A robot wrist end-side grasp slide compensation method, system and device

Abstract

The embodiment of the application provides a kind of robot wrist end side's grabbing sliding compensation method, system and equipment, method includes: the intermediate force data is extracted to feature, obtains normal force variation and tangential force variation;According to the high frequency amplitude, contact center offset distance, normal force variation and tangential force variation, determine the sliding grade of dexterous hand workpiece grabbing;According to the high frequency amplitude, contact center offset distance, normal force variation, tangential force variation and sliding grade, determine the torque compensation amount and joint posture adjustment amount of the dexterous hand;Torque compensation amount and joint posture adjustment amount including control instruction are sent to the dexterous hand motor, so that dexterous hand motor adjusts the dexterous hand of robot according to control instruction.The torque compensation amount and joint posture adjustment amount can be directly calculated by the wrist end side of the present application.

Description

Technical Field

[0001] The embodiments of the present invention relate to the field of robot control technology, and in particular to a method, system and device for gripping and sliding compensation on the wrist end of a robot. Background Technology

[0002] As robotics technology advances towards higher precision, greater flexibility, and greater adaptability, dexterous robotic hands are widely used for grasping and handling irregular, fragile, and soft workpieces. However, in actual operation, the dexterous hand at the end of the robot's wrist is prone to slipping relative to the fingertips when grasping workpieces due to factors such as uneven workpiece surface friction coefficients, load variations, sudden changes in motion posture, and external disturbances.

[0003] Existing robot grasping control solutions suffer from the following technical shortcomings: Relying solely on tactile or force sensors fails to comprehensively capture multi-dimensional characteristics of sliding, such as high-frequency vibrations, contact position shifts, and dynamic force changes, leading to missed detections and misjudgments. They often employ post-event detection and compensation, which cannot respond quickly in the early stages of sliding, resulting in delayed compensation and poor control performance. The lack of graded quantification of sliding intensity results in fixed or coarsely adjusted compensation amounts, making it difficult to match the control requirements of different sliding levels and prone to overcompensation or undercompensation. Existing solutions largely rely on host computer computation, but insufficient computing power and significant communication latency at the wrist end prevent real-time closed-loop control. Summary of the Invention

[0004] The technical problem to be solved by the embodiments of the present invention is to provide a method, system and device for gripping and sliding compensation on the wrist end of a robot, which can directly calculate the torque compensation amount and joint posture adjustment amount through the wrist end, reduce communication delay and improve real-time performance and control accuracy.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows: A method for compensating for slippage during grasping on the end of a robot wrist, applied to a control device disposed on the end of a robot wrist, wherein the control device is communicatively connected to a tactile sensor for detecting the dexterous hand, a force sensor for detecting the dexterous hand, and a motor for controlling the dexterous hand, the method comprising: Acquire the raw tactile data of the dexterous hand collected by the tactile sensor and the raw force data collected by the force sensor; The raw tactile data is preprocessed to obtain intermediate tactile data; Preprocess the raw force data to obtain intermediate force data; Feature extraction is performed on the intermediate tactile data to obtain the high-frequency amplitude and the contact center offset distance; Feature extraction is performed on the intermediate force data to obtain the changes in normal force and tangential force; Based on the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change, the sliding level of the dexterous hand gripping the workpiece is determined; Based on the high-frequency amplitude, contact center offset distance, normal force change, tangential force change, and sliding level, the torque compensation amount and joint posture adjustment amount of the dexterous hand are determined. Control commands, including torque compensation and joint posture adjustment, are sent to the dexterous hand motor, so that the dexterous hand motor adjusts the robot's dexterous hand according to the control commands.

[0006] Optionally, the raw tactile data is preprocessed to obtain intermediate tactile data, including: The original tactile data is filtered and denoised to obtain denoised tactile data; The denoised tactile data is normalized to obtain intermediate tactile data; The raw force data is preprocessed to obtain intermediate force data, including: The original force data is filtered and denoised to obtain denoised force data; The denoised intensity data is normalized to obtain intermediate intensity data.

[0007] Optionally, the original tactile data is filtered and denoised to obtain denoised tactile data, including: according to We obtain the denoised tactile data; in, T fil ( x, y, t (This refers to the tactile data after denoising.) σ These are the filter kernel parameters. i =-2, -1, 0, 1, 2, j =-2, -1, 0, 1, 2, T ( x, y, t This refers to the raw tactile data; The denoised tactile data is normalized to obtain intermediate tactile data, including: according to This yields intermediate tactile data; in, T zj ( x, y, t (This refers to intermediate tactile data.) T fil ( x, y, t (This refers to the tactile data after denoising.) T max This represents the maximum range of the tactile sensor. T minThis is the minimum value of the tactile sensor's range; The original force data is filtered and denoised to obtain denoised force data, including: according to , We obtain the denoised force data; in,{ F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data, { F a ( t ) ，F b ( t )} represents the original force data. k =-2, -1, 0, 1, 2, △t The sampling interval; The denoised intensity data is normalized to obtain intermediate intensity data, including: according to , This yields intermediate force data; in,{ F a，zj ( t ) ，F b，zj ( t )} represents intermediate force data, { F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data. T a，max This represents the maximum value of the normal vector path of the force sensor. T a，min This represents the minimum normal vector path of the force sensor. T b，max This represents the maximum value of the tangential vector path of the force sensor. T b，min This represents the minimum tangential vector path of the force sensor.

