A rice seed taking mechanism control method and system based on flexible bionics principle

Abstract

This invention discloses a control method and system for a rice seed-harvesting mechanism based on the principle of flexible bionics, belonging to the field of agricultural machinery control technology. The method includes: acquiring ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seed; identifying the drift state of the pressure sensor data; identifying changes in rice seed characteristics; if the pressure sensor data drift state indicates data drift, adjusting the current motor current value of the clamping mechanism; if the rice seed characteristic change information indicates a change in characteristics, adjusting the current pressure value of the clamping mechanism to achieve flexible seed harvesting. This invention can combine the pressure sensor data drift state and the rice seed characteristic change information to adjust the control parameters of the clamping mechanism, thereby achieving control of the rice seed-harvesting mechanism and improving accuracy and reliability.

Description

Technical Field

[0001] This invention relates to the field of agricultural machinery control technology, and in particular to a control method and system for a rice seed-harvesting mechanism based on the principle of flexible bionics. Background Technology

[0002] In the field of mechanized rice cultivation, flexible bionic seed-harvesting mechanisms are widely used due to their mimicking the grasping characteristics of organisms. These mechanisms use built-in pressure sensors to monitor the contact force between the grippers and the seeds in real time, and dynamically adjust the gripping force based on feedback data, aiming to achieve non-destructive seed harvesting. However, the field operating environment is complex and variable. Seed-harvesting mechanisms are exposed to harsh environments containing moisture, mud, and fine sand for extended periods, causing the protective coating on the pressure sensor to gradually fail under continuous physical wear. Moisture and tiny particles then penetrate the sensor's internal encapsulation structure, forming weak conductive paths near the sensitive element, causing slight and unstable drift in the output signal. This drift manifests as a persistently low pressure reading, and its fluctuation range is similar to the sensor's inherent noise and ambient temperature fluctuations, making it difficult to effectively identify using simple threshold judgments or low-pass filtering and other conventional signal processing methods. In this situation, the adaptive optimization mechanism built into the control system may incorrectly attribute phenomena such as decreased seed-harvesting efficiency and prolonged adjustment time to changes in seed batch characteristics, such as mistakenly believing that the seeds have become slipperier or harder, rather than a sensor malfunction. This mechanism then automatically and progressively increases the target gripping force setting in an attempt to compensate for perceived insufficient gripping force. This mechanism-level adjustment, combined with low sensor reading deviations, causes the actual pressure applied by the gripper to the seed to far exceed the design limit. For rice seeds with thin seed coats, exposed embryos, or low moisture content, this continuous excessive force can cause latent damage such as microscopic cracks in the seed coat and indentations inside the endosperm, significantly reducing germination rate and seedling vigor, resulting in imperceptible cumulative damage, ultimately leading to an unacceptably high damage rate, low control accuracy, and low reliability.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the Invention

[0004] The main objective of this invention is to propose a control method and system for a rice seed-harvesting mechanism based on the principle of flexible bionics. This method can combine the drift state of pressure sensor data and the information on changes in rice seed characteristics to adjust the control parameters of the clamping mechanism, thereby achieving control of the rice seed-harvesting mechanism and improving accuracy and reliability.

[0005] On one hand, embodiments of the present invention provide a control method for a rice seed-harvesting mechanism based on the principle of flexible bionics, comprising the following steps: The system acquires ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seeds. The clamping distance data is used to reflect the distance between the two clamping surfaces of the clamping mechanism. Based on the pressure data, the clamping distance data, and the motor current data, identify the pressure sensor data drift state; Based on the ambient temperature data and the ambient humidity data, identify information on changes in rice seed characteristics; If the pressure sensor data drift status indicates that data drift exists, then the current motor current value of the clamping mechanism is adjusted based on the pressure data, the motor current data, and the clamping distance data. If the rice seed characteristic change information indicates that there is a characteristic change, the current pressure value of the clamping mechanism is adjusted to achieve flexible seed extraction.

[0006] In some embodiments, identifying the pressure sensor data drift state based on the pressure data, the clamping distance data, and the motor current data includes: Obtain the target pressure value and the reference clamping distance range; Calculate the lower pressure safety limit based on the target pressure value and the pressure safety margin; Calculate the average pressure value based on the pressure data; The average pressure value is compared with the lower pressure safety limit to obtain the first pressure comparison result; The clamping distance data is compared with the reference clamping distance range to obtain the first clamping distance comparison result; Based on the first pressure comparison result and the first clamping distance comparison result, the data drift state of the pressure sensor is identified.

[0007] In some embodiments, identifying the pressure sensor data drift state based on the first pressure comparison result and the first clamping distance comparison result includes: Obtain historical average current values; Calculate the upper limit of current safety based on the historical average current value and current safety margin; Calculate the current average current value based on the motor current data; The current average current value is compared with the current safety upper limit to obtain the current comparison result; Based on the first pressure comparison result, the first clamping distance comparison result, and the current comparison result, the data drift state of the pressure sensor is identified.

[0008] In some embodiments, identifying rice seed characteristic change information based on the ambient temperature data and the ambient humidity data includes: Acquire historical temperature and humidity data; The pressure safety range is determined based on the target pressure value and the pressure safety margin; The average pressure value is compared with the pressure safety range to obtain a second pressure comparison result; Calculate the historical average temperature based on the historical temperature data; Calculate the historical average humidity based on the historical humidity data; The ambient temperature data is compared with the historical average temperature to obtain the temperature comparison result; The ambient humidity data is compared with the historical average humidity to obtain the humidity comparison result; Based on the second pressure comparison result, the current comparison result, the temperature comparison result, and the humidity comparison result, the information on the changes in the characteristics of the rice seeds is identified.

[0009] In some embodiments, identifying the rice seed characteristic change information based on the second pressure comparison result, the current comparison result, the temperature comparison result, and the humidity comparison result includes: Get the maximum clamping distance; Calculate the average clamping distance based on the clamping distance data; The average clamping distance and the upper limit of the clamping distance are compared to obtain the second clamping distance comparison result; Based on the second pressure comparison result, the current comparison result, the temperature comparison result, the humidity comparison result, and the second clamping distance comparison result, the information on the changes in the characteristics of the rice seeds is identified.

[0010] In some embodiments, adjusting the current motor current value of the clamping mechanism based on the pressure data, the motor current data, and the clamping distance data includes: Extract the current pressure value from the pressure data; Extract the current motor current value from the motor current data; Extract the current clamping distance value from the clamping distance data; The current adjustment amount is determined based on the current pressure value, the current motor current value, and the current clamping distance value. The current value of the clamping mechanism is adjusted according to the current adjustment amount.

[0011] In some embodiments, determining the current adjustment amount based on the current pressure value, the current motor current value, and the current clamping distance value includes: Calculate the expected pressure value based on the current motor current value and the current clamping distance value; Calculate the pressure deviation value based on the expected pressure value and the current pressure value; Based on the pressure deviation value, the current adjustment amount is determined by consulting the mapping table between the pressure deviation value and the current adjustment amount.

[0012] In some embodiments, adjusting the current pressure value of the clamping mechanism includes: Obtain rice seed information, including the hardness of the seed coat; Based on the hardness of the seed shell, the pressure that can be withheld is determined by consulting a mapping table between the hardness of the seed shell and the pressure that can be withheld. The current pressure value of the clamping mechanism is adjusted according to the tolerable pressure value.

