Automatic weeding device and method in transformer substation and obstacle avoidance control system

Abstract

The invention discloses an automatic weeding device and method in a transformer substation and an obstacle avoidance control system, and belongs to the technical field of weeding. The weeding device comprises a machine body, a crawler-type walking mechanism is arranged at the bottom of the machine body, and a top plate is arranged at the top of the machine body; the cutter driving mechanism is arranged on the front side of the top plate, and weeding rotary cutters are arranged below the cutter driving mechanism to drive the weeding rotary cutters to move; the sensor module is arranged on the cutter driving mechanism and is used for acquiring the distance of an obstacle; the weed recognition module is arranged on the cutter driving mechanism and is used for shooting weed images; the obstacle avoidance control system calculates the position of an obstacle based on the distance of the obstacle so as to realize the bypassing obstacle avoidance of the weeding device and the weeding avoidance of the weeding rotary knife, and plans the moving path of the weeding device and the movement of the weeding rotary knife based on the weed image information so as to realize accurate weeding; according to the weeding device, the tool bit can be better protected, precise weeding can be achieved, automatic weeding is achieved, labor is greatly saved, the labor intensity of workers is reduced, and the working efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of weeding, and in particular to an automatic weeding device and method, and an obstacle avoidance control system in a substation. Background Art

[0002] Because the outer casing of outdoor capacitor equipment in substations is charged, it is equipped with enclosed safety fences to ensure the safety of personnel and equipment. However, because the ground inside the fence is exposed to the external environment for a long time, it can gradually become damaged and broken due to the influence of external environment and weather conditions. This leads to the growth of weeds in the capacitor area during the summer. If these weeds are not promptly removed, they can damage the insulation of the equipment and reach the live parts, affecting the safety and normal operation of the capacitor. Therefore, timely cleaning is often necessary. However, most weed removers require manual operation, which is labor-intensive and inefficient.

[0003] Prior art document 1 (CN117136729A) discloses an automatic lawn mower, including a chassis, a mobile system, a control system and a mowing system, wherein the control system is connected to and outputs a driving signal to the mowing system and the mobile system. The control system is provided with a sensor system processing unit, a motor drive unit and a motor control unit, and a camera, an infrared sensor and an ultrasonic radar for collecting information about obstacles ahead are provided on the top of the control cabin. The camera, infrared sensor and ultrasonic radar are connected to and output sensing signals to the sensor system processing unit, and the sensor system processing unit is connected to and obtains the sensing signals collected by the perception system. The motor control unit is connected to the motor drive unit and exchanges information, and the motor drive unit outputs a driving signal to the motors of the mowing system and the mobile system.

[0004] The shortcomings of prior art document 1 are that the prior art cannot identify the weeding location and does not avoid weeding for the mowing system. Moreover, the ultrasonic radar set in the prior art is installed in the control cabin, which is far away from the obstacles, resulting in inaccurate ranging. At the same time, the prior art only provides the operation method and does not provide the specific structure of the lawn mower. In addition, the ultrasonic radar only serves to assist in obtaining obstacle information. The prior art mainly uses the navigation and positioning unit to plan the path of the lawn mower to achieve obstacle avoidance. However, the disadvantages of navigation and positioning are as follows:

[0005] (1) High cost: The navigation and positioning unit is expensive, which will increase the cost of the lawn mower.

[0006] (2) Accuracy is affected by the environment: Navigation positioning accuracy is greatly affected by the environment. For example, weather and terrain will affect the accuracy of navigation positioning.

[0007] (3) Unable to handle dynamic obstacles: The navigation and positioning unit cannot identify and handle dynamic obstacles, such as pedestrians, animals, etc.

[0008] Prior art document 2 (CN219919718U) discloses an automatic driving weeding robot, including a crawler walking mechanism, a bracket is installed on the top left side of the crawler walking mechanism, a camera is installed above the left side of the bracket, a connecting plate is installed on the right side of the crawler walking mechanism, and a lifting push rod is installed on the top of the right side of the connecting plate. The lifting push rod adjusts the machine to the required weeding height, automatically identifies the weeding area according to the camera, and automatically avoids weeds.

[0009] Although the prior art can use a camera to identify the weeding environment and operate the lifting push rod to achieve weeding avoidance, the disadvantage of prior art document 2 is that the camera set in the prior art is installed with a GPS positioning system, and the weeding avoidance operation is still achieved by using the navigation positioning principle, which has the following disadvantages:

[0010] 1. Dependence on GPS signals: GPS signals may be unstable or unreceivable in certain environments. The GPS signals in some substations may be poor, which may cause the weeding robot to not work properly and require manual intervention or auxiliary positioning equipment.

[0011] 2. Positioning accuracy limitations: Positioning accuracy is limited and may not meet the requirements of fine weeding operations. For example, in some cases, the robot may not be able to accurately identify small obstacles or precisely control the weeding range.

[0012] 3. Higher cost: The system and related hardware costs are high, which will increase the overall cost of the weeding robot.

[0013] In addition, its weeding avoidance operation mainly adjusts the angle and height of the weeding head by lifting the push rod. For obstacles with larger sizes, simply raising the angle of the weeding head cannot achieve effective avoidance. Summary of the Invention

[0014] In order to solve the deficiencies in the prior art, the present invention provides an automatic weeding device and method, and an obstacle avoidance control system in a substation, which realizes automated weeding, greatly saves manpower, reduces the labor intensity of workers, improves work efficiency, and solves the problems existing in the background technology.

[0015] The present invention adopts the following technical solutions.

