Parametric-control polymorphic bionic scallop treasure intelligence acquisition system

Abstract

The invention provides a parameter-control multi-state bionic scallop treasure intelligence acquisition system, which relates to the technical field of intelligent processing of aquatic products and comprises a multi-mode sensing module, a variety identification and parameter matching module, a self-adaptive parameter regulation and control module, a bionic execution module and a closed-loop feedback control module, the multi-modal sensing module is used for acquiring morphological characteristic data and mechanical state data of the scallops; the variety identification and parameter matching module determines a scallop variety type and obtains a corresponding initial parameter control parameter set; the adaptive parameter regulation and control module starts a polymorphic force regulation and control mechanism according to the adductor muscle tension signal, and dynamically adjusts execution force parameters; the bionic execution module is used for opening shells of scallops and separating the scallops from the scallops through the flexible operation unit; the closed-loop feedback control module monitors the force data and the displacement data in real time, and dynamically adjusts and executes the force parameters. The scallop harvesting device can intelligently identify multiple scallops and adaptively adjust the operation parameters, so that accurate force regulation and control and flexible lossless operation are realized, and the harvesting efficiency and the product integrity rate are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent processing technology for aquatic products, specifically to a parameter-controlled multi-state biomimetic scallop adductor intelligent extraction system. Background Technology

[0002] Scallops are an important marine economic shellfish, and their adductor muscle (commonly known as the scallop adductor muscle) has extremely high nutritional and economic value. In the scallop processing industry chain, the harvesting and separation of the scallop adductor muscle is a core technological step, directly determining the product's quality grade and economic benefits. Traditional scallop adductor muscle harvesting mainly relies on manual labor, which suffers from high labor intensity, low efficiency, and high product breakage rates. With the large-scale and standardized development of the aquatic product processing industry, there is an urgent need to automate and intelligently harvest scallop adductor muscle.

[0003] Existing scallop processing equipment faces the following technical bottlenecks: First, different scallop species (such as the Yesso scallop and the comb-shaped scallop) exhibit significant differences in morphological parameters such as shell size, shell thickness, and column size. Traditional mechanical equipment with fixed parameters struggles to adapt to and process multiple scallop species, resulting in high breakage rates and low integrity rates during processing. Second, the scallop's adductor muscle possesses strong contractile force, requiring precise force control during shell opening and column separation. Excessive force can damage the column tissue, while insufficient force fails to achieve separation. Existing equipment lacks real-time force feedback and adaptive force control mechanisms. Third, both the scallop shell and column are biological soft tissues, characterized by fragility, significant deformation, and slippery surfaces. Traditional rigid mechanical gripping methods easily cause crushing damage, making flexible and non-destructive operation impossible.

[0004] In recent years, soft robotics technology has been gradually applied in the food processing field, enabling the gentle grasping of fragile items through flexible materials and biomimetic design. However, existing soft robotic systems mostly employ open-loop control or simple force feedback control, lacking the ability to coordinate and regulate multiple parameters according to the characteristics of organisms, making it difficult to meet the precision handling requirements of complex biological objects such as scallops. Furthermore, existing systems have technological gaps in areas such as species identification, adaptive parameter matching, and multi-mode force control, failing to achieve truly intelligent and adaptive scallop harvesting.

[0005] Therefore, there is an urgent need to develop a parameter-controlled multi-state biomimetic scallop adductor intelligent harvesting system that can intelligently identify scallop species, adaptively adjust operating parameters, achieve precise multi-state intensity control, and has closed-loop feedback control capabilities, in order to solve the technical problems in the automated harvesting of multiple scallop species, improve harvesting efficiency, reduce damage rate, and ensure product quality. Summary of the Invention

[0006] The purpose of this invention is to provide a parameter-controlled multi-state biomimetic scallop adductor harvesting system to solve the technical problems in existing scallop adductor harvesting equipment, such as the inability to adaptively handle multiple scallop varieties, lack of precise force control, and high breakage rate during operation.

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] This invention provides a multimodal biomimetic scallop adductor muscle extraction system. The system collects morphological and mechanical data of scallops through a multimodal sensing module. The morphological data includes shell length, shell thickness, and adductor muscle size information, while the mechanical data includes adductor muscle tension signals. A species identification and parameter matching module determines the scallop species type based on the morphological data. If the scallop species type matches a preset range, it acquires the corresponding initial set of control parameters, which includes a force control range and operation timing parameters. An adaptive parameter control module, based on the mechanical state data and the initial control parameter set, activates a multimodal force control mechanism if the adductor muscle tension signal matches a first tension threshold range. This adjusts the execution force parameter to the first force interval within the force control range; otherwise, it adjusts it to the second force interval within the range. A biomimetic execution module drives a flexible operating unit to perform shell opening and adductor muscle separation operations based on the adjusted execution force parameters. The flexible operating unit includes a multi-segment biomimetic tentacle structure and a force feedback sensor array. The closed-loop feedback control module monitors the force and displacement data of the force feedback sensor array in real time. If the force data exceeds the safe force threshold, it generates a control command and transmits it to the adaptive parameter control module to dynamically adjust the execution force parameter.