[0008] Optionally, feature extraction is performed on the intermediate tactile data to obtain the high-frequency amplitude and contact center offset distance, including: according to A single-channel time-domain tactile signal is obtained; in, T avg (t This is a single-channel time-domain tactile signal. x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to The frequency domain tactile signal is obtained; in, T freq ( k () is a frequency domain tactile signal. k =0, 1, ..., N -1, N For the frame length of the Fast Fourier Transform, FFT For the Fast Fourier Transform function, T avg ( t This is a single-channel time-domain tactile signal; according to This yields the high-frequency amplitude. in, A h For high frequency amplitude, k = N 0, 1, ..., N -1, N 0 is the starting index for the high-frequency band. , w 0 represents the lower limit of the dominant frequency of the sliding feature. f s Sampling frequency, N For the frame length of the Fast Fourier Transform, T freq ( k () is a frequency domain tactile signal; according to , The initial contact center is obtained; in,( x 0, y 0) is the initial contact center. x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to , To obtain the current contact center; in,( x t , y t (This is the current contact center.) x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to , thus obtaining the contact center offset distance; in, d The distance of the contact center offset. p For the physical spacing of the contact sensor pixels, ( x 0, y 0) is the initial contact center, ( x t , y t () is the current contact center.

[0009] Optionally, feature extraction is performed on the intermediate force data to obtain the changes in normal force and tangential force, including: according to The change in normal force is obtained; in, △F a ( t () represents the change in normal force. F a，zj ( t () represents the normal force of the intermediate force data. △t The sampling interval; according to The change in tangential force is obtained; in, △F b ( t ) represents the change in tangential force. F b，zj ( t () represents the tangential force of the intermediate force data. △t The sampling interval is denoted as .

[0010] Optionally, based on the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change, the sliding level of the dexterous hand gripping the workpiece is determined, including: according to The excess amounts of the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change are determined. in, s 1 represents the overscaling of the high-frequency amplitude. A h For high frequency amplitude, A h0 For high-frequency amplitude threshold, s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. d The distance of the contact center offset. d 0 represents the contact center offset distance threshold. s 3 represents the excess amount of the change in normal force. △F a The change in normal force △F a0 The threshold for the change in normal force. s 4 represents the excess amount of the change in tangential force. △F b The change in tangential force △F b0 This is the threshold for the change in tangential force. according to Determine the sliding level of the dexterous hand when gripping the workpiece. in, L For the sliding level of a dexterous hand gripping a workpiece, s 1 represents the overscaling of the high-frequency amplitude. s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. s 3 represents the excess amount of the change in normal force. s 4 represents the excess amount of the change in tangential force.

[0011] Optionally, based on the high-frequency amplitude, contact center offset distance, normal force change, tangential force change, and sliding level, the torque compensation and joint posture adjustment of the dexterous hand are determined, including: according to The overall weight is obtained; in, w For comprehensive weighting, g For weighting coefficients, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 This is the threshold for the change in tangential force. according to The torque compensation amount of the dexterous hand is obtained; in, M This is the torque compensation amount for the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, A h For high frequency amplitude, w For comprehensive weighting, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 1 is the first compensation coefficient. k 2 is the second compensation coefficient. k 3 is the third compensation coefficient; according to The joint posture adjustment amount of the dexterous hand is obtained; in, θ This refers to the amount of joint posture adjustment of the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, d The distance of the contact center offset. d max For the maximum allowable offset distance, w For comprehensive weighting, △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 4 is the fourth compensation coefficient. k 5 is the fifth compensation coefficient.

[0012] Embodiments of the present invention also provide a gripping slip compensation system for the end face of a robot wrist, comprising: The acquisition module is used to acquire the raw tactile data of the dexterous hand collected by the tactile sensor and the raw force data collected by the force sensor; The processing module is used to preprocess the original tactile data to obtain intermediate tactile data; preprocess the original force data to obtain intermediate force data; extract features from the intermediate tactile data to obtain high-frequency amplitude and contact center offset distance; extract features from the intermediate force data to obtain normal force variation and tangential force variation; determine the sliding level of the dexterous hand grasping the workpiece based on the high-frequency amplitude, contact center offset distance, normal force variation, and tangential force variation; determine the torque compensation amount and joint posture adjustment amount of the dexterous hand based on the high-frequency amplitude, contact center offset distance, normal force variation, tangential force variation, and sliding level; and send control commands including torque compensation amount and joint posture adjustment amount to the dexterous hand motor, so that the dexterous hand motor adjusts the robot's dexterous hand according to the control commands.

[0013] Embodiments of the present invention also provide a computing device, comprising: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the one or more processors to perform the method as described above.

[0014] Embodiments of the present invention also provide a computing device readable storage medium storing a program that, when executed by a processor, implements the method described above.

[0015] The above-described solutions of the embodiments of the present invention have at least the following beneficial effects: The above-described solution in this invention comprehensively captures multi-dimensional sliding features by simultaneously collecting and processing tactile and force sensor data, thus solving the problems of misjudgment and missed detection caused by single sensor detection. Through preprocessing and precise feature extraction, early sliding identification is achieved, avoiding workpiece detachment or damage due to compensation lag. Based on multi-feature quantification to determine the sliding level, different graded compensation strategies can be output to match different sliding degrees, avoiding overcompensation and undercompensation. Direct calculation of torque compensation and joint posture adjustment at the wrist end reduces communication latency and improves real-time performance and control accuracy. The overall solution ensures grasping stability while improving the robot's adaptive handling capability for irregular, fragile, and soft workpieces, effectively enhancing operational accuracy, safety, and versatility. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating the gripping and sliding compensation method for the robot wrist end provided in an embodiment of the present invention.

