Robot arm anti-loosening grabbing control method and system

Abstract

The application relates to the technical field of robot grabbing control, and discloses a robot mechanical arm anti-loosening grabbing control method and system, wherein an independent finger driving mechanism is used to differentially adjust the output forces of the fingers of a gripper. The differential adjustment is one of the core innovations of the application, which breaks through the limitations of traditional fixed clamping force or overall adjustment of clamping force. For example, when it is detected that a certain part of a workpiece has a sliding trend, the system can increase the output force of the corresponding finger of the part, and adjust the output forces of the fingers of other parts correspondingly, so as to suppress the sliding and avoid excessive overall clamping force. The fine force control can effectively suppress the separation trend of the workpiece in linear and rotational directions generated during high-speed movement and posture change, and significantly improves the grabbing stability.

Description

Technical Field

[0001] This invention relates to the field of robot grasping control technology, and in particular to a robot arm anti-loosening grasping control method and system. Background Technology

[0002] In modern automated production, robotic arms are widely used in precision assembly tasks to ensure product quality and production efficiency. For example, six-axis robotic arms are often used to grip polished metal cylindrical workpieces and precisely insert them into pre-drilled holes. To accomplish such tasks, the end effector of the robotic arm is typically equipped with a two-finger gripper, with pressure sensors mounted on its inner side. Ideally, the control system calculates the minimum gripping force based on information such as the workpiece's weight, size, and surface roughness, and applies a preset clamping force slightly greater than this minimum to ensure workpiece stability during movement and assembly. The entire production process is designed to operate at high speed to meet stringent production cycle requirements.

[0003] However, the actual production environment is far more complex than theoretical assumptions. After machining, metal workpieces often have a thin, unevenly distributed layer of cutting fluid or oil film remaining on their surface. This oil film significantly alters the frictional characteristics of the workpiece surface, resulting in actual static friction force being much lower than the theoretically calculated value. Furthermore, due to the uneven distribution of the oil film, the friction force varies in different parts of the workpiece. When a robotic arm uses a preset clamping force calculated based on ideal dry friction conditions to grasp the workpiece, the workpiece may appear stable at the moment of static grasping, but the problem quickly becomes apparent during dynamic operation.

[0004] During high-speed movement, the workpiece is subjected to enormous inertial forces. The combined force of this inertial force and the workpiece's own weight can easily exceed the maximum static friction force provided by the oil film interface, causing the workpiece to slip slightly in the grippers. This slippage may initially be only a linear displacement of a few millimeters, but as the robotic arm's posture changes, such as from horizontal gripping to vertical insertion, the direction of gravity changes, and the slippage problem becomes more severe. Furthermore, due to uneven distribution of the oil film on the workpiece surface, the frictional forces at the contact surfaces between the grippers and the workpiece are unequal. When the robotic arm performs movements with angular acceleration, such as rotation or rapid turning, this unbalanced frictional force will generate additional torque on the workpiece, causing the workpiece to rotate or deflect slightly in the grippers. This posture deviation is crucial for subsequent assembly processes; even a deviation of one degree can lead to assembly failure, equipment collisions, or even production line shutdowns.

[0005] While increasing the clamping force seems like a direct solution to the above problems, for workpieces with mirror-polished surfaces and extremely high requirements for appearance quality, excessive clamping force can leave permanent indentations or scratches on the workpiece surface, rendering it a defective product, resulting in material waste and increased costs. Therefore, the control system must operate within a very narrow force range: the clamping force must be large enough to resist the various linear and rotational tendencies to disengage during movement, while simultaneously being small enough to ensure no damage to the workpiece surface. This safe force window constantly changes due to the randomness of the oil film condition on each workpiece surface, making a fixed, optimized clamping force value impractical. The challenge for the control system is no longer simply "grabbing" the workpiece, but rather how to sense and suppress any minute displacement and orientation deflection of the workpiece in the grippers throughout the entire movement process, while maintaining the clamping force within a safe range that does not damage the workpiece.

[0006] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0007] This invention provides a robotic arm anti-loosening gripping control method and system, aiming to solve the problems of slight slippage, rotation or deflection of workpieces caused by uneven friction and inertial force when robotic arms grasp workpieces with oil film on the surface, as well as the resulting assembly failure and workpiece damage.

[0008] In a first aspect, to solve the above-mentioned technical problems, the present invention provides a method for preventing the robotic arm from detaching from its gripping control, comprising:

[0009] On the contact surface of the gripper fingertip of the robotic arm, a distributed array of miniature deformation sensors is used to monitor the local deformation on the contact surface between the gripper and the workpiece in real time, and obtain the local deformation signal.

[0010] Multi-point deformation signal differential analysis is performed on the local deformation signal to identify the small relative motion trend information of the workpiece;

[0011] Based on the micro relative motion trend information, the output force of each gripper finger of the gripper is adjusted through an independent gripper finger drive mechanism to suppress the micro relative motion trend. The adjustment includes differential adjustment of the output force of each gripper finger.

[0012] The output force of each gripper finger is continuously monitored and compared with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece.

[0013] Preferably, adjusting the output force of each gripper finger of the gripper includes:

[0014] A piezoelectric material layer is integrated on the contact surface of the gripper fingertip, and the piezoelectric material layer is connected to an independent drive circuit;

[0015] Based on the micro relative motion trend information, local transient stick-slip events are identified, and reverse micropulse torques are calculated;

[0016] An electrical pulse command is sent to the piezoelectric material layer corresponding to the local transient stick-slip event to drive the piezoelectric material layer to apply a micro-pulse torque in order to decouple the stickiness effect;

[0017] Local deformation signals beneath the piezoelectric material layer are monitored at a high sampling rate;

[0018] The local deformation signal is compared with the preset workpiece damage threshold to detect local pressure pulses;

[0019] When the local pressure pulse is detected, the local pressure pulse event is recorded, and the output force of other grippers or the gripper posture is adjusted to disperse the pressure.

[0020] Preferably, the continuous monitoring of the output force of each gripping finger and the comparison with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece includes:

[0021] Within each grasping cycle, the local deformation signal is analyzed to extract local stress events that do not exceed the instantaneous damage threshold but exhibit significant fluctuations.

[0022] For each local stress event, a local stress accumulation factor for the region where the local stress event is located is calculated based on the amplitude, duration, and frequency of occurrence of the local stress event.

[0023] Set a local stress accumulation fatigue threshold;

[0024] When the local stress accumulation factor in any region exceeds the local stress accumulation fatigue threshold, an accumulation damage warning is triggered.

[0025] Based on the cumulative damage warning, adjust the gripping force distribution or gripper posture of subsequent gripping operations to avoid stress accumulation.

[0026] Preferably, adjusting the gripping force distribution or gripper posture in subsequent grasping operations based on the cumulative damage warning to avoid stress accumulation includes:

[0027] The environmental parameters within the working area of ​​the robotic arm are sensed in real time to obtain environmental parameter information;

[0028] Based on the environmental parameter information and the initial physical properties of the surface oil film of the workpiece, the physical properties of the oil film in each region of the workpiece surface are evaluated to obtain the physical property information of the oil film.

[0029] Based on the physical properties of the oil film, the current clamping force distribution, and the gripper posture, the stress distribution change trend on the contact surface between the workpiece and the gripper is predicted, and stress distribution change trend information is obtained.

[0030] When the stress distribution trend information indicates a risk of stress concentration or accelerated fatigue damage, the gripping force distribution or gripper posture of subsequent gripping operations is adjusted according to the stress distribution trend information to avoid stress concentration or accelerated fatigue damage.

[0031] Preferably, the step of triggering a cumulative damage warning when the local stress accumulation factor in any region exceeds the local stress accumulation fatigue threshold includes:

[0032] Continuously monitor the changes in the microscopic physical properties of the workpiece material to obtain information on changes in material properties;

[0033] The damage tolerance of different areas of the workpiece is dynamically evaluated to obtain damage tolerance information;

[0034] The local stress accumulation fatigue threshold is adaptively adjusted based on the material property change information and the damage tolerance information.

[0035] When the local stress accumulation factor in any region exceeds the adaptively adjusted local stress accumulation fatigue threshold, the cumulative damage warning is triggered.

[0036] Preferably, the continuous monitoring of changes in the microscopic physical properties of the workpiece material includes:

[0037] A miniature laser Doppler vibration measurement system is integrated into the gripper fingertip of the robotic arm;

[0038] The micro laser Doppler vibration measurement system emits a laser beam to irradiate the surface of the workpiece and receives the reflected laser beam from the surface of the workpiece.

[0039] By analyzing the Doppler frequency shift of the reflected laser beam, local micro-vibration information of the workpiece surface is obtained;

[0040] Based on the local micro-vibration information, the changes in the microscopic physical properties of the workpiece material are identified.

[0041] Preferably, identifying changes in the microscopic physical properties of the workpiece material based on the local micro-vibration information includes:

[0042] The workpiece surface is scanned non-contactly using the micro laser Doppler vibration measurement system to obtain local morphological information of the workpiece surface;

[0043] Based on the local morphology information, identify the micro-particles or areas with uneven oil film thickness on the surface of the workpiece.

[0044] Based on the microparticles or the uneven oil film thickness, the micro-vibration signals caused by surface foreign matter are filtered out.

[0045] Based on the filtered micro-vibration information, the changes in the microscopic physical properties of the workpiece material are identified.

[0046] Preferably, identifying changes in the microscopic physical properties of the workpiece material based on the filtered micro-vibration information includes:

[0047] When performing non-contact scanning on the surface of the workpiece, an initial scan is performed on multiple representative areas of the workpiece to establish a local micro-vibration response baseline of the workpiece in an undamaged state. The local micro-vibration response baseline includes the inherent vibration characteristics and anisotropic response of different areas.

[0048] In subsequent monitoring, the local micro-vibration information of the workpiece surface will be acquired in real time and compared with the baseline in a regional manner to identify the difference between the real-time local micro-vibration information and the baseline.

[0049] If the difference exceeds a preset difference threshold, it is determined that the microscopic physical properties of the workpiece material have changed.

