Automatic inspection navigation corn cultivator and method

Abstract

An automated inspection and navigation-based corn weeding robot and method are disclosed. The weeding robot includes: a mobile vehicle equipped with a visual perception component for identifying crops and weeds and a navigation component for planning field paths; and a rotary tillage mechanism including a rotary tillage shaft located at the rear of the mobile vehicle and components capable of driving the rotary tillage shaft to rise, fall, and rotate. The axial direction of the rotary tillage shaft is along the width direction of the mobile vehicle. Multiple blades capable of rotating are mounted on the rotary tillage shaft via a blade mounting disc. The blade tips trace on the same spindle-shaped envelope, which is adapted to the cross-section of a corn ridge that is low in the middle and high on both sides. This weeding robot and method assist in weeding and loosening the soil around corn plants, ensuring the integrity of the ridge surface and the safety of the plants during operation, thereby improving the growth quality of corn during the inter-row cultivation period.

Description

Technical Field

[0001] This invention relates to the field of weeding robot technology, specifically to an automatic inspection and navigation corn weeding robot and method. Background Technology

[0002] Intertillage management of corn is a key step in ensuring high yields. It requires soil cultivation between rows and plants, either mechanically or manually. The core objective is to create a soil environment conducive to corn root growth while suppressing weeds. The main tasks during intertillage are loosening the soil and weeding.

[0003] Loosening the soil is an important soil management measure during crop cultivation. Timely loosening of the soil can improve soil aeration, enhance soil water retention and drainage capacity, regulate soil temperature, and directly improve the soil environment. It can also promote crop root growth and development, improve water and fertilizer utilization, suppress weed growth, and increase crop yield, among many other benefits.

[0004] During the crop growth period, weeds can affect crop growth, reduce crop yield and quality, induce and spread pests and diseases, and some weeds mixed with crops may cause adverse reactions in humans. Therefore, weeding is one of the important soil management measures during crop planting.

[0005] Currently, conventional weeding methods can be broadly categorized into chemical weeding and physical weeding. Common chemical weeding typically involves the use of various herbicides. While this method is highly efficient and easy to operate, long-term use can lead to soil degradation and pollution, water pollution, and other harmful effects on the ecological environment. Furthermore, long-term, excessive use of herbicides increases chemical residues in agricultural products and can also impact the health of users. Common physical weeding methods include manual weeding and mechanical weeding. Manual weeding is labor-intensive and extremely inefficient, especially during the corn planting stage. At this stage, corn seedlings have grown to a certain height and have high leaf coverage, making manual weeding difficult due to poor visibility and reduced efficiency. Manual weeding also carries a significant risk of damaging the raised beds in standardized farmland. Mechanical weeding often uses various agricultural implements, but this method is not only dependent on the crop's growth stage but also fails to remove weeds around the crop roots and may damage shallow root systems.

[0006] To improve the growth quality of corn during the inter-cultivation period, it is necessary to propose an automated inspection and navigation corn inter-cultivation and weeding robot and method. Summary of the Invention

[0007] The purpose of this invention is to propose an automatic inspection and navigation corn weeding robot and method to assist in weeding and loosening the soil of corn plants. During the operation, the integrity of the ridge surface and the safety of the plants can be guaranteed, thereby improving the growth quality of corn during the inter-cultivation period.

[0008] The technical solution adopted in this invention is: an automatic inspection and navigation corn tillage and weeding robot, comprising: The mobile vehicle is equipped with visual perception components for identifying crops and weeds, as well as navigation components for planning field paths. A rotary tillage mechanism, comprising a rotary tillage shaft disposed at the rear of the mobile vehicle body and components capable of driving the rotary tillage shaft to lift, lower, and rotate itself; The axial direction of the rotary tiller shaft is along the width direction of the moving vehicle body. Multiple blades that can rotate are mounted on the rotary tiller shaft via a blade mounting disc. The blade tip trajectory is on the same spindle-shaped envelope surface, which is adapted to the cross-section of the cornfield ridge that is low in the middle and high on both sides. The weeding mechanism includes two weeding shovels disposed on both sides of the mobile vehicle body and a component capable of driving the two weeding shovels to move towards or away from each other in the width direction of the mobile vehicle body. An information processing terminal is configured to receive recognition information from the visual perception component during the movement of the mobile vehicle, and to: drive the mobile vehicle to the center of an adjacent ridge, so that the blades are in contact with the ridge; and The weeding mechanism is driven so that the two weeding shovels selectively avoid crops and move only to the weed area to clear it.

