Intelligent ridge mowing robot structure and autonomous navigation and mowing operation control method thereof

Abstract

The application discloses a kind of intelligent ridge mowing robot structure and its autonomous navigation, mowing operation control method, including robot mobile platform body, robot mobile platform body front end grass cutting device, robot mobile platform body both sides grass cutting device and perception decision and planning control device.The ridge mowing robot of the present application can be electrically adjusted by design grass cutting device and grass cutting device position, using autonomous navigation control algorithm, using lithium battery drive, realize narrow ridge and field wide width high-precision unmanned weeding operation, improve crop yield and quality, widen the same equipment operating environment, reduce the weeding equipment investment cost of planting personnel.Compared with traditional weeding equipment, structure is small and exquisite, operation is flexible, degree of intelligence is higher, weeding efficiency is greatly improved, using ridge mowing robot can greatly meet the weeding needs of planting personnel, completely replace time-consuming and labor-consuming manual ridge weeding operation.

Description

Technical Field

[0001] This invention belongs to the field of agricultural machinery technology, specifically relating to the structure of an intelligent field ridge mowing robot and its autonomous navigation and mowing operation control method. Background Technology

[0002] As a major traditional agricultural country, China attaches great importance to agricultural development. Field ridges are an important component of farmland, facilitating movement for farmers and serving to delineate farmland areas and retain water and fertilizer. Because most field ridges are constructed of fertile, sun-drenched soil, they are highly susceptible to weed growth. Unchecked weed growth on these ridges hinders movement, encroaches on farmland space, competes with crops for sunlight and nutrients, and its seeds fall into the field, leading to widespread weed infestation. This directly impacts crop growth, indirectly promotes the growth and reproduction of pests and diseases, disrupts the ecological balance, and ultimately results in reduced grain yields and lower quality.

[0003] Currently, weeding methods both domestically and internationally are mainly divided into chemical and non-chemical weeding. While chemical weeding is simple and efficient, it leaves large amounts of chemical residues in the soil, polluting the natural environment. Through the food chain, these residues eventually reach the human body, harming health. Furthermore, long-term use of herbicides can lead to weed resistance, reducing weeding efficiency. Non-chemical weeding primarily includes manual pulling and manually operated weeding machinery, which are labor-intensive and inefficient. This is particularly limiting in narrow field ridge environments, where weeding machinery is scarce and lacks automation; manual pulling remains the primary method. Therefore, given the current labor shortage and rising planting costs, the development of an intelligent field ridge weeding robot that can improve efficiency, reduce labor costs, and expand operating environments is urgently needed.

[0004] Although the wolfberry plant-to-plant and field-ridge mower designed in patent CN207721045 can change the angle between the rear arm and the base through the angle adjustment unit to achieve weeding operations on non-horizontal grounds such as wolfberry plantations and field ridges, it needs to be driven by a rotary tiller or tractor, has a low level of intelligence, and cannot be used in narrow field ridge mowing environments.

[0005] Although the mowing device designed in patent CN215379938 can solve the problem of mowing width and height through a mowing mechanism that can be adjusted horizontally and vertically to adapt to different mowing scenarios, it still cannot meet the needs of mowing operations on slopes in field ridge environments.

[0006] Currently, there are no intelligent lawnmower robots in China that integrate weeding operations on field ridges and in the field. Compared to traditional lawnmower robots, this invention has a wider operating range and a higher level of intelligence. This invention can remotely adjust the height of the mowing blades based on the mowing effect without stopping the mowing operation to adapt to complex field ridge environments; it can also achieve high-precision autonomous navigation for unmanned mowing operations, avoiding double-cutting and missed cutting, improving mowing efficiency, and reducing labor costs; it can also adapt to both field ridge and in-field mowing environments by remotely setting the positions of the mowing devices on both sides of the robot to adapt to mowing operations on field ridges with different slopes. When the mowing devices on both sides of the robot are horizontal, it can achieve wide-area mowing in the field, significantly improving mowing efficiency, expanding the operating environment of the same equipment, and reducing the investment costs of weeding equipment for growers. Summary of the Invention

[0007] The purpose of this invention is to design an intelligent field mowing robot with higher intelligence, wider environmental applicability, and better controllability. It is suitable for high-precision autonomous navigation while performing weeding operations, achieving human-machine separation, greatly reducing labor intensity and improving work efficiency. This invention uses lithium batteries for power, employing clean and renewable energy compared to traditional fuel-powered weeding devices, producing no exhaust emissions and helping to reduce environmental pollution. The weeding blades of this invention are equipped with electric adjustment devices, allowing the height of the blades to be remotely adjusted according to the weeding effect without stopping the mowing operation. This adapts to complex field ridges and field environments, avoiding the problem of current lawnmowers requiring stopping the mowing operation and manually rotating screws or clips to adjust the blade height, which reduces work efficiency. The structural features of this invention include a lightweight design, compact structure, and flexible operation. The positions of the weeding devices on both sides of the robot are electrically adjustable to suit narrow field ridges and wide field weeding operations. Its high controllability enables multi-scenario weeding operations, reducing the investment cost of weeding equipment for growers and increasing income.

[0008] To achieve the above objectives, the technical solution of the present invention is as follows:

[0009] A smart field mowing robot structure, characterized by comprising a robot mobile platform body, a front-end mowing device of the robot mobile platform body, mowing devices on both sides of the robot mobile platform body, and a perception, decision-making, planning, and control device: the robot mobile platform body includes a rugby ball-shaped hub motor, a rugby ball-shaped hub motor arc-shaped fixing component, a robot mobile platform top plate, a hub motor, a hub motor fixing component, an articulated motor, and a robot support frame; the rugby ball-shaped hub motor serves as the front wheel of the robot mobile platform body; the left and right sides of the rugby ball-shaped hub motor arc-shaped fixing component are each cut with an inverted U-shaped groove, and the sides of the rugby ball-shaped hub motor... The rotating shafts are respectively placed into the inverted U-shaped slots of the curved fixing parts of the rugby ball-shaped hub motors and secured with nuts; the top of the curved fixing parts of the rugby ball-shaped hub motors is polished to a flat surface, and the rotor end of the joint motor is fixed to the top of the curved fixing parts of the rugby ball-shaped hub motors through screw holes; the stator end of the joint motor is fixed to the top plate of the robot mobile platform through screw holes; there are two hub motors, which serve as the rear wheels of the robot mobile platform body; there are two hub motor fixing parts, each with an inverted U-shaped slot cut on its left and right sides, and the rotating shafts on both sides of the hub motor are respectively placed into the inverted U-shaped slots of the hub motor fixing parts. The hub motor fixing component is fixed to the top plate of the robot mobile platform via screw holes. The top plate of the robot mobile platform is an aluminum plate, located above the arc-shaped fixing component of the rugby ball hub motor and the stator end of the joint motor, and is horizontally fixed. The robot bracket is an aluminum profile, fixed to the top plate of the robot mobile platform via screw holes and corner pieces. The mowing device includes a front electric push rod, an electric weed cutter, and an electric weed cutter fixing plate. There are three front electric push rods, both ends of which are flat bases. The bottom end of the front electric push rod is fixed to the top plate of the robot mobile platform via screw holes, and the top end of the front electric push rod is fixed to the top plate of the robot mobile platform. The electric weed cutters are fixed to one end of the electric weed cutter fixing plate via screw holes. There are four electric weed cutters, equidistantly distributed at one end of the electric weed cutter fixing plate, and connected to it via screw holes. The electric weed cutter fixing plate is made of aluminum and moves up and down via the extension and retraction of a front-mounted electric push rod, thereby driving the electric weed cutters to move synchronously. There are two mowing devices, one on each side of the robot's mobile platform, including a side electric push rod, a bottom fixing plate for the side electric push rod, a top fixing plate for the side electric push rod, a hinge, a mowing motor, and a flexible mowing head. The top and bottom ends of the side electric push rods are both round hole connectors.The bottom fixing plate of the side electric push rod is made of aluminum plate. It is vertically fixed to the top plate of the robot moving platform via screw holes and corner brackets. A small aluminum plate with a round hole is vertically welded to the middle of the outer side of the bottom fixing plate. The round hole connector of the side electric push rod is coaxially connected to the round hole of the small aluminum plate on the outer side of the bottom fixing plate. A screw passes through the round hole, with a washer at each end, and is fixed with a nut. The bottom end of the side electric push rod can rotate around the screw. The top fixing plate of the side electric push rod is also made of aluminum plate. Two small aluminum plates with round holes are vertically welded to the middle of the outer side of the top fixing plate. The round hole connector of the top fixing plate is coaxially connected to the outer side of the small aluminum plate on the outer side of the top fixing plate. Small aluminum sheets are coaxially connected through circular holes. A screw passes through the holes, with a washer at each end, and is secured with nuts. The top of the side electric push rod can rotate around the screw. The hinge is made of stainless steel. One hinge is connected to the bottom fixing plate of the side electric push rod through a screw hole, and the other hinge is connected to the top fixing plate of the side electric push rod through a screw hole. The top fixing plate of the side electric push rod can rotate around the hinge axis. The stator end of the mowing motor is fixed to the top fixing plate of the side electric push rod through a screw hole, and the rotor end of the mowing motor is fixed to the flexible trimming head through a screw hole, so that the flexible trimming head rotates synchronously with the mowing motor. The flexible trimming head includes a trimming head shell and a trimming rope, and is driven by the mowing motor to trim the grass. The robot head rotates at high speed. The perception, decision-making, and planning control device includes a 3D LiDAR, a BeiDou RTK-GPS system, a depth camera assembly, a battery, a central processing unit, an inertial measurement module, and a control box. The 3D LiDAR is located at the front end of the robot's mobile platform body and is fixed parallel to the robot support via screw holes. The BeiDou RTK-GPS system includes a GNSS high-precision positioning antenna and a combined navigation receiver. The GNSS high-precision positioning antenna is located on both sides of the robot's mobile platform body and is fixed parallel to the robot support via screw holes. The depth camera assembly includes an RGB-D depth camera and a camera bracket. There are two depth camera assemblies; one depth camera assembly is fixed to the robot support via screw holes and corner brackets. At the front end of the robot mobile platform body, another depth camera assembly is fixed to the rear end of the robot mobile platform body via screw holes and corner brackets; the battery is a lithium battery, placed above a battery mounting plate, which is fixed to the top plate of the robot mobile platform via screw holes; the central processing unit is a microcomputer, with mounting holes around it, and is fixed to the top plate of the robot mobile platform via screw holes; the inertial measurement module has mounting holes around it, and is fixed to the top plate of the robot mobile platform via screw holes; the integrated navigation receiver is electrically connected to the GNSS high-precision positioning antenna and is fixed to the top plate of the robot mobile platform via screw holes; the control box is made of stainless steel and is fixed to the top plate of the robot mobile platform via screw holes.