[0050] Preferably, determining that the microscopic physical properties of the workpiece material have changed when the difference exceeds a preset difference threshold includes:

[0051] When performing non-contact scanning on the surface of the workpiece, the spectral reflectance information of the workpiece surface is obtained through a multispectral imaging system;

[0052] Based on the spectral reflectance information, identify the minute scratches or localized oxide layer areas on the surface of the workpiece.

[0053] When comparing the real-time local micro-vibration information with the baseline, the differences in the micro-scratches or the local oxide layer areas are weighted or excluded.

[0054] When the difference between the weighted or excluded real-time micro-vibration information and the baseline exceeds the preset difference threshold, it is determined that the microscopic physical properties of the workpiece material have changed.

[0055] Secondly, the present invention provides a robotic arm anti-loosening gripping control system, comprising:

[0056] The monitoring terminal is used to monitor the local deformation on the contact surface between the gripper and the workpiece in real time through a distributed array of miniature deformation sensors on the contact surface of the gripper fingertips of the robotic arm, and obtain the local deformation signal.

[0057] The adjustment end is used to perform multi-point deformation signal differential analysis on the local deformation signal to identify the small relative motion trend information of the workpiece; according to the small relative motion trend information, the output force of each gripper finger of the gripper is adjusted through the independent gripper finger drive mechanism to suppress the small relative motion trend, and the adjustment includes differential adjustment of the output force of each gripper finger;

[0058] The output end is used to continuously monitor the output force of each clamping finger and compare it with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece.

[0059] This application discloses a robotic arm anti-loosening gripping control method. By deploying a distributed array of miniature deformation sensors on the contact surface of the gripper fingertips, it is possible to monitor the local deformation on the contact surface between the gripper and the workpiece in real time and with precision, thereby obtaining information on the minute relative motion trends that may occur in the workpiece during gripping. This method further utilizes multi-point deformation signal differential analysis technology, which can accurately identify these minute motion trends. Even when there is an uneven oil film or complex stress on the workpiece surface, it can effectively capture the workpiece's tendency to slide, rotate, or deflect.

[0060] Based on the identified subtle relative motion trends, this application employs an independent gripper finger drive mechanism to differentially adjust the output force of each gripper finger. This differential adjustment is one of the core innovations of this application, overcoming the limitations of traditional fixed gripping forces or overall adjustment of gripping force. For example, when a slippage tendency is detected at a certain part of the workpiece, the system can specifically increase the output force of the corresponding gripper finger at that part, while adjusting the output force of the gripper fingers at other parts accordingly to suppress slippage and avoid excessive overall gripping force. This refined force control effectively suppresses various linear and rotational detachment tendencies of the workpiece during high-speed movement and posture changes, significantly improving gripping stability.

[0061] Furthermore, this application continuously monitors the output force of each gripping finger and compares it with a preset workpiece damage threshold to ensure that the output force is always maintained within a safe range that does not damage the workpiece. This resolves the contradiction between "being able to grip" and "not damaging" in existing technologies. Through real-time feedback and adjustment, the system can dynamically adapt to the randomness of the oil film condition on the workpiece surface, avoiding indentations or scratches on high-requirement workpieces such as mirror-polished surfaces due to excessive clamping force, thereby ensuring product quality and reducing material waste and production costs.

[0062] In summary, the robotic arm anti-loosening gripping control method of this application effectively solves the loosening and damage problems faced by robotic arms when gripping complex workpieces in the prior art through real-time and precise local deformation monitoring, intelligent differentiated force adjustment, and strict damage threshold control. It significantly improves the stability, accuracy, and safety of gripping operations and has significant practical value and technological progress. Attached Figure Description

[0063] Figure 1 This is a schematic diagram of a robotic arm anti-loosening gripping control method provided in an embodiment of the present invention;

[0064] Figure 2 This is a schematic diagram of a process for adjusting the output force of each gripper finger according to an embodiment of the present invention;

[0065] Figure 3 A flowchart of a method for continuously monitoring the output force of each pinching finger is provided in this embodiment of the invention;

[0066] Figure 4 This is a schematic diagram of a robotic arm anti-loosening gripping control system provided in an embodiment of the present invention. Detailed Implementation

[0067] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0068] Reference Figure 1 The present invention provides a flowchart of a robotic arm anti-loosening gripping control method, which includes the following steps:

[0069] S1, On the contact surface of the gripper fingertip of the robotic arm, a distributed array of miniature deformation sensors is used to monitor the local deformation on the contact surface between the gripper and the workpiece in real time, and obtain the local deformation signal.

[0070] S2, perform multi-point deformation signal differential analysis on the local deformation signal to identify the small relative motion trend information of the workpiece;

[0071] S3, based on the micro relative motion trend information, the output force of each gripper finger of the gripper is adjusted through the independent gripper finger drive mechanism to suppress the micro relative motion trend. The adjustment includes differential adjustment of the output force of each gripper finger.

[0072] S4. Continuously monitor the output force of each clamping finger and compare it with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece.

[0073] This application effectively suppresses the minute relative motion of the workpiece during the gripping process by real-time, precise local deformation monitoring and intelligent differential force adjustment, while strictly controlling the clamping force to avoid damage to the workpiece, thereby significantly improving the stability and reliability of precision assembly.

[0074] The "robotic arm" mentioned in this application refers to an automated device capable of mimicking the functions of a human arm to perform tasks such as grasping, moving, and assembling. It typically consists of multiple joints and end effectors (such as grippers). The "contact surface of the gripper fingertips" refers to the area where the gripper directly contacts the workpiece, serving as a crucial interface for force transmission and deformation sensing. The "distributed micro-deformation sensor array" refers to a network of multiple micro-deformation sensors integrated in a specific layout on the gripper contact surface, providing high spatial resolution local deformation data. These sensors can be resistance strain gauges, fiber optic sensors, or capacitive sensors, etc., and their function is to convert the minute deformations of the workpiece in contact with the gripper into measurable electrical signals. The "local deformation signal" is the raw data output by these sensors, reflecting the force and degree of deformation at various points on the contact surface. The "workpiece damage threshold" refers to the maximum stress or deformation limit that the workpiece surface can withstand during the grasping process; exceeding this threshold will result in scratches, indentations, or other forms of damage to the workpiece.

[0075] On the contact surface of the gripper fingertips of a robotic arm, a distributed array of miniature deformation sensors monitors the local deformation on the contact surface between the gripper and the workpiece in real time, obtaining local deformation signals. Specifically, a series of miniature piezoresistive sensors can be uniformly arranged on the contact surface of the gripper fingertips. When the gripper grasps the workpiece, the workpiece comes into contact with the sensor array, causing the sensors to deform and thus causing a change in resistance. These resistance changes are collected in real time and converted into electrical signals, i.e., local deformation signals. As an alternative implementation, a miniature capacitive sensor array can also be used. When the workpiece comes into contact with the sensors, the contact pressure changes the distance between the capacitor plates, thereby changing the capacitance value. These changes in capacitance value are converted into local deformation signals.

[0076] Multi-point deformation signal differential analysis is performed on local deformation signals to identify subtle relative motion trends in the workpiece. For example, local deformation signals collected by two or more adjacent sensors can be compared in real time. When the workpiece experiences slight slippage in the gripper, the sensor signals along the slippage direction exhibit dynamic and directional changes; for instance, the sensor signal at the leading edge of the slippage may increase rapidly, while the signal at the trailing edge may decrease rapidly. By calculating the differences between these signals and their rates of change, the subtle relative motion trend of the workpiece can be inferred, such as whether it is slipping to the left, to the right, or undergoing slight rotation. As another approach, machine learning algorithms can be used for pattern recognition of local deformation signals. By pre-training the model, it can identify features corresponding to different types of subtle relative motion (such as translation, rotation, and stick-slip) from complex deformation signal patterns, thereby outputting subtle relative motion trend information of the workpiece.

[0077] Based on information about minute relative motion trends, an independent gripper finger drive mechanism adjusts the output force of each gripper finger to suppress these trends. This adjustment includes differentiated adjustment of the output force of each gripper finger. Specifically, when a tendency for the workpiece to slide in a certain direction is detected, the output force of the gripper finger in that sliding direction can be increased, while the output force of other gripper fingers can be appropriately reduced or maintained to generate a reverse frictional torque or force, thereby suppressing the sliding. For example, if the workpiece tends to slide to the left, the gripping force of the left gripper finger can be increased to "push back" the workpiece. Alternatively, the independent gripper finger drive mechanism can consist of multiple independent micro servo motors, each controlling the gripping force of one gripper finger. When receiving information about minute relative motion trends, the control system sends commands to the corresponding servo motors according to a preset control strategy, causing them to precisely adjust the output force of the corresponding gripper finger. This adjustment can be continuous or pulsed to achieve rapid response and suppression of minute motion trends.

[0078] The system continuously monitors the output force of each gripper finger and compares it with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece. For example, a force sensor can be integrated into the drive mechanism of each gripper finger to measure the actual output force applied to the workpiece in real time. These force sensors can be piezoelectric sensors or strain gauge sensors. The real-time measured output force is compared with the preset workpiece damage threshold. If the output force of any gripper finger approaches or exceeds the damage threshold, the control system will immediately issue an alarm and automatically adjust the gripping force to avoid damaging the workpiece.

[0079] As another approach, a mechanical model of the workpiece material can be established, combined with finite element analysis, to predict the stress distribution on the workpiece surface under different clamping forces. In actual operation, the clamping force monitored in real time is input into this model to dynamically assess whether the maximum stress on the workpiece surface exceeds the damage threshold. Once a damage risk is predicted, the clamping force is immediately adjusted.

[0080] This application presents a robotic arm anti-loosening gripping control method that utilizes a distributed array of miniature deformation sensors deployed on the contact surface of the gripper's fingertips to achieve real-time, high-precision monitoring of local deformation on the contact surface between the gripper and the workpiece. This refined sensing capability is key to solving the problem of minute workpiece slippage caused by uneven oil film and dynamic inertial forces in traditional gripping methods. By performing multi-point differential analysis on these local deformation signals, the system can identify the minute relative motion trends of the workpiece within the gripper, including translation and rotation. This step quantifies the fuzzy concept of "slippage" into identifiable motion trend information, providing a basis for subsequent precise control.

