Servo clamping device with surface thin film force sensing and force state detection method

By introducing a thin-film pressure sensor and layered dual feedback control into the micro servo gripping device, the problems of large gripping force feedback error and contact recognition lag in the servo gripping device are solved, realizing high-precision and reliable gripping control, which is suitable for scenarios such as micro-electronic assembly and biological sample handling.

CN123103741APending Publication Date: 2026-11-17RECONOVA TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611451944.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-18
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing micro servo clamping devices suffer from strong indirectness and large errors in clamping force feedback, cannot directly reflect the true state of the clamped object, have delayed contact recognition, poor compliance, and are not suitable for clamping fragile or flexible workpieces.

Method used

A servo clamping device with surface thin-film force sensing is adopted. By arranging thin-film pressure sensors on the clamping fingers to directly collect the force on the workpiece, combined with a hierarchical dual feedback control architecture and a three-stage compliant force control algorithm, high-precision and reliable clamping control is achieved.

Benefits of technology

It achieves high-precision and reliable clamping force detection, and can accurately identify working conditions such as contact time, off-center load and micro-slippage, thereby improving the stability and adaptability of clamping operations. It is suitable for scenarios such as micro-electronic assembly and biological sample handling.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123103741A_ABST
    Figure CN123103741A_ABST
Patent Text Reader

Abstract

This invention discloses a servo clamping device with surface thin-film force sensing and a method for detecting force state. The device includes a gripper servo motor, left and right gripping fingers, a thin-film pressure sensing unit, a signal conditioning module with temperature compensation, and a main control unit. The left and right gripping fingers adopt a three-layer composite structure of a substrate, a thin-film pressure sensor, and a wear-resistant protective layer. The thin-film pressure sensor is directly attached to the working surface of the gripping fingers, and the detection path bypasses the transmission mechanism. The main control unit uses the direct force signal from the thin-film pressure sensor as the core feedback, while simultaneously incorporating the torque and angle signals from the servo motor as auxiliary feedback, constructing a layered dual-feedback control architecture. The method includes steps such as coarse position approach, precise contact recognition, three-stage compliant constant force closed-loop control, multi-dimensional state analysis, and abnormal response. This invention achieves direct and accurate detection of the actual force on the clamped workpiece and compliant clamping, suitable for high-precision clamping scenarios such as microelectronic assembly and biological sample handling.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of micro-precision clamping and force feedback control technology, and in particular to a servo clamping device with surface thin film force sensing and a method for detecting the force state. Background Technology

[0002] Miniature servo grippers, due to their small size and high control integration, are widely used in miniature gripping scenarios such as small industrial robots, desktop automation equipment, and dexterous hand end effectors. Most existing miniature servo gripping solutions generally estimate the gripping force indirectly by directly utilizing the torque feedback function built into the servo motor; that is, the gripping force of the fingers is calculated by converting the motor's output torque into the transmission ratio. However, existing servo gripping solutions have the following drawbacks: First, the force feedback is highly indirect and prone to large errors. The torque feedback of the servo motor can only reflect the load torque at the motor output shaft end. It needs to be transmitted to the working surface of the clamping fingers through transmission mechanisms such as gear sets and crank connecting rods. Mechanical friction, return clearance, elastic deformation of components, and assembly tolerances during the transmission process will all generate force loss and interference, resulting in a large deviation between the calculated clamping force and the actual force on the workpiece, which cannot meet the accuracy requirements of precision clamping.

[0003] Second, it cannot directly reflect the true state of the clamped object. Servo torque feedback can only output the macroscopic value of the overall load, and cannot sense the local pressure, off-center load state, and contact position of the contact surface between the clamping fingers and the workpiece. It is also difficult to identify fine working conditions such as micro-slippage, local contact, and multi-point contact of the workpiece, which can easily lead to problems such as workpiece crushing, slippage, and clamping deviation.

[0004] Third, contact recognition is delayed and lacks flexibility. Relying solely on torque changes to determine workpiece contact requires a significant change in motor load before recognition can occur. The contact moment is prone to rigid impact, making it unsuitable for clamping fragile or flexible workpieces.