[0013] In some embodiments, identifying the pressure sensor data drift state based on the first pressure comparison result and the first clamping distance comparison result includes: Impedance data of the clamping surface region of the clamping mechanism is acquired using an impedance spectroscopy measuring element. Vibration data of the clamping mechanism during the clamping process are collected by a vibration measuring element; The impedance data is subjected to spectral analysis to extract impedance features, which include impedance amplitude and phase angle. Perform time-domain analysis on the vibration data to calculate the vibration attenuation rate; Based on the impedance characteristics and the vibration attenuation rate, the clamping force is calculated, which represents the clamping force applied by the clamping mechanism to the rice seed. The clamping force of the clamping face and the lower limit of the pressure safety are compared to obtain the third pressure comparison result; Based on the first pressure comparison result, the first clamping distance comparison result, and the third pressure comparison result, the data drift state of the pressure sensor is identified.

[0014] On the other hand, embodiments of the present invention provide a rice seed-harvesting mechanism control system based on the principle of flexible bionics, comprising: The data acquisition module is used to acquire ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seed. The clamping distance data is used to reflect the distance between the two clamping surfaces of the clamping mechanism. The sensor data drift identification module is used to identify the pressure sensor data drift state based on the pressure data, the clamping distance data, and the motor current data. The seed characteristic change identification module is used to identify rice seed characteristic change information based on the ambient temperature data and the ambient humidity data. The motor current adjustment module is used to adjust the current motor current value of the clamping mechanism based on the pressure data, the motor current data, and the clamping distance data if the pressure sensor data drift status indicates that data drift exists. The pressure adjustment module is used to adjust the current pressure value of the clamping mechanism if the rice seed characteristic change information indicates that there is a characteristic change, so as to achieve flexible seed picking.

[0015] The embodiments of this application include at least the following beneficial effects: First, the embodiments of this application acquire ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seeds. Then, based on the pressure data, clamping distance data, and motor current data, the drift state of the pressure sensor data is identified. Next, based on the ambient temperature data and ambient humidity data, the change information of rice seed characteristics is identified. If the pressure sensor data drift state indicates that there is data drift, the current motor current value of the clamping mechanism is adjusted. If the change information of rice seed characteristics indicates that there is a change in characteristics, the current pressure value of the clamping mechanism is adjusted to achieve flexible seed picking. Thus, by combining the drift state of the pressure sensor data and the change information of rice seed characteristics, the control parameters of the clamping mechanism can be adjusted to achieve control of the rice seed picking mechanism, thereby improving accuracy and reliability.

[0016] Other features and advantages of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in the description and the drawings. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0018] Figure 1 This is a flowchart illustrating a control method for a rice seed-harvesting mechanism based on the principle of flexible bionics, according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the steps of identifying the pressure sensor data drift state based on pressure data, clamping distance data, and motor current data according to an embodiment of the present invention. Figure 3This is an overall architecture diagram of a rice seed-harvesting mechanism control method based on the flexible biomimetic principle according to an embodiment of the present invention; Figure 4 This is a schematic diagram of a rice seed-harvesting mechanism control system based on the principle of flexible bionics, according to an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments.

[0020] In related technologies, during mechanized rice cultivation, technicians employ a flexible seed-grabbing mechanism that mimics biological grasping techniques to improve sowing efficiency and accuracy. The core of this mechanism lies in its flexible grippers, which contain sensitive pressure sensors capable of detecting the contact force between the grippers and the rice seeds in real time. This sensed force information is promptly transmitted to the control system, allowing the system to flexibly adjust the gripper's closing degree according to pre-set rules. The aim is to securely grasp the rice seeds while avoiding any damage, ensuring successful grasping every time. For example, the system can finely adjust the gripping force for seeds of different sizes or shapes, ensuring a firm hold without crushing the seeds.

[0021] However, in actual farmland environments, rice cultivation typically takes place in complex and harsh conditions. Seed-harvesting mechanisms are inevitably exposed to damp, muddy environments containing fine sand during long-term, repetitive operations. Under this continuous physical wear, the outer protective coating of the pressure-sensing element inside the gripper gradually wears away. This protective coating is usually made of wear-resistant, waterproof polymer materials, and its main function is to isolate the sensitive parts of the sensing element from the external environment, preventing corrosion from moisture, mud, chemicals, etc. As the coating wears down, its protective ability decreases, allowing small amounts of moisture and tiny particles, such as extremely fine sand or plant fibers, to slowly penetrate into the tiny gaps or encapsulation structures inside the sensing element. It is worth noting that this penetration is a gradual process; initially, it does not immediately lead to complete failure of the sensing element or obvious abnormal output. Therefore, this hidden physical degradation is often difficult to detect in a timely manner during routine system self-checks or fault diagnosis.

[0022] Once moisture and microparticles penetrate the interior of a sensing element, they can create weak conductive paths near the sensitive parts of the element or its internal circuit connections. For example, in a piezoresistive sensing element, moisture and conductive particles may alter the equivalent impedance of its internal resistive network; in a capacitive sensing element, they may change the dielectric constant or create parasitic capacitance. This change causes a slight and unstable drift in the initial voltage output of the sensing element. This means that even without any physical pressure, the sensing element may output a non-zero voltage value, or its output voltage at a specific pressure may be slightly lower than its calibration value. This deviation is usually very small, and its fluctuation range may be similar to the inherent noise of the sensing element itself or the signal fluctuation caused by changes in ambient temperature. Therefore, conventional signal processing methods, such as simple thresholding or low-pass filtering, often struggle to effectively identify and distinguish this true signal drift caused by internal erosion. For example, if the sensing element should output 0V under no pressure but actually outputs 0.03V, while the normal noise range of the system is ±0.03V, then the 0.03V drift can easily be mistaken for normal noise.

[0023] Because the pressure readings relied upon by the control system are consistently low, when its "gripper closing force control rules" attempt to achieve the preset flexible gripping force target, it will instruct the drive motor to apply a greater physical force than actually required based on the "insufficient" signal fed back by the sensing element. The control system is a closed-loop feedback system that continuously adjusts its output to make the sensing element readings approach the target value. If the sensing element continues to report low pressure, the system will continuously increase the output power of the drive motor in order to reach the target reading, causing the actual physical pressure applied by the gripper to the seed to exceed the design value. For example, if the target gripping force is 2 Newtons, but the sensing element reports 1.8 Newtons, the control system will continue to increase the force until the sensing element reports 2 Newtons, at which point the actual applied force may have reached 2.2 Newtons. This "compensatory" over-force application is an inevitable reaction of the system to achieve its logical goal when information is inaccurate.

[0024] While this sustained, slightly stronger gripping force can successfully extract rice seeds in most cases, its long-term effects are significant. For certain morphologically fragile rice seeds, such as those with thinner seed coats, more exposed embryos, or lower moisture content and a more brittle texture, this additional pressure beyond the design limits can begin to cause microscopic damage. This damage may manifest as fine cracks in the seed coat, internal indentations in the endosperm, or even a potential impact on embryo viability. These damages are often difficult to detect with the naked eye but can significantly affect seed germination rates and seedling health. Simultaneously, because the system needs to apply greater force to reach its perceived target pressure, the grippers may require a longer adjustment time to stabilize during the closing process, resulting in a slightly longer extraction cycle and a decrease in overall extraction efficiency. However, the extent of this damage and efficiency decline is usually gradual and has not yet reached the system's preset critical fault alarm threshold; therefore, the system will not trigger an emergency shutdown or issue a clear fault warning.

[0025] To address various uncertainties that may arise during field operations, such as differences in the physical properties of different seed batches and variations in ambient temperature and humidity, control systems typically incorporate an adaptive optimization mechanism. This mechanism aims to automatically fine-tune operational parameters by analyzing historical seed-harvesting data, such as harvesting success rate, adjustment time per harvest, and required average driving force, to maintain or improve overall operational efficiency. When the mechanism detects a decrease in harvesting efficiency and an increase in adjustment time, it attempts to perform attribution analysis on this data. However, because hidden faults within the sensing elements are not detected by the system, the mechanism incorrectly attributes these signs of performance degradation to "changes in seed batch characteristics," such as assuming that the current batch of seeds may be slipperier, harder, or require greater force to grasp stably. This erroneous attribution prevents the mechanism from recognizing that the true root cause lies in data deviations within the sensing elements, instead attributing it to changes in the external environment or the target object.

