A mower airflow guidance design method and system suitable for slope operations

Abstract

The application relates to a mower airflow guide design method and system suitable for slope operation, and relates to the technical field of mowers, which comprises the following steps: collecting terrain information, image detection information and sound detection information, and preprocessing the image detection information to form detection image information; performing semantic segmentation on the detection image information and a preset detection category to form segmentation image information; obtaining a cutting object type and a cutting density according to the segmentation image information; performing time domain sampling, analog-digital conversion and fast Fourier transform on the sound detection information to obtain a sound spectrum graph; extracting detection spectrum features from the sound spectrum graph, updating the cutting object type according to the detection spectrum features and obtaining cutting object humidity; obtaining target airflow guide parameters according to the cutting object type, the cutting density, the cutting object humidity and the terrain information; and generating final control instructions through the target airflow guide parameters and uploading the final control instructions. The application has the effect of improving the operation capacity of the mower.

Description

Technical Field

[0001] This invention relates to the technical field of lawnmowers, and in particular to a design method and system for airflow guidance of lawnmowers suitable for slope operations. Background Technology

[0002] A lawnmower is a mechanized device used to trim grasses in lawns, gardens, and agricultural areas.

[0003] During the mowing process, the high-speed rotation of the mower blades creates a specific airflow field inside the cutter head. This airflow field has a significant impact on the mower's cutting effect, grass collection efficiency, energy consumption level, and operating noise. The airflow guidance function of the mower is usually achieved by the fixed structure of the blade itself. For example, an air lifting component can be set on the blade, the trailing edge of the blade can be processed into a wavy structure to improve the grass clipping discharge performance, or a combination structure of a flow-around blade and a flat blade can be set to reduce wind resistance, reduce noise, and achieve energy saving.

[0004] During the mowing process, the airflow guidance structure of lawnmowers is mostly a fixed design. When facing different terrains, the airflow guidance remains unchanged, which causes the grass clippings to easily accumulate on one side of the machine and cannot be effectively collected, thus reducing the working capacity of the lawnmower. Summary of the Invention

[0005] To improve the operational capabilities of lawnmowers, this invention provides a design method and system for airflow guidance in lawnmowers suitable for slope operations.

[0006] In a first aspect, the present invention provides an airflow guidance design method for lawnmowers suitable for slope operations, employing the following technical solution: A method for designing airflow guidance for lawnmowers suitable for slope operations, comprising: The system collects terrain information, image detection information, and sound detection information of the area where the lawnmower is located, and preprocesses the image detection information to form detection image information. Semantic segmentation is performed based on the detected image information and the preset detection categories to form segmented image information; The type and density of the object to be cut are obtained from the segmented image information; The sound spectrum is obtained by performing time-domain sampling, analog-to-digital conversion, and fast Fourier transform on the sound detection information; Extract detection spectral features from the sound spectrogram, update the cutting type and obtain the cutting humidity based on the detection spectral features; The target airflow guidance parameters are obtained based on the type of material being cut, cutting density, material humidity, and terrain information. The final control command is generated based on the target airflow guidance parameters and then uploaded.

[0007] By adopting the above technical solution, the lawnmower analyzes terrain information, image detection information, and sound detection information to generate final control commands and upload them to the lawnmower's actuator. This enables the lawnmower to adjust the blade height, blade angle, and airflow channel shape in real time based on slope and aspect information when operating on slopes. This compensates for the adverse effects of slope gravity on grass clipping transport, improves grass clipping collection on slopes, and allows the lawnmower to adaptively match airflow guidance with the physical properties of the cut grass when encountering different grass species, densities, or moisture levels. This changes the current situation where the airflow guidance structure of existing lawnmowers is fixed and cannot be adjusted according to real-time working conditions. It enables the lawnmower to operate with optimal airflow parameters in complex terrain and diverse grass conditions, improving the lawnmower's environmental adaptability and comprehensive operating capabilities.

[0008] Optionally, methods for obtaining the target airflow steering parameters include: Retrieve slope detection information from terrain information; The airflow adhesion effect coefficient is calculated based on the slope detection information and the preset unit blade height. The critical airflow velocity value is calculated by considering the type of material being cut, the cutting density, and the humidity of the material being cut. A comprehensive cost function is constructed by combining the critical airflow velocity value, the airflow wall effect coefficient, the preset benchmark target, and the preset unit control coefficient. The final control coefficient is calculated using a comprehensive cost function and a preset grass collection target, and is used as the target airflow guidance parameter.

[0009] Optional, also includes: Retrieve the entanglement type from the detection category; Extract the marker winding parameters of the winding type and their corresponding marker winding length from the segmented image information; The baseline length threshold and baseline probability threshold are obtained based on the final control command; When the marker wrapping length exceeds the baseline length threshold, the marker wrapping parameter is defined as an abnormal wrapping parameter, and the abnormal wrapping position of the abnormal wrapping parameter is extracted from the segmented image information; Collect data on the center position of the lawnmower blades, detect current data, and detect rotation speed data; By combining the abnormal winding location, blade center position, detection current data, detection speed data, and reference probability threshold, a clearing airflow control command is generated to update the final control command.

[0010] Optionally, methods for generating clear airflow control commands include: The abnormal distance and abnormal azimuth angle are calculated by using the blade center position and the abnormal winding position; The entanglement probability value is obtained based on the detected current data and detected rotation speed data; When the entanglement probability value exceeds the baseline probability threshold, the entanglement position and number of entanglements are calculated by segmenting the image information and abnormal entanglement parameters. The winding attitude information is obtained by combining the winding position, number of winding turns, abnormal distance value, and abnormal azimuth angle. A clearing airflow control command is generated based on the entanglement posture information.

[0011] Optionally, the method for generating clear airflow control commands also includes: Extract the effective impact cross-sectional area value of the cut object from the segmented image information; The reference frequency threshold is obtained by taking the effective impact cross-sectional area value; The baseline impact threshold and target duration are obtained based on the entanglement attitude information; Retrieve the high-frequency energy percentage corresponding to the reference frequency threshold and the rising slope of the time-domain waveform envelope from the sound spectrum diagram, and match the average quality estimate and the average velocity estimate. The impact kinetic energy is calculated by combining the effective impact cross-sectional area, the estimated average mass, and the estimated average velocity. The airflow impact energy threshold is determined based on the baseline impact threshold and the impact kinetic energy value. The target velocity value is obtained based on the airflow impact energy threshold. The target duration value and the target speed value are combined to generate a clear airflow control command.

[0012] Optionally, the method for generating clear airflow control commands also includes: The target impact coordinates are obtained based on the location of the abnormal entanglement. The deflection angle value is calculated based on the target impact coordinates and the preset initial guide angle; A pulse drive signal is generated by the target duration value and the target velocity value; The actual current data is acquired based on the pulse drive signal, and the reference current threshold is obtained through the winding posture information. When the actual current data is less than the reference current threshold for three consecutive preset sampling periods, a stop signal is output. The deflection angle value, pulse drive signal, actual current data, reference current threshold, and stop signal are used as the clear airflow control commands.

[0013] Optional, also includes: Collect images of grass clippings; Extract the centroid coordinates of grass clippings from the grass clipping detection image; The average projectile velocity and average projectile angle are calculated based on the centroid coordinates and the preset acquisition interval. The dispersion value is obtained based on the centroid coordinates and the grass clipping detection image; The grass collection efficiency index is obtained by weighting and summing the average projectile velocity, average projectile angle, and dispersion. The index deviation value is calculated by comparing the grass collection efficiency index value with the preset index threshold, and the final control command is updated based on the index deviation value.

