Flexible tactile sensor for robotic arm to sense deformability of object

Abstract

A flexible tactile sensor for a robotic arm to sense the deformability of an object, comprising a lower flexible layer (10), a spiral interdigital electrode (20), a porous piezoresistive layer (30), a middle flexible layer (40) and an upper flexible layer (60) which are sequentially connected in an attached manner. A reduced graphene oxide layer (50) is provided on the side of the upper flexible layer (60) facing the middle flexible layer (40), and the side of the reduced graphene oxide layer (50) facing away from the upper flexible layer (60) is connected to the middle flexible layer (40) in an attached manner. By arranging the porous piezoresistive layer (30) and the reduced graphene oxide layer (50) in a same tactile sensor, the simultaneous acquisition of pressure data and strain data of an object is implemented, and then the simultaneous measurement of the material hardness and the overall rigidity of the object can be supported, so that accurate object deformability measurement can be performed.

Description

A flexible tactile sensor for manipulators to sense object deformability Technical Field

[0001] The present invention relates to the field of measurement, and in particular to a flexible tactile sensor used by a manipulator to sense the deformability of an object. Background Art

[0002] As robotic applications gradually expand from industrial to everyday life, similar tactile perception is needed to achieve human-level dexterity. In tactile perception, the perception of deformation-related properties provides valuable tactile cues for distinguishing and identifying deformable objects and developing appropriate manipulation strategies. Determining an object's deformability is particularly challenging in tactile perception research because it is often influenced by multiple deformation-related properties, such as Young's modulus, Shore hardness, and stiffness. Young's modulus and Shore hardness represent the inherent deformability of a material, while stiffness describes an object's overall resistance to elastic deformation, which is influenced by various factors, including object size, material hardness, or deformable structure. Deformable objects in everyday life are often composed of a combination of deformable materials and structures (e.g., balloons) or materials of varying hardness and softness (e.g., oranges). Therefore, enabling robotic perception of the deformability of deformable objects in real-world, unstructured environments requires the coordinated perception of multiple deformation properties, such as material hardness and overall stiffness.

[0003] Currently, in the field of robot tactile perception, there are sensors that can independently measure material hardness. This sensor uses a 1×3 magnetostrictive tactile sensor to measure pressure at three locations. Since it can only measure pressure at three locations, it cannot accurately measure material hardness, nor can it use the pressure data at three locations to accurately calculate the overall stiffness of the object, so it cannot accurately measure the deformability of the object.

[0004] Therefore, the existing technology has defects and needs to be improved and developed.

[0005] Summary of the Invention

[0006] The technical problem to be solved by the present invention is that, in response to the above-mentioned defects of the prior art, a flexible tactile sensor for a manipulator to perceive the deformability of an object is provided, aiming to solve the problem in the prior art that the tactile sensor cannot measure the overall stiffness while measuring the hardness of the material, and cannot measure the deformability of the object more accurately.

[0007] The technical solutions adopted by the present invention to solve the technical problems are as follows:

[0008] A flexible tactile sensor for a robotic arm to sense the deformability of an object, comprising: a lower flexible layer, a spiral interdigitated electrode, a porous piezoresistive layer, an intermediate flexible layer, and an upper flexible layer, which are sequentially bonded and connected. A reduced graphene oxide layer is provided on the side of the upper flexible layer facing the intermediate flexible layer, and a side of the reduced graphene oxide layer facing away from the upper flexible layer is bonded and connected to the intermediate flexible layer.

[0009] Furthermore, the porous piezoresistive layer is formed by uniformly mixing a conductive filler, sacrificial template particles and an elastic polymer to prepare a piezoresistive sheet, and removing the sacrificial template particles from the piezoresistive sheet.

[0010] Furthermore, a side of the spiral interdigitated electrode facing the porous piezoresistive layer is exposed.

[0011] Furthermore, the reduced graphene oxide layer includes a reduced graphene oxide film, a first conductive metal film coated on a first end of the reduced graphene oxide film, and a second conductive metal film coated on a second end of the reduced graphene oxide film.