[0026] Based on this flawed attribution, the adaptive optimization mechanism automatically and progressively increases the initial value of the target grasping force. The logic is that if the seed is harder to grasp, then applying a "more powerful" grasp to "improve efficiency" is reasonable. For example, if the mechanism's original target grasping force was 2 Newtons, it might adjust it to 2.1 Newtons, then 2.2 Newtons, and so on. This autonomous adjustment by the mechanism further amplifies the excessive actual force applied due to the initial low reading deviation of the sensing element. As a result, the actual physical force applied by the system's grippers will far exceed its initial design limit, even exceeding the limits that the rice seed can withstand. For example, if the sensing element itself causes the actual force to be 0.2 Newtons higher than the target, and the mechanism increases the target by 0.2 Newtons, then the actual applied force will be 0.4 Newtons higher than the original design target, which could be devastating for the fragile rice seed.

[0027] Ultimately, this complex chain reaction created a subtle technical dilemma: in the flexible bionic seed-harvesting mechanism, long-term field operations caused wear to the outer protective layer of the pressure-sensing element, triggering weak and unstable signal drift within the element, resulting in persistently low pressure readings received by the control system. This hidden physical degradation was misinterpreted by the system's built-in adaptive optimization mechanism as a "change in seed batch characteristics," rather than a sensor malfunction. Based on this erroneous attribution, the adaptive mechanism triggered a flawed "self-optimization" process, attempting to "improve efficiency" by progressively increasing the initial value of the target grasping force. However, this mechanism-level adjustment, combined with the low reading deviation of the sensor itself, caused the system, in its pursuit of "higher efficiency," to continuously apply destructive physical pressure far exceeding the design limit to the rice seeds, resulting in an unacceptably high damage rate. This failure mode did not stem from inherent rigidity defects in the mechanism or simple operator error, but rather from a complex control failure caused by both hidden physical degradation and erroneous learning and feedback from the intelligent mechanism. This allowed the problem to worsen within the system, making it difficult to detect through conventional monitoring methods.

[0028] The embodiments of this application will be explained in detail below with reference to the accompanying drawings: Figure 1 This is an optional flowchart of a rice seed-harvesting mechanism control method based on the flexible biomimetic principle provided in an embodiment of this application. Figure 1 The method may include, but is not limited to, steps S101 to S105.

[0029] Step S101: Obtain ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seed. The clamping distance data is used to reflect the distance between the two clamping surfaces of the clamping mechanism. Step S102: Identify the pressure sensor data drift state based on the pressure data, clamping distance data, and motor current data; Step S103: Identify changes in rice seed characteristics based on ambient temperature and humidity data; Step S104: If the pressure sensor data drift status is that data drift exists, then adjust the current motor current value of the clamping mechanism according to the pressure data, motor current data and clamping distance data. Step S105: If the information on changes in rice seed characteristics indicates that there are changes in characteristics, the current pressure value of the clamping mechanism is adjusted to achieve flexible seed extraction.

[0030] Steps S101 to S105 as shown in the embodiments of this application can combine the drift state of pressure sensor data and the information on changes in rice seed characteristics to adjust the control parameters of the clamping mechanism, thereby realizing the control of the rice seed taking mechanism and improving accuracy and reliability.

[0031] In some embodiments, steps S101-S105 can first acquire ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seeds. The clamping distance data reflects the distance between the two clamping surfaces of the clamping mechanism. Ambient temperature and humidity can be acquired using a digital temperature and humidity sensor (such as SHT35 or AM2301). This sensor is installed inside the shield above the seed-taking mechanism to avoid direct sunlight while maintaining air circulation, and the sampling frequency is set to once every 2 seconds. Motor current data can be obtained by connecting a precision sampling resistor (such as a 0.1-ohm, 1% accuracy metal film resistor) in series in the DC motor power supply circuit and using the analog-to-digital converter (ADC) of the microcontroller to acquire the voltage drop across the resistor, with a sampling frequency of once every 10 milliseconds. Clamping distance data can be acquired by mounting a rotary encoder (such as Omron E6B2-CWZ6C, resolution 1000 pulses / revolution) on the output shaft of the stepper motor, and converting the clamping surface displacement by counting the number of pulses, with an accuracy of up to 0.01 mm. Pressure data can be obtained by attaching a half-bridge strain gauge sensor (such as the BE120-3AA type from AVIC Electromechanical Measurement & Control) to the back of the metal skeleton of the flexible gripper, and then performing 24-bit high-precision sampling in conjunction with the HX711 dedicated ADC module. It can be understood that the flexible biomimetic principle refers to constructing a gripping interface with adaptive deformation capabilities using flexible materials (such as silicone or rubber composites), so that the gripping force is evenly distributed on the seed surface rather than concentrated at a single point.

[0032] Then, based on the pressure data, clamping distance data, and motor current data, the pressure sensor data drift state is identified. In one embodiment, this identification process can be implemented through threshold comparison logic. The microcontroller compares the real-time collected pressure value with a preset reference value. If multiple consecutive samples deviate from the reference value by more than a set threshold, a preliminary determination of drift is made. This preset reference value can be calibrated. It is understood that the pressure sensor data drift state refers to a systematic deviation in the output signal caused by the degradation of the sensor's internal physical structure, manifested as a continuous shift in the output voltage value relative to the calibration reference under the same physical pressure.

[0033] Then, based on environmental temperature and humidity data, information on changes in rice seed characteristics is identified. In one embodiment, this can be achieved by establishing an empirical correlation table between environmental parameters and seed hardness. For example, when the temperature rises from 25 degrees Celsius to 35 degrees Celsius and the humidity is below 40%, the seed coat moisture content decreases, leading to an increase in hardness, and the system adjusts its control strategy accordingly. It is understood that the information on changes in rice seed characteristics refers to the actual differences in physical properties between seed batches, including changes in parameters such as seed coat hardness, surface friction coefficient, and geometric dimensions.

[0034] If the pressure sensor data drift indicates data drift, the current motor current of the clamping mechanism is adjusted based on the pressure data, motor current data, and clamping distance data. In the current adjustment embodiment, an open-loop compensation method can be used, directly reducing the duty cycle of the motor drive pulse according to the magnitude of the drift. For example, when it is determined that the pressure sensor reading is 0.5 Newtons too low, the motor drive current is reduced by 10%, causing the actual clamping force to return to the target value.

[0035] If the information indicating a change in rice seed characteristics suggests a change in characteristics, the current pressure value of the clamping mechanism is adjusted to achieve flexible seed removal. In the pressure adjustment embodiment, the target pressure setting value can be adjusted using a lookup table method. For example, when an increase in seed hardness is detected, the target pressure is increased from 2.0 Newtons to 2.2 Newtons.

[0036] Through the above technical solution, this embodiment realizes the distinction between sensor drift and changes in seed characteristics. When sensor drift is detected, control deviation is eliminated through current compensation. When a real change in seed characteristics is detected, pressure adjustment is used to adapt to the change in physical properties. This avoids the continuous force increase cycle caused by misjudgment in traditional adaptive mechanisms, significantly reduces the mechanical damage rate to rice seeds, and ensures the success rate of seed harvesting and the biological activity of seeds.

