Clamping force correction method and system based on camellia oleifera branch clamping manipulator and storage medium

Abstract

The invention provides a clamping force correction method and system based on an oil-tea tree branch clamping manipulator and a storage medium, and relates to the technical field of mechanization and automation, and the method comprises the steps of monitoring and analyzing oil-tea tree branches, an operation environment and a manipulator motion state, collecting key parameters, generating a clamping adaptability index and a clamping stability index, and correcting the clamping force of the oil-tea tree branches; the clamping adaptability index and the clamping stability index are combined, and a clamping control index is calculated; constructing a deep learning model, taking the historical key parameters as a training set, taking the corresponding historical manipulator clamping force as a label, training the model to obtain a clamping force prediction model, inputting the key parameters collected in real time into the model, outputting the predicted clamping force, combining the predicted clamping force with a clamping control index, and calculating the final clamping force. On the basis of obtaining key parameters in real time, the clamping adaptability index and the clamping stability index are generated, then the clamping control index is calculated, and the purpose of dynamically adjusting the clamping force is achieved.

Description

Technical Field

[0001] The invention relates to the field of mechanization and automation technology, and in particular to a clamping force correction method, system and storage medium based on a tea oil branch clamping manipulator. Background Art

[0002] In agricultural production, the management and picking of tea oil trees are key links that affect yield and quality. With the development of automation technology, manipulators are widely used in the clamping operations of tea oil trees. However, the existing manipulator clamping systems have many shortcomings in practical applications. Traditional clamping force control methods usually rely on fixed parameter settings and fail to fully consider the physical properties of tea oil branches, such as average diameter, hardness, and humidity. This results in the manipulator's clamping force being difficult to adapt to different branch conditions, and it is easy for insufficient clamping force to cause branches to slip, or excessive clamping force to cause branch damage. This fixed control strategy not only reduces the work efficiency of the manipulator, but also increases the risk of damage to trees during picking, affecting the growth and yield of tea oil.

[0003] In addition, current clamping force adjustment technologies often lack a real-time feedback mechanism. Most traditional systems are unable to dynamically monitor and adjust the clamping force during operation and lack the ability to adapt to environmental changes. For example, changes in ambient temperature and humidity will directly affect the physical properties of branches, but traditional control systems are unable to collect these key parameters in real time and adjust them accordingly. This lack of flexibility in design causes the manipulator to perform poorly in different operating environments and branch conditions, resulting in low production efficiency and waste of resources. Therefore, a new clamping force correction method is urgently needed to achieve intelligent and refined control of the clamping process of tea oil branches and improve the adaptability and working performance of the manipulator.

[0004] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0005] The purpose of the present invention is to provide a clamping force correction method, system and storage medium based on a tea oil branch clamping robot to solve the problems raised in the above-mentioned background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A method for correcting the clamping force of a tea tree branch clamping manipulator, the specific steps comprising:

[0008] Step 1: Monitor and analyze the tea tree branches, operating environment, and robot motion status, and collect key parameters, including the average diameter of the branches, branch hardness, branch humidity, target distance, robot gripping speed, gripping angle, and ambient temperature;

[0009] Step 2: Generate a clamping adaptability index based on the dimensionless average diameter, branch hardness, branch humidity and ambient temperature; generate a clamping stability index based on the dimensionless target distance, clamping angle and clamping speed; combine the clamping adaptability index and the clamping stability index, perform correlation analysis and calculate the clamping control index;

[0010] Step 3: Build a deep learning model, input the historical key parameters into the model as a training set, use the corresponding historical manipulator gripping force as a label, train the model, obtain a trained gripping force prediction model, input the key parameters collected in real time into the gripping force prediction model, and output the predicted gripping force;

[0011] Step 4: Combine the predicted clamping force with the clamping control index to obtain the corrected final clamping force.