[0009] As a preferred option, the mobile vehicle body has a streamlined overall shape, which is used to flexibly separate the corn plants on both sides during the inspection process.

[0010] As a preferred embodiment, the rear of the mobile vehicle is provided with a rotary tillage baffle, which at least partially surrounds the outside of the rotary tillage mechanism.

[0011] As a preferred embodiment, the visual perception component includes ToF depth cameras disposed at the head of the moving vehicle and on both sides of the moving vehicle.

[0012] As a preferred option, the navigation components include a remote communication device and a satellite positioning system.

[0013] As a preferred embodiment, the components capable of driving the rotary tiller shaft to rotate include: a cutter shaft lifting and power transmission arm support shaft rotatably mounted on the mobile vehicle body, cutter shaft lifting and power transmission arms fixed at both ends of the cutter shaft lifting and power transmission arm support shaft, a cutter shaft fixing plate fixed at the tail end of the cutter shaft lifting and power transmission arm, a cutter shaft connecting shaft rotatably mounted inside the cutter shaft fixing plate, and two cutter shaft connecting shafts fixedly connected to both ends of the rotary tiller shaft respectively. The mobile vehicle body is also equipped with a rotary tillage shaft drive seat that drives the blade shaft connecting shaft to rotate.

[0014] As a preferred option, the components capable of raising and lowering the rotary tiller shaft itself include a hydraulic lifting rod and a hydraulic lifting control system: The two ends of the hydraulic lifting rod are respectively hinged to the moving vehicle body and the support shaft of the cutting shaft lifting and power transmission arm. The input end of the hydraulic lifting control system is electrically connected to the information processing terminal, and the output end of the hydraulic lifting control system is electrically connected to the hydraulic lifting rod.

[0015] As a preferred embodiment, the component capable of driving the two weeding shovels to move towards or away from each other in the width direction of the mobile vehicle body includes: Two slide rails are disposed inside the moving vehicle body along its width direction; Two telescopic arms for weeding are slidably mounted in two slide rails, and the two weeding shovels are fixed to the outer ends of the corresponding telescopic arms. Two telescopic motors for weeding shovels are mounted on the mobile vehicle body. The output shafts of the telescopic motors for weeding shovels are equipped with gears that mesh with the toothed grooves on the corresponding telescopic arms of the weeding shovels.

[0016] As a preferred option, the two weeding shovels are arranged in an inverted "V" shape parallel to the ridge slope.

[0017] An automated inspection and navigation method for inter-row corn cultivation and weeding, the method utilizing the aforementioned automated inspection and navigation method for inter-row corn cultivation and weeding, includes the following steps: Step 1: Real-time acquisition of depth images and infrared intensity images of the field scene using a ToF depth camera mounted on a mobile vehicle; The information processing terminal filters the depth image to generate a three-dimensional point cloud, and uses a random sampling consistency algorithm to extract vegetation point cloud, which includes points of crops and weeds. The vegetation point cloud is horizontally sliced ​​along the vertical direction, and the points in each slice are clustered using the Euclidean clustering algorithm to obtain the feature point cloud. Step 2: Transform the detected crop row plane equations from the camera coordinate system to the moving vehicle coordinate system, and output the inter-row pose estimation by combining the motion model of the moving vehicle. Step 3: Centered on the mobile vehicle, generate a local occupancy grid map of a fan-shaped area based on the field of view of the ToF depth camera, and convert the local map to the global coordinate system based on the real-time global pose of the mobile vehicle inspection. Then, use Bayesian update rules to integrate the occupancy probability information of the local map into the global occupancy grid map to update the understanding of the layout of crops and weeds in the field. Step 4: Based on the generated global occupancy grid map and the coordinates of the moving vehicle, weeds, and crops, the information processing terminal determines: if the detectable area in front of the current weeding shovel is identified as a weed spot or a crop-free area, then the weeding shovel is extended to perform weeding and cultivation; if the detectable area in front of the current weeding shovel is identified as a crop point cloud, then the weeding shovel is retracted to avoid the crops.

[0018] Compared with the prior art, the beneficial effects of the present invention are: 1. The vehicle body adopts a dolphin-inspired streamlined design, which can smoothly push aside corn leaves during movement to avoid scratching the stems and leaves; at the same time, it reduces the impact of leaves on satellite signals, ensuring that the robot can move smoothly between densely planted rows during the mid-tillage period.