[0081] Once a slight relative motion trend is identified, this application utilizes an independent gripper finger drive mechanism to differentially adjust the output force of each gripper finger. This means that the gripper no longer simply applies a uniform preset gripping force, but can dynamically and locally adjust the gripping force distribution according to the actual motion trend of the workpiece. For example, when the workpiece tends to slide in a certain direction, the system can increase the gripping force of the gripper finger in that direction, while possibly decreasing or maintaining the gripping force of other gripper fingers, thereby generating a reverse force or torque, effectively suppressing the slight relative motion of the workpiece. This differential adjustment mechanism upgrades gripping control from static, global force control to dynamic, local torque and force control, greatly improving gripping stability.

[0082] Furthermore, this application continuously monitors the output force of each gripping finger and compares it with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece. This step resolves the contradiction between "grabbing firmly" and "not damaging" in traditional methods. Through real-time feedback and comparison, the system can suppress minute movements of the workpiece while maintaining the clamping force within the workpiece's safe tolerance range. This is particularly important for precision workpieces with polished surfaces and extremely high requirements for appearance quality, avoiding indentations or scratches caused by excessive clamping force.

[0083] Compared to existing technologies that rely on fixed preset clamping forces or simply increase the clamping force to address slippage issues, the solution presented in this application has significant advantages. Existing technologies often fail to perceive the minute dynamics of the workpiece within the gripper in real time, nor can they make precise local adjustments to the clamping force, and they struggle to ensure stable gripping while preventing workpiece damage. This application, through the combination of distributed sensing, intelligent analysis, and differentiated force control, forms a closed-loop, adaptive gripping control system. This system effectively addresses complex conditions such as uneven oil film on the workpiece surface and inertial forces from high-speed motion, achieving precise suppression of minute relative movements while ensuring no workpiece damage. This significantly improves the gripping accuracy, stability, and reliability of the robotic arm in precision assembly tasks.

[0084] In some embodiments described above, the output force of each gripper finger is adjusted by an independent gripper finger drive mechanism to suppress minor relative motion tendencies of the workpiece. However, in actual gripping, transient stick-slip events may occur on the contact surface between the workpiece and the gripper. These events can lead to gripping instability or cause localized, momentary damage to the workpiece, and cannot be effectively and quickly addressed by macroscopic force adjustments alone. To address this, this application further proposes the aforementioned method for adjusting the output force of each gripper finger. By introducing a piezoelectric material layer and micropulse torque, it aims to respond more precisely and quickly to and suppress these transient stick-slip phenomena, while simultaneously dispersing localized pressure, thereby improving gripping stability and safety.

[0085] For details, please refer to Figure 2 , Figure 2 This is a flowchart illustrating the process of adjusting the output force of each gripper finger according to an embodiment of the present invention, including the following steps:

[0086] S31, a piezoelectric material layer is integrated on the contact surface of the gripper fingertip, and the piezoelectric material layer is connected to an independent drive circuit;

[0087] S32, Based on the micro relative motion trend information, identify local transient stick-slip events and calculate the reverse micro-pulse torque;

[0088] S33, send an electrical pulse command to the piezoelectric material layer corresponding to the local transient stick-slip event to drive the piezoelectric material layer to apply a micro-pulse torque in order to decouple the stickiness effect;

[0089] S34, monitor the local deformation signal beneath the piezoelectric material layer with a high sampling rate;

[0090] S35, compare the local deformation signal with the preset workpiece damage threshold to detect local pressure pulses;

[0091] S36, when the local pressure pulse is detected, the local pressure pulse event is recorded, and the output force of other gripper fingers or the gripper posture is adjusted to disperse the pressure.

[0092] The piezoelectric material layer is integrated into the contact surface of the gripper fingertips. Its function is to convert electrical energy into mechanical energy, or mechanical energy into electrical energy, thereby achieving a rapid and precise mechanical response to a local area of ​​the contact surface. The piezoelectric material layer is connected to an independent drive circuit, ensuring that each piezoelectric unit can be individually controlled to achieve precise local torque application.

[0093] Furthermore, based on the minute relative motion trend information acquired by the distributed micro-deformation sensor array, the system can identify local transient stick-slip events occurring between the workpiece and the gripper. These stick-slip events typically manifest as sudden sliding and stopping in a localized area of ​​the contact surface, potentially leading to unstable gripping. Once such an event is identified, the system immediately calculates a reverse micro-pulse torque capable of counteracting the stick-slip trend.

[0094] Specifically, to decouple the viscous effect, i.e., to eliminate or reduce transient slippage caused by local viscous forces, the system sends an electrical pulse command to the piezoelectric material layer corresponding to the local area where the stick-slip event occurs. This command drives the piezoelectric material layer to rapidly apply a reverse micropulse torque. This torque has a short duration and moderate intensity, aiming to instantly break the viscous state and restore gripping stability.

[0095] Meanwhile, to ensure operational safety, the system continuously monitors local deformation signals beneath the piezoelectric material layer at a high sampling rate. The purpose of the high sampling rate is to capture any rapidly changing local pressure, thereby promptly detecting local pressure pulses that could potentially damage the workpiece.

[0096] When a localized pressure pulse is detected, the event is recorded. To prevent workpiece damage, the system immediately adjusts the output force of the other gripping fingers or the gripper posture. This adjustment aims to distribute the locally concentrated pressure over a larger contact area or have it borne by other gripping fingers, thereby reducing stress in a single area and protecting the workpiece from damage.

[0097] Through the above technical solutions, this application can significantly improve the stability and safety of robotic arms when grasping complex or slippery workpieces. Specifically, by rapidly identifying and actively suppressing local transient stick-slip events, it effectively avoids accidental loosening or positional displacement of the workpiece during the grasping process. Simultaneously, high-sampling-rate local pressure monitoring and a dynamic pressure dispersion mechanism ensure that even when dealing with transient stick-slip events, the workpiece is not damaged due to excessive local pressure. This achieves precise grasping control while effectively protecting the integrity of the workpiece, extending its service life, and broadening the application range of the robotic arm.

[0098] In some preferred embodiments, suppose a robotic arm needs to grasp a precision optical lens with a smooth surface and a small amount of oil film. In traditional grasping methods, due to the oil film on the lens surface, minute transient stick-slip can easily occur during grasping, causing the lens to shift or even fall. Using the solution of this application, when a distributed array of miniature deformation sensors at the gripper fingertips detects a minute relative movement trend on the contact surface between the lens and the gripper, the system immediately identifies a local transient stick-slip event. For example, in a specific area of ​​a gripper finger, the sensor signal shows a brief slip followed by a "sticking" effect by viscous force.

[0099] At this point, the system quickly calculates the reverse micro-pulse torque and sends an electrical pulse command to the corresponding piezoelectric material layer through an independent drive circuit. Once activated, the piezoelectric material layer instantly applies a micro-pulse torque, precisely targeting the stick-slip region to decouple the viscous effect and restore stable lens gripping in a very short time. Simultaneously, the system monitors local deformation signals beneath the piezoelectric material layer at a high sampling rate. If a local pressure pulse, potentially caused by the micro-pulse torque, is detected—for example, if the pressure at a certain point momentarily exceeds a preset threshold—the system immediately records the event and adjusts the output force of other gripping fingers or the gripper posture. This includes slightly adjusting the gripper's tilt angle or increasing the gripping force of other fingers to disperse the local pressure and ensure the lens surface is not damaged. Through this precise and rapid-response control, even when gripping delicate workpieces prone to stick-slip, high stability and safety can be achieved.

[0100] Traditional robotic arm grasping control methods primarily focus on preventing instantaneous workpiece damage, such as immediate failure caused by excessive grasping force or localized pressure pulses, when continuously monitoring the output force of the gripping fingers and comparing it with a preset workpiece damage threshold. However, for some fragile or frequently grasped workpieces, even if the mechanical parameters of each grasping operation do not exceed the instantaneous damage threshold, long-term, repeated localized stress fluctuations can still lead to material fatigue, resulting in cumulative damage and ultimately affecting the workpiece's performance or integrity. If these issues are not addressed, the workpiece may develop latent damage after multiple grasps, reducing product quality and lifespan, and this damage is difficult to detect and prevent in a single grasping operation. To address this, this application proposes a more comprehensive damage prevention mechanism that uses cumulative assessment of localized stress events to prevent fatigue damage to the workpiece.

[0101] refer to Figure 3 , Figure 3 This is a flowchart of a method for continuously monitoring the output force of each pinching finger, provided by an embodiment of the present invention.

[0102] S41, within each grasping cycle, the local deformation signal is analyzed to extract local stress events that do not exceed the instantaneous damage threshold but exhibit significant fluctuations;

[0103] S42, for each local stress event, calculate the local stress accumulation factor of the region where the local stress event is located based on the amplitude, duration and frequency of occurrence of the local stress event;

[0104] S43 sets the local stress accumulation fatigue threshold;

[0105] S44, when the local stress accumulation factor in any region exceeds the local stress accumulation fatigue threshold, an accumulation damage warning is triggered;

[0106] S45, based on the cumulative damage warning, adjust the gripping force distribution or gripper posture of subsequent gripping operations to avoid stress accumulation.

[0107] Specifically, within each gripping cycle, by conducting in-depth analysis of the local deformation signals acquired by the distributed micro-deformation sensor array, it is possible to identify local stress events that, although not reaching the instantaneous damage threshold, exhibit significant fluctuations in amplitude, duration, or frequency. These events may indicate that the workpiece surface or internal structure is undergoing microscopic stress cycling.

[0108] For each identified local stress event, a local stress accumulation factor for its region can be calculated. This accumulation factor calculation comprehensively considers the magnitude of the stress event, its duration, and its frequency. For example, these parameters can be quantified as their contribution to material fatigue damage through weighted summation or integration. Stress events with larger amplitudes, longer durations, and higher frequencies typically have higher accumulation factor values.