[0009] In one possible implementation, the multimodal perception module includes a high-precision visual acquisition unit, a mechanical sensing unit, and a morphological parameter extraction unit. The high-precision visual acquisition unit acquires 3D morphological and surface texture data of the scallop using a depth camera and a spectral imaging device. The 3D morphological data includes point cloud coordinate information. The mechanical sensing unit detects the adductor muscle tension signal in real time using a pressure sensor array, with a sampling frequency of no less than 50Hz to ensure the capture of dynamic changes in the tension signal. The morphological parameter extraction unit performs morphological analysis on the 3D morphological data, calculating the scallop shell length, shell thickness, and columnar dimensions. The morphological analysis employs edge detection and region segmentation algorithms to accurately extract the scallop's key geometric features.

[0010] In one possible implementation, the species identification and parameter matching module includes a species feature library, a species discrimination unit, and a parameter query unit. The species feature library stores standard morphological feature templates for multiple scallop species and corresponding parameter configuration tables. These scallop species include the Yesso scallop and the comb-scallop, each with a unique morphological feature distribution range and optimal operating parameters. The species discrimination unit compares the morphological feature data with the standard morphological feature templates. If the similarity is greater than a first similarity threshold (exemplarily, this threshold can be set to 85%), the scallop species type is determined to be the species corresponding to the standard morphological feature template. The parameter query unit, based on the determined scallop species type, queries and obtains an initial set of parameter configuration tables, achieving rapid matching from species to parameters.

[0011] In one possible implementation, based on extensive experimental data and biomechanical analysis, the first tension threshold range is preferably set to 8N to 15N, which covers the adductor muscle tension distribution of common scallop species in their natural state. Correspondingly, the first force range is set to 5N to 10N, suitable for cases with lower adductor muscle tension, allowing for gentle shell-opening and separation operations. The second force range is set to 10N to 18N, suitable for cases with higher adductor muscle tension, requiring greater operational force to overcome the resistance of the adductor muscle. The safe force threshold is set to 20N as the upper limit of the system's safety, preventing damage to the columnar tissue due to excessive operational force. These parameter ranges are optimized to minimize the risk of damage while ensuring a high success rate.

[0012] In one possible implementation, the adaptive parameter control module includes a tension analysis unit, a multi-mode force decision unit, and a parameter dynamic optimization unit. The tension analysis unit analyzes the adductor muscle tension signal in real time, calculating the tension change rate and tension stability index. The tension stability index reflects the degree of fluctuation in adductor muscle tension. The multi-mode force decision unit employs different force control strategies based on the tension stability index. If the tension stability index is less than the stability threshold (e.g., the standard deviation of the change rate is greater than 2 N / s), it indicates that the adductor muscle tension fluctuates significantly. In this case, a gradual force control strategy is adopted, gradually increasing the operating force to avoid sudden force changes that could cause stress in the scallop. If the tension stability index is greater than or equal to the stability threshold, it indicates that the adductor muscle tension is relatively stable. In this case, a rapid force control strategy can be adopted, directly adjusting to the target force range to improve operational efficiency. The parameter dynamic optimization unit combines the force control strategy with the force control range in the initial parameter set, calculates and outputs the execution force parameters through an optimization algorithm, achieving precise force control.

[0013] In one possible implementation, the bionic actuation module includes a multi-segment tentacle actuator, a force feedback acquisition unit, and a collaborative controller. The multi-segment tentacle actuator drives the multi-segment bionic tentacle structure to perform bending and extension movements. The multi-segment bionic tentacle structure includes at least three flexible joint segments, each composed of a pneumatic actuation unit made of silicone material. Silicone material has good flexibility and biocompatibility, and the pneumatic actuation method enables smooth force adjustment. The force feedback acquisition unit collects contact force and displacement data applied by the flexible actuation unit in real time during operation. The sampling frequency is synchronized with the force sensing unit to ensure data temporal consistency. The collaborative controller generates a drive control signal based on the actuation force parameters and contact force data to adjust the air pressure value of the pneumatic actuation unit. The air pressure value is positively correlated with the output force; by precisely controlling the air pressure, continuous adjustment of the operating force can be achieved.

[0014] In one possible implementation, the closed-loop feedback control module also performs the following functions: It filters the force and displacement data, using a low-pass filter to remove high-frequency noise interference and retain valid force and displacement signals. It calculates the time derivative of the force data to obtain the force change rate, which reflects sudden changes in force during operation. If the force change rate exceeds a preset threshold (e.g., 10 N / s), it is determined to be an operational anomaly (possibly caused by the scallop suddenly struggling or the operating unit slipping), and an emergency stop command is immediately generated and transmitted to the bionic execution module to stop the operation and prevent damage to the scallop or equipment. If the force data is within the safe force threshold range but close to the upper limit (e.g., greater than 18 N but less than 20 N), a control command is generated, which includes an adjustment amount to reduce the execution force parameter, preventing force exceedance through early intervention. This multi-layered safety protection mechanism ensures the reliability and safety of the system.