[0019] 2. The rotary tiller shaft is spindle-shaped, thicker in the middle and thinner at both ends. The trajectory of the blade tip forms a spindle-shaped envelope that is deep in the middle and shallow on both sides, which naturally fits the cross section of the furrow. During operation, it can reach deep into the bottom of the furrow to loosen the soil without damaging the ridge base on both sides, thus solving the problem of ridge collapse caused by traditional rotary tillers.

[0020] 3. By using a ToF depth camera to collect point clouds, and then fusing them with slice clustering and grid maps, the system can still stably estimate the row pose even when satellite signals are blocked. Based on the crop / weed identification results, the system can independently control the extension and retraction of the left and right weeding shovels to achieve precise weeding by extending when encountering weeds and retracting when encountering seedlings. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1 This is an axonometric schematic diagram of the robot in this invention; Figure 2 This is a schematic diagram of the weeding mechanism in this invention; Figure 3 This is a top view of the robot in this invention; Figure 4 This is a schematic diagram of one side of the robot in this invention; Figure 5 This is a schematic diagram of the structure on the other side of the robot of the present invention; Figure 6 This is a schematic diagram of the structure of the vehicle shell in this invention; Figure 7 This is a schematic diagram of the rotary tiller shaft and blades in this invention; Figure 8 This is an overall flowchart of the method of the present invention.

[0023] Reference numerals: 1. Vehicle body; 2. Rotary tiller baffle fixing support; 3. Rotary tiller baffle; 4. Remote communication device; 5. Satellite positioning system; 6. Satellite positioning system; 7. Rotary tiller shaft; 8. Blade; 9. Blade mounting plate; 10. Blade shaft fixing plate; 11. Blade shaft connecting shaft; 12A. Blade shaft lifting and power transmission arm; 12B. Blade shaft lifting and power transmission arm; 13A. Depth camera; 13B. Depth camera; 14A. Weeding shovel; 14B. Weeding shovel; 15A. Weeding shovel telescopic arm; 15B. Telescopic arm of weeding shovel; 16A. Telescopic motor of weeding shovel; 16B. Telescopic motor of weeding shovel; 17A. Wheel; 17B. Wheel; 17C. Wheel; 17D. Wheel; 18. ToF depth camera; 19A. Hydraulic lifting rod; 19B. Hydraulic lifting rod; 20. Hydraulic lifting control system; 21A. Slide rail; 21B. Slide rail; 22. Cutter shaft lifting and power transmission arm support shaft; 23. Hydraulic lifting power system; 24. Mobile chassis power system and vehicle power supply. Detailed Implementation

[0024] The present invention will now be described in detail through exemplary embodiments. However, it should be understood that, without further description, elements, structures, and features in one embodiment may be advantageously incorporated into other embodiments.

[0025] It should be noted that, unless otherwise defined, the technical or scientific terms used herein should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "a," "an," or "the," etc., used in the specification and claims of this patent application do not express a limitation on quantity, but rather indicate the presence of at least one; the terms "first," "second," and "third," as used herein, should not be considered as a limitation on the order of components, but are merely for distinguishing different components; the terms "comprising," "including," etc., indicate that the elements or objects preceding "comprising" or "including" encompass the elements or objects listed following "comprising" or "including" and their equivalents, but do not exclude other elements or objects having the same function.

[0026] To more clearly describe the automated inspection and navigation corn weeding robot and its method, in conjunction with the attached... Figure 1-8 Description of the embodiment: like Figure 1-7 As shown, this invention discloses an automatic inspection and navigation corn tillage and weeding robot, comprising a mobile vehicle body, a rotary tillage mechanism, a weeding mechanism, and an information processing terminal: The mobile vehicle body has an intelligent chassis and a body shell 1. The intelligent chassis has four wheels (17A, 17B, 17C, 17D) and a mobile chassis power system and a vehicle power supply 24 for controlling the wheels.

[0027] The four wheels are independently driven to adapt to complex terrain in the field and adverse travel conditions under varying weather conditions. The mobile chassis power system and the vehicle's power supply are laid flat at the bottom of the chassis inside the vehicle body, effectively utilizing the internal space to install a larger capacity battery, ensuring the robot's continuous working time and greater power output in the field. This layout provides a continuous and stable power output to the four drive wheels, the motors for the weeding telescopic device, the rotary tiller motor, and the hydraulic lifting system. It also effectively lowers the vehicle's center of gravity, significantly increasing the robot's anti-tipping ability in complex field conditions, ensuring the proper completion of weeding and soil loosening operations.

