Self-adaptive adjusting method for vertical clamping and conveying device of head vegetable harvester

Abstract

The invention relates to the technical field of agricultural machinery, in particular to a self-adaptive adjusting method for a vertical clamping and conveying device of a heading vegetable harvester, which comprises the following steps of: acquiring the ball diameter, weight and surface state data of cabbages in real time through a plurality of sensors, optimizing the surface state of the cabbages by adopting a deep learning image recognition technology, and determining the surface state of the cabbages; a physical characteristic model of the cabbages is generated based on a machine learning algorithm, the clamping force of the clamping device is adjusted through a self-adaptive algorithm according to the clamping force output by the model, the distribution of the clamping force and the adjusting speed are dynamically adjusted through a PID control algorithm in combination with real-time operation data, and the stability of the cabbages in the conveying process is ensured. According to the device, the clamping force is monitored in real time through the feedback loop, damage or falling of cabbages caused by excessive clamping or unstable clamping in the harvesting process is avoided, the harvesting efficiency can be improved, damage can be reduced, and the harvesting quality of the cabbages can be improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of agricultural machinery, in particular to a self-adaptive adjustment method for a vertical clamping and conveying device of a cabbages harvester. BACKGROUND

[0002] Cabbage is a kind of cabbages, and the harvesting process of cabbage as a widely planted vegetable requires higher accuracy and adaptability of equipment. Traditional cabbage harvesters usually rely on mechanical arms or clamping devices to grab cabbages, but these devices often fail to fully consider the different sizes, shapes and surface characteristics of cabbages, leading to cracking, damage or falling of cabbages during harvesting, thereby affecting the quality and efficiency of harvesting.

[0003] Although existing cabbage harvesting equipment has solved the problem of mechanized harvesting to some extent, most of the equipment still faces the problem of excessive or insufficient clamping force, and fails to accurately adjust according to the actual size and shape of cabbage. Specifically, the traditional technology relies on preset standard values or simple mechanical adjustment methods for clamping force adjustment, and fails to fully consider the dynamic changes of cabbage during harvesting, such as differences in ball diameter, weight and surface state, which leads to some cabbages being cracked or deformed due to excessive clamping force, and others being unable to be stably clamped due to insufficient clamping force, resulting in unstable clamping or even falling. SUMMARY

[0004] The present application provides a self-adaptive adjustment method for a vertical clamping and conveying device of a cabbages harvester.

[0005] The self-adaptive adjustment method for a vertical clamping and conveying device of a cabbages harvester includes the following steps: S1, sensor data acquisition: real-time acquisition of cabbage data through multiple sensors installed on the clamping device, including ball diameter, weight and surface state; S2, feature optimization processing: using deep learning image recognition technology to optimize the surface state of cabbage; S3, data processing and analysis: based on machine learning algorithm, analyzing the physical characteristics of cabbage according to the collected cabbage data, and generating a physical characteristic model of cabbage; S4, clamping force adjustment: adjusting the clamping force through adaptive algorithm according to the generated physical characteristic model, and the adjustment process includes real-time adjustment of the clamping force of the clamping device through a feedback loop, so that the clamping force matches the ball diameter and shape of the cabbage; S5, dynamic adjustment mechanism: dynamically adjusting the distribution and adjustment speed of the clamping force according to the position of the cabbage in the clamping device.

[0006] Optionally, the S1 includes: S11, ball diameter collection: Real-time measurement of the ball diameter of the cabbage through a laser ranging sensor installed on the clamping device; S12, weight collection: Real-time measurement of the weight of the cabbage by installing multiple pressure sensors on the contact area of the clamping device; S13, surface state collection: Real-time acquisition of image data of the surface of the cabbage through a visual sensor (such as a camera or infrared sensor) to evaluate the surface state of the cabbage.

[0007] Optionally, the S2 includes: S21, image data preprocessing: After acquiring the image data of the surface of the cabbage, the collected image data is preprocessed. Preprocessing includes image denoising and contrast adjustment; S22, model training: A convolutional neural network model is constructed, and the convolutional neural network model is trained through a training data set; S23, feature extraction and optimization: After training is completed, the trained convolutional neural network model is applied to real-time collected cabbage image data, and multiple features of the surface state of the cabbage are extracted through the convolutional neural network model; S24, surface state score generation: According to the features extracted by the convolutional neural network model, a comprehensive score of the surface state of each cabbage is generated.