[0010] Furthermore, the rugby ball-shaped hub motor connects its two rotating shafts to the arc-shaped fixing component via nuts. As the front wheels of the robot's mobile platform, the rugby ball-shaped hub motor increases the contact area with the ground compared to the traditional two-wheel design, making the robot more stable when mowing on narrow, uneven surfaces and reducing the risk of tipping over. Under normal driving conditions, the rugby ball-shaped hub motor is not active, only providing support and steering. When the robot travels on complex terrain and the hub motor's power is insufficient, the rugby ball-shaped hub motor activates, providing auxiliary driving force for the robot's movement. By switching the working state of the rugby ball-shaped hub motor, the robot's road passability is improved, energy consumption is reduced, and the robot's endurance is increased. The hub motor, with two shafts connected to the fixing component via nuts, serves as the rear wheels of the robot's mobile platform, providing power for the robot's movement.

[0011] Furthermore, the stator end of the joint motor is fixed to the top plate of the robot's moving platform through screw holes, and the rotor end of the joint motor is fixed to the top of the arc-shaped fixing part of the rugby ball hub motor through screw holes. By controlling the rotation angle of the joint, the rugby ball hub motor is driven to rotate synchronously, thereby realizing the turning angle control of the lawnmower robot.

[0012] Furthermore, the rugby ball-shaped hub motor arc fixing component, the robot mobile platform top plate, the electric weed cutter fixing plate, the hub motor fixing component, the side electric push rod bottom fixing plate, and the side electric push rod top fixing plate are all made of aluminum plates. This provides support for the entire intelligent field mowing robot while reducing its own weight, lowering power consumption, and improving battery life, while also enhancing the robot's movement and control flexibility. The robot bracket is made of aluminum profile and is fixed to the robot mobile platform top plate via screw holes and corner pieces. It is used to house the perception, decision-making, and planning control device. Transparent acrylic plates are installed around the robot bracket via screw holes to waterproof the robot's perception, decision-making, and planning control device.

[0013] Furthermore, the electric weeding shear fixing plate is made of aluminum and is used to hold the electric weeding shears. Four electric weeding shears are equidistantly distributed on the fixing plate and fixed by screw holes. There are three front electric push rods. The bottom end of the front electric push rod is fixed to the top plate of the robot moving platform by screw holes, and the top end of the front electric push rod is fixed to one end of the electric weeding shear fixing plate by screw holes. By controlling the extension and retraction of the front electric push rods, the electric weeding shear fixing plate moves up and down synchronously, thereby adjusting the height of the electric weeding shears from the ground to achieve variable grass depth cutting. By adding an electric adjustment device, the height of the weeding blades can be adjusted according to the cutting effect without stopping the cutting operation, so as to adapt to complex field ridges and field environments. This avoids the problem that the current lawnmower weeding device requires stopping the cutting operation and then manually rotating the screw or buckle on the weeding device to adjust the weeding device to the appropriate height, which leads to a decrease in work efficiency.

[0014] Furthermore, the mowing device is located on both sides of the robot's mobile platform body. The bottom and top ends of the side electric push rods can rotate around the screws passing through the round hole connectors. The top fixing plate of the side electric push rods can rotate around the hinge axis. By controlling the extension and retraction of the side electric push rods, the top fixing plate of the side electric push rods rotates around the hinge axis. When the side electric push rods extend, the top fixing plate of the side electric push rods rotates counterclockwise around the hinge axis; when the side electric push rods retract, the top fixing plate of the side electric push rods rotates clockwise around the hinge axis. The position of the top fixing plate of the side electric push rods can be adjusted in real time according to the slope of both sides of the field ridge, thereby realizing the mowing of the grass with varying depth on both sides of the field ridge. At the same time, by adjusting the position of the top fixing plate of the side electric push rods and the length of the grass-cutting rope in the flexible grass-cutting head, the machine can also achieve the desired grass-cutting depth. The robot performs mowing operations on field ridges with varying slopes. When the top fixing plate of the side electric push rod is level with the ground, the front mowing device and the two side mowing devices of the robot are on the same plane. This allows the robot to perform wide-area mowing, significantly improving mowing efficiency. Compared to traditional weeding equipment, it expands the operating environment of the same equipment and reduces the investment cost of mowing equipment for growers. The flexible mowing head rotates at high speed driven by the mowing motor. Compared to traditional blade-driven mowing, using the flexible mowing head is safer. Moreover, in complex field ridge environments, the flexible mowing head will not be damaged even if it hits stones or other objects, extending the service life of the mowing equipment. The top fixing plate of the side electric push rod is made of aluminum plate, used to fix the top of the side electric push rod and the mowing motor, while also preventing grass clippings and stones from hitting the flexible mowing head from flying, improving the robot's safety and reliability.

[0015] Furthermore, the GNSS high-precision positioning antenna and the integrated navigation receiver are electrically connected, and the integrated navigation receiver outputs positioning information. The central processing unit is a microcomputer that processes data collected by the 3D lidar, Beidou positioning RTK-GPS system, and inertial measurement module to achieve autonomous navigation of the intelligent field mowing robot, enabling unmanned mowing operations, avoiding double mowing and missed mowing, improving mowing efficiency, and reducing labor costs. The depth camera component can collect image data in front of and behind the mowing robot. The central processing unit trains a weed recognition model based on a convolutional neural network. Based on the image data in front of the mowing robot, it can accurately identify whether there are weeds in front of the mowing robot, control the start and stop of the mowing device and the mowing depth. The central processing unit processes the image data behind the mowing robot to obtain the mowing effect and adjusts the mowing device in real time based on the mowing effect to achieve closed-loop control.

[0016] Furthermore, the battery is a lithium battery, used to provide power to the lawnmower robot. There are two lithium batteries, which are used to supply power separately according to the power consumption of each electrical device. The left lithium battery powers one rugby ball-shaped hub motor, two hub motors, one joint motor, and the perception, decision-making, and planning control device, while the right lithium battery powers the mowing device and the lawn-cutting device. Compared with traditional fuel-powered lawnmowers, using lithium battery power is a clean and renewable energy source, which helps to reduce environmental pollution.

[0017] Furthermore, the control box includes a rugby ball-shaped hub motor controller, a hub motor controller, a side electric actuator controller, a front electric actuator controller, a mowing motor controller, a distribution board, a step-down module, and a hardware circuit drive control board. There are two hardware circuit drive control boards, both STM32F4 control boards, fixed inside the control box by nylon pillars. The upper hardware circuit drive control board is electrically connected to the joint motor, the rugby ball-shaped hub motor controller, the hub motor controller, the lower hardware circuit drive control board, the central processing unit, and the step-down module. The lower hardware circuit drive control board is electrically connected to the side electric actuator controller, the front electric actuator controller, the mowing motor controller, and the step-down module, enabling autonomous navigation and mowing operation control of the intelligent field mowing robot.

[0018] The present invention provides an autonomous navigation method for an intelligent field mowing robot, comprising the following steps:

[0019] Step 1: Use multi-sensor fusion SLAM technology for localization and mapping of the intelligent field ridge mowing robot. The 3D LiDAR, inertial measurement module, GNSS high-precision positioning antenna and integrated navigation receiver are used as information sources for fusion processing to improve the accuracy and stability of point cloud registration and obtain centimeter-level positioning accuracy in real time. This is used to solve the problem that the mowing robot cannot perform autonomous navigation mowing operations due to GNSS signal loss or LiDAR sensor positioning failure in field ridge and grassland environments.

[0020] Step 2: The path planning of the intelligent field mowing robot adopts an improved bidirectional RRT* algorithm based on dynamic end node guidance and potential field guidance sampling. During the path backtracking process after the path finding is completed, the generated path points are deredundant and the corner constraints of adjacent polyline segments are processed. At the same time, the algorithm combines the idea of ​​artificial potential field method. When sampling, the potential field guides the sampling points that fall in the obstacle area to be rejected to the obstacle-free area along the downward direction of the potential field of that point, so as to realize the reuse of sampling points. The path points are dynamically modified according to the threshold given by the corner constraint.

[0021] Step 3: The improved bidirectional RRT* algorithm used by the intelligent field mowing robot generates the mowing robot's working path after dynamic tip point guidance sampling, potential field guidance sampling, and backtracking path point processing. However, considering the path smoothness and the robot's maximum curvature constraint, a third-order quasi-uniform B-spline curve is used to optimize the trajectory after the path is generated, thus obtaining the final working trajectory of the mowing robot.