[0037] In some embodiments, such as Figure 2 As shown, in step S102, the pressure sensor data drift state is identified based on the pressure data, clamping distance data, and motor current data. This may include, but is not limited to, the following steps: Step S201: Obtain the target pressure value and the reference clamping distance range; Step S202: Calculate the lower pressure safety limit based on the target pressure value and the pressure safety margin; Step S203: Calculate the average pressure value based on the pressure data; Step S204: Compare the average pressure value with the lower limit of pressure safety to obtain the first pressure comparison result; Step S205: Compare the clamping distance data with the reference clamping distance range to obtain the first clamping distance comparison result; Step S206: Identify the pressure sensor data drift state based on the first pressure comparison result and the first clamping distance comparison result.

[0038] In some embodiments, a target pressure value and a baseline clamping distance range can be obtained first. The target pressure value is typically set between 1.5 and 2.5 Newtons, determined experimentally based on the average compressive strength of rice seeds. The baseline clamping distance range refers to the theoretical distance that should be achieved between the two clamping surfaces when normally gripping standard-sized seeds (approximately 7-9 mm in length), and is typically set to 6-8 mm.

[0039] Then, based on the target pressure value and the pressure safety margin, the lower pressure safety limit is calculated. The pressure safety margin can be calibrated and is usually taken as 10% to 15% of the target pressure value. For example, when the target pressure value is 2.0 Newtons, the lower pressure safety limit is calculated as 1.7 Newtons (2.0 Newtons minus the 15% margin).

[0040] Then, based on the pressure data, the average pressure value is calculated. A moving average filtering algorithm can be used, taking the arithmetic mean of the most recent 50 sampling points to eliminate random noise interference. In specific implementation, the microcontroller maintains a circular buffer of length 50. Whenever a new pressure sample value arrives, it is stored in the buffer, overwriting the oldest data point, and the arithmetic mean of all data points in the buffer is calculated. The time span of this sliding window is approximately 500 milliseconds (assuming a sampling period of 10 milliseconds), which can effectively filter out high-frequency noise while maintaining sensitivity to pressure changes.

[0041] The average pressure value is compared with the lower pressure safety limit to obtain the first pressure comparison result. Simultaneously, the clamping distance data is compared with the reference clamping distance range to obtain the first clamping distance comparison result. If the average pressure value remains below the lower pressure safety limit for more than three sampling cycles, the first clamping distance comparison result is generated, along with a low-pressure alarm signal. The clamping distance is monitored to ensure it is within the reference range. If the clamping distance has closed to 5 mm (less than the lower limit of 6 mm) but the pressure reading is still below the safety lower limit, it indicates potential sensor drift, and the first clamping distance comparison result is generated.

[0042] Finally, based on the comparison results of the first pressure and the first clamping distance, the data drift status of the pressure sensor is identified. When both the pressure is below the safety lower limit and the clamping distance is within the normal closed range, sensor data drift is determined to exist. By introducing the clamping distance as a spatial position verification parameter, the system effectively distinguishes between sensor zero-point drift (normal clamping distance but low pressure reading) and excessively small seed size (clamping distance not within range and low pressure). This avoids misjudging normal small seed grasping as sensor failure, improving the accuracy and reliability of drift identification.

[0043] Through the above technical solution, this embodiment can further improve the accuracy of pressure sensor data drift identification by comparing it with the lower limit of pressure safety and with the reference clamping distance range, thus avoiding the limitations of judging by a single parameter.

[0044] In some embodiments, step S206, identifying the pressure sensor data drift state based on the first pressure comparison result and the first clamping distance comparison result, may include, but is not limited to, the following steps: Obtain historical average current values; Calculate the upper limit of current safety based on historical average current values ​​and current safety margin; Calculate the current average current value based on the motor current data; The current average current value is compared with the safe current limit to obtain the current comparison result; Based on the first pressure comparison result, the first clamping distance comparison result, and the current comparison result, the pressure sensor data drift state is identified.

[0045] In some embodiments, the historical average current value can be obtained first. The historical average current value is taken from the average current consumption of the most recent 100 successful seed-taking operations and stored in non-volatile memory. Specifically, after each successful seed-taking operation (defined as the clamping mechanism completing a full cycle of closing, holding, and releasing without triggering overload protection), the microcontroller records the current value during that operation, stores this value in a circular queue, and calculates the historical average current value. When the queue is full (100 data entries), new data overwrites the oldest data, thus maintaining the timeliness of historical data. The storage medium can be the microcontroller's built-in EEPROM or an external Flash memory chip to ensure data is not lost after power failure.

[0046] Then, based on the historical average current value and the current safety margin, calculate the upper limit of the current safety limit. The current safety margin can be calibrated and is usually set to 20% of the historical average. For example, if the historical average current is 500 mA, then the safety limit is 600 mA.

[0047] Next, calculate the current average current value based on the motor current data. A sliding window averaging method can be used, taking the average current samples over the most recent second. In practice, since the current sampling frequency is 100Hz (once every 10 milliseconds), 100 data points are collected within one second, and the arithmetic mean of these 100 points is calculated as the current average current value. The current average current value is then compared with the safe current limit to obtain the current comparison result. If the current exceeds the safe limit, it indicates that the motor is in an overload state, and a current comparison result is generated.

[0048] Finally, based on the first pressure comparison result, the first clamping distance comparison result, and the current comparison result, the pressure sensor data drift status is identified. In the comprehensive judgment logic, the pressure sensor data drift is only confirmed when all three conditions are met simultaneously: the pressure is below the safety lower limit, the clamping distance is within the normal range, and the motor current exceeds the safety upper limit. In specific implementation, the microcontroller executes a three-input AND gate logic judgment: input A is the first pressure comparison result (1 indicates low pressure, 0 indicates normal), input B is the first clamping distance comparison result (1 indicates normal distance, 0 indicates abnormal distance), and input C is the current comparison result (1 indicates current exceeds the limit, 0 indicates normal current). When A=1, B=1, and C=1, the output drift status flag is 1; otherwise, it is 0.

[0049] Through the above technical solution, this embodiment utilizes motor current as a mechanical verification parameter because when a sensor reading is low, causing the control system to continuously increase force, the motor current will inevitably rise abnormally. By cross-verifying the electrical, mechanical, and mechanical parameters, interference factors such as a sudden increase in seed hardness (high current, high pressure, normal distance) or mechanical jamming (high current, abnormal distance) are effectively eliminated, significantly reducing the false judgment rate. This embodiment further enhances the robustness of drift recognition and eliminates interference caused by changes in motor load.

[0050] In some embodiments, identifying changes in rice seed characteristics based on ambient temperature and humidity data may include, but is not limited to, the following steps: Step S301: Obtain historical temperature data and historical humidity data; Step S302: Determine the pressure safety range based on the target pressure value and the pressure safety margin; Step S303: Compare the average pressure value with the pressure safety range to obtain the second pressure comparison result; Step S304: Calculate the historical average temperature based on historical temperature data; Step S305: Calculate the historical average humidity based on historical humidity data; Step S306: Compare the ambient temperature data with the historical average temperature to obtain the temperature comparison result; Step S307: Compare the ambient humidity data with the historical average humidity to obtain the humidity comparison result; Step S308: Based on the second pressure comparison result, current comparison result, temperature comparison result and humidity comparison result, identify the information on changes in rice seed characteristics.

[0051] In some embodiments, historical temperature and humidity data can be acquired first. The historical temperature and humidity data are taken from the environmental parameters of the previous work shift (e.g., the morning period). Specifically, the system sets up an environmental baseline update mechanism: whenever the equipment is started or every 4 hours, the system automatically records the environmental temperature and humidity for the current 2 hours as historical temperature and humidity data.

[0052] Then, based on the target pressure value and the pressure safety margin, the pressure safety range is determined. The pressure safety range is the interval formed by adding or subtracting the safety margin from the target pressure value, for example, 1.7 Newtons to 2.3 Newtons. The average pressure value is then compared with the pressure safety range to obtain a second pressure comparison result. If the current average pressure value is within the safety range, it indicates that the clamping force control is normal, and a second pressure comparison result is generated.