[0008] Optionally, the S3 includes: S31, data preprocessing: Preprocessing of the collected cabbage data, including missing data filling, outlier removal, and standardization processing; S32, feature extraction and feature engineering: Based on the preprocessed cabbage data, physical features are extracted, and a feature vector is constructed; S33, physical feature model construction: A physical feature model of the cabbage is established using the extracted feature vector.

[0009] Optionally, the construction step of the physical feature model includes: S331, model construction: A physical feature model is constructed using a random forest algorithm; S332, model training: The constructed physical feature model is trained using a training data set; S333, model application: The trained physical feature model is used to output the corresponding clamping force.

[0010] Optionally, the S4 includes: S41, clamping force calculation: According to the feature vector extracted from the cabbage data, a physical feature model is used to output the corresponding target clamping force; S42, clamping force adjustment: According to the target clamping force output by the physical feature model, the clamping force is adjusted in real time through a PID control algorithm; S43, clamping force feedback loop: through the feedback loop, the actual clamping force of the clamping device is monitored in real time, and a feedback signal is input to the PID control algorithm to optimize the adjustment strategy of the clamping force.

[0011] Optionally, the S5 comprises: S51, data acquisition and monitoring: through a plurality of sensors installed on the clamping device, the position data of the cabbage in the clamping device is acquired in real time; S52, dynamic adjustment of clamping force distribution and adjustment speed: combined with the acquired position data of the cabbage in the clamping device, the PID control algorithm is used to dynamically adjust the distribution and adjustment speed of the clamping force; S53, stability optimization and correction: during the clamping process, by adjusting the clamping force in real time, damage or falling of the cabbage due to excessive extrusion or unstable clamping is avoided.

[0012] Optionally, the S52 specifically comprises: S521, clamping force distribution adjustment: according to the current position of the cabbage in the clamping device, the clamping force distribution of each part of the clamping device is adjusted by the PID algorithm to ensure that the clamping force uniformly acts on the surface of the cabbage, avoiding the damage caused by concentrated clamping force or unstable clamping force caused by uneven clamping force; S522, adjustment speed control: according to the position information of the cabbage, the PID control algorithm is used to adjust the change rate of the clamping force to ensure that the clamping force adjustment process is smooth and without excessive fluctuation, preventing damage or unstable clamping caused by excessive change of the clamping force during the adjustment process.

[0013] Optionally, the S53 comprises: S531, clamping force optimization: based on the real-time position of the cabbage and the clamping force feedback, the distribution and size of the clamping force are optimized to ensure that the clamping force can firmly fix the cabbage without excessive pressure on the surface of the cabbage. By adjusting the size of the clamping force in real time, the clamping force is always kept within the range most suitable for the shape and surface state of the cabbage, reducing the mechanical damage to the cabbage during the clamping process; S532, dynamic correction: during the clamping process, in the case of uneven clamping force, excessive or insufficient clamping point force, etc., the clamping force is corrected in real time. By monitoring the position change of the cabbage in the clamping device, the force distribution of the clamping point is dynamically adjusted to ensure that the clamping force always matches the real-time state and position of the cabbage. The correction process is based on the difference between the actual clamping force and the target clamping force of the clamping device, and the operation of the clamping device is adjusted through the feedback mechanism to make the clamping force distribution optimal.

[0014] The beneficial effects of the present application are: The present application, by using deep learning image recognition technology in S2 step to optimize the surface state of cabbage, can effectively identify the possible defects, cracks, unevenness and other characteristics on the surface of cabbage, and make comprehensive evaluation. This process not only improves the accurate judgment of the surface state of cabbage, but also ensures that the clamping device can adjust the clamping force more accurately according to the specific state of cabbage, avoiding unstable clamping or cracking caused by uneven surface state, thereby significantly reducing the damage risk of cabbage during clamping process.

[0015] In S3 step of the present application, by collecting cabbage data (such as ball diameter, weight, surface state, etc.) and deep analysis based on machine learning algorithm, the physical characteristics of cabbage can be extracted, and accurate physical characteristic model can be generated. This model effectively quantifies the relationship between different characteristics, providing a reliable basis for clamping force calculation and adjustment. Compared with traditional methods, data analysis based on machine learning can automatically adapt to the changes of different cabbage varieties, sizes and shapes, improving the accuracy and wide applicability of physical characteristic model, ensuring the efficiency and accuracy of clamping force calculation.