[0109] In practical applications, a local stress accumulation fatigue threshold needs to be preset. This threshold is determined based on the mechanical properties of the workpiece material, its expected service life, and the acceptable degree of damage, representing the critical point at which the workpiece begins to show fatigue damage under accumulated stress.

[0110] When the local stress accumulation factor in any region exceeds the aforementioned local stress accumulation fatigue threshold during continuous monitoring, the system will immediately trigger a cumulative damage warning. This warning mechanism aims to promptly inform the operator or control system that the relevant area of ​​the workpiece is at risk of fatigue damage.

[0111] Therefore, based on the triggered cumulative damage warning, the system will adjust subsequent gripping operations. This adjustment may include changing the distribution of clamping force, such as appropriately reducing the clamping force in stress concentration areas or dispersing the clamping force to other undamaged areas; it may also involve adjusting the posture of the grippers to change the contact point or contact area with the workpiece, thereby redistributing stress. The ultimate goal is to prevent stress from continuously accumulating in specific areas, thereby effectively preventing fatigue damage.

[0112] This application addresses the problem in existing technologies that focus only on instantaneous damage while neglecting long-term fatigue damage by introducing an assessment of local stress accumulation factors and a cumulative damage early warning mechanism. Specifically, by finely analyzing the local deformation signals within each gripping cycle, it is possible to capture local stress events that, while not immediately causing damage, pose a potential fatigue risk. The amplitude, duration, and frequency of these events are quantified as local stress accumulation factors, enabling the system to quantitatively assess the cumulative stress borne by the workpiece during multiple gripping processes. When this accumulation factor reaches a preset fatigue threshold, an early warning is triggered, prompting the system to adjust subsequent gripping strategies, such as optimizing the gripping force distribution or changing the gripper posture. This mechanism allows the robotic arm to shift from passively avoiding instantaneous damage to actively preventing cumulative fatigue damage, thereby significantly extending the workpiece's service life and improving the safety of gripping operations.

[0113] Through the above technical solution, this application can effectively identify and prevent cumulative fatigue damage that may occur in workpieces during long-term, repeated gripping, overcoming the shortcomings of existing technologies that only focus on instantaneous damage. By introducing a local stress accumulation factor and a cumulative damage early warning mechanism, this solution enables the robotic arm's gripping control system to have a long-term assessment capability of the workpiece's health status, thereby dynamically adjusting the gripping strategy based on the workpiece's actual fatigue condition. This not only significantly improves the precision and intelligence of gripping operations, but more importantly, it effectively extends the service life of vulnerable workpieces, reduces product scrap rates due to fatigue damage, and provides a more reliable guarantee for the automated processing of high-value or precision workpieces.

[0114] As a specific implementation method, a concrete example is given below. Suppose a robotic arm needs to grasp a batch of precision optical lenses made of a specific polymer material. These lenses can withstand a certain amount of instantaneous pressure, but are very sensitive to repeated small stress fluctuations, and are prone to fatigue cracks.

[0115] During the gripping process, a distributed array of miniature deformation sensors continuously monitors local deformation signals on the contact surfaces of the gripper fingertips and the lens. The control system analyzes these signals and identifies that during each gripping and releasing process, certain areas on the lens surface, although not reaching the instantaneous damage threshold, periodically experience local stress fluctuations with small amplitudes but long durations.

[0116] For these localized stress events, the system calculates the localized stress accumulation factor for the region based on its amplitude, duration, and frequency of occurrence within the gripping cycle. For example, if the accumulation factor of a region reaches a preset localized stress accumulation fatigue threshold (which is preset based on the fatigue characteristics of the polymer material) after 100 consecutive gripping cycles, the system will immediately trigger a cumulative damage warning.

[0117] Upon receiving an early warning, the control system automatically adjusts its subsequent gripping strategy based on the warning information. For example, the system might instruct the independent gripper drive mechanism to reduce the gripping force in the fatigued area by 10% during subsequent gripping, distributing this force to other undamaged areas of the lens. Alternatively, it might fine-tune the gripping posture of the grippers, causing a slight shift in the contact point between the grippers and the lens, thereby preventing stress from continuing to concentrate and accumulate in the fatigued area. In this way, even during long-term, high-frequency gripping operations, it can effectively prevent latent damage to the lens due to fatigue, ensuring its optical performance and structural integrity.

[0118] In some embodiments described above, this application proposes adjusting the clamping force distribution or gripper posture of subsequent gripping operations based on cumulative damage warnings to avoid stress accumulation. However, in practice, adjustments based solely on cumulative damage warnings may be insufficient to address complex and variable working environments and workpiece surface conditions. For example, changes in ambient temperature and humidity may affect the lubrication performance of the oil film on the workpiece surface, thereby altering the frictional characteristics and stress distribution between the gripper and the workpiece. Without considering these dynamic factors, even with adjustments, there may still be a potential risk of stress concentration or accelerated fatigue damage, leading to poor adjustment results or failure to completely prevent workpiece damage. Therefore, this application further proposes a more refined and forward-looking adjustment strategy. By comprehensively considering environmental parameters and the characteristics of the workpiece surface oil film, it predicts the trend of stress distribution changes, thereby achieving more precise clamping force distribution or gripper posture adjustments to effectively avoid stress concentration or accelerated fatigue damage.

[0119] The above-mentioned adjustment of the gripping force distribution or gripper posture in subsequent grasping operations based on cumulative damage warning to avoid stress accumulation includes:

[0120] Real-time sensing of environmental parameters within the working area of ​​the robotic arm to obtain environmental parameter information;

[0121] Based on environmental parameter information and the initial physical properties of the oil film on the surface of the workpiece, the physical properties of the oil film in each region of the workpiece surface are evaluated to obtain the physical property information of the oil film.

[0122] Based on the physical properties of the oil film, the current clamping force distribution, and the gripper posture, the stress distribution trend on the contact surface between the workpiece and the gripper is predicted, and the stress distribution trend information is obtained.

[0123] When the stress distribution trend information indicates a risk of stress concentration or accelerated fatigue damage, the gripping force distribution or gripper posture of subsequent gripping operations should be adjusted according to the stress distribution trend information to avoid stress concentration or accelerated fatigue damage.

[0124] Specifically, real-time sensing of environmental parameters within the working area of ​​a robotic arm can be understood as continuously acquiring physical quantities in the environment through various sensors integrated into the robotic arm or its working area, such as temperature sensors, humidity sensors, and air pressure sensors, and converting them into processable environmental parameter information. The purpose is to obtain external conditions that affect the grasping process and the surface condition of the workpiece.

[0125] The evaluation of the physical properties of the oil film in different regions of the workpiece surface, based on environmental parameters and the initial physical properties of the oil film, involves using real-time acquired environmental parameters combined with pre-known initial physical properties of the oil film (such as initial viscosity, density, and coefficient of thermal expansion) to dynamically calculate and predict potential changes in the physical properties of the oil film in different regions of the workpiece surface under current environmental conditions. These changes might include viscosity, thickness, and lubrication performance. The aim is to accurately grasp the dynamic changes in the lubrication state of the workpiece surface, as the oil film properties directly affect the friction and stress transmission between the gripper and the workpiece.

[0126] In practical applications, based on the physical properties of the oil film, the current clamping force distribution, and the gripper posture, the stress distribution trend on the contact surface between the workpiece and the gripper is predicted. Specifically, this involves combining the assessed physical properties of the oil film with the current clamping force distribution of the robotic arm (i.e., the magnitude and direction of the force applied by each gripper finger) and the gripper posture (i.e., the spatial position and angle of the gripper relative to the workpiece). Through finite element analysis, contact mechanics models, or data-driven prediction models, the potential stress distribution patterns on the contact surface between the workpiece and the gripper under these conditions and their trends over time are simulated and predicted, thus obtaining information on the stress distribution trend. The purpose is to predict potential stress concentration areas or fatigue damage risks in advance.

[0127] Furthermore, when stress distribution trend information indicates a risk of stress concentration or accelerated fatigue damage, the clamping force distribution or gripper posture of subsequent gripping operations is adjusted based on this trend information to avoid stress concentration or accelerated fatigue damage. This means that once a potential risk is predicted, the system will intelligently optimize subsequent gripping strategies based on the prediction results. For example, stress can be dispersed over a larger area by redistributing the output force of each gripper finger, or the gripping angle and position of the grippers can be fine-tuned to avoid high-risk areas, thereby proactively preventing workpiece damage.

[0128] Through the above technical solution, this application enables more refined and proactive management of workpiece damage risks. Compared to adjustments only made after a cumulative damage warning is triggered, this solution, by sensing environmental parameters in real time and assessing the physical properties of the oil film, can dynamically predict stress distribution trends. This allows for preventative measures to be taken before stress concentration or accelerated fatigue damage actually occurs. This significantly improves the safety and reliability of gripping operations, effectively reduces the probability of workpiece damage due to stress concentration or fatigue accumulation in complex and variable environments, extends the service life of the workpiece, and enhances the intelligence level of the robotic arm.

[0129] In some preferred embodiments, a specific example is given below. Suppose a robotic arm is grasping a precision metal part with a surface coated with lubricant. During the grasping process, a distributed array of micro-deformation sensors continuously monitors local deformation and identifies minute relative motion trends. Based on this information, the system adjusts the gripping finger output force through independent gripping finger drive mechanisms to prevent loosening. Simultaneously, the system continuously monitors the output force and triggers a cumulative damage warning based on the local stress accumulation factor.

[0130] At this point, the solution proposed in this application will play a further role. First, temperature and humidity sensors integrated within the working area of ​​the robotic arm perceive environmental parameters in real time. For example, if an increase in ambient temperature is detected, environmental parameter information is obtained. Next, based on this environmental parameter information and the initial physical properties of the oil film on the surface of the metal part (e.g., it is known that the viscosity of the oil film decreases at high temperatures), the system assesses that under the current high-temperature environment, the viscosity of the oil film in various regions of the metal part surface has significantly decreased, and the lubrication performance has weakened, thus obtaining information on the physical properties of the oil film.