[0053] Next, calculate the historical average temperature based on historical temperature data. The arithmetic mean of the ambient temperatures over the previous two hours can be used as the historical average temperature. Specifically, the calculation method is as follows: sum all temperature samples collected over the previous two hours (collected every 2 seconds, for a total of 3600 data points), then divide by the number of sampling points (3600) to obtain the historical average temperature. Simultaneously, calculate the historical average humidity based on historical humidity data. The arithmetic mean of the ambient humidity over the previous two hours can be used as the historical average humidity. The calculation method is the same as for temperature: sum and average the humidity samples over the previous two hours to obtain the historical average humidity.

[0054] The ambient temperature data is compared with the historical average temperature to obtain the temperature comparison result. A temperature change threshold of ±5 degrees Celsius is set; values ​​exceeding this range are considered significant changes. In practice, the temperature difference between the current temperature and the historical average temperature is calculated. If the temperature difference is >5°C, the temperature comparison result is 1 (indicating a significant change); otherwise, it is 0 (indicating within the normal range). Simultaneously, the ambient humidity data is compared with the historical average humidity to obtain the humidity comparison result. A humidity change threshold of ±15% relative humidity is set; values ​​exceeding this range are considered significant changes. In practice, the humidity difference between the current humidity and the historical average humidity is calculated. If the humidity difference is >15%, the humidity comparison result is 1 (indicating a significant change); otherwise, it is 0 (indicating within the normal range).

[0055] Finally, based on the results of the second pressure comparison, current comparison, temperature comparison, and humidity comparison, information on changes in rice seed characteristics is identified. For example, when the ambient temperature and humidity change significantly, and the current consumption pattern changes even though the current is within a safe range, it is determined that a real change in seed characteristics has occurred. In specific implementation, a four-dimensional judgment matrix can be constructed: second pressure comparison result (normal is 1, abnormal is 0), current comparison result (normal is 0, abnormal is 1), temperature comparison result (significant change is 1, normal is 0), and humidity comparison result (significant change is 1, normal is 0). When the conditions are met (second pressure comparison result is 1), (current comparison result is 1), and (temperature comparison result is 1 or humidity comparison result is 1), it is determined that there has been a change in seed characteristics.

[0056] Through the above technical solution, this embodiment establishes a correlation model between environmental parameters and seed physical properties, distinguishing between changes in seed hardness caused by the environment (changes in actual characteristics) and sensor drift (abnormal pressure readings despite unchanged environment). This allows the control system to adjust its strategy according to actual environmental adaptation needs, rather than incorrectly compensating for sensor malfunctions. This embodiment can accurately identify the actual changes in the physical properties of rice seeds, avoiding misjudgments caused by environmental interference.

[0057] In some embodiments, in step S308, identifying rice seed characteristic change information based on the second pressure comparison result, current comparison result, temperature comparison result, and humidity comparison result may include, but is not limited to, the following steps: Get the maximum clamping distance; Calculate the average clamping distance based on the clamping distance data; The average clamping distance and the upper limit of the clamping distance are compared to obtain the second clamping distance comparison result; Based on the comparison results of the second pressure, current, temperature, humidity, and second clamping distance, information on changes in rice seed characteristics is identified.

[0058] In some embodiments, the upper limit of the clamping distance can be obtained first. This upper limit can be calibrated, typically set to 10 mm, corresponding to larger rice seeds. Simultaneously, the average clamping distance is calculated based on the clamping distance data. The average clamping distance can be obtained by performing an arithmetic mean on the clamping distance data.

[0059] The average clamping distance is then compared with the upper limit of the clamping distance to obtain a second clamping distance comparison result. If the average clamping distance exceeds the upper limit, it indicates that the current seed size is abnormal or that the clamping mechanism is mechanically loose. For example, if the average clamping distance is >10mm, the second clamping distance comparison result is 0 (indicating abnormality), otherwise it is 1 (indicating normal).

[0060] Then, based on the second pressure comparison result, current comparison result, temperature comparison result, humidity comparison result, and second clamping distance comparison result, information on changes in rice seed characteristics is identified. When the environmental temperature and humidity change, the voltage and pressure parameters match, and the clamping distance is within the normal range, it is confirmed that the overall characteristics of the seed batch have changed. In specific implementation, the second clamping distance comparison result is added as a constraint condition to the multi-dimensional judgment logic: it needs to satisfy (second pressure comparison result = 1) and (current comparison result = 1) and (temperature comparison result = 1 or humidity comparison result = 1) and (second clamping distance comparison result = 1).

[0061] Through the above technical solution, this embodiment effectively filters out parameter anomalies caused by the intrusion of individual extra-large seeds or foreign objects by constraining the clamping distance upper limit. This ensures that the characteristic change identification targets the overall physical property migration of the seed batch, improving the reliability of the identification results and the pertinence of control strategy adjustments. The introduction of geometric size constraints in this embodiment to eliminate interference from abnormal individuals further improves the accuracy of seed characteristic identification.

[0062] In some embodiments, in step S104, adjusting the current motor current value of the clamping mechanism based on pressure data, motor current data, and clamping distance data may include, but is not limited to, the following steps: Step S401: Extract the current pressure value from the pressure data; Step S402: Extract the current motor current value from the motor current data; Step S403: Extract the current clamping distance value from the clamping distance data; Step S404: Determine the current adjustment amount based on the current pressure value, current motor current value, and current clamping distance value; Step S405: Adjust the current motor current value of the clamping mechanism according to the current adjustment amount.

[0063] In some embodiments, if a simple fixed-ratio compensation strategy is adopted (e.g., uniformly reducing the current by a fixed percentage), it is difficult to adapt to different degrees of wear of the sensor, and new control errors may also be introduced due to the time-varying characteristics of mechanical transmission efficiency.

[0064] Therefore, the current pressure value can be extracted from the pressure data first. This extraction can be achieved by reading the conversion result of the analog-to-digital converter (ADC) after the pressure sensor output signal stabilizes (e.g., after a 50-millisecond delay following the clamping action), and then applying digital filtering (such as median filtering or moving average filtering) to eliminate high-frequency noise caused by mechanical vibration. Specifically, after detecting a stop signal from the clamping motor, the microcontroller starts a 50-millisecond timer. After the timer overflows, it triggers ADC sampling, continuously acquiring 10 pressure value samples. The maximum and minimum values ​​are then discarded, and the arithmetic mean is taken as the current pressure value.

[0065] Then, the current motor current value is extracted from the motor current data. The current motor current value is obtained by real-time monitoring of the voltage drop across a precision sampling resistor (e.g., 0.1 ohms, 1% accuracy) connected in series in the motor power supply circuit, and then converting it using Ohm's law. The sampling frequency is set to once every 5 milliseconds. Specifically, the microcontroller's built-in ADC collects the voltage across the sampling resistor at a frequency of 200Hz, and the current motor current value is obtained by dividing the voltage by the resistance.

[0066] The current clamping distance value is then extracted from the clamping distance data. This value is obtained by reading the pulse count from the rotary encoder and multiplying it by the displacement coefficient corresponding to each pulse (e.g., 0.01 mm per pulse). The current adjustment amount is determined based on the current pressure value, current motor current value, and current clamping distance value. A closed-loop control algorithm can be used, for example, by using the difference between the target pressure value and the current pressure value as input, and calculating the required current correction amount through a proportional-integral (PI) controller. The PI controller parameters can be tuned using the critical proportional gain method or a trial-and-error method: set the integral coefficient to zero, gradually increase the proportional coefficient until the system exhibits constant-amplitude oscillations, and record the proportional coefficient at this point as the critical value Kc and the oscillation period Tc; calculate the final parameters using empirical formulas (e.g., proportional coefficient Kp = 0.45*Kc, integral coefficient Ki = 0.54*Kc / Tc). In practice, current adjustment is achieved by adjusting the duty cycle of the pulse width modulation (PWM) signal of the motor driver.