[0016] By combining the generated physical characteristic model, the present application uses adaptive algorithm to adjust the clamping force in real time, which can match the clamping device with the ball diameter and shape of cabbage, so as to ensure that the clamping force is neither too large nor too small. Through real-time feedback and optimization adjustment of clamping force, it can effectively avoid the situation that the clamping force is too large to cause cabbage to be cracked or too small to cause unstable clamping. This clamping force adjustment method based on physical characteristic model and adaptive algorithm not only improves the stability of cabbage harvesting, but also reduces the damage of cabbage caused by improper clamping, ensuring efficient and low damage harvesting process. BRIEF DESCRIPTION OF DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present application or prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only a part of the present application, and other drawings can also be obtained by those skilled in the art without creative labor.

[0018] Fig. 1 The method flowchart of the embodiment of the present application is shown in the following figure; Fig. 2 The S2 flowchart of the embodiment of the present application is shown in the following figure; Fig. 3 The schematic diagram of cabbage harvester of the embodiment of the present application is shown in the following figure. DETAILED DESCRIPTION

[0019] The application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted here that in order to make the embodiments more detailed, the following embodiments are the best and preferred embodiments, and other alternative methods can also be used by those skilled in the art to implement them; and the drawings are only used to describe the embodiments in more detail, and are not intended to limit the application specifically.

[0020] As Figs. 1-3 shown, the adaptive adjustment method of the vertical clamping conveying device of the head vegetable harvester includes the following steps: S1, sensor data acquisition: real-time acquisition of cabbage data through multiple sensors installed on the clamping device, including ball diameter, weight and surface state; S2, feature optimization processing: using deep learning image recognition technology to optimize the surface state of the cabbage; S3, data processing and analysis: based on the collected cabbage data, the physical characteristics of the cabbage are analyzed based on machine learning algorithm, and the physical characteristic model of the cabbage is generated; S4, clamping force adjustment: according to the generated physical characteristic model, the clamping force is adjusted through adaptive algorithm, and the adjustment process includes real-time adjustment of the clamping force of the clamping device through feedback loop, so that the clamping force matches the ball diameter and morphology of the cabbage, avoiding excessive clamping force leading to cabbage cracking, or small clamping force leading to unstable clamping; S5, dynamic adjustment mechanism: according to the position of the cabbage in the clamping device, the distribution and adjustment speed of the clamping force are dynamically adjusted to ensure the stability of the clamping process and prevent the cabbage from being damaged or falling due to excessive extrusion, cracking or unstable clamping during the conveying process.

[0021] S1 includes: S11, ball diameter acquisition: real-time measurement of the ball diameter of the cabbage through the laser ranging sensor installed on the clamping device, the laser ranging sensor scans the outline of the cabbage at high frequency, and uses the reflection time of the laser beam to calculate the distance between the cabbage surface and the sensor, and through the measurement data of multiple angles, the diameter of the cabbage is obtained; S12, weight acquisition: real-time measurement of the weight of the cabbage through multiple pressure sensors installed on the contact area of the clamping device, when the pressure sensor contacts the cabbage, the weight and mass distribution of the cabbage are reflected through the force signal. The pressure sensor can accurately measure the weight of the cabbage, and combine with the data of other sensors (such as ball diameter, surface state) to ensure that the clamping device applies appropriate clamping force to the cabbage to avoid damage or instability; S13, surface state acquisition: through a visual sensor (such as a camera or an infrared sensor), real-time acquisition of image data of the surface of the cabbage to evaluate the surface state of the cabbage, the image data collected by the visual sensor including the smoothness, disease spot, crack or other surface damage of the surface of the cabbage.