[0028] Referring to Figure 6, the vehicle body 1 adopts a dolphin-inspired streamlined design. Its geometric features are as follows: the front of the vehicle features a rounded dolphin-snout-shaped protrusion, followed by lines that smoothly extend upwards and backwards, covering all internal components. The smooth and angleless curved surface design of the vehicle body is not only for aesthetics, but its core function is "flexible seed division." When the robot operates in the field during the corn cultivation period, the corn plants are tall and the leaves extend laterally. Traditional vehicle body designs are prone to damaging the plant leaves. The streamlined vehicle body acts as a seed divider, gently pushing the leaves extending towards the furrows to both sides, avoiding the scratching and breaking of crop stems and leaves caused by the sharp edges of traditional vehicle body designs, and protecting the crop's growth to the greatest extent.

[0029] The mobile vehicle is equipped with a visual perception component for identifying crops and weeds and a navigation component for planning field paths. The visual perception component includes a ToF depth camera 18 located at the head of the mobile vehicle and ToF depth cameras (13A, 13B) located on both sides of the mobile vehicle. The navigation component includes a remote communication device 4 and a satellite positioning system (5, 6). The remote communication device 4 can receive remote signals and control the movement of the mobile vehicle. The satellite positioning system (5, 6) is used to sense the real-time position of the mobile vehicle.

[0030] The rotary tillage mechanism is designed to loosen the soil during the inter-row cultivation of corn, solving problems such as soil compaction between rows and creating a better and healthier soil environment for corn growth. The rotary tillage mechanism includes a rotary tillage shaft 7 located at the rear of the mobile vehicle and components that drive the rotary tillage shaft 7 to rise, fall, and rotate. The axis of the rotary tillage shaft 7 is along the width of the mobile vehicle. Multiple blades 8, which can rotate with it, are mounted on the rotary tillage shaft 7 via a blade mounting disc 9. The blade tips of the blades 8 follow the same spindle-shaped envelope, which is adapted to the cross-section of the corn ridge, which is concave in the middle and raised on both sides.

[0031] See Figure 7The rotary tiller shaft 7 does not use the traditional uniform-diameter cylindrical shaft, but is designed as a spindle shape, with the largest diameter in the middle section of the shaft, gradually tapering towards both ends. The blade mounting disc 9 and the blades 8 are distributed on the surface of this spindle shape. Due to the variation in the rotary tiller shaft diameter and the matching of the blade holder height, when the rotary tiller shaft rotates, the blade tip trajectories of all blades form a concave arc envelope surface that is deep in the middle and shallow at both ends. This envelope surface matches the physical cross-section of the cornfield ridge, which is low in the middle and high at both ends. During operation, the long-cutting-radius blade in the middle penetrates into the compacted layer at the bottom of the furrow to break up and loosen the soil, while the short-cutting-radius blades at both ends only slightly touch the foot of the ridge. This effectively avoids the risk of the traditional straight rotary tiller blades cutting the ridge base on both sides, causing the ridge to collapse, and ensures the integrity of the ridge shape during plant growth.

[0032] In one specific embodiment, the components capable of driving the rotary tiller 7 to rotate include a cutter shaft lifting and power transmission arm support shaft 22 rotatably mounted on the mobile vehicle body. Cutter shaft lifting and power transmission arm (12A, 12B) are fixed at both ends of the cutter shaft lifting and power transmission arm (12A, 12B). A cutter shaft fixing plate 10 is fixed at the tail end of the cutter shaft lifting and power transmission arm (12A, 12B). A cutter shaft connecting shaft 11 is rotatably mounted inside the cutter shaft fixing plate 10. The two cutter shaft connecting shafts 11 are fixedly connected to both ends of the rotary tiller 7, respectively. The mobile vehicle body is also provided with a rotary tiller drive seat 23 that drives the cutter shaft connecting shaft 11 to rotate. The rotary tiller drive seat 23 has a built-in motor, which is connected to the cutter shaft connecting shaft 11 through a belt or chain, thereby driving the rotary tiller 7 and the blades 8 to rotate, thus completing the loosening of the soil.

