meshed with the first bevel gears adjacent thereto.

[0051] Beneficial effects: The high-speed rotating weeding disc can quickly cut and remove weeds in the orchard, greatly improving the weeding efficiency and shortening the operation time. At the same time, the centrifugal force generated by the rotation of the weeding disc can pull out the weeds by the roots, avoiding the regeneration and spread of weeds and improving the weeding effect.

[0052] Furthermore, the side wall of the driving rod is symmetrically rotated with a sleeve, the top of the sleeve is fixedly connected with a shock-absorbing spring, and one end of the shock-absorbing spring away from the sleeve is fixedly connected to the frame.

[0053] Beneficial effect: When the mechanical weeder is bumped and causes the driving rod to be subjected to external shock or vibration, the shock-absorbing spring can mitigate the shock on the driving rod, thereby increasing the service life of the entire device.

[0054] Additional aspects and advantages of the present invention will be given in part in the following description and in part will be obvious from the following description, or will be learned through practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 It is an axonometric schematic diagram of an embodiment of the intelligent all-round mechanical weeder for orchards of the present invention.

[0056] Figure 2 It is a bottom view schematic diagram of an embodiment of the intelligent all-round mechanical weeder for orchards of the present invention.

[0057] Figure 3It is a schematic top-section diagram of the first driving box in the embodiment of the intelligent all-round mechanical weeder for orchards of the present invention.

[0058] Figure 4 It is a side sectional schematic diagram of the fixing plate and the adjusting plate in the embodiment of the intelligent all-round mechanical weeder for orchard of the present invention.

[0059] Figure 5 It is a front cross-sectional schematic diagram of the fourth drive box in the embodiment of the intelligent all-round mechanical weeder for orchards of the present invention.

[0060] Figure 6 It is a structural schematic diagram of the control system in the embodiment of the orchard intelligent all-round mechanical weeder of the present invention.

[0061] The figure marks in the drawings of the specification include: 1. frame; 2. camera; 3. rechargeable battery; 4. solar panel; 5. first drive box; 6. worm; 7. worm wheel; 8. drive rod; 9. drive wheel; 10. second drive box; 11. reversing gear; 12. reversing rack; 13. slider; 14. fixed block; 15. steering wheel; 16. slide rail; 17. hinged rod; 18. temperature and humidity sensor; 19. carbon dioxide sensor; 20. fixed plate; 21. third drive box; 22. screw; 23. adjustment plate; 24. limit rod; 25. weeding plate; 26. fourth drive box; 27. weeding disc; 28. rotating shaft; 29. ​​first bevel gear; 30. second bevel gear; 31. shock-absorbing spring. DETAILED DESCRIPTION

[0062] The following is further described in detail through specific implementation methods:

[0063] Embodiment 1:

[0064] As attached Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 and Figure 6 As shown: an intelligent all-round mechanical weeder for orchards, including a frame 1, a controller is fixedly connected to the top of the frame 1 by screws, a camera 2 is fixedly connected to one side of the frame 1 by screws, the controller is used to receive image information collected by the camera 2, and an energy supply component for supplying energy to the mechanical weeder is provided on the top of the frame 1.

[0065] The energy supply assembly includes a rechargeable battery 3 fixedly connected to the top of the frame 1 by bolts, a solar panel 4 fixedly connected to the top of the rechargeable battery 3 by screws, and the solar panel 4 is electrically connected to the rechargeable battery 3.

[0066] A driving assembly for driving the mechanical weeder to move and a reversing assembly for controlling the reversing of the mechanical weeder are provided at the bottom of the frame 1 .

[0067] The driving assembly includes a first driving box 5 bolted fixedly connected to the bottom of the frame 1, and a first driving member is bolted fixedly connected to the top wall of the first driving box 5. The controller is used to control the operation of the first driving member. In this embodiment, the first driving member is a first stepper motor, and a worm 6 is coaxially fixedly connected to the output shaft of the first stepper motor. The worm 6 is meshed with a worm wheel 7, and a driving rod 8 is coaxially fixedly connected to the worm wheel 7. Both ends of the driving rod 8 pass through the first driving box 5 and are rotatably matched with the frame 1. Both ends of the driving rod 8 are symmetrically bolted and fixed with driving wheels 9.