[0022] S2 includes: S21, image data preprocessing: after acquiring the image data of the surface of the cabbage, the collected image data is preprocessed. The preprocessing includes image denoising and contrast adjustment; aiming to improve the quality of the image and reduce interference factors, so that the subsequent convolutional neural network model can better identify the surface state of the cabbage. In the denoising process, median filtering or Gaussian filtering algorithm is adopted to remove noise interference in the image and ensure the accuracy of image details; S22, model training: a convolutional neural network model is constructed, and the convolutional neural network model is trained through a training data set, the training data set including a large number of cabbage surface images labeled with different surface state features, the features including surface smoothness, disease spot and crack; S23, feature extraction and optimization: after the training is completed, the trained convolutional neural network model is applied to the real-time collected cabbage image data, and multiple features of the surface state of the cabbage are extracted through the convolutional neural network model, including surface smoothness, disease spot area and crack degree; S24, surface state score generation: according to the features extracted by the convolutional neural network model, a comprehensive score of the surface state of each cabbage is generated, the surface state score including multiple sub-scores, which is usually a weighted average of the surface smoothness, disease spot and crack indicators, and the calculation formula is: ; wherein, is the comprehensive score of the surface state, is the surface smoothness, is the disease spot area, is the crack degree, , , are the weight coefficients of the surface smoothness, disease spot area and crack degree respectively, and . The comprehensive score will serve as a reference for subsequent clamping force adjustment.

[0023] S3 includes: S31, data preprocessing: the collected cabbage data is preprocessed, including missing data filling, outlier removal and standardization processing: missing data filling: if the data is missing, the missing data is filled by interpolation according to the surrounding valid data. For example, if the ball diameter data is missing, the missing data is filled by linear interpolation according to the data at adjacent time points; Outlier removal: Detect and remove abnormal data caused by sensor failure or external interference, such as when the detected ball diameter is far beyond the actual range, remove data beyond the reasonable range; Standardization: Standardize the data collected by different sensors to ensure comparability between different data dimensions. For example, the ball diameter data can be normalized to limit its value within the interval [0, 1], which facilitates subsequent analysis; S32, feature extraction and feature engineering: Based on the pretreated cabbage data, physical features are extracted and feature vectors are constructed, including: Extract physical features: Extract representative physical features from the collected ball diameter, weight and surface state data. For example, the average value of the ball diameter, the standard deviation of the weight, and the comprehensive score of the surface state are extracted; Construct feature vector: Combine the extracted physical features to form a feature vector. For example, for a certain cabbage, the feature vector contains: the average value of the ball diameter, the standard deviation of the weight, and the comprehensive score of the surface state; S33, physical feature model construction: Use the extracted feature vector to establish the physical feature model of cabbage.

[0024] The construction steps of the physical feature model include: S331, model construction: Use the random forest algorithm to construct the physical feature model; S332, model training: Use the training data set to train the constructed physical feature model, the specific training process is: Input data: Feature vector of each cabbage , wherein, is the average value of the ball diameter, is the standard deviation of the weight, is the comprehensive score of the surface state; Output data: The output is the clamping force F corresponding to the feature vector; The training data set is , wherein, ; Input the training data set into the constructed physical feature model for training. Finally, the model outputs the clamping force prediction through the average value of all regression trees, represented as: ; Wherein, is the predicted clamping force, is the predicted clamping force of the jth regression tree for the ith sample, is the number of trees in the random forest; S333, model application: Use the trained physical feature model to output the corresponding clamping force.

[0025] S4 includes: S41, clamping force calculation: Based on the feature vector extracted from the cabbage data, the physical feature model is used to output the corresponding target clamping force; S42, clamping force adjustment: Based on the target clamping force output by the physical feature model, the clamping force is adjusted in real time through the PID control algorithm, specifically including: Error calculation: Calculate the error between the target clamping force and the actual clamping force ; ,in, is the actual clamping force fed back; Proportional term calculation: Based on the error value , calculate the proportional term ,in is the proportional gain coefficient; Integral term calculation: Calculate the integral term by accumulating the error ,in is the integral gain coefficient, which represents the correction of error accumulation; Differential term calculation: Calculate the differential term based on the error trend ,in is the differential gain coefficient; Clamping force adjustment output: Combine the proportional term, integral term and differential term to get the adjusted clamping force ; ; Adjusted clamping force The clamping force is dynamically adjusted according to the characteristics and state of the cabbage to match the target clamping force; S43, clamping force feedback loop: Through the feedback loop, the actual clamping force of the clamping device is monitored in real time The feedback signal is input into the PID control algorithm to optimize the adjustment strategy of the clamping force. This feedback loop ensures that the clamping force of the clamping device is always kept within a reasonable range to avoid damage to the cabbage during the clamping process.