[0014] Optional, also includes: Collect vibration detection information of the blades during the lawnmower cutting process; Based on vibration detection information, time-domain feature vectors and frequency-domain feature vectors are generated, and the time-domain feature vectors and frequency-domain feature vectors are combined as a comprehensive feature vector; The wear value is detected based on the comprehensive feature vector matching; Airflow deviation parameters are generated based on the detected wear values ​​and the final control commands; The flow correction compensation command is matched with the airflow deviation parameter, and when the detected wear value is greater than the preset reference wear threshold, the flow correction compensation command is output to update the final control command.

[0015] Optionally, methods for updating the final control command also include: Extract the labeled vibration amplitude of the frequency corresponding to the detected rotational speed data from the vibration detection information; The marking wave quality of the blade can be obtained based on the marking vibration amplitude value; The detection phase difference is obtained by comparing the vibration detection information with the preset reference vibration signal, and the detection phase difference corresponding to the marked vibration amplitude is used as the marked phase difference. The marking angle position of the blade is obtained based on the phase difference value of the marking; The counterweight movement parameters are obtained by combining the marked fluctuation mass, the marked angle position, and the preset counterweight specifications, and then the counterweight movement parameters are added to the final control command.

[0016] Secondly, this application provides an airflow guidance design system for lawnmowers suitable for slope operations, employing the following technical solution: An airflow guidance design system for lawnmowers suitable for slope operations includes: The acquisition module is used to acquire terrain information, image detection information, and sound detection information. A memory for storing a program for a lawnmower airflow guidance design method suitable for slope operations; The processor is used to load and execute programs stored in memory.

[0017] In summary, this application includes at least one of the following beneficial technical effects: 1. By analyzing terrain information, image detection information, and sound detection information to generate final control commands and upload them to the lawnmower's actuator, the adverse effects of slope gravity on grass clipping transport can be compensated, the grass clipping collection effect on the slope can be improved, and the lawnmower's environmental adaptability and comprehensive operation capability can be enhanced. 2. By generating a clearing airflow control command to update the final control command, it is possible to detect grass stems or foreign objects with entanglement risk in advance through image recognition, and then confirm the actual entanglement status through load fluctuation analysis of current data and speed data. After confirming entanglement, it actively generates a clearing airflow control command to improve the accuracy and timeliness of anti-entanglement control. 3. By combining the target duration value and the target speed value to generate a clearing airflow control command, the grass clippings generated during the cutting process can be used as an impact medium. This ensures that the generated airflow speed provides sufficient impact energy to clear the entangled material while avoiding excessive energy consumption. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating a method for designing airflow guidance for lawnmowers suitable for slope operations; Figure 2 This is a schematic diagram of the overall structure of a lawnmower suitable for slope operation; Figure 3 This is a bottom view of the lawnmower blade.

[0019] The parts referred to by the numbers in the above attached diagrams are as follows: 1. Lawn mower; 2. Blade disc; 3. Blade; 4. Grass collection port; 5. Deflector; 6. Counterweight. Detailed Implementation

[0020] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0021] Reference Figure 1 , Figure 2 as well as Figure 3 This application discloses an airflow guidance design method for lawnmowers suitable for slope operations, comprising the following steps: S10: Collect terrain information, image detection information, and sound detection information of the area where the lawnmower 1 is located, and preprocess the image detection information to form detection image information.

[0022] Terrain information refers to information such as the slope value and slope direction of the ground where the lawnmower 1 is located. The slope value and slope direction are collected by the tilt sensor preset on the lawnmower 1 as terrain information.

[0023] Image detection information refers to the image of the area to be cut in front of the lawnmower 1. The image detection information is obtained by capturing ground images in the direction of the lawnmower 1's movement using a camera pre-installed on the lawnmower 1.

[0024] Sound detection information refers to the sound signal generated by the interaction between the blade 3 and the cutting material during the cutting process of the lawnmower 1. The sound signal of the lawnmower 1 during the cutting process is collected by a microphone pre-installed on the lawnmower 1 as sound detection information.

[0025] Detection image information refers to the processed image detection information. The image detection information is formed by sequentially performing image denoising, contrast enhancement, and color space conversion to filter out image noise and highlight the visual features of the cut object.

[0026] S11: Perform semantic segmentation based on the detected image information and the preset detection categories to form segmented image information.

[0027] The detection categories are the types of targets that technicians need to identify from the images, including lawn categories, mowed area categories, unmowed area categories, flower categories, root and stem categories, grass stem categories, and foreign object categories.

[0028] Segmented image information refers to the images corresponding to different detection categories. By classifying each pixel in the detection image information into one of the detection categories, pixels of the same category form a continuous region, and pixel regions of different categories are distinguished by different labels to form segmented image information.

[0029] S12: Obtain the cutting type and cutting density based on the segmented image information.

[0030] The cut object type refers to the actual type of object corresponding to the category of the uncut region in the segmented image information, including grass type, flower type, and root type. This is determined by counting the total number of pixels for each category in the segmented image information, and then counting the first pixel count of the flower category and the second pixel count of the root type within the uncut region. If the ratio of the first pixel count to the total number of pixels in the uncut region is greater than a preset first percentage threshold, the cut object type is determined to be a flower type. If the ratio of the second pixel count to the total number of pixels in the uncut region is greater than a preset second percentage threshold, the cut object type is determined to be a root type. If both the ratios of the first and second pixel counts to the total number of pixels in the uncut region are less than or equal to their respective percentage thresholds, the cut object type is determined to be a grass type.

[0031] The first and second percentage thresholds are thresholds set by technical personnel to determine the percentage of different types.

[0032] Cutting density refers to the number of objects to be cut per unit area. It is calculated by dividing the number of connected components corresponding to the grass stem category within the uncut region category in the segmented image information by the area of ​​the uncut region category.

[0033] S13: Perform time-domain sampling, analog-to-digital conversion, and fast Fourier transform on the sound detection information to obtain the sound spectrum.

[0034] A sound spectrum diagram is a graph that converts sound detection information into a frequency domain representation. It is obtained by sampling the sound detection information in the time domain at a preset sampling frequency to obtain a discrete time domain signal. The amplitude of the discrete time domain signal is converted from an analog quantity to a digital quantity to obtain a digital time domain signal. Then, a fast Fourier transform is performed on the digital time domain signal to obtain a sound spectrum diagram with frequency as the horizontal axis and amplitude as the vertical axis.

[0035] The sampling frequency is set by the technicians to sample the sound data.

[0036] S14: Extract the detection spectrum features from the sound spectrum map, update the cutting type with the detection spectrum features and obtain the cutting humidity.

[0037] The detected spectral features refer to the frequency domain parameters in the sound spectrum that reflect the physical properties of the object being cut, including peak frequency, harmonic structure, and low-frequency energy proportion. Peak frequency refers to the frequency value corresponding to the frequency component with the largest amplitude in the sound spectrum. Harmonic distortion rate is the sum of the ratios of the second harmonic amplitude, the third harmonic amplitude, and the fundamental frequency amplitude in the sound spectrum. Low-frequency energy proportion refers to the ratio of the total energy in the sound spectrum at frequencies below a first preset frequency (e.g., 500Hz) to the total energy across the entire frequency range. Peak frequency, harmonic structure, and low-frequency energy proportion are extracted from the sound spectrum as the detected spectral features.

[0038] By calculating the similarity between the detected spectral features and the features of each sample in the preset sound feature library, the grass species type corresponding to the sample feature with the highest similarity is selected as the updated cutting type.

[0039] The sound feature library is a database of sound samples of different cutting materials when cut by blade 3, pre-set and stored by technicians. The library collects sound signals from grass samples of known grass species under standard cutting conditions, performs time-domain sampling, analog-to-digital conversion, and fast Fourier transform to obtain a sample spectrum. From the spectrum, the peak frequency, harmonic distortion rate, and low-frequency energy proportion are extracted as sample feature vectors. These feature vectors are then associated with and stored with the corresponding grass species.