[0005] Although flexible thin-film pressure sensors have been applied in some clamping scenarios, existing technologies have not addressed the inherent defects of servo torque feedback, nor have they clarified the collaborative mechanism of "servo torque for coarse control and surface force sensor for accurate perception of the actual force on the workpiece," thus failing to fully leverage the driving capabilities of micro servos and the direct force measurement advantages of thin-film sensors. Summary of the Invention

[0006] The purpose of this invention is to provide a servo clamping device with surface thin film force sensing that can achieve high precision and high reliability compliant clamping under miniaturized volume constraints, and a method for detecting the force state, so as to solve the problem of large indirect force measurement error of servo torque.

[0007] To achieve the above objectives, the solution of the present invention is: A servo gripping device with surface thin film force sensing includes a gripper servo motor, a gripper base, a left gripper, a right gripper, a thin film pressure sensing unit, a signal conditioning module with temperature compensation, and a main control unit. The gripper servo motor is mounted and fixed in the middle of the gripper base. The output end of the gripper servo motor is connected to the left and right grippers for driving the left and right grippers to complete the gripping action of clamping in opposite directions or opening in opposite directions. The gripper servo motor has angle feedback output capability and torque feedback output capability. The thin-film pressure sensing unit includes thin-film pressure sensors arranged on the left and right gripping fingers respectively, which are used to synchronously acquire the normal contact pressure between the two gripping fingers and the workpiece. Thin-film pressure sensors are respectively attached to the working surfaces of the left and right clamping fingers facing the clamping space. The sensing area of ​​the thin-film pressure sensor completely covers the clamping contact surface. The outer side of the thin-film pressure sensor is covered with a wear-resistant protective layer, which completely covers the sensing area of ​​the thin-film pressure sensor. The thin-film pressure sensor directly transmits force to the clamped workpiece, and its detection path does not pass through the transmission mechanism. The temperature-compensated signal conditioning module is electrically connected to the thin-film pressure sensing unit and has a built-in voltage divider conversion circuit, NTC temperature compensation network, instrumentation amplifier circuit and second-order low-pass filter circuit. The NTC temperature compensation network is matched in series with the thin-film pressure sensor to compensate for the temperature drift error of the piezoresistive sensor. The signal conditioning module is used to convert the weak resistance change signal of the thin-film pressure sensor into a stable standard voltage signal. The main control unit is bidirectionally connected to the signal conditioning module and the gripper servo motor; the main control unit has a built-in AD acquisition unit, a segmented calibration model, a three-stage compliant force control algorithm and a multi-dimensional state recognition algorithm; The main control unit uses the direct force signal from the thin-film pressure sensor as the core feedback to perform closed-loop control of the entire process, including contact recognition, compliant approach, and constant force maintenance. At the same time, the main control unit connects the torque and angle signals from the gripper servo motor as auxiliary feedback for overload monitoring and redundancy protection at the transmission end, thus constructing a hierarchical dual feedback control architecture of "core precision control + auxiliary protection".

[0008] Preferably, the gripper servo motor is a micro servo motor, and the left and right gripper fingers are connected to the output end of the gripper servo motor via a gear set or crank-connecting rod mechanism.

[0009] Preferably, the NTC temperature compensation network in the signal conditioning module and the thin-film pressure sensor form a series voltage divider structure, and the main control unit corrects the zero-point reference in real time according to the output of the NTC temperature compensation network; The segmented calibration model of the main control unit establishes a three-dimensional calibration mapping table of "voltage-temperature-force value". The main control unit obtains the corresponding force value by looking up the table or interpolation calculation based on the current ambient temperature and standard voltage signal.