[0012] Furthermore, the specific logic for collecting key parameters is as follows:

[0013] Use a laser rangefinder to measure the maximum and minimum diameters of the cross section of the branch to be clamped, and calculate the average diameter according to the following formula:

[0014]

[0015] Where D is the average diameter of the branches to be clamped, D max is the maximum diameter, D min is the minimum diameter;

[0016] Use a Vickers hardness tester to obtain the hardness parameters of the branch: select the cross section of the branch to be tested, ensure that the surface is flat and dry, place the indenter of the Vickers hardness tester perpendicular to the cross section of the branch, and apply a certain test force. After unloading, use a microscope to measure the diagonal length of the indentation, calculate the Vickers hardness value of the branch according to the following formula, and use the Vickers hardness value as the branch hardness:

[0017]

[0018] Where H is the hardness of the branch, F is the applied test force, and d is the diagonal length of the indentation;

[0019] The specific logic for obtaining the humidity of the branches is as follows: the branch to be tested is weighed by weighing, then dried in an oven to a constant weight, and weighed again. The humidity of the branch is calculated according to the following formula:

[0020]

[0021] Where M is the branch humidity, W wet is the mass of the branch in a wet state, W dry is the mass of the branch in dry state;

[0022] The target distance refers to the actual distance between the manipulator and the target object, which is calculated according to the following formula:

[0023]

[0024] Where P is the target distance, (x target ,y target ,z target ) is the coordinate of the target object in three-dimensional space, (x init ,y init ,z init ) is the initial position coordinate of the manipulator in three-dimensional space;

[0025] Using sensors, the clamping speed V and the clamping angle θ are acquired in real time, wherein the clamping angle refers to the angle between the manipulator and the horizontal plane during the clamping process;

[0026] A temperature sensor is used to collect the ambient temperature of the manipulator when it is performing a clamping operation, denoted as T.

[0027] Furthermore, the clamping adaptability index is generated based on the following formula:

[0028]

[0029] Among them, ADAP is the clamping adaptability index, D is the average diameter of the branch, H is the branch hardness, M is the branch humidity, T is the ambient temperature, and T ref is the ambient temperature under ideal conditions, e is a natural constant, k1, k2, k3 and k4 are preset proportional coefficients, and k3>k2=k1>k4>0;

[0030] The clamping stability index is generated based on the following formula:

[0031]

[0032] Where STA is the clamping stability index, V is the clamping speed, θ is the clamping angle, and θ ref is the optimal gripping angle of the manipulator, P is the target distance, k5, k6 and k7 are preset proportional coefficients, and k5>k6>k7>0;

[0033] The clamping control index is calculated based on the following formula:

[0034]

[0035] Among them, CON is the clamping control index, ADAP is the clamping adaptability index, STA is the clamping stability index, D is the average diameter of the branches, e is the natural constant, and μ is the adjustment coefficient, which is used to adjust and weigh the influence of other parameters, and μ>0.

[0036] Furthermore, the process of constructing the deep learning model specifically includes:

[0037] Collect n groups of historical key parameters and corresponding clamping forces of the manipulator's clamping operation, wherein the key parameters include the average diameter of the branch, the hardness of the branch, the humidity of the branch, the target distance, the clamping speed of the manipulator, the clamping angle and the ambient temperature, and preprocess the collected data to remove outliers and missing values. Build a deep learning model, use the historical key parameters as a training set, and use the clamping force corresponding to each group of historical key parameters as a label to train the model to obtain a trained clamping force prediction model, input the key parameters collected in real time into the clamping force prediction model, and output the predicted clamping force of the current manipulator performing the clamping operation.

[0038] Furthermore, the predicted clamping force is combined with the clamping control index to obtain the corrected final clamping force, based on the following formula:

[0039] F corrected =F predicted *(1+k f *CON)

[0040] Among them, F corrected is the final clamping force, F predicted To predict the clamping force, CON is the clamping control index, k f is the clamping force adjustment coefficient, which is determined based on actual test data.

[0041] The present invention further provides a clamping force correction system based on a tea tree branch clamping manipulator, wherein the clamping force correction system based on a tea tree branch clamping manipulator is used to execute the above-mentioned clamping force correction method based on a tea tree branch clamping manipulator, and comprises:

[0042] The data acquisition module is used to monitor and analyze the tea tree branches, operating environment and robot motion status, and collect key parameters, including the average diameter of the branches, branch hardness, branch humidity, target distance, robot gripping speed, gripping angle and ambient temperature;