[0017] Figure 2 This is a schematic diagram of the gripping and sliding compensation system on the wrist end of a robot provided in an embodiment of the present invention. Detailed Implementation

[0018] Exemplary embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the invention are shown in the drawings, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0019] like Figure 1 As shown, an embodiment of the present invention provides a grasping and sliding compensation method for the end of a robot wrist, applied to a control device disposed on the end of a robot wrist. The control device is communicatively connected to a tactile sensor for detecting the dexterous hand, a force sensor, and a motor for controlling the dexterous hand. The method includes: Step 11: Obtain the raw tactile data of the dexterous hand collected by the tactile sensor and the raw force data collected by the force sensor; Step 12: Preprocess the raw tactile data to obtain intermediate tactile data; Step 13: Preprocess the raw force data to obtain intermediate force data; Step 14: Extract features from the intermediate tactile data to obtain the high-frequency amplitude and the contact center offset distance; Step 15: Extract features from the intermediate force data to obtain the changes in normal force and tangential force; Step 16: Determine the sliding level of the dexterous hand gripping the workpiece based on the high frequency amplitude, contact center offset distance, normal force change, and tangential force change. Step 17: Determine the torque compensation amount and joint posture adjustment amount of the dexterous hand based on the high frequency amplitude, contact center offset distance, normal force change, tangential force change and sliding level. Step 18: Send control commands, including torque compensation and joint posture adjustment, to the dexterous hand motor so that the dexterous hand motor adjusts the robot's dexterous hand according to the control commands.

[0020] In this embodiment, by simultaneously acquiring and jointly processing tactile and force sensor data, multi-dimensional sliding features are comprehensively captured, solving the problems of misjudgment and missed detection caused by single sensor detection. Preprocessing and precise feature extraction enable early identification of sliding, preventing workpiece detachment or damage due to compensation lag. Based on multi-feature quantification to determine the sliding level, graded compensation strategies can be output to match different sliding degrees, avoiding overcompensation and undercompensation. Direct calculation of torque compensation and joint posture adjustment at the wrist end reduces communication latency and improves real-time performance and control accuracy. The overall solution ensures grasping stability while improving the robot's adaptive handling capability for irregular, fragile, and soft workpieces, effectively enhancing operational accuracy, safety, and versatility.

[0021] In an optional embodiment of the present invention, in step 11, the raw tactile data of the dexterous hand collected by the tactile sensor and the raw force data collected by the force sensor are obtained; specifically, the raw tactile data is... T ( x, y, t ) m×n The original force data is { F a ( t ) ，F b ( t )}; in, x ∈[1, m ], y ∈[1, n ], t Sampling time, m×n The resolution of the tactile sensor array (number of rows of pixels × number of columns). F a For normal force, F b It is a tangential force.

[0022] In this embodiment, by simultaneously acquiring array-style raw tactile data and dual-component raw force data, the spatial distribution information and force dynamic information during the sliding process can be completely preserved, avoiding data loss that could lead to feature extraction distortion. Using a unified sampling time ensures that the tactile and force signals are time-aligned, eliminating detection errors caused by asynchrony and improving the accuracy of sliding judgment.

[0023] In an optional embodiment of the present invention, step 12 involves preprocessing the original tactile data to obtain intermediate tactile data, including: Step 121: Filter and denoise the original tactile data to obtain denoised tactile data; specifically, according to... We obtain the denoised tactile data; in, Tfil ( x, y, t (This refers to the tactile data after denoising.) σ These are the filter kernel parameters. i =-2, -1, 0, 1, 2, j =-2, -1, 0, 1, 2, T ( x, y, t This refers to the raw tactile data; Step 122: Normalize the denoised tactile data to obtain intermediate tactile data; specifically, according to... This yields intermediate tactile data; in, T zj ( x, y, t (This refers to intermediate tactile data.) T fil ( x, y, t (This refers to the tactile data after denoising.) T max This represents the maximum range of the tactile sensor. T min This is the minimum value of the tactile sensor's range; In step 13, the raw force data is preprocessed to obtain intermediate force data, including: Step 131: Filter and denoise the original force data to obtain denoised force data; according to , We obtain the denoised force data; in,{ F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data, { F a ( t ) ，F b ( t )} represents the original force data. k =-2, -1, 0, 1, 2, △t The sampling interval; Step 132: Normalize the denoised intensity data to obtain intermediate intensity data; according to , This yields intermediate force data; in,{ F a，zj ( t ) ，F b，zj( t )} represents intermediate force data, { F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data. T a，max This represents the maximum value of the normal vector path of the force sensor. T a，min This represents the minimum normal vector path of the force sensor. T b，max This represents the maximum value of the tangential vector path of the force sensor. T b，min This represents the minimum tangential vector path of the force sensor.

[0024] In this embodiment, Gaussian filtering and 5-point mean filtering are employed to effectively suppress electromagnetic and vibration noise in industrial environments, preserve true sliding characteristics, and improve signal reliability. Separate filtering methods are designed for the tactile array and force signals to adapt to different data characteristics, resulting in more accurate noise reduction. Range normalization maps the data to a unified range, eliminating numerical deviations caused by differences in sensor ranges, allowing for direct quantification and comparison of features such as high-frequency amplitude, contact center offset, and force changes. The preprocessed data is stable, standardized, and comparable, significantly reducing errors in subsequent feature extraction and sliding determination, providing high-quality input for real-time and accurate compensation control.