[0131] Subsequently, the system combines this oil film physical properties information with the current clamping force distribution (e.g., a certain clamping finger applies a larger force) and the gripper posture, and calculates using a preset finite element model to predict that under the current conditions, due to insufficient oil film lubrication, the risk of stress concentration in a specific area on the contact surface between the metal part and the gripper will increase significantly and may accelerate fatigue damage, thus obtaining information on the stress distribution change trend.

[0132] Given that this stress distribution trend indicates a risk of stress concentration or accelerated fatigue damage, the system immediately adjusts subsequent gripping operations. Specifically, the system instructs the independent gripper drive mechanism to redistribute the gripping force. For example, it appropriately reduces the output force of the gripper fingers in predicted high-stress areas while increasing the output force of other gripper fingers to disperse the overall gripping force, or it fine-tunes the gripping posture of the jaws to ensure that the contact point avoids high-risk areas. Through this proactive adjustment, stress concentration or accelerated fatigue damage caused by environmental changes is effectively avoided, ensuring the integrity of precision metal components during the gripping process.

[0133] In some embodiments described above, a cumulative damage warning is triggered by setting a local stress accumulation fatigue threshold to avoid workpiece damage. However, in practical applications, the microscopic physical properties of the workpiece material may change over time, due to environmental factors, or stress history, and the damage tolerance of different regions may also vary. If the local stress accumulation fatigue threshold remains fixed, it may not accurately reflect the true fatigue state and damage risk of the workpiece, leading to untimely or false warnings, affecting the reliability and efficiency of the gripping operation. Therefore, this application further proposes a method for adaptively adjusting the local stress accumulation fatigue threshold to more accurately assess the cumulative damage risk of the workpiece.

[0134] When the local stress accumulation factor in any region exceeds the local stress accumulation fatigue threshold, a cumulative damage warning is triggered, including:

[0135] Continuously monitor the changes in the microscopic physical properties of the workpiece material to obtain information on changes in material properties;

[0136] The damage tolerance of different areas of the workpiece is dynamically evaluated to obtain damage tolerance information;

[0137] The local stress accumulation fatigue threshold is adaptively adjusted based on the material property change information and the damage tolerance information.

[0138] When the local stress accumulation factor in any region exceeds the adaptively adjusted local stress accumulation fatigue threshold, the cumulative damage warning is triggered.

[0139] Specifically, continuous monitoring of changes in the microscopic physical properties of workpiece materials refers to the real-time or periodic tracking of changes at the microscopic level, such as the internal structure, surface condition, and mechanical properties of the workpiece material. These changes may include, but are not limited to, the initiation and propagation of microcracks, localized changes in material hardness, alterations in surface roughness, and the formation of oxide or corrosion layers. By acquiring this information on changes in material properties, a more accurate understanding of the current health condition and potential damage trends of the workpiece can be obtained.

[0140] The dynamic assessment of damage tolerance in different regions of the workpiece can be understood as a real-time or near-real-time judgment of the workpiece's ability to withstand stress accumulation without failure, based on its material type, geometry, manufacturing process, historical stress conditions, and currently monitored changes in microscopic physical properties. For example, weak points, stress concentration areas, or areas with existing minor defects may have lower damage tolerance than other areas and require a lower fatigue threshold.

[0141] In practical applications, adaptively adjusting the local stress accumulation fatigue threshold based on the material property change information and the damage tolerance information means using the monitored microscopic changes in the material and the assessed regional damage tolerance as inputs, and dynamically correcting the original local stress accumulation fatigue threshold through a preset algorithm or model (e.g., based on machine learning, fuzzy logic, or expert systems). For example, when a microcrack or a decrease in material hardness is detected in a certain area, the fatigue threshold for that area will be lowered; conversely, if the material properties remain good or there are signs of self-repair, the threshold can be appropriately raised. The purpose is to enable the threshold to more accurately match the actual damage tolerance of the workpiece, thereby improving the accuracy of the early warning.

[0142] Therefore, when the local stress accumulation factor in any region exceeds the adaptively adjusted local stress accumulation fatigue threshold, the cumulative damage warning is triggered, ensuring the sensitivity and accuracy of the warning mechanism and avoiding misjudgments caused by threshold mismatch.

[0143] This application achieves a more accurate assessment of the cumulative damage risk of a workpiece because it incorporates continuous monitoring of changes in the microscopic physical properties of the workpiece material and dynamic evaluation of damage tolerance in different areas. In traditional methods, the local stress accumulation fatigue threshold is typically a preset fixed value, failing to adequately consider material aging, fatigue damage accumulation, or performance degradation caused by environmental factors that may occur during actual gripping. Because information on changes in the microscopic physical properties of the workpiece material is acquired in real time, combined with dynamic evaluation of damage tolerance in different areas of the workpiece, the local stress accumulation fatigue threshold can be adaptively adjusted according to the actual condition of the workpiece. This adjustment mechanism ensures that the threshold always matches the current health status and actual load-bearing capacity of the workpiece, thereby making the triggering of cumulative damage warnings more timely and accurate, effectively avoiding workpiece damage or reduced gripping efficiency caused by inappropriate thresholds.

[0144] Through the above technical solution, the local stress cumulative fatigue threshold is no longer a static fixed value, but can be dynamically adjusted according to the actual microscopic physical properties of the workpiece material and the damage tolerance of different areas. This adaptive threshold management mechanism significantly improves the accuracy and reliability of cumulative damage early warning, effectively avoiding premature or delayed warnings caused by improper threshold settings. Therefore, it not only more effectively protects the workpiece from cumulative fatigue damage and extends its service life, but also optimizes the gripping strategy of the robotic arm, maximizing gripping efficiency and stability while ensuring workpiece safety.

[0145] In some preferred embodiments, a specific example is illustrated below. Assume a robotic arm is grasping a batch of workpieces made of a specific alloy material. During the initial grasping phase, the local stress accumulation fatigue threshold is set to a standard value based on the new material. As the grasping operation progresses, a continuous monitoring system detects minute signs of oxidation on some workpiece surfaces, which is identified as a change in the microscopic physical properties of the workpiece material. Simultaneously, through stress history and material property analysis of these areas, a dynamic evaluation system determines that the damage tolerance of these oxidized areas has decreased. Based on this information on material property changes and damage tolerance, the local stress accumulation fatigue threshold is adaptively lowered to reflect the workpiece's current lower fatigue tolerance. When subsequent grasping operations cause the local stress accumulation factor of an oxidized area to reach this lowered threshold, a cumulative damage warning is immediately triggered. Based on this warning, the robot control system can adjust the gripping force distribution of subsequent grasping operations, for example, distributing the gripping force to unoxidized or more damage-tolerant areas, or adjusting the gripper posture to avoid applying excessive stress to oxidized areas, thereby effectively preventing actual damage to the workpieces due to cumulative fatigue.

[0146] In some embodiments described above, continuous monitoring of changes in the microscopic physical properties of workpiece materials is proposed. However, in its implementation, if the monitoring method is not precise or real-time enough, it may affect the accuracy of the adaptive adjustment of the cumulative fatigue threshold. Traditionally, monitoring the microscopic physical properties of materials may rely on contact measurements or time-consuming offline analysis. This is difficult to achieve in real-time, non-destructive, and precise monitoring in dynamic robotic gripping environments, which may lead to untimely or inaccurate adjustment of the cumulative fatigue threshold, failing to effectively prevent cumulative damage to the workpiece. Therefore, this application further proposes a specific method for continuously monitoring the microscopic physical properties of workpiece materials, aiming to provide a non-contact, high-precision, and real-time monitoring method to ensure more accurate and reliable adaptive adjustment of the cumulative fatigue threshold.

[0147] Specifically, the aforementioned continuous monitoring of changes in the microscopic physical properties of the workpiece material includes:

[0148] A miniature laser Doppler vibration measurement system is integrated into the gripper fingertip of the robotic arm;

[0149] The micro laser Doppler vibration measurement system emits a laser beam to irradiate the surface of the workpiece and receives the reflected laser beam from the surface of the workpiece.

[0150] By analyzing the Doppler frequency shift of the reflected laser beam, local micro-vibration information of the workpiece surface is obtained;

[0151] Based on the local micro-vibration information, the changes in the microscopic physical properties of the workpiece material are identified.

[0152] Specifically, a miniature laser Doppler vibration measurement system can be integrated into the fingertips of a robotic arm's gripper. This system is a non-contact measurement device that utilizes the Doppler effect of laser light to measure minute vibrations on an object's surface. The system typically includes a laser emitter, an optical receiver, and a signal processor. The laser emitter generates a laser beam and guides it to the workpiece surface. The optical receiver receives the laser beam reflected from the workpiece surface.

[0153] The process of illuminating a workpiece surface with a laser beam using a miniature laser Doppler vibration measurement system and receiving the reflected laser beam from the workpiece surface involves a slight frequency shift caused by minute vibrations on the workpiece surface when the laser beam strikes it—a Doppler frequency shift. The reflected laser beam carries this frequency shift information and is captured by an optical receiver.

[0154] In practical applications, analyzing the Doppler frequency shift of the reflected laser beam to obtain local micro-vibration information on the workpiece surface involves a signal processor demodulating and analyzing the received reflected laser beam to extract the Doppler frequency shift. This frequency shift is proportional to the vibration velocity of the workpiece surface. Through integration and other processing, local micro-vibration information, such as vibration frequency and amplitude, can be obtained. This micro-vibration information can reflect the internal structure of the workpiece material, its stress state, and the presence of microscopic damage.

[0155] Furthermore, identifying changes in the microscopic physical properties of the workpiece material based on local micro-vibration information refers to determining whether changes have occurred in the workpiece material's microscopic physical properties, such as elastic modulus, damping characteristics, and internal defects, by performing pattern recognition, spectral analysis, or comparison with a preset baseline on the acquired local micro-vibration information. For example, material fatigue, the generation of microcracks, or changes in the internal structure can all lead to specific changes in its micro-vibration response.