[0022] Step 4: Under normal driving conditions, the intelligent field mowing robot is driven by hub motors. The rugby ball-shaped hub motors do not work, only providing support and steering. When the robot travels on complex terrain and the hub motors lack sufficient power, the rugby ball-shaped hub motors work to provide auxiliary driving force for the robot's movement. The rugby ball-shaped hub motors perform independent steering control under the drive of the joint motors. To simplify the motion model of the mowing robot while reflecting the vehicle characteristics, assuming the mowing robot travels in a low-speed and slip-free scenario, the center point of the rear axle of the mowing robot's mobile platform is used as the rotation reference point to obtain the simplified motion model of the mowing robot. The rotation angle of its rugby ball-shaped hub motors is:

[0023] ;

[0024] In the formula, For the rotating corner of the rugby ball-shaped hub motor, The wheelbase is for the hub motor and the rugby ball-shaped hub motor. The wheelbase of the hub motor. The turning radius;

[0025] Step 5: Based on the pure tracking algorithm, the lawnmower robot starts from its current position. Reaching the path tracking point The trajectory of the movement can be regarded as a segment of... The trajectory of a circular arc with radius can be obtained using the law of sine: ;

[0026] After conversion and simplification, we get: ;

[0027] In the formula, Forward sight distance, This refers to the heading deviation, specifically the current orientation of the lawnmower robot relative to the target point. The angle between them;

[0028] Step 6: According to the curvature formula, the curvature of the lawnmower robot's working path can be expressed as: ;

[0029] In the formula, The curvature of the lawnmower's working path;

[0030] Step 7: The rotation angle of the rugby ball-shaped hub motor of the lawnmower robot can be represented as: ;

[0031] Combining the above formula, the control quantity of the pure tracking algorithm can be obtained. The final expression is: ;

[0032] The time domain is introduced in the formula. for Lawn-mowing robot and target point The included angle;

[0033] definition Current posture and target point of the lawnmower robot The lateral error can be obtained as follows: ;

[0034] Meanwhile, the curvature of the lawnmower robot's working path can be expressed as: ;

[0035] Step 8: Adjust the forward viewing distance Expressed as a linear function of the lawnmower robot's linear velocity, we can obtain: ;

[0036] Combining the above formula, we can obtain: ;

[0037] In the formula, This is the proportionality coefficient. The linear velocity of the lawnmower robot;

[0038] Step 9: According to the angular velocity calculation method, we can obtain: ;

[0039] In the formula, The angular velocity of the lawnmower robot;

[0040] Step 10: Adjust the scaling factor Then, the pure tracking algorithm was adjusted to improve the angular velocity of the lawnmower robot. Forward viewing distance They are inversely proportional; that is, the larger the forward sight distance, the smaller the angular velocity, and the smoother the tracked trajectory; the smaller the forward sight distance, the larger the angular velocity, and the more oscillating the tracked trajectory.

[0041] Step 11: To better determine the forward sight distance, the PSO (Particle Swarm Optimization) algorithm is introduced, and a heading error variable is introduced here. , representing the angle between the robot's body pose and the optimized mowing path, using lateral error. With heading error The sum of the root mean squares is used as the main reference for the particle swarm optimization algorithm to determine the lateral error of the lawnmower robot. The evaluation parameters are:

[0042] ;

[0043] lawnmower heading error The evaluation parameters are:

[0044] ;

[0045] Based on the above two equations, the fitness function is designed as follows: ;

[0046] In the formula, Sampling time, For working hours, The fitness function designed for weights. Able to determine the lateral error of the lawnmower robot Adjustment and The decision weights have good adaptability;

[0047] Specifically, in a pure tracking model based on the PSO particle swarm optimization algorithm, when the lateral error... When it is large, the forward sight distance Mainly through lateral error Make decisions; when lateral errors When it is small, forward sight distance Mainly through heading error Make decisions;

[0048] Therefore, design weights: , ;

[0049] In the formula, The base of the weighting function is . To adjust the parameters, This represents the maximum lateral error of the lawnmower robot.

[0050] Step 12: Using a 3D LiDAR, inertial measurement module, and BeiDou positioning RTK-GPS system, the position and attitude information of the lawnmower robot are acquired in real time to confirm the distance and approximate volume of obstacles, and to determine whether the lawnmower robot can cross the obstacles. If the obstacle is determined to be insurmountable, obstacle avoidance actions are performed to maintain a suitable safe distance from the obstacle. If the obstacle is determined to be traversable, the height of the mowing device is adjusted to cross the obstacle, realizing obstacle avoidance function in unmanned lawnmower operation, reducing equipment wear and increasing service life. Based on the optimized operation path, a pure tracking algorithm is used to control the joint motors and hub motors in coordination through the central processing unit and upper-level hardware circuit drive control board, thereby adjusting the angular velocity and speed of the lawnmower robot in real time.

[0051] The present invention provides a method for controlling the mowing operation of an intelligent field ridge mowing robot, comprising the following steps:

[0052] Step 1: Use the depth camera component at the front end of the robot's mobile platform to collect image data, and train a weed recognition model based on a convolutional neural network through the central processing unit to accurately identify whether there are weeds in front of the lawnmower robot.

[0053] Step 2: Design a remote control terminal for the lawnmower robot. The human-machine interface of the remote control terminal is located on a mobile application and communicates with the central processor of the lawnmower robot through a local area network. The depth camera components installed at the front and rear of the lawnmower robot can capture images of the area around the lawnmower robot in real time, and display the captured images in real time on the human-machine interface via high-definition image transmission.

[0054] Step 3: Based on the specific mowing scenario, select the corresponding mowing mode via the remote control terminal. If it is a field ridge mowing operation, set the initial positions of the mowing and cutting devices in the field ridge mowing operation mode. After reaching the starting position of the field ridge mowing operation, the mowing devices on both sides of the robot will adjust to the set positions under the action of the side electric push rods. When the robot reaches the end position of the field ridge mowing operation and needs to change the ridge, the mowing devices on both sides will rise under the action of the side electric push rods and adjust to a horizontal position with the ground. Then, the robot will move to the next ridge to carry out the mowing operation, and so on. If it is a field mowing operation, set the initial positions of the mowing and cutting devices in the field mowing operation mode. The front mowing device and the side mowing devices of the robot are on the same plane, which allows the robot to achieve wide-area mowing in the field, greatly improving the mowing efficiency.

[0055] Step 4: The central processing unit receives control commands from the remote control terminal via local area network communication, processes them, and sends them to the hardware circuit driver control board to control the lawn mowing robot to perform lawn mowing operations. When weeds are detected in front of the lawn mowing robot, the mowing and trimming devices are automatically activated to perform lawn mowing operations. The lawn mowing operation is analyzed through the image of the lawn mowing robot's surroundings displayed on the human-machine interface of the remote control terminal. The position of the mowing and trimming devices is adjusted remotely to achieve the expected lawn mowing effect.

[0056] Step 5: When the lawn mowing robot achieves the expected lawn mowing effect, the current positions of the mowing and cutting devices can be saved through the remote control terminal. In subsequent operations in the same scene, the robot can automatically adjust the mowing and cutting devices to the corresponding positions saved in advance.

[0057] Step 6: Set the optimal operating current of the mowing motor as the threshold current. The hardware circuit drives the control board to detect the operating current of the mowing motor in real time. If the operating current of the mowing motor is less than the threshold current, the speed of the mowing robot is increased to improve mowing efficiency. If the operating current of the mowing motor is greater than the threshold current, the speed of the mowing robot is reduced to reduce the load on the mowing motor and extend its service life. If the operating current of the mowing motor is still greater than the threshold current after the mowing robot adjusts its speed ten times, the mowing operation is automatically stopped and an alarm message is sent to the remote control terminal to prevent the mowing motor from burning out due to excessive load caused by the mowing device being blocked by grass or debris being caught in it, thus causing economic losses.

[0058] Step 7: The remote control terminal of the lawnmower robot has a higher priority than the autonomous lawnmower operation control priority. It can display the operation trajectory and robot status parameter information in real time, so as to take over the control of the lawnmower robot at any time. It can realize emergency braking and one-click return of the lawnmower robot, making the lawnmower robot more controllable and improving its safety and reliability.

[0059] The beneficial effects of this invention are:

[0060] 1) This invention relates to an intelligent field ridge weeding robot, which can cut weeds on the field ridge and its two sides in narrow field ridge environments, avoid the weeds on the field ridge from affecting crop growth, improve grain yield and quality, and increase the income of planting personnel.

[0061] 2) This invention enables high-precision autonomous navigation lawn mowing. It utilizes multi-sensor fusion SLAM technology for lawn mowing robot localization and mapping, employs an improved bidirectional RRT* algorithm based on dynamic tip node guidance and potential field guidance sampling for path planning, optimizes the path using a third-order quasi-uniform B-spline curve, and finally uses a pure tracking algorithm to avoid double-cutting and missed cutting, reducing human intervention and replacing time-consuming and labor-intensive manual weeding operations, thus lowering labor costs.

[0062] 3) The positions of the mowing blades in this invention are all electrically adjustable. The height of the mowing blades can be remotely adjusted based on the mowing effect via camera image transmission without stopping the mowing operation. This adapts to complex field ridges and environments, avoiding the current problem where adjusting the blade height of a mower requires stopping the mowing operation and manually rotating a screw or latch on the device, leading to reduced work efficiency.

[0063] 4) This invention is compatible with both field ridge and field mowing operations, expanding the operational range. The robot has a compact structure and flexible operation. By adjusting the position of the weeding devices on both sides of the mowing robot, it can achieve unmanned weeding operations on narrow field ridges and wide fields on different slopes. It has strong environmental adaptability and can realize weeding operations in multiple scenarios, reducing the investment cost of weeding equipment for growers and increasing income.

[0064] 5) This invention designs a remote control terminal. The human-machine interface of the remote control terminal is located on a mobile application and communicates with the central processor of the lawnmower robot via a local area network. The remote control terminal has a higher priority than the autonomous lawnmower operation control priority. It can display the operation path and robot status parameter information, so as to take over the control of the lawnmower robot at any time. It can realize emergency braking and one-click return of the lawnmower robot. Compared with traditional driving or remote control weeding equipment, both operability and safety reliability are greatly improved.