[0040] The humidity of the cut object refers to its humidity level. It is determined by extracting the humidity-sensitive frequency band energy ratio from the detection spectrum characteristics, and then querying the preset humidity mapping table based on the humidity-sensitive frequency band energy ratio to obtain the humidity level of the cut object.

[0041] The humidity mapping table stores humidity levels corresponding to different humidity-sensitive frequency band energy ratios. The higher the humidity-sensitive frequency band energy ratio, the higher the humidity level. The humidity mapping table is formed by collecting cutting sound signals of the same grass seed at different humidity levels, extracting the ratio of the energy value of the second preset frequency band (e.g., 1000Hz to 2000Hz) to the energy value of the third preset frequency band (e.g., 200Hz to 500Hz) in the sound spectrum of each humidity level as the humidity-sensitive energy ratio, and finally associating and storing the humidity-sensitive energy ratio with the corresponding humidity level.

[0042] S15: Obtain the target airflow guidance parameters based on the type of material being cut, cutting density, material humidity, and terrain information.

[0043] The target airflow guidance parameters refer to the blade height, blade angle, and shape and size parameters of the airflow channel set to adapt to the current working conditions. These parameters are obtained by analyzing the type of material being cut, cutting density, material humidity, and terrain information. The specific analysis method is described in S20 to S24.

[0044] S16: Generate the final control command based on the target airflow guidance parameters and upload it.

[0045] Reference Figure 2 and Figure 3 The lawnmower 1 is equipped with a guide plate 5 located inside the cutter head 2 of the lawnmower 1. The shape of the airflow channel inside the cutter head 2 can be changed by changing the angle of the guide plate 5.

[0046] The final control command refers to the set of control signals used to drive the actuator of the lawnmower 1. By converting the height value, angle value and target guide channel shape value of the blade 3 into drive signals of the blade 3 and the guide plate 5, the final drive signals are combined to form the final control command, and the final control command is uploaded to the control system of the lawnmower 1.

[0047] Methods for obtaining target airflow steering parameters include: S20: Retrieve slope detection information from terrain information.

[0048] Slope detection information refers to the slope value and slope direction in terrain information. Slope detection information is obtained by extracting the slope value and slope direction from the terrain information.

[0049] S21: Calculate the airflow adhesion effect coefficient based on slope detection information and preset unit blade height.

[0050] The unit blade height is the baseline value set by the technicians for the height of blade 3 above the ground, used to quantify the degree of influence of changes in blade 3 height on the airflow adhesion effect.

[0051] The airflow adhesion coefficient refers to the coefficient of ability of airflow to adhere and flow between the surface of the cutter head 2 and the ground. It is used to characterize the ability of the airflow generated by the blade 3 under inclined conditions to effectively lift the cut material and guide it towards the grass collection port 4. The basic adhesion coefficient is obtained by querying a preset adhesion effect mapping table using the slope value from the slope detection information. Then, the airflow adhesion coefficient is calculated based on the ratio of the current blade height to the unit blade height: Airflow adhesion coefficient = Basic adhesion coefficient × (Unit blade height / Current blade height). α α is a correction index set in advance by the technicians. The value of α ranges from 0.3 to 0.7. The value of α is obtained by the technicians through experiments based on the structure of the blade 2 of the lawnmower 1.

[0052] The base wall effect mapping table stores the base wall effect coefficients corresponding to different slope values. The larger the slope value, the smaller the base wall effect coefficient.

[0053] S22: Calculate the critical airflow velocity value based on the type of material being cut, the cutting density, and the humidity of the material being cut.

[0054] The critical airflow velocity value refers to the minimum airflow velocity value required for the cut material to be lifted from the ground and smoothly delivered into the grass collection channel. The average single plant weight value of the cut material is obtained by querying the preset cut material mapping table by the cut material type, and the density correction coefficient of the cut material is obtained by querying the preset cut material mapping table according to the cut material humidity.

[0055] By selecting grass samples of different grass species (including but not limited to Zoysia japonica, Bermuda grass, ryegrass, and Kentucky bluegrass), the average weight of a single grass plant was weighed using a precision electronic balance under standard laboratory conditions. The grass species type was then associated with and stored in relation to the corresponding average weight per plant. Next, samples of the same grass species were selected at different humidity levels (humidity levels ranging from 10% to 90%, with each 10% level representing a separate level). Their weight was measured, and the ratio to the standard weight (weight at 20% humidity) was calculated. This ratio served as the density correction coefficient for each humidity level. The humidity level and density correction coefficient were then associated and stored. Finally, the two associated data sets were integrated to form a cutting material mapping table.

[0056] The product of the average single plant weight and the density correction coefficient is used as the new single plant weight. Then, the total weight of the cut material per unit area is calculated based on the cutting density. Based on the new single plant weight and the total weight of the cut material per unit area, the critical airflow velocity is calculated using a preset airflow velocity calculation formula.

[0057] The formula for calculating airflow velocity is: Critical airflow velocity value = Preset first velocity coefficient × Square root of corrected single plant weight value + Preset second velocity coefficient × Square root of total weight of cut material per unit area.

[0058] The first and second speed coefficients are coefficient values ​​set by technicians to adapt to the current airflow speed.

[0059] S23: Construct a comprehensive cost function by combining the critical airflow velocity value, the airflow wall effect coefficient, the preset benchmark target, and the preset unit control coefficient.

[0060] The baseline targets are the target performance indicators of the lawnmower 1 under standard operating conditions set by the technicians, including standard grass collection efficiency, standard energy consumption and standard noise.

[0061] The unit control coefficients are the unit height adjustment of blade 3, the unit angle adjustment of blade 3, and the unit angle control of guide plate 5 set by the technician.

[0062] The comprehensive cost function is a mathematical function used to evaluate the comprehensive performance of different combinations of airflow guidance parameters (height of blade 3, angle of blade 3, and angle of deflector 5). The smaller the function value, the better the comprehensive performance.

[0063] By using the critical airflow velocity and the airflow adhesion coefficient as constraints, and the aforementioned combination of airflow steering parameters as independent variables, weed collection efficiency sub-functions, energy consumption sub-functions, and noise sub-functions are established respectively. The comprehensive cost function is obtained by multiplying the weed collection efficiency sub-function, energy consumption sub-function, and noise sub-function by preset weighting coefficients for weed collection efficiency, energy consumption, and noise, respectively, and then summing them.

[0064] The grass collection efficiency weighting coefficient, energy consumption weighting coefficient, and noise weighting coefficient are weighting coefficients set by technicians to comprehensively consider various target conditions. The sum of the grass collection efficiency weighting coefficient, energy consumption weighting coefficient, and noise weighting coefficient equals 1.

[0065] The value of the grass collection efficiency sub-function is related to the combination of airflow guidance parameters. Predicted grass collection efficiency values ​​for different parameter combinations are obtained by querying a pre-set airflow simulation database. A larger ground clearance for blade 3 and a gentler angle for blade 3 result in lower energy consumption. More drastic changes in the angles of blade 3 and guide vane 5 lead to greater noise.

[0066] The airflow simulation database stores predicted grass collection efficiency values ​​corresponding to different combinations of airflow guidance parameters.

[0067] S24: The final control coefficient is calculated using the integrated cost function and the preset grass collection target as the target airflow guidance parameter.

[0068] The grass collection target is the minimum acceptable grass collection efficiency of the lawnmower 1 set by the technicians.

[0069] The final control coefficient refers to the combination of the height value of blade 3, the angle value of blade 3, and the angle value of guide plate 5 when the value of the comprehensive cost function reaches its minimum under the premise of satisfying the grass collection target constraint.