[0010] The present invention can also adopt the following technical solutions: A method for detecting the servo clamping force state of a servo clamping device with a surface thin film force sensor, comprising the following steps: S1. System Initialization and Temperature Drift Self-Calibration: The device completes hardware self-test upon power-up, performs temperature drift self-calibration under no-load conditions, collects the zero-point output of the thin-film sensor at the current ambient temperature, and corrects the zero-point reference by combining the built-in temperature compensation model; multi-point gradient loading calibration is performed using a standard force gauge to establish a three-dimensional calibration mapping table of "voltage-temperature-force value" and store it in the main control unit; at the same time, the correspondence between servo torque and transmission load is calibrated, and the secondary overload protection threshold is set. S2. Servo position coarse control approach and collision warning: The main control unit sends position commands to the gripper servo motor and controls the left and right gripper fingers to open to the preset working opening through angle feedback. The gripper approaches the workpiece to be gripped with the external mechanism. At this stage, the position open-loop control is the main method and the thin film sensor operates at low frequency sampling. If an abnormal pressure change is detected, the feed will stop immediately and a collision warning will be triggered. S3. Precise contact recognition based on direct force signal: The gripper servo motor drives the gripper fingers to close at a low speed. When the pressure value detected by the thin film pressure sensors on both sides is greater than the contact threshold and is stably exceeded for N consecutive sampling cycles, it is determined that the gripper fingers have completed effective contact with the workpiece. The contact determination is based on the direct force signal of the thin film. S4. Three-stage compliant constant force closed-loop control: After entering the clamping stage, the system executes a three-stage progressive force control strategy: Soft-touch buffering phase: After contact, the closing continues at an extremely low speed, and the force value rises at a set slope to 80% of the target force to avoid rigid impact; Smooth transition phase: Gradually increase the closed-loop gain to smoothly approach the target force value and suppress overshoot; Constant force holding stage: Closed-loop adjustment is performed using an incremental PID algorithm to stabilize the actual force on the workpiece within the allowable error range of the target force; S5. Multi-dimensional Clamping Status Intelligent Judgment: Based on real-time force data from a thin-film pressure sensor, three types of working conditions are identified simultaneously: off-center load identification, micro-slip detection, and overload protection. S6. Abnormal Response and Data Traceability Storage: After identifying abnormal working conditions, the main control unit executes the corresponding control strategy and completes the operation after the clamping is stable. It automatically stores the force curve, temperature data, status events and servo auxiliary data of this clamping operation.

[0011] Preferably, in step S1, the temperature drift self-calibration supports two modes: single calibration upon power-on and periodic dynamic calibration during operation; during operation, a zero-point correction is performed once every set interval to offset the temperature drift accumulated during long-term operation; in step S3, the contact threshold is a pressure threshold adaptively set according to the workpiece material and surface characteristics.

[0012] Preferably, in step S4, the closing speed of the soft-touch buffering stage is not higher than 30% of the closing speed of the smooth transition stage, and the closed-loop gain of the smooth transition stage is not higher than 60% of the closed-loop gain of the constant force holding stage.

[0013] Preferably, in step S5, a multi-dimensional state recognition algorithm is used to identify three working conditions, including: Off-center load identification: Calculate the ratio of the difference between the force values ​​measured by the left and right clamping fingers. If the difference exceeds the set threshold, it is determined to be an off-center clamping load. Micro-slip detection: The sliding window variance algorithm is used to extract the high-frequency fluctuation characteristics of the force signal. When the variance within the window exceeds the slip threshold and the force value shows a downward trend, it is determined to be micro-slip of the workpiece. The overload protection is set with two levels of overload thresholds. The first level threshold triggers a deceleration warning, and the second level threshold triggers an immediate release protection.

[0014] Preferably, in step S5, the slip determination of the micro-slip detection combines the dual conditions of "variance mutation + force trend" to control the false judgment rate within 1%, and the window length of the sliding window variance algorithm is set to 50~100 sampling points.

[0015] Preferably, in step S6, the corresponding control strategy includes: outputting alignment correction prompts when there is an off-center load, increasing force in a stepped manner with a set step distance when there is slight slippage, and performing graded protection when there is an overload.

[0016] Preferably, the data stored in step S6 is used for subsequent quality traceability and parameter iteration optimization, and the main control unit automatically corrects the contact threshold, the slip threshold and the target force value based on historical clamping data.

[0017] Compared with the prior art, the servo clamping device with surface thin film force sensing and the force state detection method of the present invention have the following advantages: Force detection is direct and accurate, compensating for the shortcomings of servo torque feedback at its source.

[0018] This invention places thin-film pressure sensors directly on the contact surfaces of the left and right clamping fingers and the workpiece, bypassing the transmission chain to directly collect the actual force on the workpiece. It is completely unaffected by gear friction, linkage clearance, and mechanism deformation. The force detection accuracy and reliability are far higher than those of indirect estimation by servo torque. In principle, it solves the core problem of "inaccurate measurement of the actual force on the workpiece" in servo clamping.

[0019] It provides comprehensive clamping status perception, overcoming the limitations of torque feedback in recognition.

[0020] This invention relies on force sensing data from the surface of the gripper fingers to accurately identify various working conditions such as contact time, off-center load, micro-slippage, and local overload. These are details that servo torque feedback cannot effectively perceive, greatly improving the stability of the gripping operation and the adaptability to irregular and fragile workpieces.