[0016] A first aspect of the present invention provides an automatic weeding device in a substation, comprising: a body, a tool drive mechanism, a sensor module, a weed identification module and an obstacle avoidance control system. A crawler walking mechanism is provided at the bottom of the body, and a top plate is provided on the top of the body; the tool drive mechanism is provided on the front side of the top plate, and a weeding rotary blade is provided below it to drive the weeding rotary blade to move; the sensor module is provided on the tool drive mechanism for obtaining the distance of the obstacle; the weed identification module is provided on the tool drive mechanism for capturing the weed image; the obstacle avoidance control system calculates the obstacle position based on the obstacle distance to enable the weeding device to bypass the obstacle and the weeding rotary blade to avoid weeding, and plans the movement path of the weeding device and the movement of the weeding rotary blade based on the weed image information to achieve precise weeding.

[0017] According to the automatic weeding device in the substation, the sensor module includes a front ultrasonic sensor arranged at the front end of the cutter drive mechanism and two side ultrasonic sensors arranged at both sides of the cutter drive mechanism.

[0018] According to the automatic weeding device in the substation, the tool driving mechanism includes: a bracket, a light rod, a fixed frame and an adjustment seat; the brackets are arranged on both sides of the top plate; the light rod is arranged between the two brackets, and there are no less than two light rods arranged parallel to each other in the upper and lower directions between the brackets, and a fixed frame that can be moved left and right is provided on the light rod, and a left and right moving mechanism for driving the fixed frame to move is provided between the brackets; an up and down moving mechanism is provided on both sides of the fixed frame, and an adjustment seat is provided on the up and down moving mechanism for adjusting the height of the adjustment seat up and down; a weeding motor is installed on the middle part of the top of the adjustment seat through a U-shaped frame, and the output shaft of the weeding motor is connected to the weeding rotary knife.

[0019] According to the automatic weeding device in the substation, the left and right moving mechanism includes: a rack, a gear and a travel motor; a rack is arranged between the top ends of the brackets, the rack is arranged above the light rod and parallel to the light rod, the rack and the gear are engaged with each other, the transmission shaft of the gear is connected to the output end of the travel motor, and a motor seat is provided at the bottom of the travel motor, which is fixedly connected to the fixed frame.

[0020] According to the automatic weeding device in the substation, the up and down moving mechanism includes: an extension plate, an adjustment plate, a stud and a nut; extension plates extending outward along the length direction are provided on both sides of the fixing frame, and an adjustment plate distributed L-shaped therewith is provided on the side of a pair of extension plates. The adjustment seat is located between the two adjustment plates, and an oblong hole is provided on the surface of the adjustment plate in the vertical direction. Studs are provided at both ends of the adjustment seat, and the ends of the studs pass through the oblong holes and are threadedly connected with nuts.

[0021] According to the automatic weeding device in the substation, the weeding device further includes a fuzzy controller, and the fuzzy controller of the obstacle avoidance control system is connected to the motor of the crawler-type walking mechanism to drive the crawler-type walking mechanism to achieve obstacle avoidance.

[0022] A second aspect of the present invention provides an automatic weeding method in a substation, according to the automatic weeding device in the substation, comprising the following steps:

[0023] Obtaining obstacle distance and weed image information around the weed removal device;

[0024] According to the obstacle distance information, the obstacle position is calculated, and the weeding device is controlled to avoid the obstacle and the weeding rotary knife to avoid weeding;

[0025] Based on the weed image information, the location and range of the weeds are identified, the movement path of the weeding device and the movement of the weeding rotary blade are planned, and the weeding operation is performed.

[0026] According to the automatic weeding method in the substation, the obstacle position includes the front obstacle position and the obstacle positions on both sides. The obstacle avoidance control system controls the weeding rotary cutter to avoid weeding according to the obstacle positions on both sides, and controls the weeding device to bypass the obstacle through the fuzzy controller according to the front obstacle position.

[0027] According to the automatic weeding method in the substation, the weeding device is controlled by the fuzzy controller to avoid obstacles, including the following steps:

[0028] The obstacle distance measured by the ultrasonic sensor is divided into five distance fuzzy sets from near to far: C, N, M, Fa, VF;

[0029] The angle interval of the weeding device rotating left and right through the crawler walking mechanism is divided into three angle fuzzy sets from left to right: L, F, R;

[0030] According to the distance fuzzy set of the obstacle measured by the front-end ultrasonic sensor, the angle fuzzy set that the crawler walking mechanism needs to adjust is selected, the moving path of the crawler walking mechanism is automatically planned, and the weeding device executes the detour and obstacle avoidance.

[0031] According to the automatic weeding method in a substation, the step of obtaining image information of weeds around the weeding device and identifying the location and range of the weeds includes the following steps:

[0032] The obstacle avoidance control system pre-processes the collected weed images;

[0033] The obstacle avoidance control system uses the YOLO algorithm to identify weeds in the preprocessed image and determine the location and range of the weeds.

[0034] An obstacle avoidance control system for executing the above-mentioned automatic weeding method in a substation, comprising:

[0035] An information acquisition module is used to obtain obstacle distances and weed image information around the weed removal device;

[0036] The obstacle avoidance and avoidance control module is used to calculate the location of obstacles based on obstacle distance information, and control the weeding device to avoid obstacles and the weeding rotary blade to avoid weeding;

[0037] The path planning and rotary blade control module is used to identify the location and range of weeds based on weed image information, plan the movement path of the weeding device and the movement of the weeding rotary blade, and perform weeding operations.