[0068] like Figure 2 As shown, the reversing assembly includes a second drive box 10 fixedly connected to the bottom of the frame 1 by bolts, and a second drive member is fixedly connected to the side wall of the second drive box 10 by bolts. The controller is used to control the operation of the second drive member. In this embodiment, the second drive member is a second stepper motor. The output shaft of the second stepper motor passes through the bottom wall of the second drive box 10 and extends to the outside of the second drive box 10 and rotates with the second drive box 10. The output shaft of the second stepper motor is coaxially fixed with a reversing gear 11, and the reversing gear 11 is meshed with a reversing rack 12. The reversing rack 12 is integrally formed with a slider 13 on the side away from the reversing gear 11.

[0069] A fixing block 14 is integrally formed at the bottom of the vehicle frame 1 , and steering wheels 15 are symmetrically hinged on both sides of the fixing block 14 . A slide rail 16 is also integrally formed on one side of the fixing block 14 , and the slider 13 slidably cooperates with the slide rail 16 .

[0070] Articulated rods 17 are symmetrically hinged at both ends of the reversing rack 12 , and one end of the articulated rod 17 away from the reversing rack 12 is hinged to the steering wheel 15 .

[0071] A monitoring component for monitoring the current environmental conditions and an adjusting component for adjusting the weeding height are provided on one side of the frame 1 .

[0072] The monitoring component includes a temperature and humidity sensor 18 and a carbon dioxide sensor 19 fixedly connected to one side of the frame 1 by screws. The controller is used to receive the temperature and humidity signals sent by the temperature and humidity sensor 18 and the carbon dioxide concentration signal sent by the carbon dioxide sensor 19.

[0073] like Figure 4As shown in the figure, the adjustment assembly includes a fixing plate 20 welded to the top of the frame 1. The fixing plate 20 is in the shape of a "C". A third drive box 21 is bolted to the top of the fixing plate 20. A third drive member is fixedly connected to the inner side wall of the third drive box 21. The controller is used to control the operation of the third drive member. In this embodiment, the third drive member is a third stepping motor. The output shaft of the third stepping motor penetrates through the bottom wall of the third drive box 21 and the fixing plate 20 and is rotationally matched with the bottom wall of the third drive box 21 and the fixing plate 20. A screw rod 22 is coaxially bolted to the output shaft of the third stepping motor. An L-shaped adjustment plate 23 is in threaded cooperation with the screw rod 22. The adjustment plate 23 is vertically slidably matched with the inner side wall of the fixing plate 20. Two limiting rods 24 are symmetrically welded inside the fixing plate 20. The limiting rods 24 penetrate through the adjustment plate 23 and are slidably matched with the adjustment plate 23. A weeding assembly for weeding is provided on one side of the adjustment plate 23.

[0074] The weeding assembly includes a weeding plate 25 welded to one end of the adjustment plate 23 away from the limiting rod 24. A fourth drive box 26 is bolted to the top of the weeding plate 25. Two weeding discs 27 are symmetrically provided at the bottom of the weeding plate 25. Rotating shafts 28 are coaxially welded to the tops of the weeding discs 27. The rotating shafts 28 penetrate through the weeding plate 25 and the bottom wall of the fourth drive box 26 and extend into the fourth drive box 26 and are rotationally matched with the weeding plate 25 and the fourth drive box 26. One end of the rotating shaft 28 extending into the fourth drive box 26 is coaxially fixedly clamped with a first bevel gear 29.

[0075] As Figure 5 shown in the figure, a fourth drive member is bolted to the inner bottom wall of the fourth drive box 26. The controller is used to control the operation of the fourth drive member. In this embodiment, the fourth drive member is a double-headed motor. Second bevel gears 30 are coaxially fixedly clamped to the output shafts of the double-headed motor. The second bevel gears 30 are meshed with the adjacent first bevel gears 29 respectively.

[0076] The controller is provided with a control system for controlling the operation of the mechanical weeding machine, including the following modules:

[0077] An information module, which is used to establish a database based on all the data of the orchard over the years, and combine the Internet large model to sort out and deeply analyze the crop categories and weed categories in the database, and establish crop feature templates and weed feature templates of different types.

[0078] An identification module, which is used to perform image recognition and feature extraction on the current environment, compare all the extracted features with all the established crop feature templates and weed feature templates respectively, distinguish the crops and weeds in the current environment image based on the database, obtain the recognition result, and store the recognition result in the database.

[0079] The recognition module is also used to recognize the fluctuations and complexities of the current environment, adjust the weeding height according to the fluctuations of the current environment, and adjust the weeding speed according to the complexity of the environment.