[0021] Dual feedback coordination balances accuracy and reliability.

[0022] This invention uses surface thin-film force sensing as the core control basis to ensure the accuracy of clamping force control; and servo torque feedback as a secondary auxiliary to realize overload protection of the transmission end and fault monitoring of the mechanism. The two perform their respective functions and work together, which not only solves the problem of insufficient torque feedback accuracy, but also retains the original protection capabilities of the servo system.

[0023] The structure is simple and easy to integrate. The thin-film sensor is ultra-thin and flexible, and can be directly mounted without modifying the gripper substrate structure. The overall device is small in size and light in weight, and can be directly adapted to the upgrade and transformation of existing servo grippers. The integration cost is low and the difficulty of large-scale application is small. Attached Figure Description

[0024] Figure 1 This is a schematic diagram of the overall structure of the clamping and detection device of the present invention.

[0025] Figure 2 This is a schematic diagram of the exploded structure of the single-finger gripper of the present invention.

[0026] Figure 3 This is a block diagram illustrating the closed-loop control system principle of the present invention.

[0027] Labeling: 1. Gripper servo motor; 2. Left gripper finger; 3. Right gripper finger; 4. Gripper finger substrate; 5. Thin-film pressure sensor; 6. Wear-resistant protective layer. Detailed Implementation

[0028] To further explain the technical solution of the present invention, the present invention will be described in detail below through specific embodiments.

[0029] like Figure 1 As shown, the present invention provides a servo clamping device with surface thin film force sensing, which mainly includes a gripper servo motor 1, a left gripper finger 2, a right gripper finger 3, a gripper finger base 4, a thin film pressure sensing unit, a signal conditioning module with temperature compensation, and a main control unit.

[0030] The gripper servo motor 1 is mounted and fixed in the middle of the gripper base 4. Its output end is connected to the left gripper 2 and the right gripper 3 through a gear set or crank-connecting rod mechanism, driving the left gripper 2 and the right gripper 3 to complete the gripping action of clamping towards each other or opening away from each other. The gripper servo motor 1 is a micro servo motor with dual feedback output capability of angle feedback and torque. Its angle feedback is used for position closed-loop control, and its torque feedback is used for load monitoring at the transmission end.

[0031] The thin-film pressure sensing unit includes thin-film pressure sensors 5 arranged on the left gripper 2 and the right gripper 3 respectively, for synchronously acquiring the normal contact pressure between the gripper on both sides and the workpiece.

[0032] like Figure 2 As shown, thin-film pressure sensors 5 are respectively attached to the working surfaces of the left clamping finger 2 and the right clamping finger 3 facing the clamping space. The sensing area of ​​the thin-film pressure sensor 5 completely covers the clamping contact surface. The thin-film pressure sensor 5 is a piezoresistive flexible thin-film pressure sensor. The outer side of the thin-film pressure sensor 5 is covered with a wear-resistant protective layer 6, which completely covers the sensing area of ​​the thin-film pressure sensor 5. The wear-resistant protective layer 6 protects the sensor from mechanical wear and ensures the linearity of force transmission and response speed. The thin-film pressure sensor 5 directly transmits force to the clamped workpiece. Its detection path does not pass through any transmission links such as gears and connecting rods, thus physically decoupling the nonlinear force loss caused by transmission friction, return clearance, and elastic deformation of components.

[0033] The temperature-compensated signal conditioning module is electrically connected to the thin-film pressure sensing unit. The module incorporates a voltage divider conversion circuit, an NTC temperature compensation network, an instrumentation amplifier circuit, and a second-order low-pass filter circuit. The NTC temperature compensation network is connected in series with the thin-film pressure sensor 5 to form a temperature-sensitive voltage divider structure, used to compensate for the temperature drift error of the piezoresistive sensor in real time. The signal conditioning module amplifies, filters, and temperature-compensates the weak resistance change signal generated by the pressure applied to the thin-film pressure sensor 5, converting it into a stable standard voltage signal for output to the main control unit.