[0043] The index calculation module is used to generate a clamping adaptability index according to the dimensionless average diameter, branch hardness, branch humidity and ambient temperature; generate a clamping stability index according to the dimensionless target distance, clamping angle and clamping speed, combine the clamping adaptability index and the clamping stability index, perform correlation analysis and calculate the clamping control index;

[0044] The model building module is used to build a deep learning model, input the historical key parameters into the model as a training set, use the corresponding historical manipulator clamping force as a label, train the model, obtain a trained clamping force prediction model, input the key parameters collected in real time into the clamping force prediction model, and output the predicted clamping force;

[0045] The force correction module is used to combine the predicted clamping force with the clamping control index to obtain a corrected final clamping force.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention effectively solves the problem of inaccurate clamping force of traditional manipulators by monitoring and analyzing multiple factors and combining deep learning models. Based on real-time acquisition of key parameters, this method generates a clamping adaptability index and a clamping stability index, and then calculates a clamping control index, thereby achieving the goal of dynamically adjusting the clamping force. In this way, not only the reliability and stability of the clamping process are improved, unnecessary damage to the branches is avoided, but also the operating efficiency of the manipulator is improved, the precise picking and efficient management of tea oil trees are promoted, and the development of agricultural automation technology is effectively promoted. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0049] Figure 2-3 It is a three-dimensional schematic diagram of the manipulator; in the figure: base 1, connecting block 2, driving device 3, mechanical claw 4, connecting rod 5, contact 6;

[0050] Figure 4 It is a schematic diagram of the system module of the present invention. DETAILED DESCRIPTION

[0051] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0052] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0053] Example:

[0054] See also Figure 1-Figure 3 , the present invention provides a technical solution:

[0055] A method for correcting the clamping force of a tea tree branch clamping manipulator, the specific steps comprising:

[0056] Step 1: Monitor and analyze the tea tree branches, operating environment, and robot motion status, and collect key parameters, including the average diameter of the branches, branch hardness, branch humidity, target distance, robot gripping speed, gripping angle, and ambient temperature;

[0057] In this embodiment, the key parameters are collected based on the following specific logic:

[0058] Use a laser rangefinder to measure the maximum and minimum diameters of the cross section of the branch to be clamped, and calculate the average diameter according to the following formula:

[0059]

[0060] Where D is the average diameter of the branches to be clamped, D max is the maximum diameter, D min is the minimum diameter;

[0061] Use a Vickers hardness tester to obtain the hardness parameters of the branch: select the cross section of the branch to be tested, ensure that the surface is flat and dry, place the indenter of the Vickers hardness tester perpendicular to the cross section of the branch, and apply a certain test force. After unloading, use a microscope to measure the diagonal length of the indentation, calculate the Vickers hardness value of the branch according to the following formula, and use the Vickers hardness value as the branch hardness:

[0062]

[0063] Where H is the hardness of the branch, F is the applied test force, and d is the diagonal length of the indentation;

[0064] The specific logic for obtaining the humidity of the branches is as follows: the branch to be tested is weighed by weighing, then dried in an oven to a constant weight, and weighed again. The humidity of the branch is calculated according to the following formula:

[0065]

[0066] Where M is the branch humidity, W wet is the mass of the branch in a wet state, W dry is the mass of the branch in dry state;

[0067] The target distance refers to the actual distance between the manipulator and the target object, which is calculated according to the following formula:

[0068]

[0069] Where P is the target distance, (x target ,y target ,z target ) is the coordinate of the target object in three-dimensional space, (x init ,y init ,z init ) is the initial position coordinate of the manipulator in three-dimensional space;

[0070] Using sensors, the clamping speed V and the clamping angle θ are acquired in real time, wherein the clamping angle refers to the angle between the manipulator and the horizontal plane during the clamping process;

[0071] A temperature sensor is used to collect the ambient temperature of the manipulator when it is performing a clamping operation, denoted as T.

[0072] The advantage of step 1 is that it provides a scientific basis for the subsequent clamping force correction by collecting a series of key parameters, such as the average diameter of the branch, hardness, humidity, target distance, clamping speed, clamping angle and ambient temperature. This multi-dimensional data collection method can effectively reflect the changes in different environments and branch states, overcome the shortcomings of the existing technology that only relies on fixed parameters for clamping, and ensure that the dynamic adjustment of the clamping force has higher accuracy and reliability.