[0038] Compared with the prior art, the beneficial effects of the present invention include at least:

[0039] 1. The present invention can effectively avoid equipment through the obstacle avoidance control system, enabling the weeding device to better weed near the roots, realizing automated weeding, greatly saving manpower, reducing workers' labor intensity, and improving work efficiency;

[0040] 2. The obstacle avoidance control system of the present invention controls the weeding rotary cutter to avoid weeding according to the position of obstacles on both sides, and controls the weeding device to avoid obstacles by using a fuzzy controller according to the position of obstacles in front. This can not only better protect the cutter head but also achieve precise weeding.

[0041] 3. The up and down moving mechanism of the present invention enables the weeding device to better weed near the roots, achieving better weeding effect;

[0042] 4. The present invention controls the weeding device to avoid obstacles by means of a fuzzy controller, and can respond to obstacles more quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 A schematic structural diagram of the automatic weeding device provided by the present invention;

[0044] Figure 2 A front view of the automatic weeding device provided by the present invention;

[0045] Figure 3 A side view of the automatic weeding device provided by the present invention;

[0046] Figure 4 Schematic diagram of the structure of the tool drive mechanism in the present invention;

[0047] Figure 5 for Figure 4 Exploded view of

[0048] Figure 6 Schematic diagram of the structure of the fixing frame in the present invention;

[0049] Figure 7 Schematic diagram of the obstacle avoidance control system in the present invention;

[0050] Figure 8 This is a flow chart of the automatic weeding method of the present invention;

[0051] Figure 9 Module block diagram of the obstacle avoidance control system in the present invention;

[0052] In the figure: 1. Top plate; 2. Body; 3. Track; 4. Cutter drive mechanism; 5. Bracket; 6. Extension frame; 7. Weed identification module; 8. Weeding rotary knife; 9. Rack; 10. Gear; 11. Travel motor; 12. Weeding motor; 13. Adjustment seat; 14. Polish rod; 15. Fixed frame; 16. Extension plate; 17. Adjustment plate; 18. Sleeve; 19. Oblong hole; 20. Stud; 21. Ultrasonic sensor 1; 22. Ultrasonic sensor 2; 23. Ultrasonic sensor 3; 24. Ultrasonic sensor 4; 25. Ultrasonic sensor 5; 26. Motor seat. DETAILED DESCRIPTION

[0053] To make the objectives, technical solutions, and advantages of the present invention more clear, the technical solutions of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. The embodiments described in this application are only part of the embodiments of the present invention, not all of them. Based on the spirit of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0054] like Figure 1 As shown, embodiment 1 of the present invention provides an automatic weeding device in a substation, including a body 2, a tool drive mechanism 4, a sensor module 27, a weed recognition module 7 and an obstacle avoidance control system.

[0055] A crawler-type walking mechanism is provided at the bottom of the fuselage 2, and a top plate 1 is provided at the top of the fuselage 2;

[0056] The tool drive mechanism 4 is provided on the front side of the top plate 1, and a weeding rotary blade 8 is provided below the tool drive mechanism 4 to drive the weeding rotary blade 8 to move;

[0057] The sensor module is provided on the tool drive mechanism 4 and is used to obtain the distance to the obstacle.

[0058] The weed recognition module 7 is provided on the cutter drive mechanism 4 and is used to capture weed images;

[0059] The obstacle avoidance control system calculates the obstacle position based on the obstacle distance to enable the weeding device to avoid obstacles and the weeding rotor 8 to avoid weeding. Based on the weed image information, the movement path of the weeding device and the movement of the weeding rotor 8 are planned to achieve precise weeding.

[0060] The crawler-type walking mechanism includes: two crawlers 3 located on the left and right sides of the fuselage 2, and a left motor and a right motor respectively used to drive the two crawlers 3 to move. The left motor and the right motor are respectively used to control the movement of the two crawlers. Through the setting of the left motor and the right motor, it is convenient to control the movement of the fuselage 2, thereby facilitating the control of the fuselage 2 to avoid obstacles.

[0061] like Figure 2 、 Figure 3 and Figure 4 As shown, the tool drive mechanism 4 includes brackets 5 arranged on both sides of the top plate 1, and at least two light rods 14 arranged parallel to each other are provided between the brackets 5. A fixed frame 15 that can move left and right is provided on the light rod 14, and a left and right moving mechanism for driving the fixed frame 15 to move is provided between the brackets 5.

[0062] On both sides of the fixed frame 15 are provided an up-and-down moving mechanism, on which an adjustment seat 13 is provided for adjusting the height of the adjustment seat 13. A weeding motor 12 is mounted on the top middle portion of the adjustment seat 13 via a U-shaped frame, and the output shaft of the weeding motor 12 is connected to the weeding rotary blade 8.

[0063] In a further preferred but non-limiting embodiment of the present invention, the left and right movement mechanism includes: a rack 9, a gear 10 and a travel motor 11. A rack 9 is provided between the top ends of the bracket 5. The rack 9 is provided above the polished rod 14 and is parallel to the polished rod 14. The rack 9 and the gear 10 are meshed with each other. The transmission shaft of the gear 10 is connected to the output end of the travel motor 11. A motor seat 26 is provided at the bottom of the travel motor 11. The motor seat 26 is fixedly connected to the fixed frame 15. By starting the travel motor 11, the gear 10 rotates. Since the gear 10 and the rack 9 are meshed with each other, the gear 10 can move along the length direction of the rack 9. The gear 10 drives the travel motor 11, the motor seat 26 and the fixed frame 15 to move together, thereby achieving the effect of driving the left and right movement of the fixed frame 15 by the travel motor 11.