[0034] The main control unit is bidirectionally connected to the signal conditioning module and the gripper servo motor 1. The main control unit incorporates an AD acquisition unit, a segmented calibration model, a three-stage compliant force control algorithm, and a multi-dimensional state recognition algorithm. Using the direct force signal from the thin-film pressure sensor 5 as the core feedback, the main control unit achieves closed-loop control throughout the entire process, including contact recognition, compliant approach, and constant force maintenance. Simultaneously, it incorporates the torque and angle signals from the gripper servo motor 1 as auxiliary feedback for overload monitoring and redundancy protection at the transmission end, thus constructing a hierarchical dual-feedback control architecture of "core precision control + auxiliary protection." In this architecture, the thin-film force signal undertakes the core functions of precise force control and fine state recognition, while the servo torque signal undertakes the auxiliary functions of transmission chain anomaly monitoring and emergency protection. The functional boundaries of the two are clear, and their data integration is complementary.

[0035] The segmented calibration model is established as follows: During system initialization, a standard force gauge is used to perform multi-point gradient loading calibration on the thin-film pressure sensor 5. The sensor output voltage is recorded at different temperature points, and a three-dimensional calibration mapping table of "voltage-temperature-force value" is established and stored in the non-volatile memory of the main control unit. In actual operation, the main control unit obtains the corresponding force value based on the current ambient temperature and the voltage value acquired by the AD converter through table lookup or linear interpolation.

[0036] The three-stage compliant force control algorithm divides the clamping force control process into three stages: Soft-touch buffering phase: After contact recognition is completed, the main control unit controls the gripper servo motor 1 to continue closing at an extremely low speed, and the force value rises steadily to 80% of the target force at a set slope. The purpose of this phase is to avoid rigid impact at the moment of contact and protect fragile and flexible workpieces.

[0037] Smooth transition phase: The main control unit gradually increases the closed-loop gain to smoothly approach the target force value, while suppressing overshoot through gradual gain increase. The closed-loop gain in this phase does not exceed 60% of the closed-loop gain in the constant force holding phase.

[0038] Constant force maintenance stage: The main control unit uses an incremental PID algorithm for high-frequency closed-loop adjustment to stabilize the actual force on the workpiece within the allowable error range of the target force. The parameters of the incremental PID algorithm are independently tuned in the three stages, and the stage switching adopts an automatic triggering mechanism based on the force change rate, that is, when the force change rate is lower than the set threshold, it automatically enters the next stage.

[0039] The multi-dimensional state recognition algorithm, based on the real-time force data from the thin-film pressure sensor 5, simultaneously performs the following three types of working condition recognition: Off-center load identification: The main control unit calculates the ratio of the difference between the force values ​​measured by the left clamping finger 2 and the right clamping finger 3 in real time. When the ratio exceeds the set threshold, it is determined to be an off-center clamping load and an alignment correction prompt is output.

[0040] Micro-slip detection: The main control unit uses a sliding window variance algorithm to process the force signal in real time. The window length is set to 50-100 sampling points, and the variance of the force value within the window is calculated. When the variance exceeds the slip threshold and a decreasing trend in the force value is detected simultaneously, it is determined that the workpiece has micro-slipped. Dual-condition logic can effectively reduce the false judgment rate to less than 1%. After determining micro-slipping, the main control unit increases the force in a stepwise manner with a set step distance to restore stable clamping.

[0041] Overload protection: The main control unit sets two levels of overload thresholds. The first level of overload threshold triggers a deceleration warning, and the main control unit reduces the closing speed and outputs a warning signal; the second level of overload threshold triggers an immediate release protection, and the main control unit controls the gripper servo motor 1 to reverse and open, so as to avoid damage to the workpiece or the mechanism.

[0042] In addition, the main control unit performs temperature drift self-calibration, supporting two modes: single calibration upon power-on and periodic dynamic calibration during operation. Single calibration upon power-on is performed when the device starts up to establish an initial zero-point reference. Periodic dynamic calibration during operation performs zero-point correction at set intervals, collecting sensor outputs under the current no-load condition and combining them with the built-in temperature compensation model to correct the zero-point reference, thereby offsetting the cumulative temperature drift during long-term operation.

[0043] After the clamping operation is completed, the main control unit automatically stores the complete data record of this clamping operation, including force curves, temperature data, status events, and servo-assisted data. This data supports subsequent quality traceability and parameter iterative optimization. The main control unit can automatically correct the contact threshold, slip threshold, and target force value based on historical clamping data, enabling the system to adaptively evolve.

[0044] like Figure 3 As shown, the present invention also provides a servo clamping force state detection method with surface thin film force sensing, which includes the following steps: S1. System initialization and temperature drift self-calibration.