[0065] 6) This invention is powered by a lithium battery and features a lightweight design. Compared to traditional fuel-powered weeding devices, it uses clean and renewable energy, produces no exhaust emissions, and helps reduce environmental pollution. When weeds are detected in front of the weeding robot, the mowing and trimming devices are automatically activated to perform weeding operations, significantly improving the robot's range.

[0066] Due to the narrowness of field ridges and the complexity of the terrain, large weeding equipment is difficult to access. Therefore, traditional weeding on field ridges generally involves manual pulling or spraying chemical herbicides. While chemical herbicides are effective, they leave significant residues in the soil, polluting the environment. Through the food chain, these residues can eventually reach humans, harming their health. Furthermore, long-term use of chemical herbicides can lead to weed resistance, reducing weeding efficiency. Manual weeding is labor-intensive and inefficient, resulting in high labor costs for growers. Because there are few weeding devices suitable for narrow field ridge environments, most are powered by gasoline engines, causing exhaust emissions and severe environmental pollution. They also tend to result in overlapping or missed cuts, and the depth of the weeds left after weeding is uneven, leading to poor results. Therefore, manual pulling remains the primary method. In addition, weeding on field ridges requires cutting weeds on the ridges and their slopes, demanding greater flexibility in the structural design of weeding equipment compared to field weeding, presenting a greater challenge.

[0067] To address the aforementioned issues, this invention presents a more intelligent, environmentally adaptable, and controllable field mowing robot. It is suitable for high-precision autonomous navigation while automatically mowing, achieving human-machine separation, significantly reducing labor intensity and improving work efficiency. Employing a lightweight design, its compact structure and flexible operation allow for the removal of weeds from field ridges and their slopes. Furthermore, by using multi-sensor fusion SLAM technology for robot localization and mapping, and employing path planning and tracking algorithms, it can achieve high-precision mowing without human intervention, avoiding double-cutting and missed cuts, replacing time-consuming and labor-intensive manual weeding, and reducing labor costs. Simultaneously, the mowing blade positions are electrically adjustable. Even in complex field ridge environments, the blade position can be adjusted in real-time without stopping mowing, preventing uneven weed depths after removal. A remote control terminal via mobile phone is also included, displaying the work path and robot status parameters, allowing for easy takeover of control of the robot, and enabling emergency braking and one-click return. Compared to traditional driver- or remote-controlled weeding equipment, this invention offers significantly improved operability and safety. Furthermore, to broaden the mowing range, reduce the investment costs of weeding equipment for growers, and increase income, the position of the weeding devices on both sides of the robot can be adjusted to achieve high-precision, unmanned weeding operations across narrow field ridges and wide areas. Simultaneously, powered by lithium batteries, it utilizes clean and renewable energy compared to traditional fuel-powered weeding devices, producing no exhaust emissions and thus reducing environmental pollution. This invention has broad application prospects in weeding, especially in narrow field ridges and their slopes. Attached Figure Description

[0068] Figure 1 This is the left front view of the present invention;

[0069] Figure 2 This is the left rear view of the present invention;

[0070] Figure 3 This is the front view of the present invention;

[0071] Figure 4 This is a rear view of the present invention;

[0072] Figure 5 This is a top view of the present invention;

[0073] Figure 6 This is a schematic diagram of the grass-cutting device of the present invention;

[0074] Figure 7 This is a schematic diagram of grass cutting operations on narrow field ridges according to the present invention;

[0075] Figure 8 This is a schematic diagram of the wide-span mowing operation in the field according to the present invention;

[0076] Figure 9 This is a flowchart of the path planning process of the present invention;

[0077] Figure 10 This is a simplified schematic diagram of the kinematic model of the present invention;

[0078] Figure 11 This is a schematic diagram of the autonomous navigation system of the present invention;

[0079] The components are as follows: 1- Flexible mowing head; 2- Mowing motor; 3- Electric weed cutter fixing plate; 4- Electric weed cutter; 5- Rugby-shaped hub motor; 6- Front electric push rod; 7- Rugby-shaped hub motor arc fixing piece; 8- Joint motor; 9- Depth camera assembly; 10- Combined navigation receiver; 11- Control box; 12- Robot bracket; 13- 3D LiDAR; 14- GNSS high-precision positioning antenna; 15- Battery; 16- Inertial measurement module; 17- Central processing unit; 18- Robot mobile platform top plate; 19- Side electric push rod bottom fixing plate; 20- Hinge; 21- Hub motor; 22- Side electric push rod; 23- Side electric push rod top fixing plate; 24- Hub motor fixing piece; 25- Field ridge; 26- Flexible mowing head shell; 27- Mowing rope. Detailed Implementation

[0080] The following section, with reference to the accompanying drawings in the invention examples, provides a detailed description of the structure of the intelligent field ridge mowing robot and its autonomous navigation and mowing operation control methods.

[0081] like Figure 1 , Figure 2 , Figure 3 , Figure 4 and Figure 5The images shown are the left front view, left rear view, front view, rear view, and top view of an intelligent field mowing robot structure. It includes four parts: the robot's mobile platform body, the mowing device at the front end of the robot's mobile platform body, the mowing devices on both sides of the robot's mobile platform body, and the perception, decision-making, planning, and control device. The main components are: a flexible mowing head 1, a mowing motor 2, an electric mowing shear fixing plate 3, electric mowing shears 4, a rugby ball-shaped hub motor 5, a front-mounted electric push rod 6, an arc-shaped fixing piece for the rugby ball-shaped hub motor 7, an articulated motor 8, a depth camera assembly 9, a combined navigation receiver 10, a control box 11, a robot support 12, and a 3D LiDAR 13. The robot is composed of 24 components, including a GNSS high-precision positioning antenna (14), a battery (15), an inertial measurement module (16), a central processing unit (17), a top plate of the robot mobile platform (18), a bottom fixing plate of the side electric push rod (19), a hinge (20), a hub motor (21), a side electric push rod (22), a top fixing plate of the side electric push rod (23), and a hub motor fixing component (24). The rugby ball-shaped hub motor (5) has a rated voltage of 48V and a maximum power of 300W. As the front wheel of the robot mobile platform, it increases the contact area with the ground compared to the traditional two-wheel design, making the robot move more smoothly on narrow, uneven surfaces and reducing the risk of tipping over. Under normal driving conditions, the rugby ball-shaped hub motor 5 is not operational, only providing support and steering. When the robot travels on complex terrain and the hub motor 21 lacks sufficient power, the rugby ball-shaped hub motor 5 activates, providing auxiliary driving force for the robot's movement. By switching the operating state of the rugby ball-shaped hub motor 5, the robot's road clearance is improved, power consumption is reduced, and the robot's endurance is increased. The curved fixing part 7 of the rugby ball-shaped hub motor is 5mm thick, with a polished flat top. Each of the left and right sides has an inverted U-shaped slot, 20mm long and 11mm wide. The diameter of the rotating shafts on both sides of the rugby ball-shaped hub motor 5 is 10mm. Each component is placed into the inverted U-shaped slot of the rugby ball-shaped hub motor arc fixing part 7, with a 10mm inner diameter, 20mm outer diameter, and 1mm thick shim placed in the middle, and then secured with an M10 nut. The rugby ball-shaped hub motor arc fixing part 7, hub motor fixing part 24, electric weed cutter fixing plate 3, side electric push rod bottom fixing plate 19, side electric push rod top fixing plate 23, and robot moving platform top plate 18 are all made of aluminum plates. While providing support for the entire intelligent field mowing robot, they also reduce their own weight, reduce power consumption, improve battery life, and enhance the flexibility of robot movement control. Holes have been drilled at the corresponding positions during processing.The joint motor 8 has a rated voltage of 24V and a maximum power of 200W. Both its rotor and stator ends have eight M4 threaded holes evenly distributed circumferentially. The rotor end is fixed to the top of the rugby ball-shaped hub motor arc-shaped fixing part 7 using eight 12mm long M4 screws, and the stator end is fixed to the top plate 18 of the robot's mobile platform using eight 16mm long M4 screws. By controlling the rotation angle of the joint 8, the rugby ball-shaped hub motor 5 rotates synchronously, achieving robot turning angle control. There are two hub motors 21, with a rated voltage of 48V and a maximum power of 400W. These serve as the rear wheels of the robot's mobile platform, providing power for the robot's movement. The hub motor fixing part 24 is 5mm thick, with the left... On both sides, there are inverted U-shaped slots, each 20mm long and 13mm wide. The rotating shafts of the hub motor 21, each with a diameter of 12mm, are placed into the inverted U-shaped slots of the hub motor mounting bracket 24. A washer with an inner diameter of 12mm, an outer diameter of 20mm, and a thickness of 2mm is placed in the middle and secured with an M12 nut. The top of the hub motor mounting bracket 24 is secured to the top plate 18 of the robot mobile platform with four 20mm long M6 screws. The robot bracket 12 is made of European standard 3030L-2.2 aluminum profile and is secured to the top plate 18 of the robot mobile platform with M6 screw holes and corner brackets. A transparent acrylic plate is installed around it with screw holes to waterproof the robot's perception, decision-making, and planning control device. A front-mounted electric push rod 6... There are three electric weeding shears, each with a rated voltage of 24V and a maximum power of 50W. Both ends are flat bases with four M5 screw holes. The bottom of the front electric push rod 6 is fixed to the top plate 18 of the robot moving platform by four 15mm long M5 screws, and the top is fixed to one end of the electric weeding shear fixing plate 3 by four 15mm long M5 screws. There are four electric weeding shears 4, each with four M6 screw holes. The electric weeding shears 4 are evenly distributed at one end of the electric weeding shear fixing plate 3 and connected to it by 16 15mm long M6 screws. The electric weeding shear fixing plate 3 moves up and down through the extension and retraction of the front electric push rod 6, thereby driving the electric weeding shears 4 to move synchronously, achieving variable weed retention. Deep mowing, through the addition of an electric adjustment device, allows for adjustment of the mowing blade height based on the mowing effect without stopping the mowing operation. This adapts to complex field ridges and field environments, avoiding the current problem where adjusting the blade height of the mowing device requires stopping the mowing operation and then manually rotating the screw or buckle on the mowing device, which reduces work efficiency. The three-dimensional LiDAR 13 is located at the front of the robot's mobile platform body, with a rated voltage of 12V, and is fixed to the LiDAR metal base with 3M adhesive. The LiDAR metal base has four M6 threaded holes evenly distributed around the circumference, which are fixed to the robot bracket 12 in parallel with four 15mm long M6 screws.The GNSS high-precision positioning antenna 14 is located at the top of the robot's mobile platform body, with a rated voltage of 12V. It is electrically connected to the integrated navigation receiver 10 and fixed to the metal base via a strong magnetic chuck. The metal base has four M6 threaded holes evenly distributed around its circumference, which are fixed to the robot bracket 12 by four 15mm long M6 screws. The lawnmower robot is powered by a lithium battery, which is placed above a battery mounting plate. The battery mounting plate is fixed to the top plate 18 of the robot's mobile platform via screw holes. Compared to traditional fuel-powered lawnmowers, the lithium battery uses clean and renewable energy, produces no exhaust emissions, and helps reduce environmental pollution. The central processing unit 17 and the inertial measurement unit 18 are also included. The measurement module 16, the integrated navigation receiver 10, and the control box 11 are all fixed to the top plate 18 of the robot mobile platform with M6 screws of 20mm in length. The central processing unit 17 is a microcomputer that processes the data collected by the 3D LiDAR 13, the Beidou positioning RTK-GPS system, and the inertial measurement module 16 to achieve autonomous navigation of the intelligent field mowing robot, enabling unmanned mowing operations, avoiding double mowing and missed mowing, improving mowing efficiency, and reducing labor costs. There are two depth camera components 9, including an RGB-D depth camera and a camera bracket. One is fixed to the front end of the robot mobile platform body through screw holes and corner brackets, and the other is fixed to the front end of the robot mobile platform body through screw holes and corner brackets. The screw holes and corner fittings are fixed to the rear end of the robot's mobile platform body, allowing for the acquisition of image data from both the front and rear of the mowing robot. A weed recognition model based on a convolutional neural network is trained via the central processing unit 17. Based on the image data from the front of the mowing robot, it can accurately identify the presence of weeds in front of the robot, controlling the start / stop of the mowing device and the mowing depth. The central processing unit 17 processes the image data from the rear of the mowing robot to obtain the mowing effect, and adjusts the mowing device in real time based on the effect, achieving closed-loop control. The control box 11 houses the rugby ball-shaped hub motor controller, the hub motor controller, the side electric push rod controller, and the front-mounted... The system includes an electric actuator controller, a lawnmower motor controller, a distribution board, a step-down module, and a hardware circuit drive control board. There are two hardware circuit drive control boards, both STM32F4 control boards, fixed to the control box 11 via upper and lower nylon pillars. The upper hardware circuit drive control board is electrically connected to the articulated motor 8, the rugby ball-shaped hub motor controller, the hub motor controller, the lower hardware circuit drive control board, the central processing unit 17, and the step-down module. The lower hardware circuit drive control board is electrically connected to the side electric actuator controller, the front electric actuator controller, the lawnmower motor controller, and the step-down module, enabling autonomous navigation and lawnmower operation control for the intelligent field mowing robot.