[0012] The present invention also provides a method for measuring the deformability of an object, comprising:

[0013] Obtaining pressure data and strain data obtained by measuring the deformability of the object using the flexible tactile sensor as described above, and obtaining kinesthetic data of the manipulator;

[0014] Comparing the pressure data with a preset contact pressure to obtain a comparison result, and determining a current state of the manipulator according to the comparison result;

[0015] When the manipulator is in a contact state, the overall stiffness of the object to be measured is obtained according to the kinesthetic data and the pressure data, and the material hardness of the object to be measured is obtained according to the pressure data and the strain data;

[0016] Wherein, the flexible tactile sensor is attached to the surface of the manipulator.

[0017] Furthermore, the kinesthetic data includes displacements of the object to be measured at different time points, and the pressure data includes pressure values ​​at different time points; and obtaining the overall stiffness of the object to be measured based on the kinesthetic data and the pressure data includes:

[0018] With the displacement as the horizontal axis and the pressure value as the vertical axis, the displacement and pressure values ​​corresponding to different time periods are selected to generate a displacement-pressure curve;

[0019] The displacement-pressure curve is linearly fitted to obtain the displacement-pressure curve slope, and the overall stiffness of the object to be measured is obtained according to a preset corresponding relationship between the displacement-pressure curve slope and the overall stiffness.

[0020] Furthermore, the strain data includes strain values ​​at different times; and obtaining the material hardness of the object to be measured based on the pressure data and the strain data includes:

[0021] intercepting the pressure data and the strain data using a time sliding window;

[0022] Calculate the signal characteristics of each sliding time window;

[0023] The signal features of each time sliding window are input into a trained multi-layer perceptron regression model to obtain the material hardness of the object to be tested.

[0024] Furthermore, the calculating of the signal characteristics of each sliding time window includes:

[0025] With the strain value as the horizontal axis and the pressure value as the vertical axis, the strain values ​​and pressure values ​​corresponding to different time periods are selected to generate a strain-pressure curve, and a linear fit is performed on the strain-pressure curve to obtain the slope of the strain-pressure curve;

[0026] The average pressure value is obtained based on each pressure value in the sliding time window;

[0027] The average strain value is obtained based on each strain value within the sliding time window;

[0028] The slope of the strain-pressure curve, the average pressure value, and the average strain value are used as signal features in each sliding time window.

[0029] The present invention also provides a device for measuring the deformability of an object, wherein the device includes: a memory, a processor, and a measurement program for measuring the deformability of the object stored in the memory and executable on the processor, wherein the measurement program implements the steps of the measurement method described above when executed by the processor.

[0030] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program can be executed to implement the steps of the measurement method described above.

[0031] The present invention provides a flexible tactile sensor for a robotic arm to sense the deformability of an object. The sensor comprises a lower flexible layer, a spiral interdigitated electrode, a porous piezoresistive layer, an intermediate flexible layer, and an upper flexible layer, all bonded together in sequence. The upper flexible layer is provided with a reduced graphene oxide layer on the side facing the intermediate flexible layer, and the reduced graphene oxide layer is bonded to the intermediate flexible layer on the side facing away from the upper flexible layer. By placing the porous piezoresistive layer and the reduced graphene oxide layer in the same tactile sensor, the present invention enables simultaneous acquisition of pressure and strain data from an object, thereby enabling simultaneous measurement of the object's material hardness and overall stiffness, enabling relatively accurate measurement of the object's deformability. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] FIG1 is a schematic structural diagram of a preferred embodiment of a flexible tactile sensor for a manipulator to sense the deformability of an object in the present invention;

[0033] FIG2 is a schematic diagram of the insensitivity to planar tensile strain of the porous piezoresistive layer of the present invention;

[0034] FIG3 is a schematic diagram of the structure of a reduced graphene oxide layer in the present invention;

[0035] FIG4 is a schematic diagram of the pressure-insensitive characteristics of the strain sensing composite layer of the present invention;

[0036] FIG5 is a flow chart of a preferred embodiment of a measurement method for measuring deformability of an object according to the present invention;

[0037] FIG6 is a schematic diagram of a quantitative calculation process of material hardness in the present invention;

[0038] Explanation of the accompanying drawings: 10, lower flexible layer; 20, spiral interdigitated electrode; 30, porous piezoresistive layer; 40, intermediate flexible layer; 50, reduced graphene oxide layer; 60, upper flexible layer; 51, reduced graphene oxide film; 52, first conductive metal film; 53, second conductive metal film. DETAILED DESCRIPTION

[0039] The present invention provides a flexible tactile sensor for a robotic arm to sense the deformability of an object. To make the objectives, technical solutions, and effects of the present invention more clear and explicit, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention.