[0080] The monitoring module is used to monitor the current environmental data. Based on the database and the Internet big model, it analyzes the monitoring results and predicts the future weed growth trend through the decision tree algorithm based on the historical data and monitoring results. It focuses on monitoring the areas where the weed growth trend will increase significantly in the future. The environmental data include the temperature, humidity and carbon dioxide concentration of the environment.

[0081] The control module is used to formulate and execute a weed control plan based on the database, identification results and monitoring results.

[0082] The specific implementation process is as follows: the staff starts the first stepper motor through the controller. At this time, the output shaft of the first stepper motor rotates, driving the worm 6 coaxially fixed and clamped therewith to rotate, the worm 6 drives the worm wheel 7 meshing therewith to rotate, the worm wheel 7 drives the driving rod 8 coaxially fixed and clamped therewith to rotate, and the driving rod 8 drives the driving wheel 9 to rotate, thereby realizing the movement of the mechanical weeder.

[0083] like Figure 2 As shown, when a steering operation is required, the controller controls the second stepper motor to start, and the output shaft of the second stepper motor rotates, driving the reversing gear 11 coaxially fixed and connected thereto to rotate. Since the reversing gear 11 is meshed with the reversing rack 12, the reversing gear 11 will drive the reversing rack 12 to move laterally during the rotation process, and the reversing rack 12 drives the slider 13 integrally formed therewith to slide laterally on the slide rail 16. At this time, since the two ends of the articulated rod 17 are respectively articulated with the reversing rack 12 and the steering wheel 15, when the reversing rack 12 moves laterally, the articulated rod 17 moves laterally together with the reversing rack 12, and at this time the steering wheel 15 is pulled under the drive of the articulated rod 17, so that the steering wheel 15 is turned, and then the steering of the mechanical weeder is realized.

[0084] like Figure 4 As shown, when the mechanical weeder needs to adjust the weeding height, the controller starts the third stepper motor, the output shaft of the third stepper motor rotates, and the output shaft of the third stepper motor drives the screw rod 22 fixedly connected with the coaxial bolt to rotate, and the screw rod 22 drives the adjustment plate 23 matched with its thread to rise and fall, and the adjustment plate 23 is limited in the vertical direction by the limit rod 24.

[0085] The adjusting plate 23 will drive the weeding plate 25 welded thereto to rise and fall together during the lifting process. When the weeding plate 25 reaches a predetermined height, the controller starts the double-headed motor to make its output shaft rotate synchronously. Figure 5As shown, at this time, since the output shafts of the double-headed motors are coaxially fixedly connected to the second bevel gear 30, the first bevel gears 29 are meshed with the second bevel gears 30, and the first bevel gears 29 are coaxially fixedly connected to the rotating shaft 28, and the end of the rotating shaft 28 away from the first bevel gear 29 is coaxially welded to the weeding disk 27, when the output shaft of the double-headed motor drives the second bevel gear 30 to rotate, the second bevel gear 30 will drive the first bevel gear 29 to rotate, and the first bevel gear 29 will drive the rotating shaft 28 to rotate, and the rotating shaft 28 will drive the weeding disk 27 to rotate to perform a weeding operation.

[0086] During this process, the control system will develop a targeted weed control plan.

[0087] First, the information module collects all the data of the orchard over the years and establishes a database. Combined with the Internet big model, the database is sorted and deeply analyzed for crop categories and weed categories, and different types of crop feature templates and weed feature templates are established. At the same time, the recognition module uses camera 2 to perform image recognition and analysis on the environment where the mechanical weeder is located, and extracts features from the current environment image. All extracted features are compared with all established crop feature templates and weed feature templates, and weeds and crops in the current environment image are distinguished based on the database. The recognition results are obtained and stored in the database to achieve deep learning of the database. The recognition module also uses camera 2 to identify the ups and downs of the current environment and the complexity of the environment, and adjusts the weeding height according to the ups and downs of the current environment, and adjusts the weeding speed according to the complexity of the environment.

[0088] At the same time, the monitoring module monitors the temperature and humidity conditions and carbon dioxide concentration of the current environment in real time through the temperature and humidity sensor 18 and the carbon dioxide sensor 19, and summarizes the temperature and humidity signals and carbon dioxide concentration signals of the current environment. Based on the database and the Internet big model, the monitoring results are analyzed and obtained. The future weed growth trend is predicted based on the monitoring results, and key monitoring is carried out on areas where the weed growth trend will increase significantly in the future. Finally, the control module formulates and executes a suitable weed control plan based on the database, identification results and monitoring results.