[0082] like Figure 6 The diagram shown is a structural schematic of an intelligent field mowing robot, as follows: Figure 7 The diagram shows a smart field mowing robot operating on narrow field ridges. Figure 8The diagram shows a schematic of a smart field mowing robot operating in a wide-area field. The robot has two mowing devices, one on each side of its mobile platform. Each device includes a side electric push rod 22, a bottom fixing plate 19 for the side electric push rod, a top fixing plate 23 for the side electric push rod, a hinge 20, a mowing motor 2, and a flexible mowing head 1. The two side electric push rods 22 have a rated voltage of 24V and a maximum power of 100W. Both the top and bottom ends have 7mm diameter round hole connectors. The bottom fixing plate of the side electric push rod 22 is vertically fixed to the top plate 18 of the robot's mobile platform via screw holes and corner fittings. A small aluminum sheet, 80mm long, 40mm wide, and 10mm thick, with a 7mm diameter round hole, is vertically welded to the middle of the outer side. The bottom round hole connector of the side electric actuator 22 is coaxially connected to the small aluminum plate round hole on the outer side of the side electric actuator bottom fixing plate 19. A 40mm long M6 screw passes through the round hole, with a 6mm inner diameter, 12mm outer diameter, and 2mm thick washer placed at each end. It is fixed by an M6 nut, allowing the bottom of the side electric actuator 22 to rotate around the screw. Two small aluminum plates, each 80mm long, 40mm wide, and 10mm thick, with 7mm diameter round holes drilled in them, are vertically welded to the middle of the outer side of the side electric actuator top fixing plate 23. The two small aluminum plates are 20mm apart. The top round hole connector of the side electric actuator 22 is coaxially connected to the two small aluminum plates round holes on the outer side of the side electric actuator top fixing plate 23. A 60mm long M6 screw passes through the round hole. The screw passes through a round hole, with a washer of 6mm inner diameter, 12mm outer diameter, and 2mm thickness placed at each end, and is fixed by an M6 nut. The top of the side electric push rod 22 can rotate around the screw as an axis. The hinge 20 is made of stainless steel. One side of the hinge 20 is connected to the bottom fixing plate 19 of the side electric push rod by four M6 screws with a length of 15mm. The other side of the hinge 20 is connected to the top fixing plate 23 of the side electric push rod by four M6 screws with a length of 15mm. The top fixing plate 23 of the side electric push rod can rotate around the axis of the hinge 20. The flexible mowing head 1 includes a mowing head shell 26 and a mowing rope 27. The stator and rotor ends of the mowing motor 2 each have six M6 nuts distributed equidistantly in a circle. The stator end is fixed to the top fixing plate 23 of the side electric push rod with 6 M4 screws with a length of 10mm. The rotor end is fixed to the flexible grass trimmer head 1 with 6 M4 screws with a length of 15mm. This allows the flexible grass trimmer head 1 to rotate synchronously and at high speed with the grass cutting motor 2 to perform grass cutting operations. Compared with traditional blade rotation grass cutting, using the flexible grass trimmer head 1 is safer. Moreover, the field ridge environment is complex. Even if the flexible grass trimmer head 1 hits a stone or other object, it will not be damaged, thus extending the service life of the grass cutting equipment. The top fixing plate 23 of the side electric push rod is used to fix the top of the side electric push rod 22 and the grass cutting motor 2. At the same time, it can prevent grass clippings and stones hit by the flexible grass trimmer head 1 from flying, thus improving the safety and reliability of the robot.By controlling the extension and retraction of the side electric push rod 22, the top fixing plate 23 of the side electric push rod rotates around the hinge 20 axis. When the side electric push rod 22 extends, the top fixing plate 23 rotates counterclockwise around the hinge 20 axis; when the side electric push rod 22 retracts, the top fixing plate 23 rotates clockwise around the hinge 20 axis. The position of the top fixing plate 23 of the side electric push rod is adjusted in real time according to the slope on both sides of the field ridge, thereby realizing the mowing of grass with varying depth on both sides of the field ridge. At the same time, by adjusting the position of the top fixing plate 23 of the side electric push rod and the length of the mowing rope 27, the robot can also perform mowing operations on field ridges with different slopes and heights, such as... Figure 6 The diagram shows an intelligent field mowing robot operating on a narrow field ridge 25. When the top fixing plate 23 of the side electric push rod is horizontal to the ground, the front mowing device and the side mowing devices of the robot are on the same plane, allowing the robot to perform wide-area mowing. Figure 7 As shown, the efficiency of mowing is greatly improved. Compared with traditional weeding equipment, it expands the working environment of the same equipment and reduces the investment cost of mowing equipment for growers.

[0083] like Figure 9 The diagram shows a path planning flowchart for an intelligent field ridge mowing robot. It utilizes multi-sensor fusion SLAM technology for the robot's localization and mapping. A 3D LiDAR (13), inertial measurement module (16), GNSS high-precision positioning antenna (14), and integrated navigation receiver (10) are used as information sources for fusion processing, improving the accuracy and stability of point cloud registration and achieving centimeter-level positioning accuracy in real time. This addresses the problem of the robot being unable to perform autonomous navigation mowing operations due to GNSS signal loss or LiDAR sensor positioning failure in field ridge and grassland environments. The path planning of the intelligent field ridge mowing robot employs an improved bidirectional RRT* algorithm based on dynamic end-node guidance and potential field guidance sampling. During the post-path backtracking process, redundant path points are removed and adjacent polyline segment corner constraints are processed. Simultaneously, the algorithm incorporates the artificial potential field method, using the potential field to guide sampling points falling in obstacle areas to be rejected along the downward direction of the potential field at that point to the obstacle-free area, thus enabling the reuse of sampling points. The path points are dynamically modified according to the threshold given by the kinematic corner constraints of the lawnmower robot. The improved bidirectional RRT* algorithm used by the intelligent field mowing robot can generate the mowing operation path after dynamic tip point guided sampling, potential field guided sampling, and backtracking path point processing. However, considering the path smoothness and the robot's maximum curvature constraint, a third-order quasi-uniform B-spline curve is used to optimize the trajectory after the path is generated, thus obtaining the final operation trajectory of the lawnmower robot.