[0070] By using the grass collection target as a constraint, that is, requiring the value of the grass collection efficiency sub-function to be greater than or equal to the grass collection efficiency value corresponding to the grass collection target, an iterative optimization algorithm is used to solve the comprehensive cost function under the premise of satisfying this constraint. In each iteration, the height value of blade 3, the angle value of blade 3, and the angle value of guide vane 5 are adjusted, and the value of comprehensive cost function is recalculated until the value of comprehensive cost function reaches the minimum value or the number of iterations reaches the preset maximum number of iterations. The height value of blade 3, the angle value of blade 3, and the angle value of guide vane 5 when the value of comprehensive cost function reaches the minimum value are taken as the final control coefficients, and the final control coefficients are used as the target airflow guidance parameters.

[0071] Also includes: S30: Retrieve the winding type from the detection category.

[0072] The entanglement type refers to the category of the object that causes the blade 3 to become entangled in the detection category, including the grass stem category and the foreign object category. The grass stem category and the foreign object category are extracted from the detection category as the entanglement type.

[0073] S31: Extract the marker winding parameters of the winding type and their corresponding marker winding length from the segmented image information.

[0074] The entanglement parameter refers to the parameters of each connected component in the segmented image information that are labeled as entanglement type (grass stem category or foreign object category), including the position coordinates and contour information of each connected component.

[0075] The marking winding length refers to the extension length of the connected domain corresponding to each marking winding parameter along the rotation direction of the lawnmower blade 3.

[0076] By extracting all pixels belonging to the grass stem or foreign object category from the segmented image information, adjacent pixels belonging to the entanglement type are merged into connected components. The length of the long side of the minimum bounding rectangle of each connected component is calculated as the marker entanglement length, and the position coordinates of the connected component in the segmented image information are recorded as the marker entanglement parameters.

[0077] S32: Obtain the baseline length threshold and baseline probability threshold based on the final control command.

[0078] The baseline length threshold is the critical length value used to determine whether grass stems or foreign objects may wrap around the blade 3, and the baseline probability threshold is the critical probability value used to confirm whether the blade 3 is already wrapped.

[0079] The reference length threshold and reference probability threshold are matched from a preset winding map table using the final control command. The winding map table stores the reference length threshold and reference probability threshold corresponding to different final control commands. The parameters in the winding map table are set in advance by those skilled in the art based on actual conditions through experiments.

[0080] S33: When the marker wrapping length exceeds the baseline length threshold, the marker wrapping parameter is defined as an abnormal wrapping parameter, and the abnormal wrapping position of the abnormal wrapping parameter is extracted from the segmented image information.

[0081] Abnormal entanglement parameter refers to the marked entanglement parameter whose marked entanglement length exceeds the baseline length threshold. Abnormal entanglement position refers to the position coordinate of the connected component corresponding to the abnormal entanglement parameter in the segmented image information. When the marked entanglement length exceeds the baseline length threshold, it indicates that there is a situation where grass stems or foreign objects may be entangled in the lawnmower blade 3. Therefore, the marked entanglement parameter whose marked entanglement length exceeds the baseline length threshold is defined as abnormal entanglement parameter, and the position coordinate of the connected component of the abnormal entanglement parameter is extracted from the segmented image information as the abnormal entanglement position.

[0082] S34: Collect the blade center position, current data, and rotation speed data of lawnmower 1.

[0083] The blade center position refers to the spatial coordinates of the rotation axis of blade 3 of lawnmower 1 in the coordinate system of lawnmower 1, which is obtained through pre-input by the operator.

[0084] The detected current data refers to the real-time current value of the blade 3 drive motor during operation, which is acquired by a current sensor preset on the motor drive circuit.

[0085] The detected rotational speed data refers to the real-time rotational speed value of the blade 3 drive motor during operation, which is obtained by a speed sensor preset on the motor shaft or by a Hall sensor of the motor driver.

[0086] S35: Combines abnormal winding location, blade center position, detection current data, detection speed data, and reference probability threshold to generate a clear airflow control command to update the final control command.

[0087] By analyzing the abnormal winding location, blade center position, detection current data, detection rotation speed data, and reference probability threshold, a clearing airflow control command is generated. This clearing airflow control command is then added to the final control command to obtain a new final control command. The specific analysis method for generating the clearing airflow control command is described in S40 to S65.

[0088] Methods for generating purge airflow control commands include: S40: Calculate the abnormal distance value and abnormal azimuth angle by using the blade center position and the abnormal winding position.

[0089] The abnormal distance value refers to the straight-line distance between the abnormal winding position and the center position of the blade, and the abnormal azimuth angle refers to the directional angle value of the abnormal winding position relative to the center position of the blade in the plane of the cutter head 2.

[0090] A plane coordinate system for the cutter head 2 is established with the center of the blade as the origin. The abnormal entanglement location is transformed from the image coordinate system of the segmented image information to the plane coordinate system of the cutter head 2, obtaining the horizontal and vertical coordinate values ​​of the abnormal entanglement location in the plane coordinate system of the cutter head 2. The square root of the sum of the squares of the horizontal and vertical coordinate values ​​is calculated as the abnormal distance value, and the arctangent of the ratio of the vertical to the horizontal coordinate value is calculated as the abnormal azimuth angle.

[0091] S41: Obtain the entanglement probability value based on the detected current data and detected rotation speed data.

[0092] The entanglement probability value refers to the likelihood that blade 3 is currently in an entangled state. The value ranges from 0 to 1, and the larger the value, the higher the likelihood of entanglement.

[0093] Within a preset time window (e.g., 0.5 seconds), the variance of current fluctuation is calculated for multiple sampled values ​​of the detected current data, and the variance of speed fluctuation is calculated for multiple sampled values ​​of the detected speed data. Then, the variance of current fluctuation and the variance of speed fluctuation are combined into a two-dimensional feature vector.

[0094] By inputting a two-dimensional feature vector into a pre-set winding recognition model, the winding recognition model outputs a winding probability value. The winding recognition model is a binary classification model pre-trained based on a logistic regression algorithm or a support vector machine algorithm. The training samples are multiple sets of current fluctuation variance values ​​and rotation speed fluctuation variance values ​​under known winding states.

[0095] S42: When the entanglement probability value exceeds the baseline probability threshold, the entanglement position and number of entanglements are calculated by segmenting the image information and the abnormal entanglement parameters.

[0096] The winding position refers to the specific attachment position of the winding material on the blade 3, and the number of winding turns refers to the number of complete rotations of the winding material around the axis of the blade 3.

[0097] By extracting all pixels of the connected component corresponding to the abnormal winding parameters from the segmented image information, the minimum bounding rectangle of the connected component is calculated. The center point coordinates of the minimum bounding rectangle are used as the initial value for the winding position. Based on the transformation relationship between the image coordinate system and the cutter head 2 spatial coordinate system, the center point coordinates are converted into axial and radial position values ​​in the cutter head 2 space as the winding position. Then, the ratio of the long side length to the short side length of the minimum bounding rectangle is calculated as the aspect ratio. The number of winding turns is obtained by looking up the winding mapping table based on the aspect ratio; a larger aspect ratio results in more winding turns. The winding mapping table stores the number of winding turns corresponding to different aspect ratios.

[0098] S43: Combine the winding position, number of winding turns, abnormal distance value, and abnormal azimuth angle to obtain winding posture information.

[0099] The winding attitude information refers to the multi-dimensional feature data that describes the winding state of the winding material on the blade 3. The feature vector formed by fusing the winding position, number of winding turns, abnormal distance value and abnormal azimuth angle is used as the winding attitude information.

[0100] S44: Generate clear airflow control commands based on the winding posture information.

[0101] The clear airflow control command is generated by analyzing the winding posture information. The specific analysis and generation method is described in S50 to S65.

[0102] Methods for generating purge airflow control commands also include: S50: Extract the effective impact cross-sectional area value of the cut object from the segmented image information.