[0156] Through the above technical solution, this application enables non-contact, high-precision, real-time monitoring of the microscopic physical properties of workpiece materials. Compared to traditional methods relying on macroscopic mechanical responses or contact sensors, this solution can capture subtle changes occurring within the workpiece material earlier and more sensitively, such as the onset of fatigue damage or the degradation of material properties. This provides more accurate and timely input data for the adaptive adjustment of the local stress accumulation fatigue threshold, significantly improving the accuracy and reliability of cumulative damage early warning. This precise monitoring capability helps to take preventative measures before damage accumulates to an irreversible stage, thereby effectively extending the service life of the workpiece and avoiding production accidents caused by accidental loosening or damage.

[0157] In some existing implementations, when identifying changes in the microscopic physical properties of a workpiece material based on local micro-vibration information, interference from foreign objects or surface inhomogeneities on the workpiece surface may occur. For example, tiny particles or uneven oil film thickness on the workpiece surface may distort the local micro-vibration signal, thus affecting the accurate judgment of the true changes in the material's microscopic physical properties. If these problems are not addressed, it may lead to misjudgments of the workpiece's damage level, thereby affecting the effectiveness of the gripping control strategy and potentially accelerating workpiece damage. To address this, this application further proposes that when identifying changes in the microscopic physical properties of a workpiece material, non-contact scanning of the workpiece surface is used to obtain local morphology information, and this information is used to filter out micro-vibration signals caused by surface foreign objects, thereby improving the accuracy of identification.

[0158] In this regard, this application further proposes the above-mentioned method of identifying changes in the microscopic physical properties of workpiece materials based on local micro-vibration information, including:

[0159] The local morphology information of the workpiece surface is obtained by non-contact scanning of the workpiece surface using a miniature laser Doppler vibration measurement system.

[0160] Based on local morphology information, identify micro-particles or areas with uneven oil film thickness on the workpiece surface;

[0161] Based on areas with minute particles or uneven oil film thickness, filter out micro-vibration signals caused by surface foreign matter;

[0162] Based on the filtered micro-vibration information, the changes in the microscopic physical properties of the workpiece material are identified.

[0163] Specifically, by illuminating the workpiece surface with a laser beam and receiving the reflected laser beam through a miniature laser Doppler vibration measurement system, not only can the Doppler frequency shift of the reflected laser beam be analyzed to obtain local micro-vibration information, but the system can also be used to perform high-resolution non-contact scanning of the workpiece surface. This scanning process aims to acquire local topographic information of the workpiece surface, such as surface roughness, micro-texture, local depressions or protrusions, etc. This topographic information can be used to construct a three-dimensional model or a two-dimensional elevation map of the workpiece surface.

[0164] Based on the acquired local topographic information, it is possible to identify whether there are microparticles or areas with uneven oil film thickness on the workpiece surface. For example, by using image processing algorithms or deep learning models to analyze anomalies or regions in the topographic data, it can be determined whether there are micro-foreign objects attached to the workpiece surface or differences in oil film thickness in different areas. These microparticles or uneven oil film layers may generate additional micro-vibration signals that are not caused by the material itself during the gripper's grasping process.

[0165] Furthermore, once microparticles or areas of uneven oil film thickness are identified, the original local micro-vibration signals can be filtered out based on the characteristics of these areas. For example, an interference model can be established to predict the characteristics of micro-vibration signals that may be generated by these surface foreign objects or uneven oil films, and then these interfering components can be subtracted from or suppressed from the total micro-vibration signal. The purpose is to eliminate noise caused by surface foreign objects or uneven oil films, ensuring that the micro-vibration information analyzed subsequently more accurately reflects the microscopic physical properties of the workpiece material itself. Thus, after filtering out the micro-vibration signals caused by surface foreign objects, the resulting filtered micro-vibration information will be purer and more accurate. Based on this pure micro-vibration information, changes in the microscopic physical properties of the workpiece material, such as material fatigue, microcrack initiation, and hardness changes, can be identified more reliably.

[0166] This application's solution effectively solves the problem of interference from surface foreign matter or uneven oil film on the accuracy of micro-vibration signal identification in traditional methods by introducing the acquisition and analysis of local morphology information of the workpiece surface. Specifically, the miniature laser Doppler vibration measurement system can perform non-contact scanning of the workpiece surface while acquiring micro-vibration information, thereby obtaining fine local morphology information. It is precisely because this morphology information can reveal the presence of microparticles or areas of uneven oil film thickness on the workpiece surface that the system can specifically identify and quantify the potential impact of these surface foreign matter on the micro-vibration signal. By precisely filtering out these micro-vibration signals caused by surface foreign matter, it ensures that the micro-vibration information ultimately used to identify changes in the microscopic physical properties of the workpiece material purely reflects the internal state of the material, rather than being interfered with by external surface conditions. This mechanism ensures a more accurate and reliable judgment of microscopic changes such as material fatigue and damage.

[0167] Through the above technical solution, this application can significantly improve the accuracy and robustness of identifying changes in the microscopic physical properties of workpiece materials. Compared with directly using raw local micro-vibration information for identification, this application avoids misjudgments caused by surface foreign objects by pre-identifying and filtering interference signals caused by tiny particles or uneven oil film thickness on the workpiece surface. This allows the system to more accurately assess the true damage state and fatigue level of the workpiece, thus providing more reliable data support for subsequent gripping force adjustment, posture optimization, and cumulative damage early warning. Therefore, it can not only effectively prevent workpiece damage caused by misjudgment during gripping, but also extend the service life of the workpiece and improve the intelligence level of the robotic arm's gripping control.

[0168] In some preferred embodiments, it is assumed that a robotic arm is grasping a precision ceramic workpiece. During the grasping process, a miniature laser Doppler vibration measurement system continuously monitors the micro-vibrations on the workpiece surface. If the workpiece surface is inadvertently contaminated with tiny dust particles, or if a locally uneven oil film is left during the manufacturing process, these surface foreign objects may generate additional, non-material-inherent, micro-vibration signals when the grippers contact it. According to the present application, the miniature laser Doppler vibration measurement system first performs a non-contact scan of the grasping contact area of ​​the ceramic workpiece to obtain its local morphology information. By analyzing this morphology data, the system identifies areas of dust particles or uneven oil film thickness on the surface.

[0169] Subsequently, the system intelligently filters out the real-time monitored local micro-vibration signals based on the identified foreign object characteristics. For example, if dust particles resonate within a specific frequency range, the system will specifically suppress signals within that frequency range. After filtering, the obtained micro-vibration information will more accurately reflect the microstructural changes of the ceramic material itself, such as the presence of microcrack initiation or material fatigue. Based on this pure micro-vibration information, the system can more reliably determine the damage state of the ceramic workpiece and adjust the clamping force accordingly, avoiding misjudgments and over-clamping caused by surface foreign object interference, thereby effectively protecting the precision workpiece.

[0170] In some embodiments described above, the microscopic physical properties of the workpiece material are identified by analyzing local micro-vibration information on the workpiece surface and filtering out micro-vibration signals caused by surface foreign matter. However, in practical applications, even after preliminary signal processing, changes in the microscopic physical properties of the workpiece material may manifest as subtle deviations rather than significant absolute changes. Without a reliable reference benchmark, it is difficult to accurately determine whether these subtle changes truly indicate material performance degradation based solely on real-time monitoring data, potentially leading to misjudgments or delayed warnings. Therefore, this application further proposes a more accurate and reliable method for identifying changes in the microscopic physical properties of workpiece materials.

[0171] Specifically, the above-mentioned identification of changes in the microscopic physical properties of workpiece materials based on local micro-vibration information includes:

[0172] When performing non-contact scanning on the surface of the workpiece, an initial scan is performed on multiple representative areas of the workpiece to establish a local micro-vibration response baseline of the workpiece in an undamaged state. The local micro-vibration response baseline includes the inherent vibration characteristics and anisotropic response of different areas.

[0173] In subsequent monitoring, the local micro-vibration information of the workpiece surface will be acquired in real time and compared with the baseline in a regional manner to identify the difference between the real-time local micro-vibration information and the baseline.

[0174] If the difference exceeds a preset difference threshold, it is determined that the microscopic physical properties of the workpiece material have changed.

[0175] The initial scan refers to a comprehensive or sampled measurement of the vibration characteristics of the workpiece surface using a miniature laser Doppler vibration measurement system before the workpiece is put into use or when it is known to be in a healthy, undamaged state. The multiple representative regions can be selected based on the workpiece's structural characteristics, stress distribution, or potential damage risk areas; for example, they can be stress concentration areas, easily worn areas, or critical connection points. The local micro-vibration response baseline is the vibration "fingerprint" of these regions in a healthy state. It records not only the inherent vibration frequency and damping characteristics of the material itself, but also the anisotropic response caused by factors such as the material's crystal structure, internal defect distribution, or surface treatment. The purpose of establishing this baseline is to provide a precise and personalized reference standard for subsequent damage detection.

[0176] Furthermore, during subsequent monitoring, the miniature laser Doppler vibration measurement system integrated into the gripper fingertips of the robotic arm continuously acquires real-time local micro-vibration information of the workpiece surface. This real-time information is then compared regionally with a pre-established baseline of local micro-vibration response. Regional comparison means comparing the real-time data with the data of the corresponding region in the baseline to identify the differences between them. These differences can be drifts in vibration frequency, changes in amplitude, alterations in damping characteristics, or deviations in anisotropic response, etc.

[0177] When the difference between the detected real-time local micro-vibration information and the baseline exceeds a preset difference threshold, the system determines that the microscopic physical properties of the workpiece material have changed. This difference threshold is set based on experimental data, material properties, or experience, and serves as a boundary to distinguish between normal fluctuations and potential damage. For example, this threshold can be a percentage deviation or an absolute value. Once the difference exceeds this threshold, it indicates that the workpiece material may have undergone changes in its microscopic physical properties, such as fatigue, microcrack initiation, hardness changes, or internal structural rearrangement.