[0084] like Figure 10 The figure shown is a simplified schematic diagram of the kinematic model of the present invention. For the rotating corner of the rugby ball-shaped hub motor, The wheelbase is for the hub motor and the rugby ball-shaped hub motor. The wheelbase of the hub motor. The turning radius is given. Under normal driving conditions, the intelligent field mowing robot is driven by hub motor 21, while the rugby ball-shaped hub motor 5 does not work, only providing support and steering. When the robot travels on complex terrain and hub motor 21 lacks power, the rugby ball-shaped hub motor 5 works to provide auxiliary driving force for the robot's movement. The rugby ball-shaped hub motor 5 performs independent steering control under the drive of joint motor 8. To simplify the motion model of the mowing robot while reflecting the vehicle characteristics, assuming the mowing robot travels in a low-speed and slip-free scenario, the center point of the rear axle of the mowing robot's mobile platform is used as the rotation reference point to obtain the simplified motion model of the mowing robot. The rotation angle of its rugby ball-shaped hub motor is: ;

[0085] In the formula, For the rotating corner of the rugby ball-shaped hub motor, The wheelbase is for the hub motor and the rugby ball-shaped hub motor. The wheelbase of the hub motor. The turning radius;

[0086] like Figure 11 The diagram shown is a schematic representation of the autonomous navigation system of this invention. Current position For a target point on the path, For the turning radius, Forward sight distance, For heading deviation, This refers to lateral error;

[0087] Based on a pure tracking algorithm, the lawnmower robot moves from its current position... Reaching the path tracking point The trajectory of the movement can be regarded as a segment of... The trajectory of a circular arc with radius can be obtained using the law of sine: ;

[0088] After conversion and simplification, we get: ;

[0089] In the formula, Forward sight distance, This refers to the heading deviation, specifically the current orientation of the lawnmower robot relative to the target point. The angle between them;

[0090] According to the curvature formula, the curvature of the lawnmower robot's working path can be expressed as: ;

[0091] In the formula, The curvature of the lawnmower's working path;

[0092] The rotation angle of the rugby ball-shaped hub motor of the lawnmower robot can be expressed as: ;

[0093] Combining the above formula, the control quantity of the pure tracking algorithm can be obtained. The final expression is: ;

[0094] The time domain is introduced in the formula. for Lawn-mowing robot and target point The included angle;

[0095] definition Current posture and target point of the lawnmower robot The lateral error can be obtained as follows: ;

[0096] Meanwhile, the curvature of the lawnmower robot's working path can be expressed as: ;

[0097] forward sight distance Expressed as a linear function of the lawnmower robot's linear velocity, we can obtain: ;

[0098] Combining the above formula, we can obtain: ;

[0099] In the formula, This is the proportionality coefficient. The linear velocity of the lawnmower robot;

[0100] According to the method for calculating angular velocity, we can obtain: ;

[0101] In the formula, The angular velocity of the lawnmower robot;

[0102] By adjusting the scaling factor Then, the pure tracking algorithm was adjusted to improve the angular velocity of the lawnmower robot. Forward viewing distance They are inversely proportional; that is, the larger the forward sight distance, the smaller the angular velocity, and the smoother the tracked trajectory; the smaller the forward sight distance, the larger the angular velocity, and the more oscillating the tracked trajectory.

[0103] To better determine the forward sight distance, the Particle Swarm Optimization (PSO) algorithm is introduced, along with a heading error variable. , representing the angle between the robot's body pose and the optimized mowing path, using lateral error. With heading error The sum of the root mean squares is used as the main reference for the particle swarm optimization algorithm to determine the lateral error of the lawnmower robot. The evaluation parameters are:

[0104] ;

[0105] lawnmower heading error The evaluation parameters are:

[0106] ;

[0107] Based on the above two equations, the fitness function is designed as follows: ;

[0108] In the formula, Sampling time, For working hours, The fitness function designed for weights. Able to determine the lateral error of the lawnmower robot Adjustment and The decision weights have good adaptability;

[0109] Specifically, in a pure tracking model based on the PSO particle swarm optimization algorithm, when the lateral error... When it is large, the forward sight distance Mainly through lateral error Make decisions; when lateral errors When it is small, forward sight distance Mainly through heading error Make decisions;

[0110] Therefore, design weights: , ;

[0111] In the formula, The base of the weighting function is . To adjust the parameters, This represents the maximum lateral error of the lawnmower robot.

[0112] The robot uses a 3D LiDAR 13, an inertial measurement module 16, and a BeiDou RTK-GPS system to acquire the position and attitude information of the lawnmower robot in real time. This allows the robot to determine the distance and approximate size of obstacles and whether it can overcome them. If the obstacle is deemed insurmountable, the robot performs obstacle avoidance maneuvers, maintaining a safe distance from the obstacle. If the obstacle is deemed traversable, the robot adjusts the height of the mowing device to cross it, thus achieving obstacle avoidance during unmanned lawnmower operations, reducing equipment wear, and extending its service life. Based on the optimized work path, a pure tracking algorithm is used, with the central processing unit 17 and the upper-level hardware circuit driving control board coordinating the control of the joint motors and wheel hub motors, thereby adjusting the angular velocity and speed of the lawnmower robot in real time.

[0113] The technical solution of this invention is: a method for controlling the mowing operation of an intelligent field ridge mowing robot, comprising the following steps:

[0114] Step 1: Lawn mowing operation control flowchart. Image data is collected by the depth camera component 9 at the front end of the robot mobile platform. The central processing unit 17 trains a weed recognition model based on convolutional neural network, which can accurately identify whether there are weeds in front of the lawn mowing robot.

[0115] Step 2: Design a remote control terminal for the lawn mowing robot. The human-machine interface of the remote control terminal is located on a mobile application and communicates with the central processing unit 17 of the lawn mowing robot through a local area network. The depth camera components 9 installed at the front and rear of the lawn mowing robot can capture images of the area around the lawn mowing robot in real time, and display the captured images in real time on the human-machine interface via high-definition image transmission.

[0116] Step 3: Based on the specific mowing scenario, select the corresponding mowing mode via the remote control terminal. If it is a field ridge mowing operation, set the initial positions of the mowing and cutting devices in the field ridge mowing operation mode. After reaching the starting position of the field ridge mowing operation, the mowing devices on both sides of the robot are adjusted to the set positions by the side electric push rods 22. When the robot reaches the end position of the field ridge mowing operation and needs to change the ridge, the mowing devices on both sides are raised by the side electric push rods 22 and adjusted to a horizontal position with the ground. Then, the robot moves to the next ridge to perform mowing operations, and so on. If it is a field mowing operation, set the initial positions of the mowing and cutting devices in the field mowing operation mode. The front mowing device and the side mowing devices of the robot are on the same plane, which allows the robot to perform wide-area mowing in the field, greatly improving mowing efficiency.

[0117] Step 4: The central processing unit 17 receives control commands from the remote control terminal via local area network communication, processes them, and sends them to the hardware circuit driver control board to control the lawn mowing robot to perform lawn mowing operations. When weeds are detected in front of the lawn mowing robot, the mowing and trimming devices are automatically activated to perform lawn mowing operations. The lawn mowing operation is analyzed through the image of the lawn mowing robot's surroundings displayed on the human-machine interface of the remote control terminal. The position of the mowing and trimming devices is adjusted at the remote control terminal to achieve the expected lawn mowing effect.

[0118] Step 5: When the lawn mowing robot achieves the expected lawn mowing effect, the current positions of the mowing and cutting devices can be saved through the remote control terminal. In subsequent operations in the same scene, the robot can automatically adjust the mowing and cutting devices to the corresponding positions saved in advance.

[0119] Step 6: Set the optimal operating current of the mowing motor 2 as the threshold current. The hardware circuit drives the control board to detect the operating current of the mowing motor in real time. If the operating current of the mowing motor is less than the threshold current, the speed of the mowing robot is increased to improve the mowing efficiency. If the operating current of the mowing motor is greater than the threshold current, the speed of the mowing robot is reduced to reduce the load on the mowing motor and extend its service life. If the operating current of the mowing motor is still greater than the threshold current after the mowing robot adjusts its speed ten times, the mowing operation is automatically stopped and an alarm message is sent to the remote control terminal to prevent the mowing motor from burning out due to excessive load caused by the mowing device being blocked by grass or debris being caught in it, thus causing economic losses.

[0120] Step 7: The remote control terminal of the lawnmower robot has a higher priority than the autonomous lawnmower operation control priority. It can display the operation trajectory and robot status parameter information in real time, so as to take over the control of the lawnmower robot at any time. It can realize emergency braking and one-click return of the lawnmower robot, making the lawnmower robot more controllable and improving its safety and reliability.