[0033] In one specific embodiment, the components capable of raising and lowering the rotary tiller shaft 7 include a hydraulic lifting rod (19A, 19B) and a hydraulic lifting control system 20. The two ends of the hydraulic lifting rod (19A, 19B) are hinged to the mobile vehicle body and the cutter shaft lifting and power transmission arm support shaft 22, respectively. The input end of the hydraulic lifting control system 20 is electrically connected to an information processing terminal, and the output end of the hydraulic lifting control system 20 is electrically connected to the hydraulic lifting rod (19A, 19B). When the hydraulic lifting rod extends, it drives the cutter shaft lifting and power transmission arm (12A, 12B) to rotate downwards by a certain angle around the cutter shaft lifting and power transmission arm support shaft 22, causing the blade 8 to be pressed into the soil to a predetermined depth. When the hydraulic lifting rod retracts, the cutter shaft lifting and power transmission arm (12A, 12B) lifts the blade 8, preventing damage to the blade in non-working areas.

[0034] Furthermore, a rotary tillage baffle 3 is fixed to the rear of the mobile vehicle body via a rotary tillage baffle support 2. The rotary tillage baffle 3 at least partially surrounds the outside of the rotary tillage mechanism. The rotary tillage baffle 3 serves two purposes: first, it physically blocks soil clods and stones kicked up by the high-speed rotating blades, preventing them from splashing and injuring people or damaging crops behind; second, it can serve as a mounting base for installing the remote communication device 4, satellite positioning system 5, and satellite positioning system 6 onto the rotary tillage baffle 3.

[0035] The weeding mechanism includes two (14A, 14B) symmetrically arranged on both sides of the mobile vehicle body, and a component that can drive the two weeding shovels (14A, 14B) to move towards or away from each other in the width direction of the mobile vehicle body, for the purpose of precisely removing weeds on the ridge surface.

[0036] In one specific embodiment, the components that enable the two weeding shovels (14A, 14B) to move towards or away from each other in the width direction of the mobile vehicle include slide rails, weeding shovel telescopic arms, and weeding shovel telescopic motors. Two slide rails (21A, 21B) are disposed within the mobile vehicle along its width direction; two weeding shovel telescopic arms (15A, 15B) are slidably disposed within the two slide rails (21A, 21B), and the two weeding shovels (14A, 14B) are fixed to the outer ends of their respective weeding shovel telescopic arms (15A, 15B); two weeding shovel telescopic motors (16A, 16B) are mounted on the mobile vehicle, and gears are mounted on the output shafts of the weeding shovel telescopic motors (16A, 16B), which mesh with the toothed grooves on the corresponding weeding shovel telescopic arms (15A, 15B).

[0037] This weeding mechanism adopts a rack and pinion direct-drive telescopic mechanism, which has high transmission efficiency and fast response speed compared to traditional hydraulic or screw drives. When the robot detects changes in row width or row spacing deviations through the depth camera during its movement, the telescopic motors (16A, 16B) of the weeding shovel can drive the telescopic arms (15A, 15B) of the weeding shovel to extend and retract rapidly, driving the weeding shovels (14A, 14B) to perform rapid displacement correction, achieving precise row-by-row weeding with high dynamic response. At the same time, the mechanism has good rigidity and high load-bearing capacity, which can effectively overcome the soil resistance encountered by the weeding shovel when it enters the soil, prevent deformation of the mechanism, and has high anti-pollution ability, ensuring operational stability.

[0038] See Figure 2Two weeding shovels (14A, 14B) are arranged in an inverted "V" shape parallel to the ridge slope. The blades of the shovels are inclined at a preset angle to the horizontal plane, which is set to be parallel to the ridge slope angle of standardized farmland. The width of the shovels is consistent with the length of the ridge surface to ensure that during weeding operations, the entire ridge surface can be covered in all directions to remove weeds from all locations on the ridge surface. The inclined angle parallel to the ridge surface also protects the ridge shape from damage. During operation, depth cameras (13A, 13B) on the side of the vehicle scan the ridge surface on both sides in real time. Detecting the location of weeds on the ridge surface and the distance between the ridge surface and the weeding shovels, the information processing terminal immediately instructs the weeding shovel extension motors (16A, 16B) to rotate. The motors drive the extension arms to extend in a short time through gear and rack transmission, so that the weeding shovels are always parallel to the ridge surface to remove weeds from the surface layer of the ridge, ensuring thorough weeding without damaging the soil structure of the ridge. After the weeding operation is completed, the information processing terminal 18 immediately instructs the telescopic motors (16A, 16B) of the weeding shovel to rotate and retract the weeding shovel.