[0025] In an optional embodiment of the present invention, step 14, performing feature extraction on the intermediate tactile data to obtain the high-frequency amplitude and contact center offset distance, includes: Step 141, according to A single-channel time-domain tactile signal is obtained; in, T avg ( t This is a single-channel time-domain tactile signal. x ∈[1, m ], y ∈[1, n ], m×n The resolution of the tactile sensor array (number of rows of pixels × number of columns). T zj ( x, y, t (This refers to intermediate tactile data;) Step 142, according to The frequency domain tactile signal is obtained; in, T freq ( k () is a frequency domain tactile signal. k =0, 1, ...,N -1, N The frame length of the Fast Fourier Transform (the number of signal sampling points in the Fast Fourier Transform). FFT For the Fast Fourier Transform function, T avg ( t This is a single-channel time-domain tactile signal; Step 143, according to This yields the high-frequency amplitude. in, A h For high frequency amplitude, k = N 0, 1, ..., N -1, N 0 is the starting index for the high-frequency band. , w 0 represents the lower limit of the dominant frequency of the sliding feature. f s Sampling frequency, N For the frame length of the Fast Fourier Transform, T freq ( k () is a frequency domain tactile signal; Step 144, according to , The initial contact center is obtained; in,( x 0, y 0) is the initial contact center. x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) Step 145, according to , To obtain the current contact center; in,( x t , y t (This is the current contact center.) x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) Step 146, according to , thus obtaining the contact center offset distance; in, d The distance of the contact center offset. p For the physical spacing of the contact sensor pixels, ( x 0, y 0) is the initial contact center, ( x t , y t () is the current contact center.

[0026] In this embodiment, the array tactile data is first averaged in the time domain to simplify calculations and retain global vibration characteristics, adapting to real-time end-side computation. High-frequency amplitudes are extracted via FFT transformation, enabling accurate identification of early-stage micro-vibrations and achieving early-stage slip detection. The initial and current contact centers are calculated to objectively reflect changes in contact position. The offset is quantified using Euclidean distance to intuitively characterize the degree of slippage. The calculation is stable, robust, and unaffected by local noise, accurately and sensitively extracting core tactile features of slippage, providing a reliable basis for slippage level determination and compensation control.

[0027] In an optional embodiment of the present invention, step 15, performing feature extraction on the intermediate force data to obtain the change in normal force and the change in tangential force, includes: Step 151, according to The change in normal force is obtained; in, △F a ( t () represents the change in normal force. F a，zj ( t () represents the normal force of the intermediate force data. △t The sampling interval; Step 152, according to The change in tangential force is obtained; in, △F b ( t ) represents the change in tangential force. F b，zj ( t () represents the tangential force of the intermediate force data. △t The sampling interval is denoted as .

[0028] In this embodiment, normalized force data from adjacent time points are used for difference calculation, resulting in minimal computational load and suitability for real-time control at the robot wrist end. Changes in normal force quickly reflect sudden changes in gripping tightness, while changes in tangential force directly characterize slippage trends, accurately capturing the transient mechanical features at the moment of slippage. The difference calculation is unaffected by range or reference offset, exhibiting strong robustness and providing stable and reliable mechanical characteristics. This complements tactile characteristics, providing a realistic and timely dynamic force basis for slippage level determination and compensation calculation.

[0029] In an optional embodiment of the present invention, step 16, determining the sliding level of the dexterous hand gripping the workpiece based on the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change, includes: In step 161, according to The excess amounts of the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change are determined. in, s 1 represents the overscaling of the high-frequency amplitude. A h For high frequency amplitude, A h0 For high-frequency amplitude threshold, s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. d The distance of the contact center offset. d 0 represents the contact center offset distance threshold. s 3 represents the excess amount of the change in normal force. △F a The change in normal force △F a0 The threshold for the change in normal force. s 4 represents the excess amount of the change in tangential force. △F b The change in tangential force △F b0 This is the threshold for the change in tangential force. In step 162, according to Determine the sliding level of the dexterous hand when gripping the workpiece. in, L For the sliding level of a dexterous hand gripping a workpiece, s 1 represents the overscaling of the high-frequency amplitude. s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. s 3 represents the excess amount of the change in normal force. s 4 represents the excess amount of the change in tangential force.

[0030] In this embodiment, relative exceedance values ​​are used instead of absolute values ​​for judgment, eliminating judgment biases caused by differences in workpiece material, surface condition, and sensors, resulting in a more objective and stable classification. Slippage is divided into four levels (0-3) to achieve refined and quantitative differentiation of slippage degree, avoiding the coarseness of binary "yes / no" judgments. The classification combines the number of exceedance parameters with the maximum exceedance percentage, resulting in clear logic, strong robustness, and accurate differentiation of different states such as micro-slippage, light slippage, and heavy slippage. The classification results directly connect to subsequent compensation strategies, providing a clear basis for classified, precise, and adaptive compensation, significantly improving the safety and stability of grasping control.

[0031] In an optional embodiment of the present invention, step 17, determining the torque compensation amount and joint posture adjustment amount of the dexterous hand based on the high-frequency amplitude, contact center offset distance, normal force change, tangential force change, and sliding level, includes: Step 171, according to The overall weight is obtained; in, w For comprehensive weighting, g For weighting coefficients, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 This is the threshold for the change in tangential force. Step 172, according to The torque compensation amount of the dexterous hand is obtained; in, M This is the torque compensation amount for the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, A h For high frequency amplitude, w For comprehensive weighting, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 1 is the first compensation coefficient. k 2 is the second compensation coefficient. k 3 is the third compensation coefficient; Step 173, according to The joint posture adjustment amount of the dexterous hand is obtained; in, θ This refers to the amount of joint posture adjustment of the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, d The distance of the contact center offset. d max For the maximum allowable offset distance, w For comprehensive weighting, △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 4 is the fourth compensation coefficient. k 5 is the fifth compensation coefficient.