[0067] Finally, the current value of the clamping mechanism's motor is adjusted based on the current adjustment amount. If the calculated current adjustment amount is positive (indicating a need to increase the current), the PWM duty cycle is increased linearly by the corresponding proportion; if it is negative, the duty cycle is reduced, with the adjustment step size limited to no more than 5% per control cycle to prevent sudden current changes from impacting the mechanical structure.

[0068] Through the above technical solution, this embodiment achieves refined and dynamic adjustment of motor current, no longer a simple fixed compensation, but a precise calculation based on real-time pressure, current and distance feedback, ensuring that the actual clamping force can quickly and accurately return to the safe range in the event of sensor drift, significantly improving the accuracy and adaptability of compensation control, and avoiding over-compensation or under-compensation.

[0069] In some embodiments, in step S404, determining the current adjustment amount based on the current pressure value, the current motor current value, and the current clamping distance value may include, but is not limited to, the following steps: Calculate the expected pressure value based on the current motor current and current clamping distance. Calculate the pressure deviation value based on the expected pressure value and the current pressure value; Based on the pressure deviation value, the current adjustment amount is determined by consulting the mapping table between the pressure deviation value and the current adjustment amount.

[0070] In some embodiments, if a complex real-time physical model is used for calculation, it may place excessive demands on the computing power and storage resources of the microcontroller, increasing system cost and power consumption; while if a simple linear proportional relationship is used, it is difficult to accurately reflect the nonlinear coupling characteristics between the motor, transmission mechanism and clamping force, especially the complex mechanical behavior when the flexible gripper deforms.

[0071] Therefore, the expected pressure value can be calculated first based on the current motor current and current clamping distance. The expected pressure value refers to the theoretical clamping force calculated based on the current motor current and clamping distance under the ideal condition of no pressure sensor drift. This value reflects the expected mechanical output of the motor drive system. The calculation of the expected pressure value is based on a preset motor current-clamping distance-pressure mapping model. The model is established as follows: During the calibration phase when the sensor is in a healthy state (no drift), a standard weight or force gauge is used as the load. The system records the actual measured pressure values ​​under different combinations of clamping distances (e.g., 6 mm, 7 mm, 8 mm, covering the seed size range) and motor currents (e.g., 300 mA, 400 mA, 500 mA, covering the operating current range), forming a three-dimensional lookup table. During actual operation, the microcontroller substitutes the real-time collected current and distance values ​​into the model, and obtains the expected pressure value F_est through table lookup and bilinear interpolation algorithm.

[0072] Then, the pressure deviation is calculated based on the expected pressure value and the current pressure value. This deviation can be obtained by subtracting the expected pressure value from the current pressure value. This deviation quantifies the degree of reading deviation caused by sensor drift.

[0073] Then, based on the pressure deviation value, the current adjustment amount is determined by consulting a mapping table between the pressure deviation value and the current adjustment amount. This mapping table is obtained through experimental calibration. The specific process is as follows: In a laboratory environment, a precision adjustable resistor is connected in series with the output of the sensor signal conditioning circuit to simulate different degrees of zero-point drift (e.g., equivalent drifts of 0.1 N, 0.2 N, 0.3 N, 0.4 N, and 0.5 N). For each degree of drift, the motor current is manually adjusted to bring the actual clamping force (monitored by a standard force gauge) back to the target pressure value, and the required current adjustment amount is recorded. By repeating the experiment multiple times and taking the average value, a mapping table between the pressure deviation value and the current adjustment amount is established. The microcontroller queries this table based on the calculated pressure deviation value; if the pressure deviation value falls within a certain range, the corresponding current adjustment amount is selected.

[0074] Through the above technical solution, this embodiment transforms the complex nonlinear coupling relationship into a preset mapping table, which significantly reduces the complexity of real-time calculation and facilitates implementation in resource-constrained embedded systems. At the same time, by driving the adjustment through the deviation between the expected pressure value and the actual value, accurate compensation for drift is achieved. Regardless of the wear stage of the sensor, the appropriate current correction amount can be quickly determined, improving the response speed and compensation accuracy of the control system and avoiding complex floating-point operations.

[0075] In some embodiments, adjusting the current pressure value of the clamping mechanism in step S105 may include, but is not limited to, the following steps: Obtain rice seed information, including the hardness of the seed coat; Based on the hardness of the seed shell, the pressure that can be withheld is determined by consulting a mapping table between the hardness of the seed shell and the pressure that can be withheld. Adjust the current pressure value of the clamping mechanism according to the tolerable pressure value.

[0076] In some embodiments, if a uniform pressure adjustment strategy is adopted (such as a fixed increase of 0.2 Newtons), it may not be possible to balance the reliability of grasping and the safety of the seed. For hard seeds, insufficient pressure may cause them to slip, while for soft seeds, excessive pressure may cause damage.

[0077] To achieve this, rice seed information, including seed coat hardness, can be obtained first. A near-infrared spectral sensor (such as an NIR spectrometer using a gallium arsenide indium photodiode array) can be installed above the seed transport channel. By collecting the near-infrared reflectance spectrum of the seeds (wavelength range 900 nm to 1700 nm), a pre-established partial least squares regression (PLS) model can be used to predict the seed hardness value, measured in Newtons. The training process for this model is as follows: Collect 200 representative rice seed samples (covering different varieties, origins, and storage conditions). Measure the actual hardness value of each seed using a standard hardness meter (such as a texture analyzer) as a reference value. Simultaneously, collect the spectral data of each seed using an NIR spectrometer. Preprocess the spectra (e.g., standard normal variable transformation SNV, first derivative smoothing). Establish a PLS regression model using leave-one-out cross-validation, with spectral data as the independent variable and the measured hardness value as the dependent variable. Determine the optimal number of principal components through iterative calculation (usually the number of principal components that minimizes the sum of squared predicted residuals, PRESS). When the correlation coefficient R² between the model's predicted and measured values ​​is greater than 0.9 and the root mean square error (RMSE) is less than 0.5 Newtons, the model training is considered complete, and the model coefficient matrix is ​​stored in the system's non-volatile memory. In practical applications, an NIR spectrometer acquires seed spectra in real time, which are then input into the PLS model to calculate the seed coat hardness.

[0078] Then, based on the hardness of the seed coat, the permissible pressure value is determined by consulting a mapping table between seed coat hardness and tolerable pressure. After obtaining the seed coat hardness, a preset mapping table between seed coat hardness and tolerable pressure is consulted. This table is established through destructive experiments: 30 seed samples of each hardness grade (e.g., 5-8 N, 8-12 N, 12-16 N, 16-20 N, etc., measured by a hardness tester) are selected, and the clamping pressure is gradually increased until visible cracks or internal damage appear in the seeds (confirmed by microscopic observation). The critical pressure value is recorded, and 80% is taken as the safe threshold for the tolerable pressure of that hardness grade, ensuring a safety margin.

[0079] Then, based on the tolerable pressure value, the current pressure value of the clamping mechanism is adjusted. After determining the tolerable pressure value, it can be set as the new target pressure value. The control system adjusts the motor drive parameters of the clamping mechanism accordingly to make the actual clamping force approach the target value.

[0080] Through the above technical solution, this embodiment achieves an adaptive match between the seed-picking pressure and the mechanical strength of the seed itself. For seeds with hard shells, the pressure can be appropriately increased to ensure a firm grip, while for seeds with fragile shells, the pressure can be reduced to avoid damage. This truly realizes flexible seed picking based on the individual characteristics of the seed, significantly reducing the seed breakage rate and improving the sowing quality and germination rate.