[0073] Compared with the prior art, the beneficial effect of step 1 is reflected in its improved real-time and adaptability of the clamping process. Through real-time monitoring of key parameters, it is possible to respond to changes in the oil tea tree branches and the environment in a timely manner, thereby dynamically adjusting the clamping force during the clamping process, reducing the risk of damage to the branches, and improving the work efficiency of the manipulator. In the present application, the use of this step can provide a solid data foundation for the overall solution, ensuring that the subsequent generation of clamping adaptability and stability indexes is more scientific and reasonable, thereby achieving more effective clamping force correction and promoting accurate picking and efficient management of oil tea trees.

[0074] Step 2: Generate a clamping adaptability index based on the dimensionless average diameter, branch hardness, branch humidity and ambient temperature; generate a clamping stability index based on the dimensionless target distance, clamping angle and clamping speed; combine the clamping adaptability index and the clamping stability index, perform correlation analysis and calculate the clamping control index;

[0075] In this embodiment, the clamping adaptability index is generated based on the following formula:

[0076]

[0077] Among them, ADAP is the clamping adaptability index, D is the average diameter of the branch, H is the branch hardness, M is the branch humidity, T is the ambient temperature, and T ref is the ambient temperature under ideal conditions, e is a natural constant, k1, k2, k3 and k4 are preset proportional coefficients, and k3>k2=k1>k4>0. This is because the humidity of the branches directly affects their physical properties, such as toughness and elasticity. Too high or too low humidity will affect the clamping effect, so humidity is given a higher weight to ensure that it occupies a dominant position in the adaptability assessment; the diameter and hardness of the branches are both important parameters affecting the clamping stability. The diameter size affects the contact area of ​​the clamping point, while the hardness determines the deformation resistance of the branches during clamping. These two factors are equally important in the clamping process, so setting them as equal coefficients can reasonably reflect their relative importance to the clamping adaptability; the effect of temperature on the clamping adaptability is relatively small, because although temperature can affect the physical properties of the material, its impact is usually not as direct and significant as humidity, diameter and hardness. Therefore, the lowest weight given to the temperature difference reflects its secondary position in the adaptability assessment.

[0078] ADAP is used to measure the suitability of the clamping force required for the tea tree branches. The larger the ADAP, the more suitable it is to apply a larger clamping force to the tea tree branches. Larger branch diameters usually require more clamping force to ensure a stable clamping. Therefore, as the diameter D increases, it indicates that a higher clamping force correction is required, and the larger the ADAP. The greater the branch hardness H, the tea tree can withstand a larger clamping force, and a larger clamping force can ensure stable clamping. Therefore, the larger H is, the more suitable it is to apply a larger clamping force to the tea tree, and the larger the ADAP is. When |TT ref |When it increases, it means that the deviation between the ambient temperature and the ideal value increases, that is, the ambient temperature is too cold or too hot. Under overcold conditions, the surface of the tea tree will become brittle. If the clamping force applied is too large, it will cause damage to the tea tree. Therefore, it is more inappropriate to apply a large clamping force to the tea tree branches, and the smaller the ADAP. Similarly, under overheating conditions, the surface of the tea tree will dry and dehydrate. If the clamping force applied is too large, it will cause damage to the tea tree. Therefore, it is more inappropriate to apply a large clamping force to the tea tree branches, and the smaller the ADAP. When low-hardness branches are subjected to clamping force, they are easily compressed, deformed, or even damaged, thus affecting the health and subsequent growth of the tree. This may cause branch cracking, peeling, or other forms of physical damage. When the humidity of the branch is higher, it means that the surface of the branch has more moisture, and it is more likely to slide during clamping, resulting in unstable clamping. Therefore, in order to maintain a stable clamping, it is more appropriate to apply a large clamping force to the tea tree, and the larger the ADAP. This shows that D, H, and M are positively correlated with ADAP. |TT ref |Negatively correlated with ADAP.