[0103] The effective impact cross-sectional area value refers to the average cross-sectional area value of effective particles that can be used to impact the entangled object within the cut region category in the segmented image information. This is achieved by extracting all pixels corresponding to the cut region category from the segmented image information, extracting the connected components of each cut particle through connected component analysis, calculating the number of pixels in each connected component, converting the number of pixels into actual cross-sectional area values ​​according to the image resolution, sorting all the actual cross-sectional area values ​​of the cut particles from largest to smallest, taking the top 30% of the cut particles as effective impact particles, and calculating the arithmetic mean of the actual cross-sectional area values ​​of all effective impact particles as the effective impact cross-sectional area value.

[0104] In this embodiment, the image boundary between the cut area and the uncut area is the position of the blade 3 of the lawnmower 1.

[0105] S51: The reference frequency threshold is obtained by using the effective impact cross-sectional area value.

[0106] The reference frequency threshold is the cutoff frequency value used to filter the frequency components corresponding to the effective impact particles from the sound spectrum diagram. The reference frequency threshold is matched from the winding map table by the effective impact cross-sectional area value.

[0107] The entanglement mapping table stores the reference frequency thresholds corresponding to different effective impact cross-sectional area values. The larger the effective impact cross-sectional area value, the lower the reference frequency threshold. By collecting the sound signals generated when cutting particles of different cross-sectional areas impact under standard airflow conditions, the sound signals are subjected to spectrum analysis to extract the main frequency value corresponding to each cross-sectional area. The cross-sectional area and the main frequency value are associated and stored to form the entanglement mapping table.

[0108] S52: Obtain the baseline impact threshold and target duration value based on the winding posture information.

[0109] The baseline impact threshold refers to the minimum impact energy required to blow off or loosen the entanglement, while the target duration refers to the duration of continuous airflow used to remove the entanglement.

[0110] By extracting the number of winding turns from the winding posture information, the baseline impact threshold and the target duration value are matched from the winding mapping table based on the number of winding turns. The more winding turns, the larger the baseline impact threshold and the longer the target duration value.

[0111] S53: Retrieve the high-frequency energy percentage corresponding to the reference frequency threshold and the rising slope of the time-domain waveform envelope from the sound spectrum diagram, and match the average quality estimate and the average velocity estimate.

[0112] The high-frequency energy ratio refers to the ratio of the total energy of the frequency components above a reference frequency threshold in a sound spectrum to the total energy of the entire frequency band in the sound spectrum. It is obtained by calculating the sum of the energies of all frequency components above the reference frequency threshold in the sound spectrum and dividing this sum by the sum of the energies of all frequency components across the entire frequency band.

[0113] The rising edge slope refers to the slope of the rising segment of the time-domain waveform envelope of the sound detection information at the instant of impact with the cutting object. It is calculated by extracting the envelope from the time-domain waveform of the sound detection information and then determining the slope of the rising segment at the instant of impact.

[0114] The average mass estimate refers to the estimated average mass of the effective impacting particles in the cut particles, while the average velocity estimate refers to the estimated average velocity of the effective impacting particles under the influence of airflow. The average mass and average velocity estimates are matched from the winding mapping table based on the high-frequency energy ratio and the rise edge slope. A higher high-frequency energy ratio results in a lower average mass estimate. A higher rise edge slope results in a higher average velocity estimate.

[0115] By collecting the impact sound signals of cutting particles with known mass and velocity under standard conditions, the high-frequency energy ratio and rising edge slope are extracted and associated with the corresponding mass and velocity values ​​and stored in the winding mapping table.

[0116] S54: The impact kinetic energy is calculated by combining the effective impact cross-sectional area, the average mass estimate, and the average velocity estimate.

[0117] Impact kinetic energy refers to the kinetic energy of a single effective impacting particle when it impacts an entangling object. The average density of the effective impacting particles is calculated by using the effective impact cross-sectional area and the estimated average mass. Then, the impact kinetic energy is calculated according to the kinetic energy calculation formula based on the average density and the estimated average velocity.

[0118] Average density value = average mass estimate / (effective impact cross-sectional area value × preset particle thickness value), where the particle thickness value is the average thickness value of the effective impact particles set by the technician.

[0119] Impact kinetic energy = 0.5 × estimated average mass × (estimated average velocity) 2 0.5 is the coefficient value corresponding to the kinetic energy calculation formula set by the technicians.

[0120] S55: The airflow impact energy threshold is formed based on the benchmark impact threshold and the impact kinetic energy value.

[0121] The airflow impact energy threshold refers to the total impact energy value that the airflow needs to provide to remove entangled materials. The effective number of impacting particles is obtained by calculating the ratio of the baseline impact threshold to the impact kinetic energy value. The effective number of impacting particles is rounded up to obtain the minimum number of effective impacting particles required. The product of the minimum number and the impact kinetic energy value is then calculated as the airflow impact energy threshold.

[0122] S56: Obtain the target velocity value based on the airflow impact energy threshold.

[0123] The target velocity value refers to the velocity required for the airflow to clear entangled materials. It is calculated using the airflow impact energy threshold and the average mass estimate, and then calculated using the reaction energy formula: Airflow velocity value = (2 × airflow impact energy threshold / average mass estimate). 0.5 × speed correction coefficient, where the speed correction coefficient is a preset coefficient used to compensate for energy transfer loss. The value of the speed correction coefficient ranges from 1.1 to 1.5 and is obtained by technicians through experimental calibration based on the structure of the cutter head 2 and the characteristics of the airflow channel.

[0124] S57: Generate a clear airflow control command by combining the target duration value and the target speed value.

[0125] The airflow control command is generated by analyzing the target duration and target speed values. The specific generation method is described in S60 to S65.

[0126] Methods for generating purge airflow control commands also include: S60: Obtain the target impact coordinates based on the abnormal entanglement location.

[0127] The target impact coordinates refer to the coordinates of the target point location that needs to be focused and impacted by the clearing airflow in the spatial coordinate system of cutterhead 2, including axial and radial coordinates. By extracting the center point coordinates of the connected components corresponding to the abnormal winding parameters from the abnormal winding location, and referring to the transformation relationship between the image coordinate system and the cutterhead 2 spatial coordinate system in S40, the center point coordinates are converted into axial and radial coordinates in the cutterhead 2 spatial coordinate system as the target impact coordinates. When the abnormal winding parameters correspond to multiple connected components, the center point of the connected component with the largest marked winding length is selected as the target impact coordinates.

[0128] S61: Calculate the deflection angle value based on the target impact coordinates and the preset initial guide angle.

[0129] The initial airflow angle is the initial orientation angle value set by the technician for each airflow guide blade in the cutter head 2 of the lawnmower 1 when no airflow guidance adjustment has been made.

[0130] The deflection angle value refers to the rotation angle value required for each guide vane to rotate from the initial guide angle to the target angle. By taking the target impact coordinate as the focus, the airflow deflection direction required to guide the airflow to the focus from the position of each guide vane is calculated. Based on the current position and orientation of each guide vane, the rotation angle and rotation direction required for each guide vane to rotate from the initial guide angle to the direction that can guide the airflow to the target impact coordinate are calculated as the deflection angle value of each guide vane.

[0131] S62: Generates a pulse drive signal by using the target duration value and the target velocity value.

[0132] The pulse drive signal refers to the electrical signal used to control the drive motor of the guide plate 5 to work in an intermittent mode of start-continuous operation-stop, including the pulse start time, continuous operation duration, stop time and pulse repetition number.

[0133] The target duration is used as the pulse's duration. The target speed value determines the pulse's driving voltage or driving current amplitude. The driving voltage amplitude is proportional to the target speed value; the larger the target speed value, the higher the driving voltage amplitude. The pulse's start time is set to the current time, and the stop time is the start time plus the duration. The pulse repetition count is set to 1. The start time, duration, stop time, driving voltage amplitude, and pulse repetition count are combined to form the pulse driving signal.