[0178] Through the above technical solution, this application provides a more accurate and reliable method for identifying changes in the microscopic physical properties of workpiece materials. Compared to relying solely on micro-vibration information after filtering out foreign matter from the surface, introducing a local micro-vibration response baseline and performing real-time comparison significantly improves the sensitivity and accuracy of detecting early material damage and fatigue accumulation. This method can effectively distinguish between normal material fluctuations and actual microstructural changes, reducing false alarm rates and ensuring timely warnings of potential damage. Therefore, preventative measures can be taken earlier, such as adjusting the gripping strategy or performing maintenance, thereby effectively extending the service life of the workpiece and avoiding gripping failure or workpiece damage due to material performance degradation, further enhancing the reliability and safety of the robotic arm's gripping control.

[0179] In some preferred embodiments, a specific example is illustrated below. Suppose a robotic arm needs to grasp a batch of precision ceramic workpieces. Before the initial grasp, an initial scan is performed on the gripping area of ​​one undamaged ceramic workpiece. Using a miniature laser Doppler vibration measurement system, vibration measurements are taken at multiple key points on the workpiece surface (e.g., edges, center stress points, etc.), recording its natural vibration frequency, amplitude decay curve, and vibration response in different directions under undamaged conditions, thereby establishing a baseline for the local micro-vibration response of the ceramic workpiece. This baseline is stored in the control system as a health reference for this batch of workpieces.

[0180] In subsequent batch gripping operations, when the robotic arm grasps a new ceramic workpiece, the miniature laser Doppler vibration measurement system at the gripper fingertip scans the same key areas of the workpiece in real time and acquires real-time local micro-vibration information. For example, during the gripping process, the system detects that the real-time vibration frequency of a certain key area has drifted by 0.5% relative to the baseline, and the amplitude decay rate has increased by 10%. If the preset difference threshold specifies that a frequency drift exceeding 0.2% or an amplitude decay rate change exceeding 5% is considered abnormal, then the system will determine that the microscopic physical properties of the ceramic workpiece material have changed. This change may indicate the presence of microcracks or material fatigue inside the workpiece, even if it is not visible to the naked eye. Once this change is identified, the system can immediately trigger a cumulative damage warning and adjust the gripping force distribution or gripper posture of subsequent gripping operations based on the warning information. For example, it can reduce the gripping force in that area or change the gripping point to avoid further damage to the workpiece, thereby ensuring the stability of the gripping process and the integrity of the workpiece.

[0181] In some embodiments described above, the local micro-vibration information of the workpiece surface acquired in real time is compared regionally with a baseline, and the difference between the real-time local micro-vibration information and the baseline is identified. If the difference exceeds a preset difference threshold, it is determined that the microscopic physical properties of the workpiece material have changed. However, in practical applications, surface defects such as minor scratches and local oxide layers on the workpiece surface may themselves cause differences in local micro-vibration information. This means that even if the microscopic physical properties of the workpiece material itself have not changed substantially when compared with the baseline, these surface factors may lead to an incorrect judgment that the material properties have changed, thereby triggering unnecessary cumulative damage warnings or adjustments to the gripping operation, affecting the accuracy and efficiency of the system. Therefore, this application further proposes an optimized method for judging changes in the microscopic physical properties of workpiece materials. By introducing multispectral imaging technology, differences caused by surface defects are identified and processed to improve the accuracy of the judgment.

[0182] In response, this application further proposes a step for determining whether the microscopic physical properties of the workpiece material have changed when the aforementioned difference exceeds a preset difference threshold, including:

[0183] When performing non-contact scanning on the surface of the workpiece, the spectral reflectance information of the workpiece surface is obtained through a multispectral imaging system;

[0184] Based on the spectral reflectance information, identify the minute scratches or localized oxide layer areas on the surface of the workpiece.

[0185] When comparing the real-time local micro-vibration information with the baseline, the differences in the micro-scratches or the local oxide layer areas are weighted or excluded.

[0186] When the difference between the weighted or excluded real-time micro-vibration information and the baseline exceeds the preset difference threshold, it is determined that the microscopic physical properties of the workpiece material have changed.

[0187] Specifically, the multispectral imaging system is an imaging device capable of simultaneously capturing images within multiple narrow wavelength ranges. By analyzing the intensity of reflected light at different wavelengths, it can reveal information such as the chemical composition, surface structure, and defects of materials. The spectral reflectance information of the workpiece surface acquired by this system refers to the reflectance characteristics of the workpiece surface under illumination at different wavelengths. This data reflects the microstructure and chemical composition of the workpiece surface material. For example, micro-scratches or localized oxide layer areas, due to their different surface morphology and chemical composition compared to the surrounding normal areas, will exhibit unique reflectance characteristics in specific spectral bands. Therefore, these surface defects can be identified by analyzing the spectral reflectance information.

[0188] The identification of minute scratches or localized oxide layer regions on the workpiece surface can be understood as using the characteristic differences in spectral reflectance information, through image processing and pattern recognition algorithms, to distinguish these surface defects from normal surfaces. For example, scratches may lead to enhanced light scattering, while oxide layers may exhibit absorption peaks or decreased reflectance at specific wavelengths.

[0189] In practical applications, when comparing the real-time local micro-vibration information with the baseline regionally, weighting or excluding differences in the micro-scratches or local oxide layer areas refers to correcting the micro-vibration difference data of these areas based on the identified surface defect type and degree. For example, micro-vibration differences caused by micro-scratches can be assigned a lower weight to reduce their impact on the final judgment; or, if the difference is determined to be entirely caused by surface defects and unrelated to the material's intrinsic properties, it can be directly excluded from the comparison range. This processing method aims to eliminate or reduce the interference of surface defects on the judgment of the material's micro-physical properties. Therefore, only when the difference between the weighted or excluded real-time micro-vibration information and the baseline exceeds the preset difference threshold is it finally determined that the micro-physical properties of the workpiece material have changed. This means that only when significant micro-vibration differences still exist after excluding the influence of surface defects is it considered that the material's intrinsic micro-physical properties have changed.

[0190] This application's solution, by introducing a multispectral imaging system, can acquire more comprehensive surface information, particularly spectral reflectance information, during non-contact scanning of the workpiece surface. It is precisely because different types of surface defects (such as micro-scratches and localized oxide layers) possess unique fingerprint characteristics in spectral reflectance that the system can accurately identify these defect areas. By weighting or excluding the micro-vibration differences in these defect areas, this application effectively avoids the interference of surface defects on the judgment of the material's microscopic physical properties. This mechanism ensures that only when the internal structure or chemical composition of the workpiece material undergoes actual changes will it be accurately identified, thereby improving the accuracy and reliability of the judgment.

[0191] Through the above technical solution, this application can significantly improve the accuracy of judging changes in the microscopic physical properties of workpiece materials. Compared with methods that rely solely on micro-vibration information for comparison, this application can effectively distinguish between true changes in microscopic physical properties caused by material fatigue, damage, etc., and false differences caused by surface defects such as surface scratches and oxidation. Therefore, misjudgments caused by surface defects can be avoided, unnecessary cumulative damage warnings and gripping operation adjustments can be reduced, enabling the robotic arm's gripping control system to respond more accurately to the actual state of the workpiece, extend the workpiece's service life, and improve the overall reliability and efficiency of the gripping operation.

[0192] refer to Figure 4 , Figure 4 This is a schematic diagram of a robotic arm anti-loosening gripping control system disclosed in this application. The system includes:

[0193] The monitoring terminal is used to monitor the local deformation on the contact surface between the gripper and the workpiece in real time through a distributed array of miniature deformation sensors on the contact surface of the gripper fingertips of the robotic arm, and obtain the local deformation signal.

[0194] The adjustment end is used to perform multi-point deformation signal differential analysis on the local deformation signal to identify the small relative motion trend information of the workpiece; according to the small relative motion trend information, the output force of each gripper finger of the gripper is adjusted through the independent gripper finger drive mechanism to suppress the small relative motion trend, and the adjustment includes differential adjustment of the output force of each gripper finger;

[0195] The output end is used to continuously monitor the output force of each clamping finger and compare it with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece.

[0196] The robotic arm anti-loosening gripping control system proposed in this application aims to solve the problems of workpieces easily slipping, rotating, or deflecting at high speeds in traditional gripping methods, as well as workpiece surface damage caused by excessive clamping force. This system achieves both stability and safety assurance during the workpiece gripping process through refined sensing, intelligent analysis, and differentiated force control.

[0197] Specifically, the system in this application includes a monitoring terminal, an adjustment terminal, and an output terminal.

[0198] The monitoring terminal is configured to monitor the local deformation of the contact surface between the gripper and the workpiece in real time via a distributed array of miniature deformation sensors on the contact surface of the robotic arm's gripper fingertips, thereby obtaining local deformation signals. The specific implementation and working principle of the distributed miniature deformation sensor array have already been described in the above embodiments and will not be repeated here. It is important to emphasize that the monitoring terminal, as the sensing core of the system, functions to continuously and accurately acquire dynamic deformation information of the gripper-workpiece contact interface. This monitoring terminal may include one or more microcontrollers or dedicated signal processing units to receive analog or digital signals from the sensor array and convert them into local deformation signals suitable for subsequent processing. For example, the monitoring terminal may integrate an analog-to-digital converter (ADC) and a data buffer to ensure the real-time performance and integrity of the deformation signals.

[0199] The adjustment end is configured to perform multi-point deformation signal differential analysis on the local deformation signal to identify the workpiece's minute relative motion trend information; and based on the minute relative motion trend information, adjust the output force of each gripper finger of the gripper through an independent gripper finger drive mechanism to suppress the minute relative motion trend. This adjustment includes differentiated adjustment of the output force of each gripper finger. The specific implementation methods and working principles of multi-point deformation signal differential analysis, identification of minute relative motion trend information, and differentiated adjustment of gripper finger output force have been described in the above embodiments and will not be repeated here. It is important to emphasize that the adjustment end, as the decision-making and execution core of the system, functions to intelligently analyze the raw data acquired by the monitoring end and drive the gripper to perform a precise mechanical response based on the analysis results. This adjustment end may include a high-performance central processing unit (CPU) or digital signal processor (DSP) for executing complex differential analysis algorithms and control strategies. The independent gripper finger drive mechanism may consist of multiple independent servo motors, stepper motors, or hydraulic / pneumatic actuators, with each driver precisely controlling the output force of one gripper finger. The adjustment end ensures that the output force of each gripper finger can dynamically and differentiatedly respond to the minute movement trends of the workpiece through real-time calculation and command transmission.