[0079] The logarithmic form is used to deal with nonlinear relationships, which slows down the drastic changes caused by the increase in diameter D, making the clamping adaptability index respond more smoothly to the diameter and avoiding excessive exponential growth; the square form emphasizes the significant effect of branch hardness H on clamping stability. The square form makes the improvement of the adaptability index more obvious when the hardness increases, reflecting the weighted effect of hardness on the clamping effect; the exponential form reflects the significance of branch humidity M on clamping adaptability. The increase in humidity will significantly improve the clamping ability, and the exponential form effectively amplifies this effect; the square root form smoothes the effect of temperature difference on the adaptability index, avoiding excessive improvement of the adaptability index when the temperature is close to the ideal value, and reasonably reflecting the effect of ambient temperature on the clamping effect.

[0080] The clamping stability index is generated based on the following formula:

[0081]

[0082] Where STA is the clamping stability index, V is the clamping speed, θ is the clamping angle, and θ refis the optimal clamping angle of the manipulator, P is the target distance, k5, k6 and k7 are preset proportional coefficients, and k5>k6>k7>0. This is because a higher clamping speed may lead to unstable clamping, especially in the process of contacting and clamping objects, rapid clamping actions may increase the risk of object sliding or displacement. Therefore, the clamping speed has the most significant effect on clamping stability; changes in the clamping angle will affect the distribution of the clamping force, and thus affect the clamping stability. Although the influence of the clamping angle is important, its influence on clamping stability is usually weaker than that of the clamping speed, so it is given a medium weight; the larger the target distance, the greater the difficulty and uncertainty of clamping the object, but compared with the clamping speed and clamping angle, its influence on stability is usually smaller. STA is used to measure the stability of the clamping process. The smaller STA is, the lower the clamping stability is, and the clamping force needs to be increased to ensure the stability of the clamping. When the clamping speed V increases, the clamping action may not be smooth enough, increasing the risk of object sliding, thus affecting the stability of clamping and reducing STA. A larger deviation angle may lead to uneven distribution of clamping force, increasing the risk of clamping failure and reducing the stability of clamping. Therefore, when |θ-θ ref | increases, STA will decrease; a longer target distance may lead to increased uncertainty in clamping, increase the difficulty of clamping, and thus affect the stability of clamping. Therefore, when the target distance P increases, STA will decrease accordingly; that is, V, |θ-θ ref |, P and STA are negatively correlated.

[0083] The influence of the clamping speed V on stability is inversely proportional, so the square root form can smooth the impact of speed changes and avoid a sharp drop in stability caused by high speed; the cube root form can effectively reduce the impact of angle difference on stability, reflecting that small deviations significantly improve stability, while large deviations have little effect; the impact of the target distance P is nonlinear, and the use of logarithmic form can reduce the impact of distance on stability, ensuring that as the distance increases, the stability gradually decreases but the rate slows down.

[0084] The clamping control index is calculated based on the following formula:

[0085]

[0086] Among them, CON is the clamping control index, ADAP is the clamping adaptability index, STA is the clamping stability index, D is the average diameter of the branch, e is the natural constant, and μ is the adjustment coefficient, which is used to adjust and weigh the influence of other parameters, μ∈[0.1,2]. CON is used to correct the predicted clamping force output by the model, and the larger the CON, the more it is necessary to increase the clamping force on the basis of the predicted clamping force to ensure the stability of the clamping. A high clamping adaptability index indicates that the manipulator can better adjust its clamping strategy under different clamping conditions, so as to complete the clamping task more effectively, that is, when ADAP increases, CON will increase accordingly; higher stability means that it is not easy to slip or lose stability during the clamping process, which enhances the reliability of the clamping and improves the control ability. Therefore, the smaller the STA, the lower the clamping stability, the more it needs to increase the clamping force, and the larger the CON; larger branch diameters usually require more clamping force to ensure stable clamping, so when D increases, CON will increase; that is, ADAP, D and CON are positively correlated, and STA and CON are negatively correlated.

[0087] The advantage of step 2 is that by dimensionlessly processing the key parameters collected, the clamping adaptability index and the clamping stability index are generated, thereby quantifying the multi-dimensional information required for clamping force correction. This method can comprehensively consider the relationship between the key parameters to ensure that the adjustment of the clamping force is more scientific and accurate, especially in complex environments and various tree branch states, and can effectively avoid clamping force errors caused by fluctuations in a single parameter.