[0134] S63: Acquire actual current data based on pulse drive signal, and obtain reference current threshold through winding posture information.

[0135] The actual current data refers to the actual operating current value of the drive motor of the guide plate 5 under the action of the pulse drive signal. When the guide plate 5 runs according to the pulse drive signal, it is collected in real time by the current sensor preset on the drive circuit of the drive motor of the guide plate 5.

[0136] The reference current threshold is a critical current value used to determine whether the entangled material has been successfully removed. The number of winding turns is retrieved from the winding posture information, and the reference current threshold is matched against the winding mapping table based on this number. The more winding turns, the lower the reference current threshold. By collecting the current values ​​of the guide plate 5 drive motor under different winding turn states during normal operation, the number of winding turns and the corresponding current values ​​are associated and stored in the winding mapping table.

[0137] S64: When the actual current data is less than the reference current threshold for three consecutive preset sampling periods, output a stop signal.

[0138] The sampling period is the time interval for sampling current data set by the technicians.

[0139] The stop signal is a command signal set by the technician to indicate that the drive motor of the guide plate 5 should stop running. When the actual current data is less than the reference current threshold for three consecutive sampling cycles, it means that the entanglement has been successfully removed (the load of the drive motor of the guide plate 5 has been reduced to a normal level). At this time, the stop signal is output to terminate the pulse drive signal in advance.

[0140] S65: Uses the deflection angle value, pulse drive signal, actual current data, reference current threshold, and stop signal as the clear airflow control command.

[0141] The deflection angle value, pulse drive signal, actual current data, reference current threshold, and stop signal are used as control commands for clearing airflow.

[0142] Also includes: S70: Acquire images of grass clippings.

[0143] The grass clipping detection image refers to the image of grass clippings thrown out from the first row of grass clippings or the entrance of the grass collection bag of the lawnmower. It is obtained by taking pictures of grass clippings through a camera preset at the first row of grass clippings or the entrance of the grass collection bag of the lawnmower, and is used to analyze the throwing state of grass clippings to evaluate grass collection efficiency.

[0144] S71: Extract the centroid coordinates of grass clippings from the grass clipping detection image.

[0145] Centroid coordinates refer to the position coordinates of the center point of each grass clipping target area in the image coordinate system.

[0146] Background subtraction is performed on the grass clipping detection image to extract the target grass clipping region. For each target grass clipping region, the average of the horizontal coordinates of all pixels within the region is calculated as the horizontal pixel coordinate of the centroid, and the average of the vertical coordinates of all pixels within the region is calculated as the vertical pixel coordinate of the centroid. The horizontal and vertical pixel coordinates are combined to obtain the centroid coordinate of the target grass clipping.

[0147] S72: Calculate the average projectile velocity and average projectile angle based on the centroid coordinates and the preset acquisition interval.

[0148] The acquisition interval is the time interval set by the technicians for the camera to acquire images of grass clippings.

[0149] The average projectile velocity value refers to the average speed of the grass clippings when they are thrown, and the average projectile angle value refers to the angle between the average direction of the grass clippings when they are thrown and the horizontal plane.

[0150] For each grass clipping target in the grass clipping detection image, the instantaneous velocity and instantaneous angle values ​​of the target between two adjacent frames are calculated based on the change in its centroid coordinates and the acquisition interval. The median of the instantaneous velocity values ​​calculated for the same target across all adjacent frames is taken to obtain the projectile velocity value. The average projectile velocity value is then obtained by averaging the projectile velocity values ​​of all grass clipping targets. Similarly, the median of the instantaneous angle values ​​calculated for the same target across all adjacent frames is taken to obtain the projectile angle value. The average projectile angle value is then obtained by averaging the projectile angle values ​​of all grass clipping targets.

[0151] S73: The dispersion value is obtained based on the centroid coordinates and the grass clipping detection image.

[0152] The dispersion value is a quantitative indicator of the spread of grass clippings after they are thrown out. It is used to characterize whether the grass clippings are thrown out in a concentrated manner or in a dispersed manner. The larger the dispersion value, the more dispersed the grass clippings are and the lower the grass collection efficiency.

[0153] By calculating the covariance matrix of the centroid coordinates of all grass clipping target areas, the first and second eigenvalues ​​of the covariance matrix are extracted. The product of the first and second eigenvalues ​​is calculated as the scattered area value of the grass clippings in that frame of the image. Then, the average scattered area value is obtained by averaging the scattered area values ​​of all frames of the image. The quotient of the average scattered area value and the preset standard scattered area value is calculated as the dispersion value.

[0154] The standard distribution area value is the baseline value of the distribution area of ​​grass clippings under ideal grass collection conditions, set by technicians.

[0155] S74: The grass collection efficiency index is obtained by weighting and summing the average projectile velocity, average projectile angle, and dispersion value.

[0156] The grass collection efficiency index is a comprehensive index value used to quantitatively evaluate the current grass collection efficiency. The value ranges from 0 to 100, and the higher the value, the higher the grass collection efficiency.

[0157] The grass collection efficiency index is calculated by multiplying the average projectile velocity, average projectile angle, and dispersion value by preset weighting coefficients for velocity, angle, and dispersion, respectively, and then summing the results. All three weighting coefficients are positive numbers, and their sum equals 1. The velocity weighting coefficient ranges from 0.3 to 0.5, the angle weighting coefficient from 0.2 to 0.4, and the dispersion weighting coefficient from 0.2 to 0.4. The specific values ​​for each weighting coefficient are determined experimentally by technicians based on the structure of the mower's blade 2 and the characteristics of the grass collection channel.

[0158] S75: Compare the grass collection efficiency index value with the preset index threshold to calculate the index deviation value, and update the final control command based on the index deviation value.

[0159] The index threshold is the minimum acceptable value for grass collection efficiency set by technicians, ranging from 60 to 80.

[0160] The index deviation value refers to the deviation between the grass collection efficiency index value and the index threshold. When the grass collection efficiency index value is lower than the index threshold, it indicates that the current grass collection efficiency is insufficient, and the airflow guidance parameters need to be adjusted. Therefore, the difference between the grass collection efficiency index value and the index threshold is calculated as the index deviation value.

[0161] The absolute value of the exponential deviation is used to match the blade 3 height adjustment step value, blade 3 angle adjustment step value, and guide vane 5 angle adjustment step value from the preset control mapping table. Each adjustment step value is proportional to the absolute value of the exponential deviation. Based on each adjustment step value, the adjusted target blade 3 height value, target blade 3 angle value, and target guide vane 5 angle value are generated. The adjusted target values ​​are used to replace the corresponding original target values ​​in the final control command to obtain the new final control command.

[0162] Also includes: S80: Collect vibration detection information of blade 3 during the cutting process of lawnmower 1.

[0163] Vibration detection information refers to the mechanical vibration signals generated by the blade 3 of the lawnmower 1 during the rotating cutting process, including vibration acceleration values, vibration velocity values, or vibration displacement values. Vibration signals during the operation of the blade 3 are collected in real time by vibration sensors (such as piezoelectric accelerometers or microelectromechanical system accelerometers) preset on the cutter head 2 of the lawnmower 1 or near the drive motor of the blade 3 as vibration detection information.

[0164] S81: Generate time-domain feature vectors and frequency-domain feature vectors based on vibration detection information, and combine the time-domain feature vectors and frequency-domain feature vectors as a comprehensive feature vector.

[0165] A time-domain feature vector is a combination of multiple time-domain indices extracted from the time-domain waveform of vibration detection information to describe the statistical characteristics of the vibration signal. Within a preset time window, the vibration detection information is sampled to obtain a discrete time-domain signal sequence. The root mean square value, peak factor, and kurtosis value of the discrete time-domain signal sequence are calculated, and the root mean square value, peak factor, and kurtosis value are combined into a time-domain feature vector.