[0032] In this embodiment, a comprehensive weighting of normal and tangential forces is introduced, which dynamically allocates compensation emphasis based on the actual force applied, adapting to the sliding characteristics under different working conditions. The torque compensation amount integrates the sliding level, high-frequency amplitude, and force variation, outputting a clamping torque precisely matched to the degree of sliding, avoiding overcompensation that damages the workpiece or undercompensation that exacerbates sliding. The joint posture adjustment amount combines offset distance and tangential force variation to achieve adaptive posture correction, suppressing the sliding tendency at the position level. Real-time calculation can be completed at the wrist end, achieving precise compensation through a dual closed loop of torque and posture, significantly improving gripping stability and operational safety.

[0033] In an optional embodiment of the present invention, step 18, sending a control command including a torque compensation amount and a joint posture adjustment amount to the dexterous hand motor, so that the dexterous hand motor adjusts the robot's dexterous hand according to the control command, may include: The calculated torque compensation and joint posture adjustment are encapsulated into a bus control frame and sent to each dexterous hand motor (each joint drive unit of the dexterous hand) by addressing the motor ID. After receiving the instruction, the motor controller adds the torque compensation to the current closed-loop torque loop and corrects the target joint angle with the joint posture adjustment, thus completing the torque compensation and posture fine-tuning simultaneously. After execution, the status feedback is sent back.

[0034] In this embodiment, the bus control frame is encapsulated according to the motor ID, ensuring efficient and reliable command transmission. It is compatible with synchronous control of multi-joint dexterous hands, avoiding command errors. Torque compensation is directly superimposed into the torque loop, and attitude adjustment is used to correct joint angles in real time, achieving synchronous execution of compensation and fine-tuning. This results in fast response, minimal lag, and rapid suppression of slippage. After execution, the motor sends back status feedback, forming a complete control closed loop, facilitating system verification of compensation effectiveness and initiation of the next round of adjustment.

[0035] Example 1 System parameter presets: tactile sensor array resolution: m×n=16×16; Sampling frequency: f s =1000Hz; Fast Fourier Transform frame length: N =256; Lower limit of sliding characteristic main frequency: w 0 = 50Hz; Sensor pixel physical spacing: p =0.5mm; Weighting coefficient: g =0.5; Compensation coefficient: k 1 = 0.4 k 2 = 0.3, k 3 = 0.3 k 4 = 1.5 k 5 = 0.8; Threshold parameter: A h0 =0.5, d 0 = 2mm, △F a0 =1.0N, △F b =1.2N; Maximum allowable offset distance: d max =10mm; Sampling interval: △t =1ms; Gaussian filter kernel parameters: σ =0.8; Specific implementation process: Step 21: Synchronously acquire raw data: The control device synchronously acquires raw tactile data from the tactile sensor array. T ( x, y, t ), raw force data from the force sensor, including: Raw data of normal force F a ( t =15N, raw data of tangential force F b ( t =8N; Step 22: Data Preprocessing: Filtering and Denoising: The original tactile data is subjected to a 5×5 Gaussian filter to obtain the denoised tactile data. T fil ( x, y, t ); 5-point mean filtering is applied to the original force data to obtain F a ( t ) ，F b ( t ); Normalization: Normalize the data according to the sensor range to obtain intermediate tactile data. T zj ( x, y, tIntermediate strength data F a，zj ( t =0.75 F b，zj ( t =0.64; Step 23: Feature Extraction Tactile characteristics: Calculation of single-channel time-domain tactile signals T avg ( t The frequency domain signal is obtained by FFT. T freq ( k ), calculate high-frequency amplitude A h =0.82; Calculate the initial contact center ( x 0, y 0) = (245, 248), Current contact center ( x t , y t ) = (251, 253), which gives the contact center offset distance. d =3.2mm; Force characteristics: Calculation of the change in normal force △F a ( t =1.2N, change in tangential force △F a ( t =1.8N; Step 14: Determine the sliding level: Calculate the out-of-scalar values ​​for each parameter: s 1 = 64%, s 2=60%, s 3 = 20%, s 4=50%; 3 parameters exceed the standard, determine the sliding level. L =2; Step 25: Compensation Calculation: Calculate the overall weight w =0.75; Torque compensation M =1.26N m; Joint posture adjustment θ =2.1°; Step 26: Send control commands: Torque compensation M With joint posture adjustment θEncapsulated as a bus control frame, it is sent to each dexterous hand motor (drive unit of each joint of the dexterous hand) by addressing the motor ID; after receiving the instruction, the motor controller adds the torque compensation amount to the current closed-loop torque loop, corrects the target joint angle with the joint attitude adjustment amount, and synchronously completes torque compensation and attitude fine adjustment, and sends back status feedback after execution.

[0036] This invention employs synchronous timing acquisition of tactile and force sensors to ensure data alignment and complete preservation of multi-dimensional sliding information, thus avoiding detection errors at the source. Targeted filtering and range normalization preprocessing suppress on-site noise and eliminate range differences, providing high-quality standard data for subsequent processing. Four core features—high-frequency amplitude, contact center offset, and normal / tangential force changes—are extracted in parallel from tactile and force signals to achieve accurate early-stage sliding identification, balancing sensitivity and robustness. Sliding levels are categorized based on relative overscalar values ​​and multi-level rules, quantifying and distinguishing the degree of sliding, ensuring stable and reliable judgment unaffected by workpiece or environmental interference. Based on the sliding level and multiple features, torque compensation and joint posture adjustment are dynamically calculated to achieve graded and precise compensation, avoiding over-compensation or under-compensation. Finally, through bus addressing, dual-loop synchronous execution, and real-time status feedback, a real-time closed-loop control is formed at the end side, characterized by fast response, low latency, and high reliability.