[0040] The technical solution described in this invention is based on the application of robots. Robots are now widely used in everyday life, and their tactile perception has become an important research area. Some current tactile sensors can measure material hardness, such as V-shaped piezoelectric sensors. These sensors use different driving modes of two piezoelectric bimorphs to calibrate hardness values ​​measured at different angles. Essentially, they measure hardness based on the resonant frequency of the piezoelectric sensor. However, in addition to the measurement angle, object size also affects the resonant frequency, thereby affecting the hardness measurement. Therefore, these V-shaped piezoelectric sensors cannot accurately measure material hardness. Furthermore, these V-shaped piezoelectric sensors cannot be well integrated with robotic arms, and mounting them on robotic arms can affect their operation. Researchers in this field have also proposed a method for measuring material hardness by pressing a probe into a specified position and then using an optical fiber sensor to measure the pressure on the probe at that position to infer the material's hardness. This method requires pre-setting the position and cannot achieve adaptive sensing. Furthermore, to ensure that the measurement results are not affected by object size, the probe in this method needs to be sufficiently small, which is not compatible with the size of currently commonly used robotic arms. None of the above technical solutions can achieve simultaneous measurement of the hardness and overall stiffness of an object. Therefore, when the hardness and overall stiffness of two objects are opposite, it is impossible to accurately determine the deformability of the objects.

[0041] The measurement method of the present invention can simultaneously measure an object's material hardness and overall stiffness. It does not require a preset force or position for measuring the material hardness, and is highly adaptable to the size of the object. It can also self-decouple the measurement of pressure and tensile strain within the contact plane at the same contact point. Furthermore, the tactile sensor of the present invention is made of flexible materials, resulting in overall compliance and a certain degree of impact resistance. It can be well adhered to the fingertips of humanoid robots. Its simple structure and manufacturing process effectively reduce the processing cost of the sensor.

[0042] The technical solution of the present invention provides a flexible tactile sensor for a robotic arm to sense the deformability of an object, as shown in Figure 1, which includes: a lower flexible layer 10, a spiral interdigitated electrode 20, a porous piezoresistive layer 30, an intermediate flexible layer 40 and an upper flexible layer 60 that are sequentially bonded and connected, wherein a reduced graphene oxide layer 50 is provided on a side of the upper flexible layer 60 facing the intermediate flexible layer 40, and a side of the reduced graphene oxide layer 50 facing away from the upper flexible layer 60 is bonded and connected to the intermediate flexible layer 40.

[0043] The pressure sensing layer, composed of a lower flexible layer 10, spiral interdigitated electrodes 20, and a porous piezoresistive layer 30, measures the pressure of an object. The strain sensing layer, composed of an intermediate flexible layer 40, a reduced graphene oxide layer 50, and an upper flexible layer 60, measures the strain of an object. By applying an object to the tactile sensor of the present invention, the pressure sensing layer and the strain sensing layer simultaneously measure the object's pressure and strain. Combined with the displacement of the manipulator, the object's material hardness and overall stiffness can be simultaneously determined, enabling relatively accurate measurements of the object's deformability.

[0044] The position of the pure reduced graphene oxide film in the reduced graphene oxide layer 50 determines the plane position of the strain sensing layer for strain sensing. The position of the spiral interdigitated electrodes 20 determines the plane position of the pressure sensing layer for pressure sensing. The plane positions of strain sensing and pressure sensing overlap.

[0045] In one embodiment of the present invention, the porous piezoresistive layer 30 is formed by removing the sacrificial template particles from a piezoresistive sheet material obtained by uniformly mixing a conductive filler, sacrificial template particles, and an elastic polymer.