[0166] Wherein, the root mean square value is the square root of the average of the squares of each sample value in the discrete-time signal sequence, the peak factor is the ratio of the maximum absolute value in the discrete-time signal sequence to the root mean square value, and the kurtosis is the ratio of the fourth central moment of the discrete-time signal sequence to the fourth power of the root mean square value.

[0167] Frequency domain feature vectors refer to a combination of multiple frequency domain indices extracted from the spectrum of vibration detection information to describe the frequency domain characteristics of vibration signals.

[0168] The vibration spectrum is obtained by performing a fast Fourier transform on the discrete time domain signal sequence. The amplitude of the frequency component with the same rotation frequency as blade 3 is extracted from the vibration spectrum as the rotation frequency amplitude. The amplitude of the frequency component with integer multiples of the rotation frequency is extracted as the harmonic amplitude. The ratio of the total energy in the preset frequency band to the total energy of the entire frequency band is calculated as the frequency band energy ratio. Finally, the rotation frequency amplitude, harmonic amplitude and frequency band energy ratio are combined into a frequency domain feature vector.

[0169] The preset frequency bands include a first preset frequency band, a second preset frequency band, and a third preset frequency band. The frequency range of the first preset frequency band is 1 to 2 times the rotation frequency of the blade 3, and it is used to characterize the imbalance characteristics of the blade 3. The frequency range of the second preset frequency band is 3 to 10 times the rotation frequency of the blade 3, and it is used to characterize the wear characteristics of the blade 3. The frequency range of the third preset frequency band is 1000Hz to 3000Hz, and it is used to characterize the high-frequency impact characteristics of the blade 3.

[0170] A composite feature vector is a joint feature vector obtained by concatenating and fusing time-domain feature vectors and frequency-domain feature vectors. The composite feature vector is formed by concatenating and fusing time-domain feature vectors and frequency-domain feature vectors.

[0171] S82: Detects wear values ​​based on comprehensive feature vector matching.

[0172] The detected wear value refers to the quantitative value of the current wear level of blade 3. The detected wear value is output by inputting the comprehensive feature vector into the preset wear level evaluation model.

[0173] The wear assessment model is a pre-trained multi-class regression model based on support vector machine or neural network algorithms. The training samples consist of multiple sets of comprehensive feature vector samples with known wear levels. Each set includes a comprehensive feature vector formed by fusing time-domain and frequency-domain feature vectors collected at different wear stages (e.g., 0%, 10%, 20%, ..., 100%) of blade 3, along with the corresponding wear level label. The wear assessment model predicts the wear level of unknown samples by learning the mapping relationship between the comprehensive feature vector and the wear level.

[0174] S83: Based on the detected wear value and the final control command, an airflow deviation parameter is generated.

[0175] The airflow deviation parameter refers to the deviation between the actual airflow characteristics and the designed airflow characteristics caused by the wear of blade 3.

[0176] The airflow direction deviation value and airflow velocity deviation value obtained by querying the preset wear airflow deviation mapping table through the wear value detection are used as airflow deviation parameters.

[0177] The wear airflow deviation mapping table stores experimental values ​​of airflow direction deviation and airflow velocity deviation corresponding to different detected wear values. The method for establishing the wear airflow deviation mapping table is as follows: Under standard operating conditions, the actual airflow direction angle and actual airflow velocity values ​​inside the cutter head 2 are collected at different wear levels (e.g., 0%, 10%, 20%, ..., 100%) of the blade 3. The difference between the actual airflow direction angle and the designed airflow direction angle under each wear level is calculated as the airflow direction deviation value. The difference between the actual airflow velocity value and the designed airflow velocity value under each wear level is calculated as the airflow velocity deviation value. The detected wear values ​​are associated with and stored with the corresponding airflow direction deviation values ​​and airflow velocity deviation values ​​to form the wear airflow deviation mapping table.

[0178] S84: Matches the flow compensation command with the airflow deviation parameter, and outputs the flow compensation command to update the final control command when the detected wear value is greater than the preset reference wear threshold.

[0179] The guide vane compensation command is a control signal used to adjust the angle of the guide vanes to compensate for airflow deviation caused by wear of blade 3. The guide vane compensation command is matched from the control mapping table by the airflow deviation parameter; the larger the airflow deviation parameter, the larger the guide vane compensation command.

[0180] The reference wear threshold is the maximum allowable wear of blade 3 that is set by the technician without affecting the airflow inside the lawnmower 1. When the detected wear value is greater than the reference wear threshold, it means that the wear of blade 3 has caused airflow deviation. In this case, the airflow compensation command is added to the final control command to form a new final control command.

[0181] Methods for updating the final control command also include: S90: Extract the marked vibration amplitude of the frequency corresponding to the detected rotational speed data from the vibration detection information.

[0182] The marked vibration amplitude refers to the amplitude of the fundamental frequency component in the vibration spectrum diagram of S81 that is the same as the rotational frequency of the blade 3. From the vibration spectrum diagram obtained from S81, the rotational frequency value corresponding to the detected rotational speed data of the blade 3 is found and extracted as the marked vibration amplitude.

[0183] S91: The marking wave quality of blade 3 is obtained based on the marking vibration amplitude value.

[0184] The marked wave mass refers to the unbalanced mass value generated by the uneven mass distribution of blade 3. The quotient of the marked vibration amplitude and the preset vibration mass conversion coefficient is used as the marked wave mass.

[0185] The vibration-mass conversion coefficient is a conversion parameter obtained by technicians through a standard mass block calibration experiment. By installing a standard unbalanced mass block of known mass on the cutter head 2, measuring the resulting vibration amplitude, and calculating the ratio of the vibration amplitude to the unbalanced mass value as the vibration-mass conversion coefficient.

[0186] S92: Obtain the detection phase difference value by comparing the vibration detection information with the preset reference vibration signal, and use the detection phase difference value corresponding to the marked vibration amplitude as the marked phase difference value.

[0187] The reference vibration signal is the standard vibration signal set by the technician when the blade 3 is operating in a fully balanced state.

[0188] The detected phase difference value refers to the phase difference angle between the actual vibration detection information and the reference vibration signal at the same frequency component. The phase difference value at the fundamental frequency component is extracted from the cross power spectrum between the spectrum of the vibration detection information and the spectrum of the reference vibration signal as the detected phase difference value.

[0189] The marker phase difference refers to the detection phase difference corresponding to the marker vibration amplitude. The detection phase difference corresponding to the marker vibration amplitude is used as the marker phase difference.

[0190] S93: The marking angle position of blade 3 is obtained based on the marking phase difference.

[0191] The marked angle position refers to the angle value of the unbalanced mass on the blade 3 in the plane coordinate system of the cutter head 2, which is obtained by taking the phase difference value of the mark as the marked angle position.

[0192] S94: Combine the marked wave mass, marked angle position and preset counterweight specifications to obtain the counterweight movement parameters, and add the counterweight movement parameters to the final control command.

[0193] Reference Figure 3 The counterweight specifications are the mass parameters and installation position parameters of the movable counterweights 6 set on the cutter head 2 of the lawnmower 1, as set by the technicians. These parameters include the mass value of each counterweight 6, the initial installation position of each counterweight 6 on the cutter head 2, and the movable range of each counterweight 6.

[0194] The counterweight movement parameters refer to the distance and direction values ​​that each counterweight 6 needs to move in order to make the combined compensating centrifugal force generated after each counterweight 6 moves cancel each other out with the unbalanced centrifugal force generated by the marked wave mass.

[0195] The unbalanced centrifugal force vector is calculated by marking the fluctuating mass and the marked angular position. The compensation centrifugal force vector generated when each counterweight 6 moves a unit distance is calculated based on the counterweight specifications of each counterweight 6 (including the mass value and initial installation position of each counterweight 6).