[0039] The two weeding shovels (14A, 14B) are independently controlled by two weeding shovel extension motors (16A, 16B). The weeding operations on the left and right sides of the ridges of the robot do not affect each other. Moreover, the extension and retraction control of the weeding shovels is automatically controlled by the information processing terminal, which can ensure that the extension and retraction of the weeding shovels are timely and accurate, thus ensuring the efficiency and quality of the robot's weeding operations.

[0040] An information processing terminal (not shown in the figure) is integrated inside the mobile vehicle body. It is configured to receive recognition information from the visual perception component during the movement of the mobile vehicle body, and: drive the mobile vehicle body to the center of the adjacent field ridge, so that the blade 8 is in contact with the field ridge; and drive the weeding mechanism so that the two weeding shovels (14A, 14B) selectively avoid the crops and move only to the weed area to clear them.

[0041] like Figure 8 As shown, this invention also discloses an automatic inspection and navigation method for inter-row cultivation and weeding of corn. This method utilizes the aforementioned automatic inspection and navigation corn inter-row cultivation and weeding robot and includes the following steps: Step 1: Real-time acquisition of depth images and infrared intensity images of the field scene using a ToF depth camera mounted on the front of the mobile vehicle; The information processing terminal filters the depth image to generate a 3D point cloud, and uses a random sampling consistency algorithm to extract vegetation point cloud, which includes points of crops and weeds. The vegetation point cloud is horizontally sliced ​​along the vertical direction, and the points in each slice are clustered using the Euclidean clustering algorithm to obtain the feature point cloud. Step 2: Transform the detected crop row plane equations from the camera coordinate system to the moving vehicle coordinate system, and output the inter-row pose estimation by combining the motion model of the moving vehicle. Step 3: Centered on the moving vehicle, generate a local occupancy grid map of a fan-shaped area based on the field of view of the ToF depth camera. Then, based on the real-time global pose of the moving vehicle inspection, convert the local map to the global coordinate system. Use Bayesian update rules to integrate the occupancy probability information of the local map into the global occupancy grid map, continuously updating and refining the understanding of the layout of crops and weeds in the field. Step 4: Based on the generated global grid map and the coordinates of the moving vehicle, weeds, and crops, the information processing terminal determines: if the detectable area in front of the current weeding shovel is identified as a weed spot or a crop-free area, then the weeding shovel is extended to perform weeding and tillage; if the detectable area in front of the current weeding shovel is identified as a crop point cloud, then the weeding shovel is retracted to avoid the crops and prevent damage to the corn plants.

[0042] The parts not described in detail in the above embodiments are existing technologies.

[0043] It should be noted that although the present invention has been described through the above embodiments, the present invention may have many other embodiments. Without departing from the spirit and scope of the present invention, those skilled in the art can obviously make various corresponding changes and modifications to the present invention, but all such changes and modifications should fall within the scope of protection of the appended claims and their equivalents.

Claims

1. An automatic inspection and navigation corn tillage and weeding robot, characterized in that, include: The mobile vehicle is equipped with visual perception components for identifying crops and weeds, as well as navigation components for planning field paths. The rotary tillage mechanism includes a rotary tillage shaft (7) located at the rear of the mobile vehicle body and components capable of driving the rotary tillage shaft (7) to lift and rotate itself. The axial direction of the rotary tiller (7) is along the width direction of the moving vehicle body. Multiple blades (8) that can rotate are mounted on the rotary tiller (7) via the blade mounting plate (9). The blade tip trajectory of the blades (8) is on the same spindle-shaped envelope surface, which is compatible with the cross section of the cornfield ridge that is low in the middle and high on both sides. The weeding mechanism includes two weeding shovels disposed on both sides of the mobile vehicle body and a component capable of driving the two weeding shovels to move towards or away from each other in the width direction of the mobile vehicle body. An information processing terminal is configured to receive the recognition information of the visual perception component during the movement of the mobile vehicle, and: drive the mobile vehicle to travel at the center of the adjacent ridge, so that the blade (8) fits into the ridge; as well as The weeding mechanism is driven so that the two weeding shovels selectively avoid crops and move only to the weed area to clear it.

2. The automatic inspection and navigation corn tillage and weeding robot according to claim 1, characterized in that: The vehicle body (1) of the mobile vehicle is streamlined and is used to flexibly separate the corn plants on both sides during the inspection process.