[0046] The porous piezoresistive layer 30 of the present invention has a porous structure and can adapt to shape. The porous piezoresistive layer 30 is a three-dimensional percolation conductive network that can convert pressure into an electrical signal. As shown in Figure 2, as the pressure increases, the relative conductivity change also increases, that is, the resistance of the porous piezoresistive layer 30 becomes smaller and smaller, showing a negative piezoresistive effect. The porous piezoresistive layer 30 has the mechanical properties of low modulus (a few tens of times the modulus of the flexible layer) and low Poisson's ratio (about 0). During compression, no lateral strain is generated to increase the resistance of the porous piezoresistive layer 30 and interfere with pressure sensing, so that the pressure combination sensing layer appears to be plane strain insensitive, and the pressure value can be measured more accurately.

[0047] In a preferred embodiment of the present invention, the side of the spiral interdigitated electrode 20 facing the porous piezoresistive layer 30 is exposed.

[0048] Specifically, the spiral interdigitated electrode 20 is a filament-like electrode with a spiral shape, which is used to derive the resistance change of the porous piezoresistive layer 30. The spiral interdigitated electrode 20 is a hollow structure to ensure good contact between the electrode and the porous piezoresistive layer when the sensor is assembled. At the same time, the side of the spiral interdigitated electrode 20 facing the porous piezoresistive layer 30 is exposed and directly contacts the porous piezoresistive layer 30, so that the spiral interdigitated electrode 20 and the porous piezoresistive layer 30 are in closer contact, the signal transmission is more efficient, and the problems of contact resistance and poor contact can be reduced, which helps to improve the stability and reliability of the signal, reduce the measurement error caused by contact problems, and save the space of the sensor.

[0049] In a preferred embodiment of the present invention, as shown in FIG3 , the reduced graphene oxide layer 50 includes a reduced graphene oxide film 51 , a first conductive metal film 52 coated on a first end of the reduced graphene oxide film, and a second conductive metal film 53 coated on a second end of the reduced graphene oxide film.

[0050] Specifically, the first conductive metal film 52 is a silver nanowire (AgNWs) electrode, and the second conductive metal film 53 is a silver nanowire (AgNWs) electrode. The reduced graphene oxide layer 50 in the present invention is achieved by reducing the cracks in the graphene oxide film to achieve high-sensitivity strain sensing. Strain can cause the separation and expansion of the crack edges in the reduced graphene oxide film (rGO film), thereby increasing the resistance. Normal pressure can cause normal contact between the graphene oxide sheets, thereby reducing the resistance. However, the resistance increase of the former is much greater than the resistance decrease of the latter. As shown in Figure 4, in the strain sensing composite layer, as the pressure increases, the relative resistance changes relatively steadily, without obvious changes, so the strain sensing composite layer appears to be pressure-insensitive and can measure strain values ​​more accurately.

[0051] In one embodiment of the present invention, the reduced graphene oxide layer 50 is semi-embedded in the upper flexible layer 60 .

[0052] In a preferred embodiment of the present invention, the method for preparing the reduced graphene oxide film comprises:

[0053] The graphene oxide gel is drop-coated onto a rigid substrate to obtain a graphene film;

[0054] oxidizing the graphene film by laser direct writing to obtain an initial reduced graphene oxide film;

[0055] Silver nanowires are drop-coated on both ends of the initial reduced graphene oxide film to obtain a reduced graphene oxide film.

[0056] Specifically, the initial reduced graphene oxide film is a conductive reduced graphene oxide film having a preset pattern.

[0057] The present invention also provides a measurement method for measuring the deformability of an object, as shown in FIG5 . The steps of implementing the measurement method are as follows:

[0058] S100, obtaining pressure data and strain data obtained by measuring the deformability of an object using the flexible tactile sensor as described above, and obtaining kinesthetic data of the manipulator;