[0088] Compared with the prior art, the beneficial effect of step 2 is that it introduces two indexes, clamping adaptability and stability, so that the control of the clamping force does not only rely on simple fixed standards, but is dynamically adjusted based on real-time data. This innovation significantly improves the flexibility and adaptability of the manipulator in practical applications, reduces the risk of damage to branches, and enhances picking efficiency. In the present application, the use of this step can provide a more refined control mechanism for the overall solution, thereby promoting the intelligence of the clamping force correction process and improving the accuracy and safety of tea oil tree picking.

[0089] Step 3: Build a deep learning model, input the historical key parameters into the model as a training set, use the corresponding historical manipulator gripping force as a label, train the model, obtain a trained gripping force prediction model, input the key parameters collected in real time into the gripping force prediction model, and output the predicted gripping force;

[0090] In this embodiment, the process of building a deep learning model specifically includes:

[0091] Collect n groups of historical key parameters and corresponding clamping forces of the manipulator's clamping operation, wherein the key parameters include the average diameter of the branch, the hardness of the branch, the humidity of the branch, the target distance, the clamping speed of the manipulator, the clamping angle and the ambient temperature, and preprocess the collected data to remove outliers and missing values. Build a deep learning model, use the historical key parameters as a training set, and use the clamping force corresponding to each group of historical key parameters as a label to train the model to obtain a trained clamping force prediction model, input the key parameters collected in real time into the clamping force prediction model, and output the predicted clamping force of the current manipulator performing the clamping operation.

[0092] When obtaining historical key parameters and corresponding clamping forces, key parameters and clamping forces under conditions of stable clamping and no damage to the tea oil tree are obtained.

[0093] The advantage of step 3 is that by building a deep learning model, historical key parameters are associated with the corresponding clamping force, thereby realizing intelligent prediction of clamping force. This process enables the manipulator to quickly and accurately adjust the clamping force according to the real-time collected data to adapt to different branch characteristics and environmental conditions. This intelligent processing method significantly improves the accuracy and efficiency of clamping force correction and reduces the need for manual intervention.

[0094] Compared with the prior art, the beneficial effect of step 3 is that traditional methods often rely on experience or fixed parameters to adjust the clamping force, which is prone to mistakes or inadaptability. Through the training of the deep learning model, it is possible to automatically learn the optimal clamping force adjustment strategy based on a large amount of historical data, thereby improving the autonomous adaptability of the manipulator. This data-driven approach makes the adjustment of the clamping force more flexible and precise, significantly reducing the risk of damage to the branches. In the present application scheme, the use of this step can provide a higher level of intelligent support for the overall scheme, and realize the automation and optimization of the clamping force correction process. This not only improves the efficiency and safety of tea oil tree picking, but also lays a solid foundation for the future development of smart agriculture and promotes the process of agricultural mechanization and intelligence.

[0095] Step 4: Combine the predicted clamping force with the clamping control index to obtain the corrected final clamping force;

[0096] In this embodiment, the predicted clamping force is combined with the clamping control index to obtain the corrected final clamping force, and the formula based on it is as follows:

[0097] F corrected =F pred ted *(1+k f *CON)

[0098] Among them, F cprrected is the final clamping force, F ptedictedTo predict the clamping force, CON is the clamping control index, k f is the clamping force adjustment coefficient, k f The specific value is dynamically updated according to the clamping situation. Specifically, k is updated according to the actual situation of each clamping operation. f Update the value:

[0099] If the clamping is not stable, increase k f value, adjust it to the range of 0.3 to 0.5 to increase the final clamping force and ensure the stability of clamping;

[0100] If the tea tree has clamping damage, reduce k f value, adjust it to a value between 0.1 and 0.2 to reduce the final clamping force and reduce damage to the plants.

[0101] The advantage of step 4 is that the predicted clamping force is combined with the clamping control index to form the corrected final clamping force. This process makes effective use of real-time data, so that the clamping force can be flexibly adjusted according to specific operating conditions and environmental changes. This dynamic adjustment mechanism significantly improves the clamping accuracy of the manipulator, ensures that the optimal clamping force can be provided under different conditions, and reduces the risk of damage to the tea tree branches.