[0037] like Figure 2 As shown, an embodiment of the present invention provides a gripping slip compensation system 20 for the end of a robot wrist, comprising: The acquisition module 21 is used to acquire the raw tactile data of the dexterous hand collected by the tactile sensor and the raw force data collected by the force sensor; The processing module 22 is used to preprocess the original tactile data to obtain intermediate tactile data; preprocess the original force data to obtain intermediate force data; extract features from the intermediate tactile data to obtain high-frequency amplitude and contact center offset distance; extract features from the intermediate force data to obtain normal force variation and tangential force variation; determine the sliding level of the dexterous hand grasping the workpiece based on the high-frequency amplitude, contact center offset distance, normal force variation, and tangential force variation; determine the torque compensation amount and joint posture adjustment amount of the dexterous hand based on the high-frequency amplitude, contact center offset distance, normal force variation, tangential force variation, and sliding level; and send control commands including torque compensation amount and joint posture adjustment amount to the dexterous hand motor, so that the dexterous hand motor adjusts the robot's dexterous hand according to the control commands.

[0038] Optionally, the raw tactile data is preprocessed to obtain intermediate tactile data, including: The original tactile data is filtered and denoised to obtain denoised tactile data; The denoised tactile data is normalized to obtain intermediate tactile data; The raw force data is preprocessed to obtain intermediate force data, including: The original force data is filtered and denoised to obtain denoised force data; The denoised intensity data is normalized to obtain intermediate intensity data.

[0039] Optionally, the original tactile data is filtered and denoised to obtain denoised tactile data, including: according to We obtain the denoised tactile data; in, T fil ( x, y, t (This refers to the tactile data after denoising.) σ These are the filter kernel parameters. i =-2, -1, 0, 1, 2, j =-2, -1, 0, 1, 2, T ( x, y, t This refers to the raw tactile data; The denoised tactile data is normalized to obtain intermediate tactile data, including: according to This yields intermediate tactile data; in, T zj ( x, y, t (This refers to intermediate tactile data.) T fil ( x, y, t (This refers to the tactile data after denoising.) T max This represents the maximum range of the tactile sensor. T min This is the minimum value of the tactile sensor's range; The original force data is filtered and denoised to obtain denoised force data, including: according to , We obtain the denoised force data; in,{ F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data, { F a ( t ) ，F b ( t )} represents the original force data. k=-2, -1, 0, 1, 2, △t The sampling interval; The denoised intensity data is normalized to obtain intermediate intensity data, including: according to , This yields intermediate force data; in,{ F a，zj ( t ) ，F b，zj ( t )} represents intermediate force data, { F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data. T a，max This represents the maximum value of the normal vector path of the force sensor. T a，min This represents the minimum normal vector path of the force sensor. T b，max This represents the maximum value of the tangential vector path of the force sensor. T b，min This represents the minimum tangential vector path of the force sensor.

[0040] Optionally, feature extraction is performed on the intermediate tactile data to obtain the high-frequency amplitude and contact center offset distance, including: according to A single-channel time-domain tactile signal is obtained; in, T avg ( t This is a single-channel time-domain tactile signal. x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to The frequency domain tactile signal is obtained; in, T freq ( k () is a frequency domain tactile signal. k =0, 1, ..., N -1, N For the frame length of the Fast Fourier Transform, FFTFor the Fast Fourier Transform function, T avg ( t This is a single-channel time-domain tactile signal; according to This yields the high-frequency amplitude. in, A h For high frequency amplitude, k = N 0, 1, ..., N -1, N 0 is the starting index for the high-frequency band. , w 0 represents the lower limit of the dominant frequency of the sliding feature. f s Sampling frequency, N For the frame length of the Fast Fourier Transform, T freq ( k () is a frequency domain tactile signal; according to , The initial contact center is obtained; in,( x 0, y 0) is the initial contact center. x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to , To obtain the current contact center; in,( x t , y t (This is the current contact center.) x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to , thus obtaining the contact center offset distance; in, d The distance of the contact center offset.p For the physical spacing of the contact sensor pixels, ( x 0, y 0) is the initial contact center, ( x t , y t () is the current contact center.

[0041] Optionally, feature extraction is performed on the intermediate force data to obtain the changes in normal force and tangential force, including: according to The change in normal force is obtained; in, △F a ( t () represents the change in normal force. F a，zj ( t () represents the normal force of the intermediate force data. △t The sampling interval; according to The change in tangential force is obtained; in, △F b ( t ) represents the change in tangential force. F b，zj ( t () represents the tangential force of the intermediate force data. △t The sampling interval is denoted as .

[0042] Optionally, based on the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change, the sliding level of the dexterous hand gripping the workpiece is determined, including: according to The excess amounts of the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change are determined. in, s 1 represents the overscaling of the high-frequency amplitude. A h For high frequency amplitude, A h0 For high-frequency amplitude threshold, s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. d The distance of the contact center offset. d 0 represents the contact center offset distance threshold. s 3 represents the excess amount of the change in normal force. △F a The change in normal force △F a0 The threshold for the change in normal force. s 4 represents the excess amount of the change in tangential force.△F b The change in tangential force △F b0 This is the threshold for the change in tangential force. according to Determine the sliding level of the dexterous hand when gripping the workpiece. in, L For the sliding level of a dexterous hand gripping a workpiece, s 1 represents the overscaling of the high-frequency amplitude. s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. s 3 represents the excess amount of the change in normal force. s 4 represents the excess amount of the change in tangential force.