[0015] In one possible implementation, the system further includes a thermal control module connected to the multimodal sensing module and the biomimetic execution module. The thermal control module microwaves the scallop before the biomimetic execution module performs the shell-opening operation. Based on studies of the thermodynamic properties of biological tissues, microwave heating selectively activates heat-sensitive proteins in the adductor muscle, causing slight contraction followed by relaxation of the muscle fibers, thereby reducing the adductor muscle tension signal. The heating temperature range is preferably set to 48°C to 52°C, which effectively reduces adductor muscle tension without causing protein denaturation or a decrease in column quality. The heating time is preferably set to 8 to 12 seconds; too short a heating time has little effect, while too long a heating time may affect product quality. Through thermal pretreatment, the force required for subsequent shell-opening operations can be reduced by 30% to 50%, significantly improving the success rate of the operation and the integrity rate of the column.

[0016] In one possible implementation, the system further includes a vibration separation auxiliary module connected to the biomimetic execution module. During the separation of the scallop shell and the scallop, the vibration separation auxiliary module applies a specific vibration signal to assist in the separation of the scallop shell and the scallop. The vibration frequency is preferably set to 20Hz to 30Hz, a range close to the natural frequency of the connecting tissue between the scallop shell and the scallop, effectively reducing the adhesion force of the connecting tissue. The vibration amplitude is preferably set to 2mm to 4mm, a range that produces sufficient separation effect without causing mechanical damage to the scallop. The introduction of the vibration signal simulates the biomimetic mechanism by which some marine organisms use vibration for predation or escape, achieving a gentle and efficient separation through the resonance effect.

[0017] In one possible implementation, the system is applied to the automated harvesting of adductor muscles from multiple scallop species. The multimodal perception module uses a depth camera with a resolution of no less than 1920×1080 pixels and a frame rate of no less than 30fps to ensure the capture of fine features and dynamic changes in scallop morphology. The system's processing response latency is no more than 100ms, and the entire process from scallop input to adductor muscle separation can be achieved on a high-speed production line, meeting the efficiency requirements of industrial production. After extensive practical testing, the bionic execution module achieves an adductor muscle integrity rate of over 95%, a significant improvement compared to the 80% to 90% integrity rate of traditional manual operations, while reducing the breakage rate by more than 60%, greatly increasing the economic value of the product.

[0018] The beneficial effects of this invention are as follows:

[0019] First, this invention achieves intelligent identification and adaptive parameter matching of multiple scallop species through a species identification and parameter matching module. It can automatically adapt to differences in morphology and mechanical properties among different scallop species without requiring manual adjustment of equipment parameters, truly realizing intelligent operation with "one machine, multiple functions." The species identification accuracy based on morphological features can reach over 95%, and the parameter matching time is less than 50ms, meeting real-time processing requirements.

[0020] Secondly, this invention creatively proposes a multi-modal force control mechanism. Through an adaptive parameter control module, different force ranges and control strategies are dynamically selected based on the real-time state of the adductor muscle tension signal. Compared to traditional fixed-force operation methods, this approach can minimize the risk of breakage while ensuring a high success rate. Multi-modal force control allows the system to adapt to individual differences and state changes in scallops, significantly improving the robustness and reliability of the operation.

[0021] Third, this invention adopts a biomimetic design concept, using a multi-segment biomimetic tentacle structure and flexible materials to achieve gentle and non-destructive handling of scallops, solving the problem of easy crushing and damage caused by traditional rigid mechanical grasping. The number and material of the flexible joint segments have been optimized to ensure sufficient operating force while also being able to adapt to the shape of the scallop, achieving dexterous handling capabilities similar to those of a human hand.

[0022] Fourth, this invention establishes a complete closed-loop feedback control mechanism. Through a force feedback sensor array, it monitors the operating force and displacement in real time. Combined with multi-layered safety protection strategies, it can promptly detect and handle operational anomalies, preventing product breakage and equipment damage caused by equipment malfunctions or operational errors. The closed-loop control response time is less than 20ms, enabling rapid response to emergencies.