[0059] Specifically, the present invention utilizes the flexible tactile sensor as described above to measure the deformability of an object, applies the object to be measured to the flexible tactile sensor, and simultaneously measures the material hardness and overall stiffness of the object to be measured. When the object to be measured is applied to the flexible tactile sensor, the signal conditioning and data acquisition module controls the tactile sensor to collect data, obtains pressure data and strain data, and transmits the pressure data and strain data to a computer terminal. The signal conditioning and data acquisition module includes a signal acquisition submodule and a data acquisition submodule. The signal conditioning submodule is used to convert the resistance signal of the tactile sensor into a voltage signal, and the data acquisition module is used to provide voltage to the tactile sensor and transmit the pressure data and strain data of the tactile sensor to the computer terminal. The computer terminal sends a control signal to the manipulator controller, and the manipulator controller controls the manipulator to move. The manipulator is provided with an encoder that can calculate the displacement of the output end, that is, obtain the displacement information of the manipulator. The manipulator controller obtains kinesthetic data based on the displacement information and transmits the kinesthetic data to the computer terminal. The computer terminal simultaneously obtains the pressure data, strain data of the flexible tactile sensor and the kinesthetic data of the manipulator for subsequent judgment of the contact state and calculation of the overall stiffness and material hardness.

[0060] S200, comparing the pressure data with a preset contact pressure to obtain a comparison result, and determining a current state of the manipulator according to the comparison result;

[0061] Specifically, the computer terminal compares the pressure data with a preset contact pressure to obtain a comparison result. If the comparison result shows that the pressure data is greater than the preset contact pressure, the manipulator is confirmed to be in a contact state; if the comparison result shows that the pressure data is less than the preset contact pressure, the manipulator is confirmed to be in a non-contact state.

[0062] S300, when the manipulator is in a contact state, obtaining the overall stiffness of the object to be measured according to the kinesthetic data and the pressure data, and obtaining the material hardness of the object to be measured according to the pressure data and the strain data;

[0063] Wherein, the flexible tactile sensor is attached to the surface of the manipulator.

[0064] Specifically, the computer terminal can calculate the overall stiffness of the object to be measured based on the acquired kinesthetic data and pressure data, and simultaneously obtain the material hardness of the object to be measured based on the acquired pressure data and strain data.

[0065] In one implementation, after step S200, the method further includes:

[0066] When the manipulator is in a non-contact state, the pressure data, the strain data, and the kinesthetic data are not processed.

[0067] In one implementation, the upper flexible layer of the flexible tactile sensor is attached to the surface of the manipulator.

[0068] In one implementation, the kinesthetic data includes displacements of the object to be measured at different time points, and the pressure data includes pressure values ​​at different time points; and obtaining the overall stiffness of the object to be measured based on the kinesthetic data and the pressure data includes:

[0069] With the displacement as the horizontal axis and the pressure value as the vertical axis, the displacement and pressure values ​​corresponding to different time periods are selected to generate a displacement-pressure curve;

[0070] The displacement-pressure curve is linearly fitted to obtain the displacement-pressure curve slope, and the overall stiffness of the object to be measured is obtained according to a preset corresponding relationship between the displacement-pressure curve slope and the overall stiffness.

[0071] Specifically, the smaller the slope of the displacement-pressure curve is, the greater the overall stiffness of the object to be measured is.

[0072] In one implementation, the strain data includes strain values ​​at different times; and obtaining the material hardness of the object to be measured based on the pressure data and the strain data includes:

[0073] intercepting the pressure data and the strain data using a time sliding window;

[0074] Calculate the signal characteristics of each sliding time window;

[0075] The signal features of each time sliding window are input into a trained multi-layer perceptron regression model to obtain the material hardness of the object to be tested.

[0076] Specifically, the present invention utilizes machine learning to map tactile sensor signals to corresponding material hardness. The present invention utilizes the multilayer perceptron regression model (MLP) to quantitatively calculate material hardness. The pressure and strain data are captured using a sliding time window. Signal features within each sliding time window are then calculated as input to the MLP model, which then outputs the corresponding material hardness value, achieving a quantitative estimate of material hardness.

[0077] In one implementation, before using the flexible tactile sensor, the multilayer perceptron regression model needs to be trained to obtain a trained multilayer perceptron regression model, which can be regarded as calibrating the material hardness. The steps of training the multilayer perceptron regression model include:

[0078] Preparation of cylindrical samples with different hardness;

[0079] Using a hardness tester to measure the material hardness of the cylindrical sample to obtain a first material hardness;

[0080] A manipulator equipped with sensors is used to press cylindrical samples at different indentation displacements, and the pressure and strain signals of each press are collected.

[0081] Alternatively, a robot equipped with a sensor can be used to press a cylindrical sample with different applied forces, and the pressure and strain signals of each press can be collected.