[0064] In a further preferred but non-limiting embodiment of the present invention, a sliding sleeve 18 is provided on the back of the fixing frame 15, which is slidably connected to the polished rod 14. The length of the fixing frame 15 is less than the distance between the two brackets 5. The provision of the two polished rods 14 provides auxiliary limiting support for the fixing frame 15, improving the stability of the structure. When the travel motor 11 is activated, the fixing frame 15 can move along the length of the polished rod 14.

[0065] In one embodiment, two polished rods 14 are provided between the brackets 5 and are arranged in parallel up and down. Four sliding sleeves 18 slidably connected to the polished rods 14 are provided on the back of the fixing frame 15 .

[0066] like Figure 5 、 Figure 6 As shown, the vertical movement mechanism includes an extension plate 16, an adjustment plate 17, a stud 20, and a nut. Extension plates 16 are provided on both sides of the fixed frame 15, extending outward along its length. Adjustment plates 17 are provided on the sides of a pair of extension plates 16, forming an L-shaped arrangement with the extension plates 16. The adjustment seat 13 is located between the two adjustment plates 17. The surfaces of the adjustment plates 17 are provided with oblong holes 19 in the vertical direction. Studs 20 are provided at both ends of the adjustment seat 13. The ends of the studs 20 extend through the oblong holes 19 and are threadedly connected to nuts. By loosening the nuts, the adjustment seat 13 can be adjusted up and down, thereby facilitating control of the distance between the bottom of the weeding blade 8 and the ground, and thus the height of the weeding, providing great flexibility.

[0067] Furthermore, the bracket 5 includes a horizontal plate and a vertical plate connected vertically, so that the bracket 5 has an L-shaped structure, and the end of the horizontal plate away from the vertical plate is connected to the top plate, and a rack 9 and a light rod 14 are provided between the two vertical plates.

[0068] The automatic weeding device also includes a motor drive module, a fuzzy controller and a power supply module; the power supply module is used to provide power to the power system; the motor drive module is electrically connected to the left motor and the right motor of the crawler walking mechanism through the fuzzy controller, and the obstacle avoidance control system drives the crawler walking mechanism through the fuzzy controller to achieve obstacle avoidance.

[0069] The sensor module includes: a front ultrasonic sensor arranged at the front end of the tool driving mechanism 4 and two side ultrasonic sensors arranged at both sides of the tool driving mechanism 4. The front ultrasonic sensor and the two side ultrasonic sensors are used to obtain the distance of the obstacle.

[0070] The front-end ultrasonic sensor obtains obstacle information in front of and on both sides of the weeding device, which is used to enable the weeding device to bypass and avoid obstacles. The substation equipment obstacle information obtained by the ultrasonic sensors on both sides, that is, the equipment obstacle information between the adjustment plates 17 on the movement trajectory of the weeding device, is used to avoid weeding by the weeding rotary blade 8, ensuring that the weeding rotary blade 8 moves safely in the equipment gap.

[0071] Further, combined with Figure 5As shown, the front ultrasonic sensor comprises ultrasonic sensor 1 21, ultrasonic sensor 2 22, and ultrasonic sensor 3 23, which are evenly spaced and arranged at the front end of the tool drive mechanism 4. The side ultrasonic sensors comprise ultrasonic sensor 4 24 and ultrasonic sensor 5 25, which are respectively arranged on the left and right sides of the tool drive mechanism 4.

[0072] Furthermore, the sensor module is internally equipped with a circuit module comprising an ultrasonic transmitter circuit, an echo signal processing circuit, and an ultrasonic transducer. The ultrasonic transmitter circuit is connected to each ultrasonic sensor, the echo signal processing circuit, the ultrasonic transducer, and the obstacle avoidance control system. The ultrasonic transmitter circuit outputs a pulse signal to the ultrasonic transducer, which uses the inverse piezoelectric effect to convert the signal into ultrasonic vibrations, propagating the ultrasound waves into the air. The echo signal processing circuit processes the echo signal received by the ultrasonic sensor so that it can be recognized and calculated by the processor, thereby obtaining accurate distance information.

[0073] The processing method of the echo signal processing circuit mainly includes the following steps:

[0074] 1. Amplification: Because ultrasound gradually attenuates during propagation, the amplitude of the echo signal needs to be amplified to a level that the processor can recognize. Amplification circuits are usually implemented using devices such as operational amplifiers.

[0075] 2. Filtering: The echo signal may contain various noise and interference, such as environmental noise, electromagnetic interference, etc. The filtering circuit is used to filter out these noise and interference to improve signal quality.

[0076] 3. Shaping: The echo signal is usually an analog signal and needs to be converted into a digital signal before it can be processed by the processor. The shaping circuit is used to convert the analog signal into a square wave signal to facilitate subsequent processing.

[0077] 4. Modulation and Demodulation: To improve ranging accuracy, ultrasonic sensors typically use modulation methods such as frequency modulated continuous wave (FMCW) or pulse echo (PE). The modulation and demodulation circuit is used to demodulate the modulated signal back to the original signal.

[0078] 5. Delay measurement: Ultrasonic sensors calculate distance by measuring the time difference between the transmitted signal and the echo signal. The delay measurement circuit is used to measure this time difference and convert it into a distance value.

[0079] 6. Temperature compensation: The propagation speed of ultrasound varies with temperature. The temperature compensation circuit is used to correct the ranging results according to the ambient temperature to improve ranging accuracy.