[0045] After the device is powered on, the main control unit first completes a hardware self-test to confirm that the thin-film pressure sensor 5, signal conditioning module, and gripper servo motor 1 are all in normal condition. Then, under no-load conditions, it performs temperature drift self-calibration, acquiring the zero-point output of the thin-film sensor at the current ambient temperature and correcting the zero-point reference using the built-in temperature compensation model. A standard force gauge is used for multi-point gradient loading calibration, establishing and storing a three-dimensional calibration mapping table of "voltage-temperature-force value". Simultaneously, the correspondence between servo torque and transmission load is calibrated, and a secondary overload protection threshold is set.

[0046] S2, Servo position coarse control approach and collision warning.

[0047] The main control unit sends position commands to the gripper servo motor 1, and controls the left gripper finger 2 and right gripper finger 3 to open to the preset working angle through angle feedback. The gripper moves closer to the workpiece to be gripped along with the external robotic arm or linear module. At this stage, position open-loop control is the main method, and the thin-film sensor operates at low frequency sampling to monitor for abnormal pressure changes. If an unexpected pressure change is detected, the feed is immediately stopped and a collision warning is triggered to prevent accidental damage.

[0048] S3. Precise contact identification based on direct force signals.

[0049] The gripper servo motor 1 drives the gripper fingers to close at a low speed. When the pressure values ​​detected by the left and right membrane pressure sensors 5 simultaneously exceed the contact threshold, and the pressure values ​​consistently exceed the threshold for N consecutive sampling cycles, it is determined that the gripper fingers have completed effective contact with the workpiece. The contact determination is based entirely on the direct force signal of the membrane. Since the detection path skips the transmission mechanism, the response delay can be controlled within 20ms, which is more than 60% shorter than the traditional torque determination method.

[0050] S4, three-stage compliant constant force closed-loop control.

[0051] After contact recognition is completed, the system enters a three-stage progressive force control strategy: Soft-touch buffering phase: After contact recognition is completed, the main control unit controls the gripper servo motor 1 to continue closing at an extremely low speed, and the force value rises steadily to 80% of the target force at a set slope. The purpose of this phase is to avoid rigid impact at the moment of contact and protect fragile and flexible workpieces.

[0052] Smooth transition phase: The main control unit gradually increases the closed-loop gain to smoothly approach the target force value, while suppressing overshoot through gradual gain increase. The closed-loop gain in this phase does not exceed 60% of the closed-loop gain in the constant force holding phase.

[0053] Constant force maintenance stage: The main control unit uses an incremental PID algorithm for high-frequency closed-loop adjustment to stabilize the actual force on the workpiece within the allowable error range of the target force. The parameters of the incremental PID algorithm are independently tuned in the three stages, and the stage switching adopts an automatic triggering mechanism based on the force change rate, that is, when the force change rate is lower than the set threshold, it automatically enters the next stage.

[0054] S5, intelligent judgment of multi-dimensional clamping status.

[0055] During the constant force maintenance phase, the system continuously runs a multi-dimensional state recognition algorithm, based on the real-time force data from the thin-film pressure sensor 5, to monitor three types of abnormal operating conditions in real time: off-center load, micro-slippage, and overload. Off-center load identification: The main control unit calculates the percentage difference between the force values ​​measured by the left clamp finger 2 and the right clamp finger 3 in real time.

[0056] Micro-slip detection: The main control unit uses a sliding window variance algorithm to process the force signal in real time. The window length is set to 50-100 sampling points, and the variance of the force value within the window is calculated. When the variance exceeds the slip threshold and a decreasing trend in the force value is detected simultaneously, it is determined that the workpiece has micro-slipped. Dual-condition logic can effectively reduce the false positive rate to less than 1%.

[0057] Overload protection: The main control unit is equipped with two levels of overload thresholds. The first level of overload threshold triggers a deceleration warning, and the second level of overload threshold triggers immediate release protection.

[0058] S6. Anomaly response and data traceability storage.

[0059] Upon detecting an abnormal operating condition, the main control unit executes the corresponding control strategy: When the load is off-center, an alignment correction prompt is output for adjustment by the host computer or operator; During micro-slippage, force is increased in a stepped manner with a set step distance; In case of overload, graded protection is implemented. After the clamping is stable, the operation is completed, and the main control unit automatically stores the complete data record of this clamping operation, supporting quality traceability and parameter iteration.