3. The automatic inspection and navigation corn weeding robot according to claim 1, characterized in that: The rear of the mobile vehicle is provided with a rotary tillage baffle (3), which at least partially surrounds the outside of the rotary tillage mechanism.

4. The automatic inspection and navigation corn tillage and weeding robot according to claim 1, characterized in that: The visual perception component includes ToF depth cameras located at the head of the moving vehicle and on both sides of the moving vehicle.

5. The automatic inspection and navigation corn tillage and weeding robot according to claim 1, characterized in that: The navigation components include a remote communication device (4) and a satellite positioning system.

6. The automatic inspection and navigation corn tillage and weeding robot according to claim 1, characterized in that: The components that can drive the rotary tiller shaft (7) to rotate include: a cutter shaft lifting and power transmission arm support shaft (22) rotatably mounted on the mobile vehicle body, cutter shaft lifting and power transmission arms fixed at both ends of the cutter shaft lifting and power transmission arm support shaft (22), a cutter shaft fixing plate (10) fixed at the tail end of the cutter shaft lifting and power transmission arm, a cutter shaft connecting shaft (11) rotatably mounted inside the cutter shaft fixing plate (10), and two cutter shaft connecting shafts (11) fixedly connected to both ends of the rotary tiller shaft (7) respectively; The mobile vehicle body is also equipped with a rotary tillage shaft drive seat (23) that drives the blade shaft connecting shaft (11) to rotate.

7. The automatic inspection and navigation corn tillage and weeding robot according to claim 6, characterized in that: The components that can drive the rotary tiller shaft (7) to lift itself include the hydraulic lifting rod and the hydraulic lifting control system (20). The two ends of the hydraulic lifting rod are respectively hinged to the moving vehicle body and the support shaft (22) of the lifting and power transmission arm of the cutter shaft. The input end of the hydraulic lifting control system (20) is electrically connected to the information processing terminal, and the output end of the hydraulic lifting control system (20) is electrically connected to the hydraulic lifting rod.

8. The automatic inspection and navigation corn tillage and weeding robot according to claim 1, characterized in that: The components capable of driving the two weeding shovels to move towards or away from each other in the width direction of the mobile vehicle body include: Two slide rails are disposed inside the moving vehicle body along its width direction; Two telescopic arms for weeding are slidably mounted in two slide rails, and the two weeding shovels are fixed to the outer ends of the corresponding telescopic arms. Two telescopic motors for weeding shovels are mounted on the mobile vehicle body. The output shafts of the telescopic motors for weeding shovels are equipped with gears that mesh with the toothed grooves on the corresponding telescopic arms of the weeding shovels.

9. The automatic inspection and navigation corn tillage and weeding robot according to claim 1, characterized in that: The two weeding shovels are arranged in an inverted "V" shape parallel to the ridge slope.

10. An automatic inspection and navigation method for corn inter-row cultivation and weeding, characterized in that, The method utilizes the automatic inspection and navigation corn weeding robot according to any one of claims 1 to 9, and includes the following steps: Step 1: Real-time acquisition of depth images and infrared intensity images of the field scene using a ToF depth camera mounted on a mobile vehicle; The depth image is filtered by the information processing terminal to generate a three-dimensional point cloud, and a random sampling consistency algorithm is used to extract the vegetation point cloud, which includes points of crops and weeds. The vegetation point cloud is horizontally sliced ​​along the vertical direction, and the points in each slice are clustered using the Euclidean clustering algorithm to obtain the feature point cloud. Step 2: Transform the detected crop row plane equations from the camera coordinate system to the moving vehicle coordinate system, and output the inter-row pose estimation by combining the motion model of the moving vehicle. Step 3: Centered on the mobile vehicle, generate a local occupancy grid map of a fan-shaped area based on the field of view of the ToF depth camera, and convert the local map to the global coordinate system based on the real-time global pose of the mobile vehicle inspection. Then, use Bayesian update rules to integrate the occupancy probability information of the local map into the global occupancy grid map to update the understanding of the layout of crops and weeds in the field. Step 4: Based on the generated global occupancy grid map and the coordinates of the moving vehicle, weeds, and crops, the information processing terminal determines: if the detectable area in front of the current weeding shovel is identified as a weed spot or a crop-free area, then the weeding shovel is extended to perform weeding and cultivation; if the detectable area in front of the current weeding shovel is identified as a crop point cloud, then the weeding shovel is retracted to avoid the crops.

Images