[0080] When ultrasonic waves propagate, they can be attenuated by environmental interference and other factors, necessitating signal processing through amplification and filtering. Mechanical vibrations from the tool drive mechanism can increase detection errors and reduce accuracy, so a bandpass filter is placed after the amplifier circuit to improve accuracy.

[0081] The weed recognition module 7 is mounted on the top plate 1 via the extension frame 6. The weed recognition module 7 is mainly used to recognize the image information of weeds around the fuselage and upload the image information to the obstacle avoidance control system for more accurate weeding operations.

[0082] The obstacle avoidance control system is connected to the sensor module and the weed recognition module 7; the fuzzy controller uses the distance between each ultrasonic sensor and the obstacle fed back by the sensor module to calculate the position of the obstacle, and sends instructions to the motor control module according to the position of the obstacle, driving the crawler walking mechanism to move around to avoid the weeding device, drive the weeding rotary blade to move left and right to avoid weeding, and move the weeding rotary blade 8 to the root of the grass to be removed to perform the weeding operation.

[0083] like Figure 7 As shown, the obstacle avoidance control system uses ultrasonic sensor ranging technology for both obstacle avoidance and weeding avoidance. The primary obstacle to be avoided is substation equipment, so the weeding device's housing is made of insulating material. To avoid weeding between devices, the weed recognition module 7 captures weed image information, which is fed back to the obstacle avoidance control system by ultrasonic sensors on both sides. The obstacle avoidance control system identifies the location and range of weeds based on the weed image information, plans the weeding device's movement path, and controls the movement of the weeding blade 8, achieving precise weeding. The obstacle avoidance control system uses obstacle information captured by the ultrasonic sensors on both sides to control the left and right movement of the weeding blade 8 to avoid the equipment. The weeding device avoids obstacles by using front-end ultrasonic sensors to obtain obstacle distances and feed them back to the obstacle avoidance control system. The obstacle avoidance control system then uses a fuzzy controller to issue commands to the left and right motors, controlling the crawler-type walking platform to avoid obstacles.

[0084] The ultrasonic sensor is placed on the tool drive mechanism 4 instead of the machine body so that the obstacle avoidance system can conveniently collect environmental information and make corresponding obstacle avoidance decisions when the weeding rotary cutter 8 is in different positions, and can also directly detect equipment obstacles.

[0085] In a preferred but non-limiting embodiment of the present invention, the ultrasonic sensor uses a model AJ-SR04M ultrasonic module, which has high stability and sensitivity performance and an integrated closed waterproof probe. The ultrasonic module can accurately detect the distance between a flat object and a flat object within a range of 20cm to 800cm, and can accurately detect pedestrians within a range of 20cm to 250cm. The divergence angle can be freely adjusted, making it suitable for various measurement situations in wet and harsh environments. It is an integrated transceiver ultrasonic sensor with simple operation and an integrated circuit, which is easy to integrate and achieve miniaturization. The motor drive module uses power MOS transistors to form an H-bridge to drive the motors because the left and right motors require high power to drive the crawler-type walking platform. By turning on a pair of MOS transistors at opposite corners of the H-bridge and turning off the other pair, the forward and reverse rotation of the motors is controlled. The power supply voltage of the power module is 12V, the power supply voltage of the DC motor is also 12V, and the power supply voltage of the STM32F103C8T6 processor chip is 3.3V, while the ultrasonic sensor requires a 5V power supply. Therefore, the obstacle avoidance control system requires a variety of different voltages for power supply. To save costs and simplify the circuit, a DC12V to 5V / 3.3V conversion module (HW-DYU02) is used to generate the required 5V and 3.3V voltages, respectively. The power conversion module uses two AMS1117 chips: one is AMS1117.5.0, which outputs 5V, and the other is AMS1117, which outputs 3.3V.

[0086] like Figure 8 As shown, embodiment 2 of the present invention provides an automatic weeding method in a substation, according to the above-mentioned automatic weeding device in a substation, comprising the following steps:

[0087] Step 1: Obtain obstacle distances and weed image information around the weed removal device;

[0088] Step 2: Calculate the location of the obstacle based on the obstacle distance information, and control the weeding device to avoid the obstacle and the weeding blade to avoid weeding;

[0089] Step 3: Identify the location and range of the weeds based on the weed image information, plan the movement path of the weeding device and the movement of the weeding blade 8, and perform the weeding operation.

[0090] Furthermore, the obstacle position includes the obstacle position in front and the obstacle position on both sides. The obstacle avoidance control system controls the weeding rotary blade to avoid weeding according to the obstacle position on both sides, and controls the weeding device to bypass the obstacle through the fuzzy controller according to the obstacle position in front.

[0091] Furthermore, the fuzzy controller is used to control the weeding device to avoid obstacles, including the following steps:

[0092] The obstacle distance measured by the ultrasonic sensor is divided into five distance fuzzy sets from near to far: C, N, M, Fa, VF;

[0093] The angle interval of the weeding device rotating left and right through the crawler walking mechanism is divided into three angle fuzzy sets from left to right: L, F, R;

[0094] According to the distance fuzzy set of the obstacle measured by the front-end ultrasonic sensor, the angle fuzzy set that the crawler walking mechanism needs to adjust is selected, the moving path of the crawler walking mechanism is automatically planned, and the weeding device executes the detour and obstacle avoidance.

[0095] The specific steps include:

[0096] The distances to obstacles detected by ultrasonic sensor 1 21 , ultrasonic sensor 2 22 , and ultrasonic sensor 3 23 are represented by LD, FD, and RD, respectively, and the distances to obstacles detected are divided into five distance fuzzy sets: C (very close), N (near), M (medium), Fa (far), and VF (very far).