[0060] As a further optimization, a pre-contact monitoring step can be set between steps S2 and S3: when the membrane pressure sensor 5 detects that the pressure value reaches 20% to 50% of the pre-contact threshold, the main control unit reduces the closing speed of the gripper servo motor 1 in advance to prepare for entering the precise contact recognition stage and further reduce contact impact.

[0061] In summary, this invention directly measures force using a thin-film pressure sensor on the working surface of the gripper, combined with a layered dual-feedback control architecture and a three-stage progressive compliant force control method. This fundamentally solves the problems of inaccurate force measurement, lack of state perception, and poor compliance caused by the reliance on indirect torque estimation in existing servo gripping devices. It is suitable for applications with high requirements for gripping force accuracy, such as microelectronic assembly, biological sample handling, and precision component handling.

[0062] The above embodiments and figures are not intended to limit the product form and style of the present invention. Any appropriate changes or modifications made by those skilled in the art should be considered as not departing from the patent scope of the present invention.

Claims

1. A servo clamping device with surface thin-film force sensing, characterized in that, It includes a gripper servo motor, gripper base, left gripper, right gripper, thin-film pressure sensing unit, signal conditioning module with temperature compensation, and main control unit; The gripper servo motor is mounted and fixed in the middle of the gripper base. The output end of the gripper servo motor is connected to the left and right grippers for driving the left and right grippers to complete the gripping action of clamping in opposite directions or opening in opposite directions. The gripper servo motor has angle feedback output capability and torque feedback output capability. The thin-film pressure sensing unit includes thin-film pressure sensors arranged on the left and right gripping fingers respectively, which are used to synchronously acquire the normal contact pressure between the two gripping fingers and the workpiece. Thin-film pressure sensors are respectively attached to the working surfaces of the left and right clamping fingers facing the clamping space. The sensing area of ​​the thin-film pressure sensor completely covers the clamping contact surface. The outer side of the thin-film pressure sensor is covered with a wear-resistant protective layer, which completely covers the sensing area of ​​the thin-film pressure sensor. The thin-film pressure sensor directly transmits force to the clamped workpiece, and its detection path does not pass through the transmission mechanism. The temperature-compensated signal conditioning module is electrically connected to the thin-film pressure sensing unit and has a built-in voltage divider conversion circuit, NTC temperature compensation network, instrumentation amplifier circuit and second-order low-pass filter circuit. The NTC temperature compensation network is matched in series with the thin-film pressure sensor to compensate for the temperature drift error of the piezoresistive sensor. The signal conditioning module is used to convert the weak resistance change signal of the thin-film pressure sensor into a stable standard voltage signal. The main control unit is bidirectionally connected to the signal conditioning module and the gripper servo motor; the main control unit has a built-in AD acquisition unit, a segmented calibration model, a three-stage compliant force control algorithm and a multi-dimensional state recognition algorithm; The main control unit uses the direct force signal from the thin-film pressure sensor as the core feedback to perform closed-loop control of the entire process, including contact recognition, compliant approach, and constant force maintenance. At the same time, the main control unit connects the torque and angle signals from the gripper servo motor as auxiliary feedback for overload monitoring and redundancy protection at the transmission end, thus constructing a hierarchical dual feedback control architecture of "core precision control + auxiliary protection".

2. The servo clamping device with surface thin film force sensing according to claim 1, characterized in that: The gripper servo motor is a miniature servo motor, and the left and right gripper fingers are connected to the output end of the gripper servo motor via a gear set or crank-connecting rod mechanism.

3. The servo clamping device with surface thin film force sensing according to claim 1, characterized in that: The NTC temperature compensation network in the signal conditioning module and the thin-film pressure sensor form a series voltage divider structure. The main control unit corrects the zero-point reference in real time according to the output of the NTC temperature compensation network. The segmented calibration model of the main control unit establishes a three-dimensional calibration mapping table of "voltage-temperature-force value". The main control unit obtains the corresponding force value by looking up the table or interpolation calculation based on the current ambient temperature and standard voltage signal.