[0196] Using the moving distance of each counterweight 6 as the independent variable, and the vector difference between the combined compensating centrifugal force vector and the unbalanced centrifugal force vector generated by each counterweight 6 as the objective function, a set of moving distance values ​​that minimizes the objective function is solved. When multiple solutions exist, the solution with the smallest total moving distance of each counterweight 6 is selected as the counterweight moving parameter, and the counterweight moving parameter is added to the final control command to control the drive motor of each counterweight 6 to move each counterweight 6 to the target position.

[0197] Unbalanced centrifugal force vector = marked wave mass × (rotational angular velocity of blade 3) 2 The direction is the direction of the marked angular position. The rotational angular velocity of blade 3 is the angular velocity value converted from the detected rotational speed data.

[0198] Based on the same inventive concept, embodiments of the present invention provide an airflow guidance design system for lawnmowers suitable for slope operations, comprising: The acquisition module is used to acquire terrain information, image detection information, and sound detection information. A memory for storing a program for a lawnmower airflow guidance design method suitable for slope operations; The processor is used to load and execute programs stored in memory.

[0199] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0200] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principle of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A method for designing airflow guidance for lawnmowers suitable for slope operations, characterized in that, include: Collect terrain information, image detection information and sound detection information of the area where the lawnmower (1) is located, and preprocess the image detection information to form detection image information; Semantic segmentation is performed based on the detected image information and the preset detection categories to form segmented image information; The type and density of the object to be cut are obtained from the segmented image information; The sound spectrum is obtained by performing time-domain sampling, analog-to-digital conversion, and fast Fourier transform on the sound detection information; Extract detection spectral features from the sound spectrogram, update the cutting type and obtain the cutting humidity based on the detection spectral features; The target airflow guidance parameters are obtained based on the type of material being cut, cutting density, material humidity, and terrain information. The final control command is generated based on the target airflow guidance parameters and then uploaded.

2. The airflow guidance design method for lawnmowers suitable for slope operations according to claim 1, characterized in that, Methods for obtaining target airflow steering parameters include: Retrieve slope detection information from terrain information; The airflow adhesion effect coefficient is calculated based on the slope detection information and the preset unit blade height. The critical airflow velocity value is calculated by considering the type of material being cut, the cutting density, and the humidity of the material being cut. A comprehensive cost function is constructed by combining the critical airflow velocity value, the airflow wall effect coefficient, the preset benchmark target, and the preset unit control coefficient. The final control coefficient is calculated using a comprehensive cost function and a preset grass collection target, and is used as the target airflow guidance parameter.

3. The airflow guidance design method for lawnmowers suitable for slope operations according to claim 2, characterized in that, Also includes: Retrieve the entanglement type from the detection category; Extract the marker winding parameters of the winding type and their corresponding marker winding length from the segmented image information; The baseline length threshold and baseline probability threshold are obtained based on the final control command; When the marker wrapping length exceeds the baseline length threshold, the marker wrapping parameter is defined as an abnormal wrapping parameter, and the abnormal wrapping position of the abnormal wrapping parameter is extracted from the segmented image information; Collect the blade center position, current data and rotation speed data of the lawnmower (1); By combining the abnormal winding location, blade center position, detection current data, detection speed data, and reference probability threshold, a clearing airflow control command is generated to update the final control command.

4. The airflow guidance design method for lawnmowers suitable for slope operations according to claim 3, characterized in that, Methods for generating purge airflow control commands include: The abnormal distance and abnormal azimuth angle are calculated by using the blade center position and the abnormal winding position; The entanglement probability value is obtained based on the detected current data and detected rotation speed data; When the entanglement probability value exceeds the baseline probability threshold, the entanglement position and number of entanglements are calculated by segmenting the image information and abnormal entanglement parameters. The winding attitude information is obtained by combining the winding position, number of winding turns, abnormal distance value, and abnormal azimuth angle. A clearing airflow control command is generated based on the entanglement posture information.

5. The airflow guidance design method for a lawnmower suitable for slope operation according to claim 4, characterized in that, Methods for generating purge airflow control commands also include: Extract the effective impact cross-sectional area value of the cut object from the segmented image information; The reference frequency threshold is obtained by taking the effective impact cross-sectional area value; The baseline impact threshold and target duration are obtained based on the entanglement attitude information; Retrieve the high-frequency energy percentage corresponding to the reference frequency threshold and the rising slope of the time-domain waveform envelope from the sound spectrum diagram, and match the average quality estimate and the average velocity estimate. The impact kinetic energy is calculated by combining the effective impact cross-sectional area, the estimated average mass, and the estimated average velocity. The airflow impact energy threshold is determined based on the baseline impact threshold and the impact kinetic energy value. The target velocity value is obtained based on the airflow impact energy threshold. The target duration value and the target speed value are combined to generate a clear airflow control command.

6. The airflow guidance design method for a lawnmower suitable for slope operation according to claim 5, characterized in that, Methods for generating purge airflow control commands also include: The target impact coordinates are obtained based on the location of the abnormal entanglement. The deflection angle value is calculated based on the target impact coordinates and the preset initial guide angle; A pulse drive signal is generated by the target duration value and the target velocity value; The actual current data is acquired based on the pulse drive signal, and the reference current threshold is obtained through the winding posture information. When the actual current data is less than the reference current threshold for three consecutive preset sampling periods, a stop signal is output. The deflection angle value, pulse drive signal, actual current data, reference current threshold, and stop signal are used as the clear airflow control commands.

7. The airflow guidance design method for a lawnmower suitable for slope operation according to claim 6, characterized in that, Also includes: Collect images of grass clippings; Extract the centroid coordinates of grass clippings from the grass clipping detection image; The average projectile velocity and average projectile angle are calculated based on the centroid coordinates and the preset acquisition interval. The dispersion value is obtained based on the centroid coordinates and the grass clipping detection image; The grass collection efficiency index is obtained by weighting and summing the average projectile velocity, average projectile angle, and dispersion. The index deviation value is calculated by comparing the grass collection efficiency index value with the preset index threshold, and the final control command is updated based on the index deviation value.

8. The airflow guidance design method for a lawnmower suitable for slope operation according to claim 7, characterized in that, Also includes: Vibration detection information of the blade (3) during the cutting process of the lawnmower (1); Based on vibration detection information, time-domain feature vectors and frequency-domain feature vectors are generated, and the time-domain feature vectors and frequency-domain feature vectors are combined as a comprehensive feature vector; The wear value is detected based on the comprehensive feature vector matching; Airflow deviation parameters are generated based on the detected wear values ​​and the final control commands; The flow correction compensation command is matched with the airflow deviation parameter, and when the detected wear value is greater than the preset reference wear threshold, the flow correction compensation command is output to update the final control command.

9. A method for designing airflow guidance for lawnmowers suitable for slope operations according to claim 8, characterized in that, Methods for updating the final control command also include: Extract the labeled vibration amplitude of the frequency corresponding to the detected rotational speed data from the vibration detection information; The marked wave mass of the blade (3) is obtained based on the marked vibration amplitude value; The detection phase difference is obtained by comparing the vibration detection information with the preset reference vibration signal, and the detection phase difference corresponding to the marked vibration amplitude is used as the marked phase difference. The marking angle position of the blade (3) is obtained based on the marking phase difference value; The counterweight movement parameters are obtained by combining the marked fluctuation mass, the marked angle position, and the preset counterweight specifications, and then the counterweight movement parameters are added to the final control command.

10. An airflow guidance design system for lawnmowers suitable for slope operations, characterized in that, include: The acquisition module is used to acquire terrain information, image detection information, and sound detection information. A memory for storing a program that implements a lawnmower airflow guidance design method suitable for slope operations as described in any one of claims 1 to 9; The processor is used to load and execute programs stored in memory.

Images