[0081] In some embodiments, step S206, identifying the pressure sensor data drift state based on the first pressure comparison result and the first clamping distance comparison result, may include, but is not limited to, the following steps: Impedance data of the clamping surface region of the clamping mechanism is acquired using an impedance spectroscopy measuring element; Vibration data of the clamping mechanism during the clamping process is collected using vibration measuring elements; Spectral analysis is performed on impedance data to extract impedance characteristics, including impedance amplitude and phase angle. Perform time-domain analysis on the vibration data to calculate the vibration attenuation rate; Based on impedance characteristics and vibration attenuation rate, the clamping force of the clamping surface is calculated. The clamping force of the clamping surface is used to represent the clamping force applied to the rice seed by the clamping surface of the clamping mechanism. The clamping force and the lower limit of pressure safety are compared to obtain the third pressure comparison result; Based on the first pressure comparison result, the first clamping distance comparison result, and the third pressure comparison result, the pressure sensor data drift state is identified.

[0082] In some embodiments, impedance data of the clamping surface region of the clamping mechanism can be acquired first using an impedance spectroscopy measuring element. The impedance spectroscopy measuring element can be a pair of embedded electrodes (e.g., gold-plated copper electrodes, 2 mm in diameter and 0.1 mm thick), symmetrically mounted inside the contact surface of the flexible gripper (approximately 0.5 mm from the surface), and impedance data is acquired through an electrochemical impedance spectroscopy (EIS) measurement circuit. The basic principle of EIS measurement is to apply a small AC excitation signal (frequency range 0.1 Hz to 100 kHz, amplitude 10 mV, logarithmic scan of 10 frequency points) to the electrodes during clamping, and measure the impedance response of the gripper-seed contact interface. As the clamping force increases, the contact area increases, the interface double-layer capacitance increases, and the contact resistance decreases, resulting in a decrease in impedance amplitude and a phase angle shift. The specific measurement circuit includes a signal generator (e.g., a direct digital frequency synthesizer chip AD9833), a potentiostat, and a current detection circuit. The microcontroller controls the signal generator to output sine waves of different frequencies, measures the amplitude ratio and phase difference of the response current to the excitation voltage, and calculates the complex impedance as impedance data.

[0083] Then, vibration data of the clamping mechanism during the clamping process is collected using a vibration measurement element. The vibration measurement element can be a miniature piezoelectric accelerometer (e.g., with a range of ±3g and a sensitivity of 300mV / g) installed at the rigid connection point of the gripper drive arm to collect vibration data. At the instant the clamping action is completed (the moment the motor stops driving), a brief mechanical pulse excitation (lasting 5 milliseconds, frequency 200 Hz) is applied to the gripper using the inverse piezoelectric effect of the piezoelectric ceramic, and the free-dampening vibration response of the gripper is measured.

[0084] Next, perform spectral analysis on the impedance data to extract impedance characteristics, including impedance amplitude and phase angle. When performing spectral analysis on the impedance data, the Fast Fourier Transform (FFT) algorithm can be used to process the time-domain signal and extract the impedance amplitude and phase angle at each frequency point. Special attention can be paid to the impedance amplitude at the 1 kHz frequency point, as the impedance at this frequency is most sensitive to changes in contact area.

[0085] Time-domain analysis is performed on the vibration data to calculate the vibration attenuation rate. When performing time-domain analysis on vibration data, the vibration attenuation rate can be obtained by calculating the logarithmic attenuation rate of adjacent vibration peak amplitudes. The specific calculation method is as follows: identify the first three peak values ​​A1, A2, and A3 of the vibration signal, and calculate the vibration attenuation rate. = ln(A1 / A2) = ln(A2 / A3), and take the average value as the vibration attenuation rate for this measurement. The vibration attenuation rate reflects the energy dissipation characteristics of the contact interface and is closely related to the contact pressure and friction coefficient.

[0086] Based on impedance characteristics and vibration attenuation rate, the clamping force is calculated. This clamping force represents the force applied by the clamping mechanism to the rice seed. Calculated using a pre-defined multiple regression model: ,in The impedance amplitude measured at a frequency of 1 kHz. This is the current clamping distance. The vibration attenuation rate, , , , These are the regression coefficients determined through calibration experiments. The model was established as follows: During the calibration phase, a standard force gauge was used to simultaneously measure the actual clamping force, and the impedance amplitude, vibration attenuation rate, and clamping distance were recorded. At least 50 sets of data under different forces were collected, and the regression coefficients were obtained by fitting the data using a multiple linear regression algorithm. , , , The model is based on contact mechanics and electrochemical theory: the reciprocal of the impedance amplitude reflects the actual contact area (positively correlated with force), and the vibration attenuation rate reflects the damping characteristics of the contact interface (positively correlated with contact pressure).

[0087] Finally, the clamping force and the lower limit of pressure are compared to obtain the third pressure comparison result. The calculated clamping force is compared with the lower limit of pressure (e.g., 1.7 Newtons). If the clamping force is greater than or equal to the lower limit of pressure but the pressure sensor reading is lower than the lower limit of pressure, the third pressure comparison result indicates that the pressure reading is too low. Based on the first pressure comparison result, the first clamping distance comparison result, and the third pressure comparison result, the pressure sensor data drift state is identified. For example, when the first pressure comparison result (based on sensor reading) indicates low pressure, the first clamping distance comparison result indicates normal distance, and the third pressure comparison result (based on independent physical measurement) indicates normal actual pressure, all three conditions are met simultaneously, confirming that the pressure sensor has data drift.

[0088] Through the above technical solution, this embodiment introduces an independent physical measurement method based on electrochemical impedance spectroscopy and mechanical vibration to directly calculate the actual clamping force. This complements and redundantly verifies the indirect inference based on motor current, significantly improving the accuracy and reliability of drift identification. In particular, even when the sensor drifts slightly and the motor current is disturbed by power supply fluctuations, it can still effectively identify the true drift state, providing a more solid basis for subsequent current compensation control and avoiding erroneous compensation due to misjudgment.

[0089] Figure 3 This diagram presents the overall architecture of a control method for a rice seed-harvesting mechanism based on the principle of flexible bionics. The entire seed-harvesting cycle reveals a continuous and coordinated relationship between the various technical components. Environmental temperature and humidity data are used to determine external conditions affecting changes in seed physical properties. Pressure data, clamping distance data, and motor current data together form the internal evidence chain for identifying pressure sensor data drift. After drift is detected, the system calculates the current adjustment based on the current pressure, motor current, and clamping distance, replacing the control path that relies solely on the pressure sensor. Upon identifying changes in seed characteristics, the system updates the target pressure value through the mapping relationship between seed shell hardness and tolerable pressure, thereby altering the clamping control benchmark. These components do not substitute for each other but work collaboratively according to the logic of "first identifying the source of the fault, then performing the corresponding adjustment." In this embodiment, the various technical features form a processing chain within the same control framework: identification, then diversion, and then adjustment. This allows the system to simultaneously block excessive force caused by sensor underreporting and avoid omitting or misjudging actual changes in seed characteristics, thus achieving flexible seed-harvesting control of the rice seed-harvesting mechanism under complex operating conditions.

[0090] The beneficial effects of implementing the embodiments of the present invention include: First, the embodiments of this application acquire ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seeds. Then, based on the pressure data, clamping distance data, and motor current data, the pressure sensor data drift state is identified. Next, based on the ambient temperature data and ambient humidity data, information on changes in rice seed characteristics is identified. If the pressure sensor data drift state indicates data drift, the current motor current value of the clamping mechanism is adjusted. If the information on changes in rice seed characteristics indicates changes in characteristics, the current pressure value of the clamping mechanism is adjusted to achieve flexible seed extraction. Thus, by combining the pressure sensor data drift state and information on changes in rice seed characteristics, the control parameters of the clamping mechanism can be adjusted to achieve control of the rice seed extraction mechanism, thereby improving accuracy and reliability.