[0043] Optionally, based on the high-frequency amplitude, contact center offset distance, normal force change, tangential force change, and sliding level, the torque compensation and joint posture adjustment of the dexterous hand are determined, including: according to The overall weight is obtained; in, w For comprehensive weighting, g For weighting coefficients, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 This is the threshold for the change in tangential force. according to The torque compensation amount of the dexterous hand is obtained; in, M This is the torque compensation amount for the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, A h For high frequency amplitude, w For comprehensive weighting, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 1 is the first compensation coefficient. k 2 is the second compensation coefficient. k 3 is the third compensation coefficient; according to The joint posture adjustment amount of the dexterous hand is obtained; in, θ This refers to the amount of joint posture adjustment of the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, d The distance of the contact center offset. d max For the maximum allowable offset distance, w For comprehensive weighting, △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 4 is the fourth compensation coefficient. k 5 is the fifth compensation coefficient.

[0044] It should be noted that this device is a device corresponding to the above method. All implementation methods in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0045] Embodiments of the present invention also provide a computing device, including: one or more processors; and a storage device for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors implement the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0046] Embodiments of the present invention also provide a computing device readable storage medium storing instructions that, when executed on a computing device, cause the computing device to perform the method described above. All implementations in the above method embodiments are applicable to this embodiment and can achieve the same technical effect.

[0047] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computing device software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0048] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0049] In the embodiments provided by this invention, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0050] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0051] In addition, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0052] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computing device-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computing device software product is stored in a storage medium and includes several instructions to cause a computing device (which may be a personal computing device, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0053] Furthermore, it should be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent solutions of the present invention. Moreover, the steps performing the above-described series of processes can naturally be executed in the order described, but are not necessarily required to be executed in chronological order; some steps can be executed in parallel or independently of each other. Those skilled in the art will understand that all or any step or component of the method and apparatus of the present invention can be implemented in any computing device (including processors, storage media, etc.) or network of computing devices, in hardware, firmware, software, or a combination thereof. This is something that those skilled in the art can achieve using basic programming skills after reading the description of the present invention.

[0054] Therefore, the object of the present invention can also be achieved by running a program or a set of programs on any computing device. The computing device can be a known general-purpose device. Therefore, the object of the present invention can also be achieved simply by providing a program product containing program code implementing the method or apparatus. That is, such a program product also constitutes the present invention, and the storage medium storing such a program product also constitutes the present invention. Obviously, the storage medium can be any known storage medium or any storage medium developed in the future. It should also be noted that in the apparatus and method of the present invention, it is obvious that the components or steps can be decomposed and / or recombined. These decompositions and / or recombinations should be considered equivalent to the present invention. Furthermore, the steps performing the above series of processes can naturally be performed in the order described, but are not necessarily required to be performed in chronological order. Some steps can be performed in parallel or independently of each other.

[0055] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for compensating for slippage during grasping on the wrist end of a robot, characterized in that, A control device is applied to the end of a robot's wrist, the control device being communicatively connected to a tactile sensor for detecting the dexterous hand, a force sensor, and a motor for controlling the dexterous hand, the method comprising: Acquire the raw tactile data of the dexterous hand collected by the tactile sensor and the raw force data collected by the force sensor; The raw tactile data is preprocessed to obtain intermediate tactile data; Preprocess the raw force data to obtain intermediate force data; Feature extraction is performed on the intermediate tactile data to obtain the high-frequency amplitude and the contact center offset distance; Feature extraction is performed on the intermediate force data to obtain the changes in normal force and tangential force; Based on the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change, the sliding level of the dexterous hand gripping the workpiece is determined; Based on the high-frequency amplitude, contact center offset distance, normal force change, tangential force change, and sliding level, the torque compensation amount and joint posture adjustment amount of the dexterous hand are determined. Control commands, including torque compensation and joint posture adjustment, are sent to the dexterous hand motor, so that the dexterous hand motor adjusts the robot's dexterous hand according to the control commands.

2. The gripping sliding compensation method for the end face of the robot wrist according to claim 1, characterized in that, The raw tactile data is preprocessed to obtain intermediate tactile data, including: The original tactile data is filtered and denoised to obtain denoised tactile data; The denoised tactile data is normalized to obtain intermediate tactile data; The raw force data is preprocessed to obtain intermediate force data, including: The original force data is filtered and denoised to obtain denoised force data; The denoised intensity data is normalized to obtain intermediate intensity data.

3. The gripping sliding compensation method for the end face of the robot wrist according to claim 2, characterized in that, The original tactile data is filtered and denoised to obtain denoised tactile data, including: according to We obtain the denoised tactile data; in, T fil ( x, y, t (This refers to the denoised tactile data.) σ These are the filter kernel parameters. i =-2, -1, 0, 1, 2, j =-2, -1, 0, 1, 2, T ( x, y, t This refers to the raw tactile data; The denoised tactile data is normalized to obtain intermediate tactile data, including: according to This yields intermediate tactile data; in, T zj ( x, y, t (This refers to intermediate tactile data.) T fil ( x, y, t (This refers to the denoised tactile data.) T max This represents the maximum range of the tactile sensor. T min This is the minimum value of the tactile sensor's range; The original force data is filtered and denoised to obtain denoised force data, including: according to , We obtain the denoised force data; in,{ F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data, { F a ( t ) ，F b ( t )} represents the original force data. k =-2, -1, 0, 1, 2, △t The sampling interval; The denoised intensity data is normalized to obtain intermediate intensity data, including: according to , This yields intermediate force data; in,{ F a，zj ( t ) ，F b，zj ( t )} represents intermediate force data, { F a，fil ( t ) ，F b，fil ( t )} represents the denoised intensity data. T a，max This represents the maximum value of the normal vector path of the force sensor. T a，min This represents the minimum normal vector path of the force sensor. T b，max This represents the maximum value of the tangential vector path of the force sensor. T b，min This represents the minimum tangential vector path of the force sensor.