[0082] The pressure signal and strain signal collected each time are extracted using a sliding time window;

[0083] Calculate the characteristic value of the signal extracted by the sliding time window to obtain the signal characteristic value;

[0084] The signal characteristic value is used as input data of a multilayer perceptron regression model, and the hardness of the first material is used as a label to train the multilayer perceptron regression model until a stopping condition is met, thereby obtaining a trained multilayer perceptron regression model.

[0085] In one implementation, calculating the signal characteristics of each sliding time window includes:

[0086] With the strain value as the horizontal axis and the pressure value as the vertical axis, the strain values ​​and pressure values ​​corresponding to different time periods are selected to generate a strain-pressure curve, and a linear fit is performed on the strain-pressure curve to obtain the slope of the strain-pressure curve;

[0087] The average pressure value is obtained based on each pressure value in the sliding time window;

[0088] The average strain value is obtained based on each strain value within the sliding time window;

[0089] The slope of the strain-pressure curve, the average pressure value, and the average strain value are used as signal features in each sliding time window.

[0090] Specifically, as shown in FIG6 , a sliding time window is used for signal extraction, and each pressure value P in the sliding time window is i Get the average pressure P avg , and according to each strain value S in the sliding time window i Get the average strain S avg The slope of the strain-pressure curve Slope and the average pressure P avg And the average strain S avgThe signal characteristics of each sliding time window are used. The smaller the slope of the strain-pressure curve, the harder the material hardness of the object to be measured. The signal characteristics of each sliding time window are input into a trained multi-layer perceptron regression model to obtain the material hardness of the object to be measured. The present invention can achieve quantitative calculation of material hardness by using a multi-layer perceptron regression model, reduce the interference of object size on measurement results, realize adaptive perception of object deformability, and thus achieve accurate measurement of object deformability.

[0091] In one implementation, the step S300 further includes: adjusting the contact pressure according to the material hardness and the overall stiffness.

[0092] Specifically, the present invention performs sensing feedback operations through a flexible tactile sensor, a manipulator and its controller, a signal conditioning and data acquisition module, and a computer terminal. The computer terminal receives in real time the relative change value of the resistance of the pressure sensing combination layer in the tactile sensor transmitted by the data acquisition module. When the relative change value of the resistance of the pressure sensing combination layer exceeds a certain threshold, the computer terminal determines that contact has occurred. At this time, the computer terminal transmits a control signal to the manipulator controller to control the manipulator to press into an arbitrary displacement, which is used to sense the deformable dual properties of the object (i.e., material hardness and overall stiffness). The computer terminal calculates the material hardness and overall stiffness based on the pressure data, strain data of the flexible tactile sensor and the displacement of the manipulator, plans the final appropriate contact pressure with the object, and automatically adjusts to achieve the target contact pressure. The present invention combines the kinesthetic signal of the manipulator and the dual-modal signal (i.e., pressure signal and strain signal) of the flexible tactile sensor to simultaneously measure the material hardness and overall stiffness of the object, which can achieve accurate measurement of the variability of the object.

[0093] The present invention also provides a device for measuring the deformability of an object, wherein the device includes: a memory, a processor, and a measurement program for measuring the deformability of an object stored in the memory and executable on the processor, wherein the measurement program implements the steps of the measurement method described above when executed by the processor; specifically as described above.

[0094] The present invention also provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program can be executed to implement the steps of the measurement method described above; specifically as described above.

[0095] In summary, the present invention provides a flexible tactile sensor for a robotic arm to sense the deformability of an object, comprising: a lower flexible layer, a spiral interdigitated electrode, a porous piezoresistive layer, an intermediate flexible layer, and an upper flexible layer, all bonded together in sequence. A reduced graphene oxide layer is provided on the side of the upper flexible layer facing the intermediate flexible layer, and the reduced graphene oxide layer is bonded together with the intermediate flexible layer on the side facing away from the upper flexible layer. By placing the porous piezoresistive layer and the reduced graphene oxide layer in the same tactile sensor, the present invention enables simultaneous acquisition of pressure and strain data from an object, thereby supporting the simultaneous measurement of the object's material hardness and overall stiffness, enabling relatively accurate measurement of the object's deformability.