[0023] Fifth, by introducing a thermodynamic control module and a vibration separation auxiliary module, this invention further optimizes the shell opening and separation process. Through cross-domain technology transfer (thermodynamics and vibration dynamics), it achieves process innovation, reduces operational difficulty, improves product integrity and production efficiency, and provides new technical ideas for intelligent processing of aquatic products. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall architecture of the parameter-controlled multi-state biomimetic scallop adductor intelligent extraction system of the present invention;

[0025] Figure 2 This is a schematic diagram of the structural composition of the multimodal sensing module of the present invention;

[0026] Figure 3 This is a schematic diagram illustrating the workflow of the variety identification and parameter matching module of the present invention;

[0027] Figure 4 This is a schematic diagram of the multi-state force decision-making mechanism of the adaptive parameter control module of the present invention;

[0028] Figure 5 This is a schematic diagram of the flexible operation unit structure of the biomimetic execution module of the present invention;

[0029] Figure 6 This is a schematic diagram of the real-time monitoring and control process of the closed-loop feedback control module of the present invention;

[0030] Figure 7 This is a schematic diagram of data flow in the practical application of the system of the present invention. Detailed Implementation

[0031] Please refer to the attached document. Figures 1-7To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions in the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0032] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0033] like Figure 1 As shown, this invention provides a parameter-controlled multimodal biomimetic scallop adductor intelligent harvesting system. This system includes a multimodal sensing module 1, a species identification and parameter matching module 2, an adaptive parameter control module 3, a biomimetic execution module 4, and a closed-loop feedback control module 5. The modules are connected and communicate with each other through data interfaces and control signals, forming a complete closed-loop control system.

[0034] The multimodal sensing module 1 is responsible for collecting multi-dimensional feature information of scallops, providing a data foundation for subsequent species identification and parameter control. For example... Figure 2 As shown, the multimodal perception module 1 includes a high-precision visual acquisition unit 11, a mechanical sensing unit 12, and a morphological parameter extraction unit 13.

[0035] The high-precision visual acquisition unit 11 employs a depth camera and a spectral imaging device to perform non-contact scanning of scallops. Based on structured light or time-of-flight principles, the depth camera acquires three-dimensional point cloud data of the scallop surface, achieving a point cloud density of over 1000 points per square centimeter, ensuring precise characterization of morphological features. The spectral imaging device collects the spectral reflectance characteristics of the scallop surface. Different scallop species exhibit different spectral characteristics in specific wavelength ranges (such as the 400nm to 700nm visible light range and the 700nm to 900nm near-infrared range) due to differences in shell composition and microstructure, providing supplementary information for species identification. The working distance of the depth camera is preferably set to 30cm to 50cm, within which both field of view coverage and depth accuracy can be ensured simultaneously.

[0036] The mechanical sensing unit 12 employs a pressure sensor array to detect the tension signal of the scallop's adductor muscle in real time. The pressure sensor array includes at least 16 miniature pressure sensors distributed at the contact points between the upper and lower shells of the scallop, capable of detecting the pressure distribution generated during adductor muscle contraction. Each sensor has a measurement range of 0N to 50N, an accuracy better than 0.1N, and a sampling frequency set to 50Hz to 100Hz, enabling it to capture the dynamic changes in adductor muscle tension. The analog signals output by the sensors are amplified and converted from analog to digital before being transmitted to the data processing unit for real-time analysis.

[0037] The morphological parameter extraction unit 13 performs morphological analysis on the acquired 3D morphological data. First, a point cloud filtering algorithm is used to remove noise points and outliers, retaining the effective point cloud of the scallop. Then, an edge detection algorithm (such as the Canny operator or the Sobel operator) is used to identify the contour boundary of the scallop shell, and the shell length and shell thickness parameters are calculated through a fitting algorithm. The shell length is defined as the maximum span of the scallop shell, and the shell thickness is defined as the average thickness of the upper and lower shells. The extraction of column size information requires combining depth and texture information. The column region is separated from the background through a region segmentation algorithm, and the diameter and height of the column are calculated. The processing time for morphological parameter extraction is preferably controlled within 30ms to meet real-time requirements.

[0038] The species identification and parameter matching module 2, based on the morphological feature data collected by the multimodal sensing module 1, achieves intelligent identification of scallop species and rapid matching of corresponding control parameters. For example... Figure 3 As shown, this module includes a variety feature library 21, a variety discrimination unit 22, and a parameter query unit 23.

[0039] The Variety Characteristic Library 21 is the system's knowledge base, storing standard morphological characteristic templates for multiple scallop varieties and corresponding parameter configuration tables. Taking the Yesso scallop and the comb scallop as examples, the typical morphological characteristics of the Yesso scallop are a shell length ranging from 80mm to 120mm, a shell thickness ranging from 12mm to 18mm, a column diameter ranging from 20mm to 30mm, and a spectral characteristic with a distinct reflection peak in the 550nm band. The typical morphological characteristics of the comb scallop are a shell length ranging from 60mm to 90mm, a shell thickness ranging from 8mm to 12mm, a column diameter ranging from 15mm to 22mm, and a spectral characteristic with a distinct reflection peak in the 600nm band. Each variety is associated with a set of parameter configurations, including recommended intensity control ranges, operation timing parameters, and thermodynamic control temperatures.