4. The gripping sliding compensation method for the end face of the robot wrist according to claim 1, characterized in that, Feature extraction is performed on the intermediate tactile data to obtain the high-frequency amplitude and contact center offset distance, including: according to A single-channel time-domain tactile signal is obtained; in, T avg ( t This is a single-channel time-domain tactile signal. x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to The frequency domain tactile signal is obtained; in, T freq ( k () is a frequency domain tactile signal. k =0, 1, ..., N -1, N For the frame length of the Fast Fourier Transform, FFT For the Fast Fourier Transform function, T avg ( t This is a single-channel time-domain tactile signal; according to This yields the high-frequency amplitude. in, A h For high frequency amplitude, k = N 0, 1, ..., N -1, N 0 is the starting index for the high-frequency band. , w 0 represents the lower limit of the dominant frequency of the sliding feature. f s Sampling frequency, N For the frame length of the Fast Fourier Transform, T freq ( k () is a frequency domain tactile signal; according to , The initial contact center is obtained; in,( x 0, y 0) is the initial contact center. x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to , To obtain the current contact center; in,( x t , y t (This is the current contact center.) x ∈[1, m ], y ∈[1, n ], m×n For the resolution of the tactile sensor array, T zj ( x, y, t (This refers to intermediate tactile data;) according to , thus obtaining the contact center offset distance; in, d The distance of the contact center offset. p For the physical spacing of the contact sensor pixels, ( x 0, y 0) is the initial contact center, ( x t , y t () is the current contact center.

5. The gripping sliding compensation method for the end face of the robot wrist according to claim 1, characterized in that, Feature extraction is performed on the intermediate force data to obtain the changes in normal force and tangential force, including: according to The change in normal force is obtained; in, △F a ( t () represents the change in normal force. F a，zj ( t () represents the normal force of the intermediate force data. △t The sampling interval; according to The change in tangential force is obtained; in, △F b ( t ) represents the change in tangential force. F b，zj ( t () represents the tangential force of the intermediate force data. △t The sampling interval is denoted as .

6. The gripping sliding compensation method for the end face of the robot wrist according to claim 1, characterized in that, Based on the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change, the sliding level of the dexterous hand gripping the workpiece is determined, including: according to The excess amounts of the high-frequency amplitude, contact center offset distance, normal force change, and tangential force change are determined. in, s 1 represents the overscaling of the high-frequency amplitude. A h For high frequency amplitude, A h0 For high-frequency amplitude threshold, s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. d The distance of the contact center offset. d 0 represents the contact center offset distance threshold. s 3 represents the excess amount of the change in normal force. △F a The change in normal force △F a0 The threshold for the change in normal force. s 4 represents the excess amount of the change in tangential force. △F b The change in tangential force △F b0 This is the threshold for the change in tangential force. according to Determine the sliding level of the dexterous hand when gripping the workpiece. in, L For the sliding level of a dexterous hand gripping a workpiece, s 1 represents the overscaling of the high-frequency amplitude. s 2 represents the amount by which the contact center offset distance exceeds the scalar limit. s 3 represents the excess amount of the change in normal force. s 4 represents the excess amount of the change in tangential force.

7. The gripping sliding compensation method for the end face of a robot wrist according to claim 1, characterized in that, Based on the high-frequency amplitude, contact center offset distance, normal force change, tangential force change, and sliding level, the torque compensation and joint posture adjustment of the dexterous hand are determined, including: according to The overall weight is obtained; in, w For comprehensive weighting, g For weighting coefficients, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 This is the threshold for the change in tangential force. according to The torque compensation amount of the dexterous hand is obtained; in, M This is the torque compensation amount for the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, A h For high frequency amplitude, w For comprehensive weighting, △F a The change in normal force △F a0 The threshold for the change in normal force. △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 1 is the first compensation coefficient. k 2 is the second compensation coefficient. k 3 is the third compensation coefficient; according to The joint posture adjustment amount of the dexterous hand is obtained; in, θ This refers to the amount of joint posture adjustment of the dexterous hand. L For the sliding level of a dexterous hand gripping a workpiece, d The distance of the contact center offset. d max For the maximum allowable offset distance, w For comprehensive weighting, △F b The change in tangential force △F b0 The threshold for the change in tangential force. k 4 is the fourth compensation coefficient. k 5 is the fifth compensation coefficient.

8. A gripping sliding compensation system for the end face of a robot wrist, characterized in that, include: The acquisition module is used to acquire the raw tactile data of the dexterous hand collected by the tactile sensor and the raw force data collected by the force sensor; The processing module is used to preprocess the raw tactile data to obtain intermediate tactile data; The raw force data is preprocessed to obtain intermediate force data; features are extracted from the intermediate tactile data to obtain high-frequency amplitude and contact center offset distance; features are further extracted from the intermediate force data to obtain normal force variation and tangential force variation; based on the high-frequency amplitude, contact center offset distance, normal force variation, and tangential force variation, the sliding level of the dexterous hand grasping the workpiece is determined; based on the high-frequency amplitude, contact center offset distance, normal force variation, tangential force variation, and sliding level, the torque compensation amount and joint posture adjustment amount of the dexterous hand are determined; control commands including torque compensation amount and joint posture adjustment amount are sent to the dexterous hand motor, so that the dexterous hand motor adjusts the robot's dexterous hand according to the control commands.

9. A computing device, characterized in that, include: One or more processors; A storage device for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to implement the method as described in any one of claims 1 to 7.

10. A computing device readable storage medium, characterized in that, The computing device readable storage medium stores a program that, when executed by a processor, implements the method as described in any one of claims 1 to 7.

Images