4. A method for detecting the force state of a servo clamping device with surface thin-film force sensing as described in any one of claims 1 to 3, characterized in that, Includes the following steps: S1. System Initialization and Temperature Drift Self-Calibration: The device completes hardware self-test upon power-up, performs temperature drift self-calibration under no-load conditions, collects the zero-point output of the thin-film sensor at the current ambient temperature, and corrects the zero-point reference by combining the built-in temperature compensation model; multi-point gradient loading calibration is performed using a standard force gauge to establish a three-dimensional calibration mapping table of "voltage-temperature-force value" and store it in the main control unit; at the same time, the correspondence between servo torque and transmission load is calibrated, and the secondary overload protection threshold is set. S2. Servo position coarse control approach and collision warning: The main control unit sends position commands to the gripper servo motor and controls the left and right gripper fingers to open to the preset working opening through angle feedback. The gripper approaches the workpiece to be gripped with the external mechanism. At this stage, the position open-loop control is the main method and the thin film sensor operates at low frequency sampling. If an abnormal pressure change is detected, the feed will stop immediately and a collision warning will be triggered. S3. Precise contact recognition based on direct force signal: The gripper servo motor drives the gripper fingers to close at a low speed. When the pressure value detected by the thin film pressure sensors on both sides is greater than the contact threshold and is stably exceeded for N consecutive sampling cycles, it is determined that the gripper fingers have completed effective contact with the workpiece. The contact determination is based on the direct force signal of the thin film. S4. Three-stage compliant constant force closed-loop control: After entering the clamping stage, the system executes a three-stage progressive force control strategy: Soft-touch buffering phase: After contact, the closing continues at an extremely low speed, and the force value rises at a set slope to 80% of the target force to avoid rigid impact; Smooth transition phase: Gradually increase the closed-loop gain to smoothly approach the target force value and suppress overshoot; Constant force holding stage: Closed-loop adjustment is performed using an incremental PID algorithm to stabilize the actual force on the workpiece within the allowable error range of the target force; S5. Multi-dimensional Clamping Status Intelligent Judgment: Based on real-time force data from a thin-film pressure sensor, three types of working conditions are identified simultaneously: off-center load identification, micro-slip detection, and overload protection. S6. Abnormal Response and Data Traceability Storage: After identifying abnormal working conditions, the main control unit executes the corresponding control strategy and completes the operation after the clamping is stable. It automatically stores the force curve, temperature data, status events and servo auxiliary data of this clamping operation.

5. The servo clamping force state detection method according to claim 4, characterized in that: In step S1, the temperature drift self-calibration supports two modes: single calibration upon power-on and periodic dynamic calibration during operation; during operation, a zero-point correction is performed once every set interval to offset the temperature drift accumulated during long-term operation; in step S3, the contact threshold is a pressure threshold adaptively set according to the workpiece material and surface characteristics.

6. The servo clamping force state detection method according to claim 4, characterized in that: In step S4, the closing speed of the soft-touch buffering stage is no higher than 30% of the closing speed of the smooth transition stage, and the closed-loop gain of the smooth transition stage is no higher than 60% of the closed-loop gain of the constant force holding stage.

7. The servo clamping force state detection method according to claim 4, characterized in that: In step S5, a multi-dimensional state recognition algorithm is used to identify three working conditions, including: Off-center load identification: Calculate the ratio of the difference between the force values ​​measured by the left and right clamping fingers. If the difference exceeds the set threshold, it is determined to be an off-center clamping load. Micro-slip detection: The sliding window variance algorithm is used to extract the high-frequency fluctuation characteristics of the force signal. When the variance within the window exceeds the slip threshold and the force value shows a downward trend, it is determined to be micro-slip of the workpiece. The overload protection is set with two levels of overload thresholds. The first level threshold triggers a deceleration warning, and the second level threshold triggers an immediate release protection.

8. The servo clamping force state detection method according to claim 7, characterized in that: In step S5, the slip determination of the micro-slip detection combines the dual conditions of "variance mutation + force trend" to control the false judgment rate within 1%, and the window length of the sliding window variance algorithm is set to 50~100 sampling points.

9. The servo clamping force state detection method according to claim 4, characterized in that: In step S6, the corresponding control strategies include: outputting alignment correction prompts when there is an off-center load, increasing force in a stepped manner with a set step distance when there is slight slippage, and performing graded protection when there is an overload.

10. The servo clamping force state detection method according to claim 4, characterized in that: The data stored in step S6 is used for subsequent quality traceability and parameter iteration optimization. The main control unit automatically corrects the contact threshold, the slip threshold and the target force value based on historical clamping data.