[0097] In a preferred but non-limiting embodiment of the present invention, the parameter ranges represented by the five fuzzy sets are as follows:

[0098] C (very close): LD / FD / RD < 20 cm;

[0099] N (near): 20cm≤LD / FD / RD<50cm;

[0100] M (medium): 50cm≤LD / FD / RD<100cm

[0101] Fa (far): 100cm≤LD / FD / RD<200cm

[0102] VF (very far): 200cm≤LD / FD / RD.

[0103] Because the Gaussian membership function curve is relatively smooth, which improves the accuracy and simplicity of the system, the membership function of the fuzzy obstacle avoidance algorithm of the present invention adopts a Gaussian type. Because the body 2 relies on two left and right motors to control movement, in order to simplify the fuzzy control movement system, the track turning angle Angle caused by the different speeds of the two motors is used as the controller output. Moving parallel to the direction of the device is a zero-degree turning angle, moving to the left of the vertical direction is a negative angle, and a positive turning angle is the opposite. The domain is [-30°, 30°], which is divided into three angle fuzzy sets: L (left), F (front), and R (right).

[0104] The L (left) turning angle range can be set from -30° to -10°.

[0105] The turning angle range of F(Front) can be set from -10° to 10°.

[0106] The turning angle range of R (right) can be set from 10° to 30°.

[0107] According to the distance fuzzy set of obstacles measured by three ultrasonic sensors, the fuzzy set of angles that the crawler mechanism needs to adjust is selected, and the moving path of the crawler mechanism is automatically planned.

[0108] The crawler-type walking mechanism moves along the moving path, and the weeding device is executed to bypass obstacles.

[0109] like Figure 8 As shown, when the ultrasonic sensor 1 21 at the front end of the adjustment seat 13 measures the obstacle distance as C, and the ultrasonic sensor 2 22 and the ultrasonic sensor 3 23 measure the obstacle distance as F and VF, that is, the obstacle is very close to the left side of the machine, the control system controls the crawler walking platform to deflect to the right, and the deflection angle is R.

[0110] When the ultrasonic sensor 1 21 at the front end of the adjustment seat 13 measures the obstacle distance as N, and the ultrasonic sensor 2 22 and the ultrasonic sensor 3 23 measure the obstacle distance as M, F, and VF, that is, the obstacle is close to the left side of the machine, the control system controls the crawler walking platform to deflect to the right, and the deflection angle is R.

[0111] When the ultrasonic sensor 1 21 at the front end of the adjustment seat 13 measures the obstacle distance as M, and the ultrasonic sensor 2 22 and the ultrasonic sensor 3 23 measure the obstacle distance as N, that is, the obstacle is close to the right side of the machine, the obstacle avoidance control system controls the crawler walking platform to deflect to the left, and the deflection angle is L; when the ultrasonic sensor 2 22 and the ultrasonic sensor 3 23 measure the obstacle distance as M, at this time the obstacle distance is located in the middle of the machine, the obstacle avoidance control system controls the crawler walking platform to move forward, and the deflection angle is F; when the ultrasonic sensor 2 22 and the ultrasonic sensor 3 23 measure the obstacle distance as F, at this time the obstacle distance is located on the left side of the machine, the obstacle avoidance control system controls the crawler walking platform to deflect to the right, and the deflection angle is R.

[0112] When the ultrasonic sensor 1 21 at the front end of the adjustment seat 13 measures the obstacle distance as F, and the ultrasonic sensor 2 22 and the ultrasonic sensor 3 23 measure the obstacle distance as C, N, and M, that is, the obstacle is close to the right side of the machine, the obstacle avoidance control system controls the crawler walking platform to deflect to the left, and the deflection angle is L.

[0113] When the ultrasonic sensor 1 21 at the front end of the adjustment seat 13 measures the obstacle distance as VF, and the ultrasonic sensor 2 22 and the ultrasonic sensor 3 23 measure the obstacle distance as C and N, that is, the obstacle is close to the right side of the machine, the obstacle avoidance control system controls the crawler walking platform to deflect to the left, and the deflection angle is L.

[0114] In a further preferred but non-limiting embodiment of the present invention, the weed recognition module 7 is used to obtain weed image information around the automatic weed removal device, including the following steps:

[0115] The weed recognition module 7 is arranged on the top of the cutter drive mechanism 4, and the weed recognition module 7 is used to obtain the image information of the weeds around the fuselage, and the image information is uploaded to the obstacle avoidance control system;

[0116] The obstacle avoidance control system uses the YOLO (You Only Look Once) algorithm for weed detection. The YOLO algorithm is a deep learning-based object detection algorithm with fast detection speed and high accuracy, and can effectively identify weed areas in images.

[0117] The method of obtaining image information of weeds around the weed removal device and identifying the location and range of the weeds includes the following steps:

[0118] The obstacle avoidance control system pre-processes the collected weed images;

[0119] The obstacle avoidance control system uses the YOLO algorithm to identify weeds in the preprocessed image and determine the location and range of the weeds.

[0120] The method for weed identification using the YOLO algorithm in the present invention specifically includes the following steps:

[0121] Data preprocessing: Preprocess the collected image data, including image enhancement, noise filtering, etc., to improve image quality.

[0122] Model training: Use a labeled weed image dataset to train the YOLO algorithm. During the training process, network parameters need to be adjusted to improve detection accuracy.

[0123] Model testing: Use the test set to test the trained YOLO algorithm to evaluate the detection accuracy and speed of the algorithm.