[0089] For example, when the temperature and humidity sensor 18 and the carbon dioxide sensor 19 complete the monitoring of the temperature and humidity conditions and carbon dioxide concentration conditions of the current environment, the monitoring module will judge the temperature and humidity conditions and carbon dioxide concentration conditions of the environment through the database and the Internet big model, and obtain the type of weeds that can grow appropriately in the current environment, and judge the growth ability of this type of weeds. At the same time, the monitoring module predicts the future temperature and humidity change trends and carbon dioxide concentration change trends of the current environment based on the Internet big model, judges the future weed growth trend, and divides and focuses on monitoring the areas where the future weed growth trend will increase significantly.

[0090] When formulating a weeding plan, the control module uses the data in the database as a basis and combines the recognition results of the recognition module to judge the weeds and plants in the environment. When the weeds that can grow in the current environment have a strong growth ability, the control module will adopt a weeding plan of multiple weedings, and control the mechanical weeder to move back and forth in the area to weed, so as to ensure that the weeds can be removed. During the weeding process, the control module identifies the lushness of the weeds in the area according to the recognition module, and intelligently adjusts the rotation speed of the weeding disc 27 according to the lushness of the weeds in the area, so that the rotation speed of the weeding disc 27 is proportional to the lushness of the weeds.

[0091] When formulating a weeding plan, the control module will also control the mechanical weeder to patrol the area where the monitoring module predicts a significant increase in weed growth trend in the future after weeding is completed, so as to achieve key monitoring of the area.

[0092] Embodiment 2:

[0093] The difference from Example 1 is that when the recognition module performs image recognition and feature extraction, the image is first subjected to grayscale processing. The grayscale processing formula is as follows:

[0094]

[0095] Among them, Gray represents the grayscale value, R represents the red component of the pixel in the image, G represents the green component of the pixel in the image, and B represents the blue component of the pixel in the image.

[0096] After the grayscale processing is completed, the recognition module uses the energy gradient formula to clarify the image. The energy gradient formula is as follows:

[0097]

[0098] The pixel at the lower left corner of the image is taken as the origin, x is the horizontal distance from the pixel to the origin, and y is the vertical distance from the pixel to the origin.

[0099] The specific implementation process is as follows: the recognition module first performs grayscale processing on the collected image using the grayscale processing formula, and after the grayscale processing is completed, the image is clarified using the energy gradient formula.

[0100] Embodiment 3:

[0101] The difference from Example 2 is that the calculation formula of the decision tree algorithm used by the monitoring module when predicting the weed growth trend is as follows:

[0102] Information entropy calculation formula:

[0103] Info(D)=-∑i=1kpilog2(pi)Info(D) (3).

[0104] Among them, D represents the data set, k represents the number of categories in the data set, and pi represents the ratio of the i-th category in the data set.

[0105] Information gain calculation formula:

[0106] Gain(A)=Info(D)-Info(D) Gain(A)=Info(D)-InfoA(D) (4).

[0107] Among them, Gain(A) represents the information gain that can be obtained by using feature A for partitioning, Info(D) represents the information entropy of data set D, and Info_A(D) represents the weighted sum of the information entropies of each subset after data set D is partitioned using feature A.

[0108] Information gain ratio calculation formula:

[0109] gr(D,A)=Gain(D,A)HA(D)g_r(D,A)=HA(D)Gain(D,A) (5).

[0110] Among them, Gain(D,A) represents the information gain that can be brought by using feature A for partitioning on dataset D, and H_A(D) represents the inherent information entropy of feature A.

[0111] The specific implementation process is as follows: the monitoring module first calculates the information entropy of each data through the information entropy calculation formula to measure the uncertainty or confusion of the data, and then calculates the information gain of each data through the information gain calculation formula to obtain the degree of reduction in the uncertainty of the data set after using a certain feature for division. Finally, the information gain ratio calculation formula is used to calculate the impact of the number of feature values ​​on the information gain to avoid selecting features with too many values.

[0112] Embodiment 4:

[0113] As attached Figure 2 As shown, the difference from Example 3 is that sleeves are symmetrically rotated at both ends of the driving rod 8, and the top of the sleeve is bolted and fixedly connected with a shock-absorbing spring 31, and the end of the shock-absorbing spring 31 away from the sleeve is bolted and fixedly connected to the frame 1.