[0096] It should be understood that the application of the present invention is not limited to the above examples. For those skilled in the art, improvements or changes can be made based on the above description. All these improvements and changes should fall within the scope of protection of the claims attached to the present invention.

Claims

1. A flexible tactile sensor for a manipulator to sense the deformability of an object, characterized in that: include: A lower flexible layer, a spiral interdigitated electrode, a porous piezoresistive layer, an intermediate flexible layer and an upper flexible layer are sequentially bonded and connected, a reduced graphene oxide layer is provided on a side of the upper flexible layer facing the intermediate flexible layer, and a side of the reduced graphene oxide layer facing away from the upper flexible layer is bonded and connected to the intermediate flexible layer.

2. The flexible tactile sensor for sensing the deformability of an object by a manipulator according to claim 1, characterized in that: The porous piezoresistive layer is formed by uniformly mixing conductive fillers, sacrificial template particles and elastic polymer to obtain a piezoresistive sheet, and removing the sacrificial template particles in the piezoresistive sheet.

3. The flexible tactile sensor for a manipulator to sense the deformability of an object according to claim 1, characterized in that: The side of the spiral interdigitated electrode facing the porous piezoresistive layer is in a bare state.

4. The flexible tactile sensor for sensing the deformability of an object by a manipulator according to claim 1, characterized in that: The reduced graphene oxide layer includes a reduced graphene oxide film, a first conductive metal film coated on a first end of the reduced graphene oxide film, and a second conductive metal film coated on a second end of the reduced graphene oxide film.

5. A method for measuring the deformability of an object, characterized in that: include: Acquire pressure data and strain data obtained by measuring the deformability of an object using the flexible tactile sensor as described in any one of claims 1 to 4, and acquire kinesthetic data of the manipulator; Comparing the pressure data with a preset contact pressure to obtain a comparison result, and determining a current state of the manipulator according to the comparison result; When the manipulator is in contact state, the overall stiffness of the object to be measured is obtained according to the kinesthetic data and the pressure data, and the material hardness of the object to be measured is obtained according to the pressure data and the strain data; Wherein, the flexible tactile sensor is attached to the surface of the manipulator.

6. The measuring method according to claim 5, characterized in that: The kinesthetic data includes displacements of the object to be measured at different time points, and the pressure data includes pressure values ​​at different time points; and obtaining the overall stiffness of the object to be measured according to the kinesthetic data and the pressure data includes: Taking the displacement as the horizontal axis and the pressure value as the vertical axis, selecting the displacement and pressure values ​​corresponding to different time periods to generate a displacement-pressure curve; The displacement-pressure is linearly fitted to obtain the displacement-pressure curve slope. The corresponding relationship between the slope of the curve and the overall stiffness can be used to obtain the overall stiffness of the object to be tested.

7. The measuring method according to claim 6, characterized in that: The strain data includes strain values ​​at different times; and obtaining the material hardness of the object to be measured according to the pressure data and the strain data includes: intercepting the pressure data and the strain data using a time sliding window; Calculate the signal features for each sliding time window; The signal features of each time sliding window are input into a trained multi-layer perceptron regression model to obtain the material hardness of the object to be tested.

8. The measuring method according to claim 7, characterized in that: The calculating of the signal characteristics of each sliding time window comprises: Taking the strain value as the horizontal axis and the pressure value as the vertical axis, selecting the strain values ​​and pressure values ​​corresponding to different time periods, generating a strain-pressure curve, and performing linear fitting on the strain-pressure curve to obtain the slope of the strain-pressure curve; The pressure average value is obtained based on each pressure value in the sliding time window; The strain average value is obtained according to each strain value in the sliding time window; The slope of the strain-pressure curve, the average pressure value, and the average strain value are used as signal features in each sliding time window.

9. A device for measuring the deformability of an object, characterized in that: The device comprises: a memory, a processor, and a measurement program for measuring the deformability of an object which is stored in the memory and can be run on the processor. When the measurement program is executed by the processor, the steps of the measurement method according to any one of claims 5 to 8 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, which can be executed to implement the steps of the measurement method according to any one of claims 5 to 8.

Images