[0040] The variety discrimination unit 22 implements the core algorithm for variety identification. It employs a feature vector matching method to construct feature vectors from the measured morphological feature data. ,in This refers to the shell length, in mm. Shell thickness, in mm; The diameter of the cylinder is in mm. This represents the spectral reflectance of the characteristic band. For each standard template in the variety characteristic library... Its eigenvector is denoted as Similarity is calculated using normalized Euclidean distance:

[0041] ,

[0042] in: For the measured eigenvector and the first The similarity of the templates, with values ​​ranging from 1 to 2. ; Represents the Euclidean norm, i.e. ; These are the measured feature vectors; For the first The standard feature vector of a variety template.

[0043] After calculating the similarity of all templates, the template with the highest similarity is selected as the candidate variety. A first similarity threshold is set. (Typical value is 0.85), if the maximum similarity If so, then the scallop species type is determined to be the species corresponding to the template; if This indicates that the measured scallops do not match the known species templates well, and may belong to unknown species or abnormal samples. The system issues a warning and uses the default parameter configuration.

[0044] The parameter query unit 23, based on the scallop species type determined by the species identification unit 22, queries the corresponding initial parameter set from the parameter configuration table in the species characteristic database 21. Taking the Yesso scallop as an example, its recommended initial parameter set includes: a strength control range of... The first intensity range is The second intensity range is The first tension threshold range is Thermodynamic temperature control is Heating time is The vibration frequency is The vibration amplitude is These parameters have been optimized through extensive experimentation to minimize damage while ensuring a high success rate for operations.

[0045] The adaptive parameter control module 3 is the core decision-making module of the system, responsible for dynamically adjusting the execution force parameters based on real-time mechanical state data. For example... Figure 4 As shown, the module includes a tension analysis unit 31, a multi-state force decision unit 32, and a parameter dynamic optimization unit 33.

[0046] The tension analysis unit 31 performs real-time analysis on the adductor muscle tension signal acquired by the mechanical sensing unit 12. The adductor muscle tension signal is denoted as... , for time The function of tension analysis includes the following aspects: First, calculating the average tension. This value reflects the average contractile force of the adductor muscle. The calculation formula is:

[0047] ,

[0048] in: This represents the average adductor muscle tension within the time window, expressed in N. This represents the number of sampling points; For the first The tension value at each sampling time point, in N; For the first Each sampling time is measured in seconds (s).

[0049] Secondly, calculate the rate of change of tension. This value reflects the rate of change of adductor muscle tension over time. It is calculated using the difference method.

[0050] ,

[0051] in: For the first The rate of change of tension at each moment, in N / s; The sampling time interval is expressed in seconds (s). and These are the tension values ​​at the current moment and the previous moment, respectively, in N.

[0052] Finally, the tension stability index is calculated. This indicator measures the degree of tension fluctuation. It is calculated using the standard deviation.

[0053] ,

[0054] in: This is a tensile stability index (standard deviation), with units of N; This represents the number of sampling points; For the first The tension value at each sampling time point, in N; This represents the average tension, expressed in N.

[0055] The multi-state force decision unit 32 employs different force control strategies based on tension analysis results. A stability threshold is set. (Typical value is 2N), the judgment logic is as follows: If This indicates that the adductor muscle tension fluctuates significantly and is in an unstable state, possibly due to stress response or individual differences in the scallops. In this case, a gradual force control strategy is adopted, with the initial force set at the lower limit of the force control range. Then in fixed increments (Typical value is 1N) Gradually increase the force, holding the position for a period of time (e.g., 0.5s) after each increase and observing the tension response, until the operation is successful or the maximum force limit is reached. A gradual approach can avoid sudden impacts that could cause further stress to the scallops, thus increasing the success rate of the operation.

[0056] like This indicates that the adductor muscle tension is relatively stable, allowing for a rapid force adjustment strategy. Based on the average tension value... With the first tension threshold range The relationship directly determines the target strength range. If This indicates that the adductor muscle tension is within the normal range, so the first force range should be selected. As the target interval. or This indicates abnormal adductor muscle tone (too low or too high), so the second force range should be selected. As the target range, the rapid intensity control strategy has a short response time, which can improve processing efficiency.

[0057] The parameter dynamic optimization unit 33 calculates the specific execution force parameters based on the force control strategy and the target force range. For a gradual strategy, the intensity of implementation should be in accordance with... Gradually increase, among which To adjust the number of times. For fast strategies, the execution strength is directly set to the median of the target interval, such as... In practical applications, feedback control algorithms (such as PID controllers) can be introduced to finely adjust the execution force, making the actual applied force closer to the ideal value.

[0058] The bionic actuation module 4 is responsible for converting the actuation force parameters into actual mechanical operations. For example... Figure 5 As shown, the module includes a multi-segment hand actuator 41, a force feedback acquisition unit 42, and a collaborative controller 43.