[0026] S5 includes: S51, data collection and monitoring: Multiple sensors installed on the clamping device collect real-time position data of the cabbage in the clamping device, specifically including: Position acquisition: Use position sensors (such as lidar, ultrasonic sensors, etc.) to obtain the position data of the cabbage in the clamping device in real time, ensuring that the clamping device can be accurately adjusted according to the real-time position of the cabbage in space. S52, dynamically adjusting the distribution of the clamping force and the adjustment speed: combining the collected position data of the cabbage in the clamping device, using a PID control algorithm to dynamically adjust the distribution of the clamping force and the adjustment speed; S53, stability optimization and correction: During the clamping process, the clamping force is adjusted in real time to prevent the cabbage from being damaged or falling due to excessive squeezing or unstable clamping.

[0027] S52 specifically includes: S521, clamping force distribution adjustment: Based on the current position of the cabbage in the clamping device, the clamping force distribution of each part of the clamping device is adjusted using a PID algorithm to ensure that the clamping force acts evenly on the surface of the cabbage, avoiding fracturing caused by concentrated clamping force or unstable clamping caused by uneven clamping force. The specific steps are as follows: Based on the position data, the target clamping force of each clamping point of the clamping device is calculated, and the output force of each point is adjusted in real time to adapt to the different shapes and sizes of cabbages; S522, speed control: Based on the position information of the cabbage, the PID control algorithm is used to adjust the rate of change of the clamping force to ensure that the clamping force adjustment process is smooth and without excessive fluctuations, and to prevent damage or unstable clamping caused by excessive changes in the clamping force during the adjustment process. The specific steps are as follows: The PID algorithm is used to calculate the increment of the clamping force adjustment and control the adjustment speed to adapt to the different position changes of the cabbage in the clamping device, ensuring that the clamping force transitions smoothly and remains consistent with the morphological changes of the cabbage.

[0028] S53 includes: S531, Clamping Force Optimization: Based on the cabbage's real-time position and clamping force feedback, the distribution and magnitude of the clamping force are optimized to ensure that the clamping force securely holds the cabbage without excessively compressing the cabbage surface. By adjusting the clamping force in real time, the clamping force is always maintained within the range that best suits the cabbage's shape and surface condition, minimizing mechanical damage to the cabbage during the clamping process. S532, Dynamic Correction: During the clamping process, the system makes real-time corrections to the clamping force, addressing unstable conditions such as uneven clamping force, excessive or insufficient clamping force, and other instabilities. By monitoring the cabbage's position within the clamping device, the system dynamically adjusts the force distribution at the clamping point to ensure the clamping force always matches the cabbage's real-time state and position. Correction is based on the difference between the clamping device's actual clamping force and the target clamping force. Through a feedback mechanism, the system adjusts the clamping device's operation to achieve the optimal clamping force distribution.

[0029] The present application encompasses any alternatives, modifications, equivalent methods and solutions made to the essence and scope of the present application. In order to make the public have a thorough understanding of the present application, specific details are described in the following preferred embodiments of the present application, and the present application can also be fully understood without the description of these details to those skilled in the art. In addition, in order to avoid unnecessary confusion to the essence of the present application, well-known methods, processes, procedures, elements and circuits, etc. are not described in detail.

[0030] The above is only the preferred embodiment of the present application, and it should be pointed out that for ordinary skilled in the art, without departing from the principles of the present application, a number of improvements and refinements can also be made, which should be considered as the protection scope of the present application.

Claims

1. A self-adaptive adjustment method for a vertical clamping and conveying device of a head vegetable harvester, characterized in that: The following steps are involved: S1, sensor data acquisition: multiple sensors installed on the clamping device collect cabbage data in real time, including cabbage diameter, weight and surface condition; S2, feature optimization processing: using deep learning image recognition technology to optimize the surface state of cabbage; S3, data processing and analysis: Based on the collected cabbage data, the physical characteristics of the cabbage are analyzed using a machine learning algorithm to generate a physical characteristic model of the cabbage; S4, clamping force adjustment: Based on the generated physical feature model, the clamping force is adjusted through an adaptive algorithm. The adjustment process includes real-time adjustment of the clamping force of the clamping device through a feedback loop to match the clamping force with the cabbage ball diameter and shape; S5, dynamic adjustment mechanism: Dynamically adjust the distribution of clamping force and the adjustment speed according to the position of the cabbage in the clamping device.