[0114] The specific implementation process is as follows: when the mechanical weeder is subjected to a bump, the shock absorbing spring 31 can buffer the bump suffered by the mechanical weeder.

[0115] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived therefrom are still within the scope of protection of the invention.

Claims

1. An intelligent all-round mechanical weeder for orchards, comprising a frame (1), a controller fixedly connected to the top of the frame (1), a camera (2) fixedly connected to one side of the frame (1), the controller being used to receive image information collected by the camera (2), characterized in that: The top of the vehicle frame (1) is provided with an energy supply component for supplying energy to the mechanical weeder, the bottom of the vehicle frame (1) is provided with a driving component for driving the mechanical weeder to move and a reversing component for controlling the reversal of the mechanical weeder, and one side of the vehicle frame (1) is provided with a monitoring component for monitoring the current environmental conditions and an adjusting component for adjusting the weeding height; The controller is equipped with a control system for controlling the operation of the mechanical weeder, including the following modules: The information module is used to establish a database based on all the data of the orchard over the years, and to organize and deeply analyze the database by crop and weed categories in combination with the Internet big model, and to establish different types of crop feature templates and weed feature templates; The recognition module is used to perform image recognition and feature extraction on the current environment, compare all extracted features with all established crop feature templates and weed feature templates, distinguish crops and weeds in the current environment image based on the database, obtain recognition results, and store the recognition results in the database; The recognition module is also used to recognize the fluctuations and complexities of the current environment, adjust the weeding height according to the fluctuations of the current environment, and adjust the weeding speed according to the complexities of the environment; The monitoring module is used to monitor the current environmental data, analyze the monitoring results based on the database and the Internet big model, and predict the future weed growth trend through the decision tree algorithm based on the historical data and monitoring results, and focus on monitoring the areas where the weed growth trend will increase significantly in the future. The environmental data includes the temperature, humidity and carbon dioxide concentration of the environment; The control module is used to formulate and execute a weed control plan based on the database, identification results and monitoring results.

2. The orchard intelligent all-round mechanical weeder according to claim 1, characterized in that: When the recognition module performs image recognition and feature extraction, it first performs grayscale processing on the image. The grayscale processing formula is as follows: Among them, Gray represents the gray value, R represents the red component of the pixel in the image, G represents the green component of the pixel in the image, and B represents the blue component of the pixel in the image; After the grayscale processing is completed, the recognition module uses the energy gradient formula to clarify the image. The energy gradient formula is as follows: The pixel at the lower left corner of the image is taken as the origin, x is the horizontal distance from the pixel to the origin, and y is the vertical distance from the pixel to the origin.

3. The orchard intelligent all-round mechanical weeder according to claim 2, characterized in that: The calculation formula of the decision tree algorithm used by the monitoring module to predict the weed growth trend is as follows: Information entropy calculation formula: Info(D)=-∑i=1kpilog2(pi)Info(D)(3); Where D represents the data set, k represents the number of categories in the data set, and pi represents the ratio of the i-th category in the data set; Information gain calculation formula: Gain(A)=Info(D)-InfoA(D) Gain(A)=Info(D)-InfoA(D)(4); Where Gain(A) represents the information gain that can be obtained by using feature A for partitioning, Info(D) represents the information entropy of data set D, and Info A(D) represents the weighted sum of the information entropies of each subset after data set D is partitioned using feature A. Information gain ratio calculation formula: $g_r(D, A)=\frac{Gain(D, A)}{H_A(D)}$ (5); Among them, $Gain(D, A)$ represents the information gain that can be brought by using feature $A$ for partitioning on dataset $D$, and $H_A(D)$ represents the inherent information entropy of feature $A$.

4. The orchard intelligent all-round mechanical weeder according to claim 3, characterized in that: The power supply component includes a rechargeable battery (3) fixedly connected to the top of the vehicle frame (1). A solar panel (4) is fixedly connected to the top of the rechargeable battery (3), and the solar panel (4) is electrically connected to the rechargeable battery (3).

5. The orchard intelligent all-round mechanical weeder according to claim 4, characterized in that: The driving component includes a first driving box (5) fixedly connected to the bottom of the vehicle frame (1). A first driving member is fixedly connected to the inner top wall of the first driving box (5), and the controller is used to control the operation of the first driving member; A worm (6) is fixedly connected to the output shaft of the first driving member. The worm (6) meshes with a worm wheel (7). A driving rod (8) is coaxially fixedly connected to the worm wheel (7). Both ends of the driving rod (8) penetrate through the first driving box (5) and are rotatably matched with the vehicle frame (1). Driving wheels (9) are symmetrically fixedly connected to both ends of the driving rod (8).