Claims

1. An intelligent fence mowing robot, characterized in that, The robot mobile platform body includes a rugby-shaped wheel hub motor (5), a rugby-shaped wheel hub motor arc-shaped fixing piece (7), a robot mobile platform top plate (18), a wheel hub motor (21), a wheel hub motor fixing piece (24), a joint motor (8) and a robot support (12); the rugby-shaped wheel hub motor (5) is used as a front wheel of the robot mobile platform body; the rugby-shaped wheel hub motor arc-shaped fixing piece (7) is cut with a reverse U-shaped notch on the left and right sides, the two rotating shafts of the rugby-shaped wheel hub motor (5) are respectively placed into the reverse U-shaped notches of the rugby-shaped wheel hub motor arc-shaped fixing piece (7) and fixed through nuts; the top end of the rugby-shaped wheel hub motor arc-shaped fixing piece (7) is polished to be a plane, the rotor end of the joint motor (8) is fixed with the top end of the rugby-shaped wheel hub motor arc-shaped fixing piece (7) through a screw hole; the stator end of the joint motor (8) is fixed with the robot mobile platform top plate (18) through a screw hole; the wheel hub motor (21) has two, and the wheel hub motor (21) is used as a rear wheel of the robot mobile platform body; the wheel hub motor fixing piece (24) has two, the wheel hub motor fixing piece (24) is cut with a reverse U-shaped notch on the left and right sides, the two rotating shafts of the wheel hub motor (21) are respectively placed into the reverse U-shaped notches of the wheel hub motor fixing piece (24) and fixed through nuts; the top end of the wheel hub motor fixing piece (24) is fixed with the robot mobile platform top plate (18) through a screw hole; the robot mobile platform top plate (18) is an aluminum plate, located above the rugby-shaped wheel hub motor arc-shaped fixing piece (7) and the stator end of the joint motor (8) and fixedly connected in a horizontal manner; the robot support (12) is fixed with the robot mobile platform top plate (18) through a screw hole and an angle piece; The mowing device includes a front electric push rod (6), an electric weed trimmer (4) and an electric weed trimmer fixing plate (3); the front electric push rod (6) has a flat base at both ends, the bottom end of the front electric push rod (6) is fixed with the robot mobile platform top plate (18) through a screw hole, and the top end of the front electric push rod (6) is fixed with one end of the electric weed trimmer fixing plate (3) through a screw hole; the electric weed trimmer (4) has four, is distributed equidistantly at one end of the electric weed trimmer fixing plate (3) and connected with the electric weed trimmer fixing plate (3) through a screw hole; the electric weed trimmer fixing plate (3) makes lifting movement through the telescopic movement of the front electric push rod (6), thereby driving the electric weed trimmer (4) to move synchronously. ​ The two mowing devices are arranged on both sides of the robot moving platform body, and each of the mowing devices comprises a side electric push rod (22), a side electric push rod bottom end fixed plate (19), a side electric push rod top end fixed plate (23), a hinge (20), a mowing motor (2) and a flexible mowing head (1); the top end and the bottom end of the side electric push rod (22) are both round hole joints, the side electric push rod bottom end fixed plate is fixedly connected with the robot moving platform top plate (18) through screw holes and an angle piece, a small aluminum sheet with a round hole is vertically welded at the middle position of the outer side of the side electric push rod bottom end fixed plate (19), the side electric push rod bottom end round hole joint is coaxially connected with the small aluminum sheet round hole on the outer side of the side electric push rod bottom end fixed plate (19), a screw rod passes through the round holes, a gasket is arranged at each end of the screw rod, and the side electric push rod bottom end can rotate around the screw rod through a nut; two small aluminum sheets with round holes are vertically welded at the middle position of the outer side of the side electric push rod top end fixed plate (23), the side electric push rod top end round hole joint is coaxially connected with the small aluminum sheet round hole on the outer side of the side electric push rod top end fixed plate (23), a screw rod passes through the round holes, a gasket is arranged at each end of the screw rod, and the side electric push rod top end can rotate around the screw rod through a nut; the hinge (20) is made of stainless steel, one leaf of the hinge (20) is connected with the side electric push rod bottom end fixed plate (19) through screw holes, the other leaf of the hinge (20) is connected with the side electric push rod top end fixed plate (23) through screw holes, and the side electric push rod top end fixed plate (23) can rotate around the hinge (20) shaft; the stator end of the mowing motor (2) is fixed with the side electric push rod top end fixed plate (23) through screw holes, and the rotor end of the mowing motor (2) is fixed with the flexible mowing head (1) through screw holes, so that the flexible mowing head (1) and the mowing motor (2) rotate synchronously; the flexible mowing head (1) comprises a mowing head shell (26) and a mowing rope (27), and the flexible mowing head (1) is driven by the mowing motor (2) to rotate at a high speed; The perception decision and planning control device comprises a three-dimensional laser radar (13), a Beidou positioning RTK-GPS system, a depth camera assembly (9), a battery (15), a central processor (17), an inertial measurement module (16) and a control box (11); the three-dimensional laser radar (13) is located at the front end of the robot mobile platform body and is fixed in parallel above the robot support (12) through a screw hole; the Beidou positioning RTK-GPS system comprises a GNSS high-precision positioning antenna (14) and a combined navigation receiver (10); the GNSS high-precision positioning antenna (14) is located on both sides of the robot mobile platform body and is fixed in parallel above the robot support (12) through a screw hole; the depth camera assembly (9) comprises an RGB-D depth camera and a camera support, and there are two depth camera assemblies (9), one of which is fixed at the front end of the robot mobile platform body through a screw hole and an angle piece, and the other is fixed at the rear end of the robot mobile platform body through a screw hole and an angle piece; the battery (15) is a lithium battery and is placed above a battery fixing plate, and the battery fixing plate is fixed with the robot mobile platform top plate (18) through a screw hole; the central processor (17) is a microcomputer, has a mounting hole around, and is fixed with the robot mobile platform top plate (18) through a screw hole; the inertial measurement module (16) has a mounting hole around and is fixed with the robot mobile platform top plate (18) through a screw hole; the combined navigation receiver (10) is electrically connected with the GNSS high-precision positioning antenna (14) and is fixed with the robot mobile platform top plate (18) through a screw hole; the control box (11) is made of stainless steel and is fixed with the robot mobile platform top plate (18) through a screw hole.

2. The intelligent fence mowing robot of claim 1, wherein, Under normal driving conditions, the rugby-shaped wheel hub motor (5) does not work and only provides support and steering effect, when the robot drives to a complex road surface and the wheel hub motor (21) is insufficient, the rugby-shaped wheel hub motor (5) works to provide auxiliary driving force for the robot movement, and the working state of the rugby-shaped wheel hub motor (5) is switched; the wheel hub motor (21) provides power for the robot to drive.

3. The intelligent fence mowing robot of claim 1, wherein, The rotation angle of the joint (8) is controlled to drive the rugby-shaped wheel hub motor (5) to rotate synchronously, so as to realize the turning angle control of the robot.

4. The intelligent fence mowing robot of claim 1, wherein, The rugby-shaped wheel hub motor arc-shaped fixing piece (7), the robot mobile platform top plate (18), the electric weeding shear fixing plate (3), the wheel hub motor fixing piece (24), the side electric push rod bottom end fixing plate (19) and the side electric push rod top end fixing plate (23) are aluminum plates, the robot support (12) is an aluminum profile and is fixed with the robot mobile platform top plate (18) through a screw hole and an angle piece, is used for placing the perception decision and planning control device, and transparent acrylic plates are installed around the robot support (12) through screw holes to waterproof the robot perception decision and planning control device.

5. The intelligent fence mowing robot of claim 1, wherein, The side electric push rod (22) bottom end and the side electric push rod (22) top end can rotate around the screw rod penetrating the round hole joint as the axis, the side electric push rod (22) is controlled to do the extension and contraction movement, the side electric push rod top end fixed plate (23) rotates around the hinge hinge (20) shaft as the axis, when the side electric push rod (22) is extended, the side electric push rod top end fixed plate (23) rotates around the hinge hinge (20) shaft as the axis counterclockwise, when the side electric push rod (22) is retracted, the side electric push rod top end fixed plate (23) rotates around the hinge hinge (20) shaft as the axis clockwise, the position of the side electric push rod top end fixed plate (23) is adjusted according to the slope of the two sides of the ridge, and then the variable grass depth mowing of the slope of the two sides of the ridge is realized, and through the adjustment of the position of the side electric push rod top end fixed plate (23) and the length of the flexible grass hitting rope in the flexible grass hitting head (1), the robot can also be used for mowing work on the ridge with different slopes, when the side electric push rod top end fixed plate (23) is horizontal to the ground, the front mowing device and the two side mowing devices of the mowing robot are in the same plane, the flexible grass hitting head (1) rotates at high speed under the driving of the mowing motor (2) to mow grass, and the side electric push rod top end fixed plate (23) is used for fixing the top end of the side electric push rod (22) and the mowing motor (2).

6. The intelligent fence mowing robot of claim 1, wherein, The GNSS high-precision positioning antenna (14) and the integrated navigation receiver (10) are electrically connected, and the positioning information is output through the integrated navigation receiver (10); the central processor (17) is a microcomputer, which processes the data collected by the three-dimensional laser radar (13), the Beidou positioning RTK-GPS system and the inertial measurement module (16), realizes the autonomous navigation of the intelligent ridge mowing robot, carries out unmanned mowing work, avoids over-mowing and missed mowing, improves the mowing efficiency and reduces the labor cost; the depth camera assembly (9) can collect image data in front of and behind the mowing robot, a weed recognition model based on a convolutional neural network is trained through the central processor (17), whether weeds exist in front of the mowing robot can be accurately recognized according to the image data in front of the mowing robot, the start and stop of the mowing device and the mowing depth are controlled, the mowing effect is obtained by processing the image data behind the mowing robot through the central processor (17), and the mowing device is adjusted in real time according to the mowing effect, so that closed-loop control is realized.

7. The intelligent fence mowing robot of claim 1, wherein, The battery (15) is a lithium battery, which is used for providing electric energy for the mowing robot, and there are two lithium batteries in total, which are separately powered according to the power consumption of each electric device, one lithium battery on the left side is used for supplying power for one football-shaped hub motor (5), two hub motors (21), one joint motor (8) and the sensing decision and planning control device, and one lithium battery on the right side is used for supplying power for the mowing device and the mowing device.

8. The intelligent fence mowing robot of claim 1, wherein, The control box (11) comprises a rugby-shaped wheel hub motor controller, a wheel hub motor controller, a side electric push rod controller, a front electric push rod controller, a mowing motor controller, a distribution board, a step-down module and a hardware circuit driving control board; the hardware circuit driving control board has two pieces, both of which are STM32F4 control boards, fixed in the control box (11) through nylon columns, the upper hardware circuit driving control board is electrically connected with the joint motor (8), the rugby-shaped wheel hub motor controller, the wheel hub motor controller, the lower hardware circuit driving control board, the central processing unit (17) and the step-down module, and the lower hardware circuit driving control board is electrically connected with the side electric push rod controller, the front electric push rod controller, the mowing motor controller and the step-down module, so as to realize the autonomous navigation and mowing operation control of the intelligent ridge mowing robot.