[0059] The multi-segment tentacle actuator 41 employs a pneumatic drive to control the movement of the multi-segment bionic tentacle structure. Inspired by octopus tentacles, the multi-segment bionic tentacle structure exhibits excellent flexibility and adaptability. Each tentacle consists of three flexible joint segments connected in series, each segment ranging from 50mm to 80mm in length. The internal structure is a hollow air bladder covered with a 5mm thick silicone layer. The silicone material is preferably selected with a Shore A hardness of 30 to 50, ensuring sufficient softness to adapt to the irregular shape of the scallop surface while also possessing sufficient rigidity to transmit operational force. The air bladder's inflation pressure is controlled by an electromagnetic proportional valve, ranging from 0kPa to 300kPa, with the pressure positively correlated with the joint bending angle. By independently controlling the air pressure of the three joint segments, complex three-dimensional motion trajectories of the tentacle can be achieved.

[0060] Force feedback acquisition unit 42 collects contact force and displacement information between the flexible manipulator and the scallop. Miniature force sensors are installed at the end and middle of each tentacle, totaling six sensors (three per tentacle, assuming the system uses two tentacles), forming a force feedback sensing array. The force sensors are thin-film pressure sensors, less than 1 mm thick, which can be embedded in the silicone layer without affecting the tentacle's flexibility. Simultaneously, displacement sensors are installed on the tentacle base, employing magnetostriction or optical scale principles to measure the tentacles' extension, contraction, displacement, and angular changes in real time. Force and displacement data are transmitted to the data processing unit via a high-speed data acquisition card (sampling frequency 100 Hz) to provide real-time feedback for closed-loop control.

[0061] The coordinating controller 43 is the core control unit of the bionic actuation module, responsible for coordinating the movements and force outputs of multiple tentacles. The coordinating controller receives actuation force parameters. and real-time contact force data Calculate the required air pressure for each pneumatic drive unit. The relationship between air pressure and output force can be established through calibration experiments, and it is usually approximately linear.

[0062] ,

[0063] in: For the first The air pressure value of each pneumatic drive unit, in kPa; Base pressure, in kPa; This is the pressure-to-force conversion factor, with units of kPa / N; For the first The target output force of each drive unit is expressed in N.

[0064] Target output force Based on the execution force parameter The task allocation for each tentacle is determined. Assuming two tentacles are used, the total execution force is evenly distributed between the two tentacles, and the force is distributed among the three joint segments of each tentacle according to a certain ratio (e.g., 3:2:1) to achieve a reasonable force distribution. The collaborative controller also synchronizes the movements of multiple tentacles, ensuring that the two tentacles move at the same speed and rhythm, avoiding uneven force distribution on the scallop due to uncoordinated movements.

[0065] The closed-loop feedback control module 5 enables real-time monitoring and dynamic adjustment of the entire operation process. For example... Figure 6 As shown, this module continuously analyzes the force and displacement data collected by the force feedback sensor array, promptly detects anomalies, and takes corresponding measures.

[0066] The force data was filtered using a low-pass filter with a cutoff frequency of 20Hz. This effectively removed high-frequency interference caused by sensor noise, mechanical vibration, etc., while retaining the valid force signal. The filtered force data is denoted as... .

[0067] The method for calculating the rate of change of force is as follows:

[0068] ,

[0069] in: For the first The rate of change of contact force at each instant, in N / s; and These are the filtered intensity values ​​for the current time and the previous time, respectively, in N. The sampling time interval is expressed in seconds (s).

[0070] Set preset rate threshold (Typical value is 10 N / s), if The system is flagged as malfunctioning. This malfunction could be caused by various factors, such as a sudden, violent struggle by the scallop leading to a sharp increase in tension, slippage of the tentacles causing a sudden decrease in contact force, or mechanical failure resulting in loss of control over the force. Regardless of the cause, the system immediately generates an emergency stop command, rapidly closing the pneumatic valve via the co-controller to halt the tentacles' movement and prevent further damage.

[0071] For situations approaching the safety limit, set a warning threshold. (Typical value is 18N), if and This indicates that the force has approached but not yet exceeded the safe force threshold. (Typical value is 20N), the system generates a control command, instructing the adaptive parameter control module to reduce the execution intensity parameter. The control command includes a suggested adjustment amount. The calculation formula is:

[0072] ,

[0073] in: This represents the intensity adjustment (negative values ​​indicate a decrease), measured in N. The adjustment coefficient has a range of values. ; This is the current filtered intensity value, in N. This is the warning threshold, expressed in N units.

[0074] After receiving the control command, the adaptive parameter control module subtracts the adjustment amount from the execution force parameter, and the updated execution force parameter is transmitted to the bionic execution module to achieve dynamic adjustment of the force. This proactive intervention mechanism can effectively prevent the force from exceeding the limit and improve the system's safety.

[0075] To further improve the system's performance and applicability, in a preferred embodiment, the system also includes a thermal control module and a vibration separation auxiliary module.