[0102] Compared with the prior art, the beneficial effects of step 4 are reflected in its comprehensiveness and adaptability. Traditional methods usually rely on fixed clamping force standards, which are often unable to cope with complex environments or changes in branch characteristics. By introducing the concept of clamping control index, the correction of the clamping force can take into account a variety of influencing factors, ensuring that the adjustment of the clamping force is more scientific and reasonable, and improving the performance and reliability of the manipulator in practical applications. In the present application scheme, the use of this step can provide a higher level of flexibility and intelligent support for the overall scheme, and promote the automation and refinement of the clamping force correction process. This not only improves the picking efficiency and safety of tea oil trees, but also provides a solid technical foundation for the promotion of manipulators in other agricultural applications, which helps to achieve a wider range of smart agricultural development.

[0103] The manipulator includes a base 1 for fixing, a driving device 3 connected to the base 1 via a connecting block 2, and a mechanical claw 4 for clamping an object to be clamped. The driving device 3 drives the mechanical claw 4 to clamp via a connecting rod 5, and the mechanical claw 4 has a plurality of contacts 6 on one side for clamping the object to be clamped.

[0104] See also Figure 4 , a clamping force correction system based on a tea tree branch clamping manipulator, comprising:

[0105] The data acquisition module is used to monitor and analyze the tea tree branches, operating environment and robot motion status, and collect key parameters, including the average diameter of the branches, branch hardness, branch humidity, target distance, robot gripping speed, gripping angle and ambient temperature;

[0106] The index calculation module is used to generate a clamping adaptability index according to the dimensionless average diameter, branch hardness, branch humidity and ambient temperature; generate a clamping stability index according to the dimensionless target distance, clamping angle and clamping speed, combine the clamping adaptability index and the clamping stability index, perform correlation analysis and calculate the clamping control index;

[0107] The model building module is used to build a deep learning model, input the historical key parameters into the model as a training set, use the corresponding historical manipulator clamping force as a label, train the model, obtain a trained clamping force prediction model, input the key parameters collected in real time into the clamping force prediction model, and output the predicted clamping force;

[0108] The force correction module is used to combine the predicted clamping force with the clamping control index to obtain a corrected final clamping force.

[0109] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0110] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0111] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0112] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for correcting the clamping force of a tea tree branch clamping manipulator, characterized in that: The specific steps include: Step 1: Monitor and analyze the tea tree branches, operating environment, and robot motion status, and collect key parameters, including the average diameter of the branches, branch hardness, branch humidity, target distance, robot gripping speed, gripping angle, and ambient temperature; Step 2: Generate a clamping adaptability index based on the dimensionless average diameter, branch hardness, branch humidity and ambient temperature; generate a clamping stability index based on the dimensionless target distance, clamping angle and clamping speed; combine the clamping adaptability index and the clamping stability index, perform correlation analysis and calculate the clamping control index; Step 3: Build a deep learning model, input the historical key parameters into the model as a training set, use the corresponding historical manipulator gripping force as a label, train the model, obtain a trained gripping force prediction model, input the key parameters collected in real time into the gripping force prediction model, and output the predicted gripping force; Step 4: Combine the predicted clamping force with the clamping control index to obtain the corrected final clamping force.

2. According to claim 1, a method for correcting the clamping force of a tea tree branch clamping manipulator is characterized in that: The specific logic for collecting key parameters is as follows: Use a laser rangefinder to measure the maximum and minimum diameters of the cross section of the branch to be clamped, and calculate the average diameter according to the following formula: Where D is the average diameter of the branches to be clamped, D max is the maximum diameter, D min is the minimum diameter; Use a Vickers hardness tester to obtain the hardness parameters of the branch: select the cross section of the branch to be tested, ensure that the surface is flat and dry, place the indenter of the Vickers hardness tester perpendicular to the cross section of the branch, and apply a certain test force. After unloading, use a microscope to measure the diagonal length of the indentation, calculate the Vickers hardness value of the branch according to the following formula, and use the Vickers hardness value as the branch hardness: Where H is the hardness of the branch, F is the applied test force, and d is the diagonal length of the indentation; The specific logic for obtaining the humidity of the branches is as follows: the branch to be tested is weighed by weighing method, then dried in an oven to constant weight, and weighed again. The humidity of the branch is calculated according to the following formula: Where M is the branch humidity, W wet is the mass of the branch in a wet state, W dry is the mass of the branch in dry state; The target distance refers to the actual distance between the manipulator and the target object, which is calculated according to the following formula: Where P is the target distance, (x target ,y target ,z target ) is the coordinate of the target object in three-dimensional space, (x init ,y init ,z init ) is the initial position coordinate of the manipulator in three-dimensional space; Using sensors, the clamping speed V and the clamping angle θ are acquired in real time, wherein the clamping angle refers to the angle between the manipulator and the horizontal plane during the clamping process; A temperature sensor is used to collect the ambient temperature of the manipulator when it is performing a clamping operation, denoted as T.