6. The orchard intelligent all-round mechanical weeder according to claim 5, characterized in that: The commutation component includes a second driving box (10) fixedly connected to the bottom of the vehicle frame (1). A second driving member is fixedly connected to the side wall of the second driving box (10), and the controller is used to control the operation of the second driving member; The output shaft of the second driving member penetrates through the bottom wall of the second driving box (10) and extends outside the second driving box (10) and is rotatably matched with the second driving box (10). A commutation gear (11) is coaxially fixedly connected to the output shaft of the second driving member. The commutation gear (11) meshes with a commutation rack (12). A slider (13) is fixedly connected to the side of the commutation rack (12) away from the commutation gear (11); A fixed block (14) is fixedly connected to the bottom of the vehicle frame (1). Steering wheels (15) are symmetrically hinged to both sides of the fixed block (14). A slide rail (16) is also fixedly connected to one side of the fixed block (14). The slider (13) is slidably matched with the slide rail (16); Hinge rods (17) are symmetrically hinged to both ends of the commutation rack (12). One end of each hinge rod (17) away from the commutation rack (12) is hinged to the steering wheel (15).

7. The intelligent all-round mechanical weeder for orchards according to claim 6, characterized in that: The monitoring component includes a temperature and humidity sensor (18) and a carbon dioxide sensor (19) fixedly connected to one side of the vehicle frame (1). The controller is used to receive the temperature and humidity signal sent by the temperature and humidity sensor (18) and the carbon dioxide concentration signal sent by the carbon dioxide sensor (19).

8. The orchard intelligent omnidirectional mechanical weeder according to claim 7, characterized in that: The adjustment component includes a fixing plate (20) fixedly connected to the top of the vehicle frame (1). The shape of the fixing plate (20) is "C". A third driving box (21) is fixedly connected to the top of the fixing plate (20). A third driving member is fixedly connected to the inner side wall of the third driving box (21), and the controller is used to control the operation of the third driving member; The output shaft of the third driving member penetrates the bottom wall of the third driving box (21) and the fixed plate (20) and is rotatably matched with the bottom wall of the third driving box (21) and the fixed plate (20); the output shaft of the third driving member is coaxially fixedly connected with a screw rod (22), the screw rod (22) is threadedly matched with an inverted L-shaped adjustment plate (23), and the adjustment plate (23) is vertically slidably matched with the inner side wall of the fixed plate (20); the fixed plate (20) is symmetrically fixedly connected with a limit rod (24), the limit rod (24) penetrates the adjustment plate (23) and is slidably matched with the adjustment plate (23); a weeding component for weeding is provided on one side of the adjustment plate (23).

9. The orchard intelligent omnidirectional mechanical weeder according to claim 8, characterized in that: The weeding assembly comprises a weeding plate (25) fixedly connected to an end of the adjusting plate (23) away from the limiting rod (24), and a fourth driving box (26) is fixedly connected to the top of the weeding plate (25); A weeding disc (27) is symmetrically arranged at the bottom of the weeding plate (25), and a rotating shaft (28) is coaxially fixedly connected to the top of the weeding disc (27). The rotating shaft (28) penetrates the weeding plate (25) and the bottom wall of the fourth drive box (26) and extends into the fourth drive box (26) and is rotatably matched with the weeding plate (25) and the fourth drive box (26). One end of the rotating shaft (28) extending into the fourth drive box (26) is coaxially fixedly connected to a first bevel gear (29); A fourth driving member is fixedly connected to the inner bottom wall of the fourth driving box (26), and the controller is used to control the operation of the fourth driving member; The output shafts of the fourth driving members are all coaxially fixedly connected with the second bevel gears (30), and the second bevel gears (30) are all meshed with the first bevel gears (29) adjacent thereto.

10. The orchard intelligent all-round mechanical weeder according to claim 9, characterized in that: The side wall of the driving rod (8) is symmetrically rotated with a sleeve, the top of the sleeve is fixedly connected with a shock absorbing spring (31), and the end of the shock absorbing spring (31) away from the sleeve is fixedly connected to the vehicle frame (1).

Images