[0076] The thermal control module employs microwave heating technology to pre-treat the scallops. The microwave generator operates at a frequency of 2.45 GHz, with an adjustable output power range of 100W to 500W. The scallops are placed in the microwave cavity, and by precisely controlling the microwave power and heating time, the temperature of the scallop's adductor muscle is raised to a preset range (48℃ to 52℃). Temperature monitoring uses an infrared temperature sensor to measure the surface temperature of the adductor muscle in real time, and feedback control ensures that the temperature does not exceed the set range. The optimal heating time is 10 seconds, which allows the adductor muscle to be fully heated without affecting the quality of the scallop body. After heating, the tension of the adductor muscle can be reduced by 30% to 50%, significantly reducing the difficulty of subsequent shell-opening operations.

[0077] The vibration separation auxiliary module uses a piezoelectric ceramic vibrator to generate high-frequency vibration. The vibrator is mounted on the base of the tentacle, and the vibration direction is consistent with the operating direction of the tentacle. The vibration frequency is controlled by a function generator, preferably set to 25Hz, and the amplitude is adjusted by the driving voltage, preferably set to 3mm. During the column separation operation, the tentacle applies a constant tensile force while superimposing a vibration signal. The periodic stress generated by the vibration can effectively reduce the adhesion force between the scallop shell and the column connecting tissue, promoting the separation process. Experiments show that after introducing vibration assistance, the average force required for separation can be reduced by 40%, the separation time can be shortened by 50%, and the column integrity rate can be increased by about 5 percentage points.

[0078] In practical applications, the system's workflow is as follows: Figure 7As shown in the diagram, the scallops are first transported to the working area of ​​the multimodal sensing module via a conveyor. A high-precision visual acquisition unit scans the scallops to acquire three-dimensional morphological and spectral data. The mechanical sensing unit then monitors the adductor muscle tension signal. The morphological parameter extraction unit processes the visual data to extract key parameters such as shell length, shell thickness, and columnar dimensions.

[0079] Secondly, the species identification and parameter matching module identifies the species based on morphological feature data, matches it with standard templates in the species feature library, calculates the similarity, and determines the scallop species type. Based on the species type, it queries the corresponding initial set of control parameters from the parameter configuration table, including recommended intensity control range, tension threshold, and thermodynamic control parameters.

[0080] Then, if the system is equipped with a thermal control module, the scallops are pretreated with microwave heating to reduce the tension of the adductor muscle. The adaptive parameter control module performs tension analysis based on the real-time adductor muscle tension signal and the initial set of control parameters, determines whether it meets the first tension threshold range, selects an appropriate intensity range and control strategy, and calculates the execution intensity parameters.

[0081] Next, the bionic actuation module drives the flexible operating unit to perform the shell-opening and scallop separation operations based on the actuation force parameters. A multi-segment tentacle actuator controls the pneumatic drive unit, adjusting the air pressure at each joint segment to ensure the tentacle contacts the scallop with appropriate posture and force. During operation, if a vibration separation auxiliary module is configured, vibration signals are superimposed to assist separation. A force feedback acquisition unit collects contact force and displacement data in real time and transmits it to the closed-loop feedback control module.

[0082] Finally, the closed-loop feedback control module filters and analyzes the force data, calculates the rate of change of force, and monitors for any abnormalities or approaching the safety limit. If an abnormality is detected, an emergency stop command is immediately generated; if the force is approaching the limit, an adjustment command is generated to notify the adaptive parameter control module to adjust the execution force parameters. This real-time closed-loop control ensures the safety and reliability of the operation process.

[0083] After the scallop adductor muscles are separated from the adductor muscles, they are conveyed to the next process (such as cleaning, grading, and packaging) by a conveyor device, while the scallop shells enter the waste disposal process. The system records the processing data for each scallop, including species type, morphological parameters, force parameters, processing time, and integrity rate, for subsequent data analysis and system optimization.

[0084] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art should understand that modifications or equivalent substitutions to the technical solutions of the present invention do not depart from the spirit and scope of the technical solutions of the present invention and should be covered within the scope of the claims of the present invention.

Claims

1. A control polymorphic bionic scallop columnar intelligence taking system, characterized in that , comprising: A multi-modal perception module for collecting morphological feature data and mechanical state data of the scallop, the morphological feature data including shell length, shell thickness and column size information, and the mechanical state data including a closed muscle tension signal; A variety identification and parameter matching module connected with the multi-modal perception module, configured to determine a scallop variety type based on the morphological feature data, and if the scallop variety type meets a preset variety range, obtain a corresponding initial parameter control parameter set, the initial parameter control parameter set including a force regulation range and an operation timing parameter; An adaptive parameter regulation module connected with the variety identification and parameter matching module and the multi-modal perception module, configured to, according to the mechanical state data and the initial parameter control parameter set, if the closed muscle tension signal meets a first tension threshold range, start a multi-state force regulation mechanism, and adjust an execution force parameter to a first force interval in the force regulation range, and if the closed muscle tension signal does not meet the first tension threshold range, adjust the execution force parameter to a second force interval in the force regulation range; A bionic execution module connected with the adaptive parameter regulation module, configured to drive a flexible operation unit to perform scallop opening and column separation operations according to the adjusted execution force parameter, the flexible operation unit including a multi-section bionic tentacle structure and a force feedback sensor array; A closed-loop feedback control module connected with the bionic execution module and the adaptive parameter regulation module, configured to monitor force data and displacement data of the force feedback sensor array in real time, and if the force data exceeds a safe force threshold, generate a regulation instruction and transmit it to the adaptive parameter regulation module to dynamically adjust the execution force parameter.