3. The method for correcting the clamping force of the oil-tea tree branch clamping manipulator according to claim 2 is characterized in that: The clamping adaptability index is generated based on the following formula: Among them, ADAP is the clamping adaptability index, D is the average diameter of the branch, H is the branch hardness, M is the branch humidity, T is the ambient temperature, and T ref is the ambient temperature under ideal conditions, e is a natural constant, k1, k2, k3 and k4 are preset proportional coefficients, and k3>k2=k1>k4>0; The clamping stability index is generated based on the following formula: Where STA is the clamping stability index, V is the clamping speed, θ is the clamping angle, and θ ref is the optimal gripping angle of the manipulator, P is the target distance, k5, k6 and k7 are preset proportional coefficients, and k5>k6>k7>0; The clamping control index is calculated based on the following formula: Among them, CON is the clamping control index, ADAP is the clamping adaptability index, STA is the clamping stability index, D is the average diameter of the branches, e is the natural constant, and μ is the adjustment coefficient, which is used to adjust and weigh the influence of other parameters, and μ>0.

4. The method for correcting the clamping force of a tea tree branch clamping manipulator according to claim 1 is characterized in that: The process of building a deep learning model specifically includes: Collect n groups of historical key parameters and corresponding clamping forces of the manipulator's clamping operation, wherein the key parameters include the average diameter of the branch, the hardness of the branch, the humidity of the branch, the target distance, the clamping speed of the manipulator, the clamping angle and the ambient temperature, and preprocess the collected data to remove outliers and missing values. Build a deep learning model, use the historical key parameters as a training set, and use the clamping force corresponding to each group of historical key parameters as a label to train the model to obtain a trained clamping force prediction model, input the key parameters collected in real time into the clamping force prediction model, and output the predicted clamping force of the current manipulator performing the clamping operation.

5. The method for correcting the clamping force of the oil-tea tree branch clamping manipulator according to claim 4 is characterized in that: The predicted clamping force is combined with the clamping control index to obtain the corrected final clamping force based on the following formula: F corrected =F predicted *(1+k f *CON) Among them, F corrected is the final clamping force, F predicted To predict the clamping force, CON is the clamping control index, k f is the clamping force adjustment coefficient, which is determined based on actual test data.

6. A clamping force correction system based on a tea tree branch clamping manipulator, characterized in that: The clamping force correction system based on the oil-tea branch clamping manipulator is used to execute the clamping force correction method based on the oil-tea branch clamping manipulator according to any one of claims 1 to 5, comprising: The data acquisition module is used to monitor and analyze the tea tree branches, operating environment and robot motion status, and collect key parameters, including the average diameter of the branches, branch hardness, branch humidity, target distance, robot gripping speed, gripping angle and ambient temperature; The index calculation module is used to generate a clamping adaptability index according to the dimensionless average diameter, branch hardness, branch humidity and ambient temperature; generate a clamping stability index according to the dimensionless target distance, clamping angle and clamping speed, combine the clamping adaptability index and the clamping stability index, perform correlation analysis and calculate the clamping control index; The model building module is used to build a deep learning model, input the historical key parameters into the model as a training set, use the corresponding historical manipulator clamping force as a label, train the model, obtain a trained clamping force prediction model, input the key parameters collected in real time into the clamping force prediction model, and output the predicted clamping force; The force correction module is used to combine the predicted clamping force with the clamping control index to obtain a corrected final clamping force.

7. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by the processor, it implements the clamping force correction method based on the oil-tea tree branch clamping robot as described in any one of claims 1 to 5.

Images