9. A method for autonomous navigation of an intelligent hedge trimmer robot, the method comprising the steps of: providing an intelligent hedge trimmer robot according to claim 1, and autonomously navigating the intelligent hedge trimmer robot. The steps are as follows: Step 1: using multi-sensor fusion SLAM (Simultaneous Localization and Mapping) technology for positioning and mapping of the intelligent ridge mowing robot, using three-dimensional laser radar (13), inertial measurement module (16), GNSS high-precision positioning antenna (14) and integrated navigation receiver (10) as information sources for fusion processing, improving the accuracy and stability of point cloud registration, obtaining real-time centimeter-level positioning accuracy, and solving the problem that the mowing robot cannot perform autonomous navigation and mowing operation due to loss of GNSS signal or positioning failure of laser sensor in the ridge and field grassland environment; Step 2: the path planning of the intelligent ridge mowing robot adopts an improved bidirectional RRT* algorithm based on dynamic end node guidance and potential field guidance sampling, removes redundant path points and adjacent polyline segment corner constraints during path backtracking after the path is found, and combines the artificial potential field method to repel the sampling points in the obstacle area to the obstacle-free area along the potential field descending direction of the point to realize the reuse of the sampling points and dynamically modify the path points according to the threshold given by the corner constraint; Step 3: the improved bidirectional RRT* algorithm adopted by the intelligent ridge mowing robot can generate the mowing robot operation path after dynamic end point guidance sampling, potential field guidance sampling and backtracking path point processing, but considering the path smoothness and the maximum curvature constraint of the robot, a third-order quasi-uniform B-spline curve is used to optimize the trajectory after the path is generated, so as to obtain the final operation trajectory of the mowing robot. Step 4: In the normal driving condition of the intelligent fence mowing robot, the wheel hub motor (21) is driven, the rugby-shaped wheel hub motor (5) does not work, and only provides support and steering effect. When the robot drives to a complex road surface and the wheel hub motor (21) is insufficient, the rugby-shaped wheel hub motor (5) works to provide auxiliary driving force for the robot movement. The rugby-shaped wheel hub motor (5) is independently controlled for steering under the driving of the joint motor (8). In order to simplify the motion model of the mowing robot and reflect the vehicle characteristics at the same time, it is assumed that the mowing robot drives in the low-speed and no-slip scene. Taking the center point of the rear axle of the mowing robot mobile platform body as the rotation reference point, the simplified motion model of the mowing robot is obtained, and the turning angle of the rugby-shaped wheel hub motor of the mowing robot is: ​ wherein is the steering angle of the hub motor, is the wheelbase of the hub motor and the rugby-shaped hub motor, is the wheel track of the hub motor, is the turning radius; Step 5: According to the pure pursuit algorithm, the mowing robot moves from the current position to the path tracking point The motion trajectory of the mowing robot can be regarded as a circular arc trajectory with as the radius, and according to the sine theorem, the following can be obtained: ​ After conversion and simplification, we get: ​ In the formula, is the front distance, is the heading deviation, i.e. the included angle between the current mower robot body pose and the target point ; Step 6: According to the curvature formula, the mowing robot work path curvature can be expressed as: ​ In the formula, is the mowing robot work path curvature; Step 7: The Lawnmower Robot Rugby Wheel Motor Turn Angle can be expressed as: ​ In combination with the above formula, the final expression of the pure tracking algorithm control quantity is: ; In the formula, a time domain is introduced, For The angle between the time instant and the target point Of the mowing robot Definitions The lateral error of the current pose of the mowing robot vehicle body and the target point can be obtained as: ​ Meanwhile, the mowing robot operation path curvature can be expressed as: ​ Step 8: The forward distance The linear function of the forward distance in terms of the linear velocity of the mowing robot is given by: ​ With the above formula, we have: ​ In the formula, is a proportionality coefficient, is the linear speed of the mowing robot movement; Step 9: According to the angular velocity calculation method, we have: ​ In the formula, is the angular velocity of the mowing robot; Step 10: Adjust the scaling factor Then, the pure tracking algorithm was adjusted to improve the angular velocity of the lawnmower robot. Forward viewing distance They are inversely proportional; that is, the larger the forward sight distance, the smaller the angular velocity, and the smoother the tracked trajectory; the smaller the forward sight distance, the larger the angular velocity, and the more oscillating the tracked trajectory. Step 11: In order to better determine the forward-looking distance, the PSO particle swarm algorithm is introduced, and the heading error variable is introduced here , which represents the angle between the body pose of the mowing robot and the optimized mowing operation path, and the sum of the lateral error and the root mean square of the heading error is used as the main reference for the particle swarm algorithm optimization, and the evaluation parameter of the lateral error of the mowing robot is: ； Lawn mower robot heading error The evaluation parameter is: ​ The fitness function is designed based on the above two equations as: ​ In the formula, is a sampling time, is a working time, is a weight, the designed fitness function can be adjusted according to the lateral error of the mowing robot adjust and the decision weight, has good adaptability; Specifically, in the pure pursuit model based on PSO particle swarm algorithm, when the lateral error is large, the look-ahead distance is mainly determined by the lateral error ; when the lateral error is small, the look-ahead distance is mainly determined by the heading error ; Thus, the weights are designed as: , ; wherein is the base of the weight function, is the adjustment parameter, is the maximum lateral error of the mowing robot; Step 12: Real-time acquisition of the position and attitude information of the mowing robot by using the three-dimensional laser radar (13), the inertial measurement module (16), and the Beidou positioning RTK-GPS system, confirmation of the distance and estimated volume of the obstacle, judgment of whether the mowing robot can cross the obstacle, if the obstacle cannot be crossed, the obstacle avoidance action is performed, and the appropriate safety distance from the obstacle is maintained, if the obstacle can be crossed, the height of the mowing device is adjusted, the obstacle is crossed, and the obstacle avoidance function in the unmanned mowing process is realized, reducing equipment wear and tear and improving service life; using the pure tracking algorithm, the central processing unit (17) and the upper hardware circuit drive control board cooperatively control the joint motor and the hub motor, and then the angular velocity and speed of the mowing robot are adjusted in real time.

10. A method for controlling the mowing operation of an intelligent fence mowing robot, using an intelligent fence mowing robot as claimed in claim 1, characterized in that, The steps are as follows: Step 1: Collect image data using the robot mobile platform body front-end depth camera assembly (9), train the weed recognition model based on the convolutional neural network through the central processing unit (17), and accurately identify whether there are weeds in front of the mowing robot; Step 2: Design a remote control end for the mowing robot, the human-computer interaction interface of the remote control end is arranged on the mobile phone application program end, and the central processing unit (17) of the mowing robot communicates through the local area network, the depth camera assemblies (9) installed in front and back of the mowing robot can capture images around the mowing robot in real time, and the captured images are displayed in real time on the human-computer interaction interface through the high-definition image transmission; Step 3: According to the specific mowing operation scene, select the corresponding mowing operation mode through the remote control end; if it is a ridge mowing operation, set the initial positions of the mowing device and the mowing device in the ridge mowing operation mode, after reaching the ridge mowing operation starting position, the left and right side mowing devices are adjusted to the set position under the drive of the side electric push rod (22), when reaching the ridge mowing operation endpoint position and needing to replace the ridge, the left and right side mowing devices are raised and adjusted to the horizontal position with the ground under the drive of the side electric push rod (22), and then the mowing robot drives to the next ridge for mowing operation, and so on; if it is a field mowing operation, set the initial positions of the mowing device and the mowing device in the field mowing operation mode, the front mowing device and the two side mowing devices of the mowing robot are in the same plane, the robot can realize field wide mowing, and the mowing efficiency is greatly improved; Step 4: The central processing unit (17) receives the control instructions from the remote control end through the local area network communication, processes and then issues to the hardware circuit drive control board, and then controls the mowing robot to perform the mowing operation; when weeds in front of the mowing robot are identified, the mowing device and the mowing device are automatically started to perform the mowing operation; The image around the mowing robot displayed on the human-computer interaction interface of the remote control end is analyzed to analyze the mowing operation situation, and the positions of the mowing device and the mowing device are adjusted on the remote control end to achieve the expected mowing operation effect; Step 5: When the mowing robot reaches the expected mowing effect, the current position of the cutting device and the mowing device can be saved through the remote control terminal. In subsequent operations in the same scenario, the robot can automatically adjust the cutting device and the mowing device to the corresponding pre-saved positions. Step 6: Set the optimal working current of the mowing motor (2) as the threshold current. Real-time detection of the mowing motor working current is performed through the hardware circuit drive control board. If the mowing motor working current is less than the threshold current, the mowing robot driving speed is increased to improve the mowing efficiency. If the mowing motor working current is greater than the threshold current, the mowing robot driving speed is reduced to reduce the load of the mowing motor and prolong the service life. If the mowing motor working current is still greater than the threshold current after ten times of adjustment of the mowing robot driving speed, the mowing operation is automatically stopped and an alarm message is sent to the remote control terminal to prevent the mowing motor from being burned out due to excessive load caused by grass or debris being wrapped into the mowing device, thereby causing economic losses. Step 7: The priority of the remote control terminal of the mowing robot is higher than that of the autonomous mowing operation control priority. The operation trajectory and robot state parameter information can be displayed in real time to facilitate taking over the control authority of the mowing robot at any time. The mowing robot can realize emergency braking and one-key return, so that the mowing robot has better controllability, safety and reliability.

Images