[0200] The output terminal is configured to continuously monitor the output force of each clamping finger and compare it with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece. The specific implementation and working principle of continuously monitoring the clamping finger output force and comparing it with the workpiece damage threshold have been described in the above embodiments and will not be repeated here. It is important to emphasize that the output terminal, as the core of the system's safety assurance, functions to monitor and assess the clamping force in real time. This output terminal can integrate a high-precision force sensor (e.g., a piezoelectric sensor or strain gauge sensor) into each clamping finger or its drive mechanism to measure the actual applied force in real time. The output terminal also includes a comparison unit for comparing the real-time force value with a preset workpiece damage threshold. Once the force value is detected to be close to or exceed the threshold, the output terminal can send a feedback signal to the adjustment terminal, triggering automatic adjustment of the clamping force, thereby effectively avoiding damage to the workpiece.

[0201] The robotic arm anti-loosening gripping control system of this application organically combines the monitoring end, adjustment end, and output end to form a closed-loop, adaptive gripping control architecture. Compared with existing technologies that rely on fixed preset gripping forces or simply increase the gripping force to deal with slippage problems, the system of this application has significant advantages. Existing technologies often cannot sense the minute dynamics of the workpiece in the gripper in real time, nor can they make fine local adjustments to the gripping force, and it is even more difficult to ensure stable gripping while avoiding workpiece damage. The system of this application, through the combination of distributed sensing hardware, intelligent analysis software, and differentiated force control actuators, can effectively cope with complex working conditions such as uneven oil film on the workpiece surface and high-speed motion inertial forces, and achieves precise suppression of minute relative movements while ensuring that the workpiece is not damaged, thereby significantly improving the gripping accuracy, stability, and reliability of the robotic arm in precision assembly tasks.

[0202] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A method for preventing the robotic arm from detaching during gripping, characterized in that, The method includes: On the contact surface of the gripper fingertip of the robotic arm, a distributed array of miniature deformation sensors is used to monitor the local deformation on the contact surface between the gripper and the workpiece in real time, and obtain the local deformation signal. Multi-point deformation signal differential analysis is performed on the local deformation signal to identify the small relative motion trend information of the workpiece; Based on the micro relative motion trend information, the output force of each gripper finger of the gripper is adjusted through an independent gripper finger drive mechanism to suppress the micro relative motion trend. The adjustment includes differential adjustment of the output force of each gripper finger. The output force of each gripper finger is continuously monitored and compared with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece. The adjustment of the output force of each gripper finger of the gripper includes: A piezoelectric material layer is integrated on the contact surface of the gripper fingertip, and the piezoelectric material layer is connected to an independent drive circuit; Based on the micro relative motion trend information, local transient stick-slip events are identified, and reverse micropulse torques are calculated; An electrical pulse command is sent to the piezoelectric material layer corresponding to the local transient stick-slip event to drive the piezoelectric material layer to apply a micro-pulse torque in order to decouple the stickiness effect; Local deformation signals beneath the piezoelectric material layer are monitored at a high sampling rate; The local deformation signal is compared with the preset workpiece damage threshold to detect local pressure pulses; When the local pressure pulse is detected, the local pressure pulse event is recorded, and the output force of other grippers or the gripper posture is adjusted to disperse the pressure.

2. The method for preventing detachment during gripping of a robotic arm according to claim 1, characterized in that, The continuous monitoring of the output force of each gripper finger and comparison with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece includes: Within each grasping cycle, the local deformation signal is analyzed to extract local stress events that do not exceed the instantaneous damage threshold but exhibit significant fluctuations. For each local stress event, a local stress accumulation factor for the region where the local stress event is located is calculated based on the amplitude, duration, and frequency of occurrence of the local stress event. Set a local stress accumulation fatigue threshold; When the local stress accumulation factor in any region exceeds the local stress accumulation fatigue threshold, an accumulation damage warning is triggered. Based on the cumulative damage warning, adjust the gripping force distribution or gripper posture of subsequent gripping operations to avoid stress accumulation.

3. The method for preventing detachment of a robotic arm's gripping control according to claim 2, characterized in that, The step of adjusting the gripping force distribution or gripper posture in subsequent grasping operations based on the cumulative damage warning to avoid stress accumulation includes: The environmental parameters within the working area of ​​the robotic arm are sensed in real time to obtain environmental parameter information; Based on the environmental parameter information and the initial physical properties of the surface oil film of the workpiece, the physical properties of the oil film in each region of the workpiece surface are evaluated to obtain the physical property information of the oil film. Based on the physical properties of the oil film, the current clamping force distribution, and the gripper posture, the stress distribution change trend on the contact surface between the workpiece and the gripper is predicted, and stress distribution change trend information is obtained. When the stress distribution trend information indicates a risk of stress concentration or accelerated fatigue damage, the gripping force distribution or gripper posture of subsequent gripping operations is adjusted according to the stress distribution trend information to avoid stress concentration or accelerated fatigue damage.

4. The method for preventing detachment of a robotic arm's gripping control according to claim 2, characterized in that, When the local stress accumulation factor in any region exceeds the local stress accumulation fatigue threshold, a cumulative damage warning is triggered, including: Continuously monitor the changes in the microscopic physical properties of the workpiece material to obtain information on changes in material properties; The damage tolerance of different areas of the workpiece is dynamically evaluated to obtain damage tolerance information; The local stress accumulation fatigue threshold is adaptively adjusted based on the material property change information and the damage tolerance information. When the local stress accumulation factor in any region exceeds the adaptively adjusted local stress accumulation fatigue threshold, the cumulative damage warning is triggered.

5. The robotic arm anti-loosening gripping control method according to claim 4, characterized in that, The continuous monitoring of changes in the microscopic physical properties of the workpiece material includes: A miniature laser Doppler vibration measurement system is integrated into the gripper fingertip of the robotic arm; The micro laser Doppler vibration measurement system emits a laser beam to irradiate the surface of the workpiece and receives the reflected laser beam from the surface of the workpiece. By analyzing the Doppler frequency shift of the reflected laser beam, local micro-vibration information of the workpiece surface is obtained; Based on the local micro-vibration information, the changes in the microscopic physical properties of the workpiece material are identified.

6. The robotic arm anti-loosening gripping control method according to claim 5, characterized in that, The step of identifying changes in the microscopic physical properties of the workpiece material based on the local micro-vibration information includes: The workpiece surface is scanned non-contactly using the micro laser Doppler vibration measurement system to obtain local morphological information of the workpiece surface; Based on the local morphology information, identify the micro-particles or areas with uneven oil film thickness on the surface of the workpiece; Based on the microparticles or the uneven oil film thickness, the micro-vibration signals caused by surface foreign matter are filtered out. Based on the filtered micro-vibration information, the changes in the microscopic physical properties of the workpiece material are identified.

7. The robotic arm anti-loosening gripping control method according to claim 6, characterized in that, The step of identifying changes in the microscopic physical properties of the workpiece material based on the filtered micro-vibration information includes: When performing non-contact scanning on the surface of the workpiece, an initial scan is performed on multiple representative areas of the workpiece to establish a local micro-vibration response baseline of the workpiece in an undamaged state. The local micro-vibration response baseline includes the inherent vibration characteristics and anisotropic response of different areas. In subsequent monitoring, the local micro-vibration information of the workpiece surface will be acquired in real time and compared with the baseline in a regional manner to identify the difference between the real-time local micro-vibration information and the baseline. If the difference exceeds a preset difference threshold, it is determined that the microscopic physical properties of the workpiece material have changed.

8. The method for preventing detachment of a robotic arm's gripping control according to claim 7, characterized in that, If the difference exceeds a preset difference threshold, then determining that the microscopic physical properties of the workpiece material have changed includes: When performing non-contact scanning on the surface of the workpiece, the spectral reflectance information of the workpiece surface is obtained through a multispectral imaging system; Based on the spectral reflectance information, identify the micro-scratches or localized oxide layer areas on the surface of the workpiece; When comparing the real-time local micro-vibration information with the baseline, the differences in the micro-scratches or the local oxide layer areas are weighted or excluded. When the difference between the weighted or excluded real-time micro-vibration information and the baseline exceeds the preset difference threshold, it is determined that the microscopic physical properties of the workpiece material have changed.

9. A robotic arm anti-loosening gripping control system, characterized in that, The system includes: The monitoring terminal is used to monitor the local deformation on the contact surface between the gripper and the workpiece in real time through a distributed array of miniature deformation sensors on the contact surface of the gripper fingertips of the robotic arm, and obtain the local deformation signal. The adjustment end is used to perform multi-point deformation signal differential analysis on the local deformation signal to identify the small relative motion trend information of the workpiece; according to the small relative motion trend information, the output force of each gripper finger of the gripper is adjusted through the independent gripper finger drive mechanism to suppress the small relative motion trend, and the adjustment includes differential adjustment of the output force of each gripper finger; The output end is used to continuously monitor the output force of each clamping finger and compare it with a preset workpiece damage threshold to ensure that the output force does not damage the workpiece. The adjustment end is used for: A piezoelectric material layer is integrated on the contact surface of the gripper fingertip, and the piezoelectric material layer is connected to an independent drive circuit; Based on the micro relative motion trend information, local transient stick-slip events are identified, and reverse micropulse torques are calculated; An electrical pulse command is sent to the piezoelectric material layer corresponding to the local transient stick-slip event to drive the piezoelectric material layer to apply a micro-pulse torque in order to decouple the stickiness effect; Local deformation signals beneath the piezoelectric material layer are monitored at a high sampling rate; The local deformation signal is compared with the preset workpiece damage threshold to detect local pressure pulses; When the local pressure pulse is detected, the local pressure pulse event is recorded, and the output force of other grippers or the gripper posture is adjusted to disperse the pressure.

Images