[0091] like Figure 4 As shown, this embodiment of the invention also provides a control system for a rice seed-harvesting mechanism based on the principle of flexible bionics, comprising: The data acquisition module 501 is used to acquire ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seed. The clamping distance data is used to reflect the distance between the two clamping surfaces of the clamping mechanism. The sensor data drift identification module 502 is used to identify the pressure sensor data drift state based on pressure data, clamping distance data and motor current data. The seed characteristic change identification module 503 is used to identify rice seed characteristic change information based on ambient temperature data and ambient humidity data. The motor current adjustment module 504 is used to adjust the current motor current value of the clamping mechanism based on the pressure data, motor current data and clamping distance data if the pressure sensor data drift status is that data drift exists. The pressure adjustment module 505 is used to adjust the current pressure value of the clamping mechanism if the rice seed characteristic change information indicates that there is a characteristic change, so as to achieve flexible seed picking.

[0092] The content of the above method embodiments is applicable to this system embodiment. The specific functions implemented in this system embodiment are the same as those in the above method embodiments, and the beneficial effects achieved are also the same as those achieved in the above method embodiments.

[0093] The embodiments described in this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided by the embodiments of this application. As those skilled in the art will know, with the evolution of technology and the emergence of new application scenarios, the technical solutions provided by the embodiments of this application are also applicable to similar technical problems.

Claims

1. A control method for a rice seed-harvesting mechanism based on the principle of flexible bionics, characterized in that, Includes the following steps: The system acquires ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seeds. The clamping distance data is used to reflect the distance between the two clamping surfaces of the clamping mechanism. Based on the pressure data, the clamping distance data, and the motor current data, identify the pressure sensor data drift state; Based on the ambient temperature data and the ambient humidity data, identify information on changes in rice seed characteristics; If the pressure sensor data drift status indicates that data drift exists, then the current motor current value of the clamping mechanism is adjusted based on the pressure data, the motor current data, and the clamping distance data. If the rice seed characteristic change information indicates that there is a characteristic change, the current pressure value of the clamping mechanism is adjusted to achieve flexible seed extraction.

2. The method according to claim 1, characterized in that, The step of identifying the pressure sensor data drift state based on the pressure data, the clamping distance data, and the motor current data includes: Obtain the target pressure value and the reference clamping distance range; Calculate the lower pressure safety limit based on the target pressure value and the pressure safety margin; Calculate the average pressure value based on the pressure data; The average pressure value is compared with the lower pressure safety limit to obtain the first pressure comparison result; The clamping distance data is compared with the reference clamping distance range to obtain the first clamping distance comparison result; Based on the first pressure comparison result and the first clamping distance comparison result, the data drift state of the pressure sensor is identified.

3. The method according to claim 2, characterized in that, The step of identifying the pressure sensor data drift state based on the first pressure comparison result and the first clamping distance comparison result includes: Obtain historical average current values; Calculate the upper limit of current safety based on the historical average current value and current safety margin; Calculate the current average current value based on the motor current data; The current average current value is compared with the current safety upper limit to obtain the current comparison result; Based on the first pressure comparison result, the first clamping distance comparison result, and the current comparison result, the data drift state of the pressure sensor is identified.

4. The method according to claim 3, characterized in that, Based on the ambient temperature data and the ambient humidity data, identify information on changes in rice seed characteristics, including: Acquire historical temperature and humidity data; The pressure safety range is determined based on the target pressure value and the pressure safety margin; The average pressure value is compared with the pressure safety range to obtain a second pressure comparison result; Calculate the historical average temperature based on the historical temperature data; Calculate the historical average humidity based on the historical humidity data; The ambient temperature data is compared with the historical average temperature to obtain the temperature comparison result; The ambient humidity data is compared with the historical average humidity to obtain the humidity comparison result; Based on the second pressure comparison result, the current comparison result, the temperature comparison result, and the humidity comparison result, the information on the changes in the characteristics of the rice seeds is identified.

5. The method according to claim 4, characterized in that, The step of identifying the rice seed characteristic change information based on the second pressure comparison result, the current comparison result, the temperature comparison result, and the humidity comparison result includes: Get the maximum clamping distance; Calculate the average clamping distance based on the clamping distance data; The average clamping distance and the upper limit of the clamping distance are compared to obtain the second clamping distance comparison result; Based on the second pressure comparison result, the current comparison result, the temperature comparison result, the humidity comparison result, and the second clamping distance comparison result, the information on the changes in the characteristics of the rice seeds is identified.

6. The method according to claim 1, characterized in that, The step of adjusting the current motor current value of the clamping mechanism based on the pressure data, the motor current data, and the clamping distance data includes: Extract the current pressure value from the pressure data; Extract the current motor current value from the motor current data; Extract the current clamping distance value from the clamping distance data; The current adjustment amount is determined based on the current pressure value, the current motor current value, and the current clamping distance value. The current value of the clamping mechanism is adjusted according to the current adjustment amount.

7. The method according to claim 6, characterized in that, The step of determining the current adjustment amount based on the current pressure value, the current motor current value, and the current clamping distance value includes: Calculate the expected pressure value based on the current motor current value and the current clamping distance value; Calculate the pressure deviation value based on the expected pressure value and the current pressure value; Based on the pressure deviation value, the current adjustment amount is determined by consulting the mapping table between the pressure deviation value and the current adjustment amount.

8. The method according to claim 1, characterized in that, Adjusting the current pressure value of the clamping mechanism includes: Obtain rice seed information, including the hardness of the seed coat; Based on the hardness of the seed shell, the pressure that can be withheld is determined by consulting a mapping table between the hardness of the seed shell and the pressure that can be withheld. The current pressure value of the clamping mechanism is adjusted according to the tolerable pressure value.

9. The method according to claim 2, characterized in that, The step of identifying the pressure sensor data drift state based on the first pressure comparison result and the first clamping distance comparison result includes: Impedance data of the clamping surface region of the clamping mechanism is acquired using an impedance spectroscopy measuring element. Vibration data of the clamping mechanism during the clamping process are collected by a vibration measuring element; The impedance data is subjected to spectral analysis to extract impedance features, which include impedance amplitude and phase angle. Perform time-domain analysis on the vibration data to calculate the vibration attenuation rate; Based on the impedance characteristics and the vibration attenuation rate, the clamping force is calculated, which represents the clamping force applied by the clamping mechanism to the rice seed. The clamping force of the clamping face and the lower limit of the pressure safety are compared to obtain the third pressure comparison result; Based on the first pressure comparison result, the first clamping distance comparison result, and the third pressure comparison result, the data drift state of the pressure sensor is identified.

10. A control system for a rice seed-harvesting mechanism based on the principle of flexible bionics, characterized in that, include: The data acquisition module is used to acquire ambient temperature data, ambient humidity data, motor current data, clamping distance data of the clamping mechanism, and pressure data between the clamping mechanism and the rice seed. The clamping distance data is used to reflect the distance between the two clamping surfaces of the clamping mechanism. The sensor data drift identification module is used to identify the pressure sensor data drift state based on the pressure data, the clamping distance data, and the motor current data. The seed characteristic change identification module is used to identify rice seed characteristic change information based on the ambient temperature data and the ambient humidity data. The motor current adjustment module is used to adjust the current motor current value of the clamping mechanism based on the pressure data, the motor current data, and the clamping distance data if the pressure sensor data drift status indicates that data drift exists. The pressure adjustment module is used to adjust the current pressure value of the clamping mechanism if the rice seed characteristic change information indicates that there is a characteristic change, so as to achieve flexible seed picking.

Images