2. The system of claim 1, wherein The multi-modal perception module comprises: A high-precision visual acquisition unit for collecting three-dimensional morphological data and surface texture data of the scallop through a depth camera and a spectral imaging device, the three-dimensional morphological data including point cloud coordinate information; A mechanical sensing unit for detecting the closed muscle tension signal in real time through a pressure sensor array, with a sampling frequency of no less than 50 Hz; A morphological parameter extraction unit connected with the high-precision visual acquisition unit, configured to perform morphological analysis on the three-dimensional morphological data, and calculate the shell length, the shell thickness and the column size information, the morphological analysis adopting an edge detection and region segmentation algorithm.

3. The system of claim 1, wherein The variety identification and parameter matching module comprises: A variety feature library for storing standard morphological feature templates of a plurality of scallop varieties and corresponding parameter control parameter configuration tables, the plurality of scallop varieties including Patinopecten and Chlamys; A variety discrimination unit connected with the variety feature library, configured to compare the morphological feature data with the standard morphological feature templates in terms of similarity, and if the similarity is greater than a first similarity threshold, determine the scallop variety type as a variety corresponding to the standard morphological feature template; A parameter query unit connected with the variety feature library and the variety discrimination unit, configured to, according to the determined scallop variety type, query and obtain the initial parameter control parameter set from the parameter control parameter configuration table.

4. The system of claim 1 or 2, wherein The first tension threshold range is 8N to 15N, the first force range is 5N to 10N, the second force range is 10N to 18N, and the safe force threshold is 20N.

5. The system of claim 1, wherein The adaptive parameter control module includes: The tension analysis unit is used to analyze the adductor muscle tension signal in real time and calculate the tension change rate and tension stability index. A multi-state force decision unit, connected to the tension analysis unit, is used to, based on the tension stability index, employ a gradual force control strategy if the tension stability index is less than a stability threshold, and employ a rapid force control strategy if the tension stability index is greater than or equal to the stability threshold. The parameter dynamic optimization unit, connected to the polymorphic force decision unit, is used to calculate and output the execution force parameter based on the force control strategy and the force control range in the initial parameter control parameter set.

6. The system of claim 1, wherein The biomimetic execution module includes: A multi-segment tentacle actuator is used to drive the multi-segment bionic tentacle structure to perform bending and extension movements. The multi-segment bionic tentacle structure includes at least three flexible joint segments, each of which is composed of a pneumatic drive unit made of silicone material. A force feedback acquisition unit, connected to the force feedback sensor array, is used to acquire in real time the contact force data and displacement data applied by the flexible operating unit during operation; A collaborative controller, connected to the multi-segment hand actuator and the force feedback acquisition unit, is used to generate a drive control signal to adjust the air pressure value of the pneumatic drive unit based on the execution force parameter and the contact force data.

7. The system of claim 1, wherein The closed-loop feedback control module is also used for: The force data and displacement data are filtered to remove high-frequency noise interference; Calculate the time derivative of the force data to obtain the rate of change of force; If the rate of change of force exceeds a preset rate threshold, it is determined to be an operational abnormality, and an emergency stop command is immediately generated and transmitted to the bionic execution module to stop the operation. If the force data is within the safe force threshold range but close to the upper limit, the control instruction is generated, which includes an adjustment amount that reduces the execution force parameter.

8. The system of claim 1, wherein The system also includes: A thermal control module, connected to the multimodal sensing module and the bionic execution module, is used to microwave heat the scallop before the bionic execution module performs the shell-opening operation. The heating temperature range of the microwave heating treatment is 48°C to 52°C, and the heating time is 8 seconds to 12 seconds, so as to activate the contraction of the adductor muscle and reduce the tension signal of the adductor muscle.

9. The system of claim 1, wherein The system also includes: The vibration separation auxiliary module, connected to the biomimetic execution module, is used to apply a vibration signal with a frequency of 20Hz to 30Hz and an amplitude of 2mm to 4mm during the separation of the scallop shell and the scallop.

10. The system of claim 1, wherein The system is applied to the automatic precious column collection scene of multi-species scallops, the resolution of the depth camera used by the multi-modal perception module is not less than 1920*1080 pixels, the frame rate is not less than 30 fps, the processing response delay of the system is not more than 100 ms, and the integrity rate of the bionic execution module to the precious column of scallops reaches more than 95%.

Images