[0124] Model deployment: The trained YOLO algorithm is deployed on the automatic weed removal device in the substation to achieve real-time weed identification.

[0125] In order to further improve the accuracy of weed recognition, the present invention makes the following improvements to the YOLO algorithm:

[0126] Data augmentation: Perform data augmentation on the training dataset, such as random cropping, rotation, scaling, etc., to increase the generalization ability of the model.

[0127] Feature fusion: Different features are integrated into the YOLO algorithm, such as color features, texture features, etc., to improve the accuracy of target detection.

[0128] Model optimization: Optimize the network structure of the YOLO algorithm, such as using a deeper network, more advanced activation functions, etc., to improve the performance of the model.

[0129] Weed Identification and Control Process:

[0130] Image acquisition: The weed recognition module 7 continuously collects image information of the substation's surrounding environment and transmits the image data to the obstacle avoidance control system.

[0131] Image preprocessing: The obstacle avoidance control system preprocesses the collected images, including:

[0132] Image enhancement: By adjusting image parameters such as brightness and contrast, the image quality is improved to make the weed features more prominent.

[0133] Noise filtering: Use filtering algorithms to remove noise from images and improve image clarity.

[0134] Weed recognition: The obstacle avoidance control system uses the YOLO algorithm to identify weeds in pre-processed images. The recognition process includes:

[0135] Feature extraction: The YOLO algorithm uses a convolutional neural network (CNN) to extract features from images, such as color, texture, shape, etc.

[0136] Object detection: The YOLO algorithm divides the image into multiple grids, each of which is responsible for predicting the target category and location within the grid, and outputs the target category and location information.

[0137] Weed positioning: The obstacle avoidance control system determines the location and range of weeds based on the target location information output by the YOLO algorithm.

[0138] In a preferred but non-limiting embodiment of the present invention, the crawler-type walking mechanism is controlled to move according to the acquired weed image information and obstacle position information, and the weed removal device is executed to bypass the obstacle;

[0139] When the automatic weeding device is close to the weeding location, the weeding rotary blade 8 is controlled to move so that the weeding rotary blade 8 is close to the root of the weeds to be weeded, and the weeding operation is performed, including the following steps:

[0140] The obstacle avoidance control system plans the weed removal device's travel path based on the location and range of weeds to ensure that all weed areas are covered. The obstacle avoidance control system controls the weed removal device's travel motor to avoid obstacles and move to the location where weeds need to be removed based on the planned path.

[0141] When the automatic weeding device moves to the position where weeding is required, the weeding motor is controlled to drive the weeding rotary blade 8 to accurately approach the roots of the weeds to be weeded to perform the weeding operation.

[0142] Furthermore, the automatic weeding method further includes: adjusting the distance between the weeding rotary blade 8 and the ground by moving the adjustment base 13 up and down.

[0143] like Figure 9 As shown, embodiment 3 of the present invention provides an obstacle avoidance control system for executing the above-mentioned automatic weeding method in a substation, including:

[0144] An information acquisition module is used to obtain obstacle distances and weed image information around the weed removal device;

[0145] The obstacle avoidance and avoidance control module is used to calculate the location of obstacles based on obstacle distance information, and control the weeding device to avoid obstacles and the weeding rotary blade to avoid weeding;

[0146] The path planning and rotary blade control module is used to identify the location and range of weeds based on weed image information, plan the movement path of the weeding device and the movement of the weeding rotary blade, and perform weeding operations.

[0147] The obstacle avoidance control system obtains environmental information in real time by using sensor modules and cameras. The obstacle avoidance control system controls the left motor and the right motor according to the obtained environmental information, thereby controlling the two tracks to move, so that the weeding device can bypass obstacles and send commands to the walking motor 11 to control the weeding blade 8 to avoid obstacles left and right, so as to effectively avoid the equipment and enable the weeding device to better weed near the roots; by loosening the nut, the adjustment seat can be adjusted up and down to facilitate the control of the distance between the bottom of the weeding blade 8 and the ground, thereby facilitating the control of the weeding height and better achieving the weeding effect.

[0148] The present invention utilizes ultrasonic sensors and cameras to enable the obstacle avoidance control system to obtain environmental information in real time and then quickly issue commands to the walking motor that controls the left and right movement of the swing structure, so as to effectively avoid the equipment and enable the weeding device to better weed near the roots. The weeding device's detour and obstacle avoidance is suitable for substations with a large number of equipment. The obstacle avoidance principle is that the three ultrasonic sensors at the front end of the adjustment seat feed back the acquired environmental signals to the single-chip computer obstacle avoidance control system. The obstacle avoidance control system determines the direction of the obstacle, and then uses the fuzzy controller to perform obstacle avoidance processing and adjust the heading of the walking platform to avoid the obstacle, thereby realizing automated weeding, greatly saving labor, reducing the labor intensity of workers, and improving work efficiency.

[0149] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. An automatic weeding device in a substation, comprising: The machine body (2), the tool drive mechanism (4), the sensor module, the weed recognition module (7) and the obstacle avoidance control system are characterized by: A crawler-type walking mechanism is provided at the bottom of the fuselage (2), and a top plate (1) is provided at the top of the fuselage (2); The tool drive mechanism (4) is arranged on the front side of the top plate (1), and a weeding rotary knife (8) is arranged below the tool drive mechanism (4) to drive the weeding rotary knife (8) to move; The sensor module is arranged on the tool driving mechanism (4) and is used to obtain the distance of the obstacle; A weed recognition module (7) is provided on the cutter drive mechanism (4) and is used to capture images of weeds; The obstacle avoidance control system calculates the obstacle position based on the obstacle distance to achieve the weeding device to avoid the obstacle and the weeding rotary blade (8) to avoid weeding, and plans the movement path of the weeding device and the movement of the weeding rotary blade (8) based on the weed image information to achieve accurate weeding.