2. The adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 1, characterized in that: Said S1 comprises: S11, ball diameter collection: The ball diameter of the cabbage is measured in real time using a laser ranging sensor installed on the clamping device; S12, weight acquisition: The weight of the cabbage is measured in real time by installing multiple pressure sensors in the contact area of ​​the clamping device; S13, surface state acquisition: using a visual sensor (such as a camera or an infrared sensor) to acquire image data of the cabbage surface in real time to evaluate the surface state of the cabbage.

3. The self-adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 2, characterized in that: The S2 includes: S21, image data preprocessing: after obtaining the cabbage surface image data, preprocessing the collected image data, the preprocessing includes image denoising and contrast adjustment; S22, model training: build a convolutional neural network model and train the convolutional neural network model using a training dataset; S23, feature extraction and optimization: After the training is completed, the trained convolutional neural network model is applied to the cabbage image data collected in real time, and multiple features of the cabbage surface state are extracted through the convolutional neural network model; S24, surface state score generation: Based on the features extracted by the convolutional neural network model, a comprehensive score is given to the surface state of each cabbage.

4. The self-adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 3, characterized in that: The S3 includes: S31, data preprocessing: preprocessing the collected cabbage data, including missing data filling, outlier removal and standardization; S32, Feature Extraction and Feature Engineering: Based on the preprocessed cabbage data, physical features are extracted and feature vectors are constructed; S33, physical characteristic model construction: using the extracted feature vectors to establish a physical characteristic model of cabbage.

5. The self-adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 4, characterized in that: The steps of constructing the physical feature model include: S331, Model Construction: Use random forest algorithm to build physical feature model; S332, model training: using the training data set to train the constructed physical feature model; S333, model application: Use the trained physical feature model to output the corresponding clamping force.

6. The self-adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 5, characterized in that: The S4 includes: S41, clamping force calculation: Based on the feature vector extracted from the cabbage data, the physical feature model is used to output the corresponding target clamping force; S42, clamping force adjustment: according to the target clamping force output by the physical characteristic model, the clamping force is adjusted in real time through the PID control algorithm; S43, clamping force feedback loop: Through the feedback loop, the actual clamping force of the clamping device is monitored in real time, and the feedback signal is input into the PID control algorithm to optimize the clamping force adjustment strategy.

7. The self-adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 6, characterized in that: The S5 includes: S51, data collection and monitoring: using multiple sensors installed on the clamping device to collect the position data of the cabbage in the clamping device in real time; S52, dynamically adjusting the distribution of the clamping force and the adjustment speed: combining the collected position data of the cabbage in the clamping device, using a PID control algorithm to dynamically adjust the distribution of the clamping force and the adjustment speed; S53, stability optimization and correction: During the clamping process, the clamping force is adjusted in real time to prevent the cabbage from being damaged or falling due to excessive squeezing or unstable clamping.

8. The self-adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 7, characterized in that: The S52 specifically includes: S521, clamping force distribution adjustment: Based on the current position of the cabbage in the clamping device, the clamping force distribution of each part of the clamping device is adjusted using a PID algorithm to avoid fracturing caused by concentrated clamping force or unstable clamping caused by uneven clamping force; S522, adjusting speed control: Based on the position information of the cabbage, a PID control algorithm is used to adjust the rate of change of the clamping force to prevent damage or unstable clamping caused by excessive changes in the clamping force during the adjustment process.

9. The self-adaptive adjustment method of the vertical clamping and conveying device of the head vegetable harvester according to claim 8, characterized in that: The S53 includes: S531, Clamping Force Optimization: Based on the cabbage's real-time position and clamping force feedback, the clamping force is adjusted in real time to reduce mechanical damage to the cabbage during the clamping process. S532, dynamic correction: By monitoring the position changes of the cabbage in the clamping device, the force distribution at the clamping point is dynamically adjusted to ensure that the clamping force always matches the real-time state and position of the cabbage.

Images