2. The automatic weeding device in a substation according to claim 1, characterized in that: The sensor module comprises a front ultrasonic sensor arranged at the front end of the tool drive mechanism (4) and two side ultrasonic sensors arranged at both sides of the tool drive mechanism (4).

3. The automatic weeding device in a substation according to any one of claims 1-2, characterized in that: The tool driving mechanism (4) comprises: a bracket (5), a polished rod (14), a fixing frame (15) and an adjusting seat (13); The brackets (5) are arranged on both sides of the top plate (1); The polished rod (14) is arranged between the two brackets (5), and at least two polished rods (14) are arranged in parallel up and down between the brackets (5). A fixed frame (15) that can move left and right is provided on the polished rod (14), and a left and right moving mechanism for driving the fixed frame (15) to move is provided between the brackets (5); Up and down moving mechanisms are arranged on both sides of the fixed frame (15), and an adjusting seat (13) is provided on the up and down moving mechanism for adjusting the height of the adjusting seat (13) up and down; a weeding motor (12) is installed at the middle part of the top of the adjusting seat (13) through a U-shaped frame, and the output shaft of the weeding motor (12) is connected to the weeding rotary knife (8).

4. The automatic weeding device in a substation according to claim 3, characterized in that: The left-right moving mechanism comprises: a rack (9), a gear (10) and a travel motor (11); A rack (9) is provided between the top ends of the brackets (5), the rack (9) is provided above the polished rod (14) and is provided in parallel with the polished rod (14), the rack (9) and the gear (10) are meshed with each other, the transmission shaft of the gear (10) is connected to the output end of the travel motor (11), a motor seat (26) is provided at the bottom of the travel motor (11), and the motor seat (26) is fixedly connected to the fixing frame (15).

5. The automatic weeding device in a substation according to claim 3, characterized in that: The up-and-down moving mechanism comprises: an extension plate (16), an adjustment plate (17), a stud (20) and a nut; Extension plates (16) extending outward along the length direction of the fixing frame (15) are provided on both sides of the fixing frame (15), and adjustment plates (17) distributed in an L shape with the extension plates (16) are provided on the sides of the pair of extension plates (16). The adjustment seat (13) is located between the two adjustment plates (17), and the surface of the adjustment plate (17) is provided with an oblong hole (19) in the vertical direction. Both ends of the adjustment seat (13) are provided with studs (20), and the ends of the studs (20) pass through the oblong holes (19) and are threadedly connected with nuts.

6. The automatic weeding device in a substation according to claim 1, characterized in that: The weeding device further includes a fuzzy controller electrically connected to the motor of the crawler-type walking mechanism; The obstacle avoidance control system drives the crawler-type walking mechanism through a fuzzy controller to avoid obstacles.

7. An automatic weeding method in a substation, according to the automatic weeding device in a substation according to any one of claims 1 to 6, characterized in that: The following steps are involved: Obtaining obstacle distance and weed image information around the weed removal device; According to the obstacle distance information, the obstacle position is calculated, and the weeding device is controlled to avoid the obstacle and the weeding rotary knife to avoid weeding; Based on the weed image information, the location and range of the weeds are identified, the movement path of the weeding device and the movement of the weeding rotary blade are planned, and the weeding operation is performed.

8. The automatic weeding method in a substation according to claim 7, characterized in that: The obstacle positions include the front obstacle position and the two side obstacle positions. The obstacle avoidance control system controls the weeding rotary blade to avoid weeding according to the two side obstacle positions, and controls the weeding device to bypass the obstacle through the fuzzy controller according to the front obstacle position.

9. The automatic weeding method in a substation according to claim 8, characterized in that: The method of controlling the weeding device to avoid obstacles by using a fuzzy controller includes the following steps: The obstacle distance measured by the ultrasonic sensor is divided into five distance fuzzy sets from near to far: C, N, M, Fa, VF; The angle interval of the weeding device rotating left and right through the crawler walking mechanism is divided into three angle fuzzy sets from left to right: L, F, R; According to the distance fuzzy set of the obstacle measured by the front-end ultrasonic sensor, the angle fuzzy set that the crawler walking mechanism needs to adjust is selected, the moving path of the crawler walking mechanism is automatically planned, and the weeding device executes the detour and obstacle avoidance.

10. The automatic weeding method in a substation according to claim 7, characterized in that: The method of obtaining image information of weeds around the weed removal device and identifying the location and range of the weeds includes the following steps: The obstacle avoidance control system pre-processes the collected weed images; The obstacle avoidance control system uses the YOLO algorithm to identify weeds in the preprocessed image and determine the location and range of the weeds.

11. An obstacle avoidance control system for executing the automatic weeding method in a substation according to any one of claims 7 to 10, characterized in that: include: An information acquisition module is used to obtain obstacle distances and weed image information around the weed removal device; The obstacle avoidance and avoidance control module is used to calculate the location of obstacles based on obstacle distance information, and control the weeding device to avoid obstacles and the weeding rotary blade to avoid weeding; The path planning and rotary blade control module is used to identify the location and range of weeds based on weed image information, plan the movement path of the weeding device and the movement of the weeding rotary blade, and perform weeding operations.

Images