Self-sensing software machine with large-area sensing function

Abstract

The invention relates to a self-sensing soft machine with a large-area sensing function, which can detect the state of the soft machine in a deformation process and can detect a loading signal from an external environment in an area with any shape and a larger range (not less than 100cm < 2 >). And according to the detection result of the global information, the parameters of the motion control module of the software machine are adjusted, so that the self-sensing software machine with a large-area sensing function is realized. The detection electrodes are only arranged on the edge of the sensing area, the interior of the sensing area is continuous, no blind area exists, and the large-area continuous sensing function can be achieved; the shape of the sensing area and the position of the boundary electrode are not limited, and the requirement for the to-be-detected area in an irregular shape can be met. Different from a conventional self-sensing software machine relying on a single and isolated sensing unit, the integrated full-flexible integration of a software driving body and large-area sensing is realized, and the problems of acquiring the whole body deformation state and sensing global environment information of the software machine are solved.

Description

Technical Field

[0001] The invention belongs to the field of flexible sensing and driving technology, and specifically is a self-sensing software machine with large-area sensing function. Background Art

[0002] Soft machines, with their soft, flexible and deformable characteristics, can perform tasks in environments where traditional rigid machines are difficult to operate, and show broad application prospects and development potential in fields such as medical surgery, rescue operations and human-computer interaction. However, the main structure of soft machines is mostly composed of hyperelastic materials, which theoretically have infinite degrees of freedom of deformation, making traditional rigid theory unsuitable for the kinematic and dynamic analysis of soft machines. Therefore, giving soft machines flexible sensing functions and realizing their own state estimation and environmental perception in non-structural environments is the key to accurately controlling the movement of soft machines.

[0003] At present, self-sensing soft machines are limited by the flexibility of the sensing material and the range of the detectable sensing area. Patent application CN114670214A implants strain gauges and metal sheets in a pneumatic soft robot to measure the curvature of the robot. Although this solution can measure changes in the length direction of the soft robot, the stiffness of the sensor changes the motion characteristics of the soft machine itself, resulting in the detected signal being unable to truly reflect the deformation characteristics when the sensor is not embedded. In order to improve the flexibility of the sensor, flexible sensing materials such as flexible triboelectric materials are used to greatly improve the flexibility of the sensing components of the self-sensing soft machine. For example, patent application CN109278050A embeds a flexible triboelectric sensor in the soft machine, allowing the soft machine to sense touch or pressure signals.

[0004] In terms of increasing the detection range of the sensing area, the existing technology includes splicing multiple identical single sensing units to expand the detection range. For example, patent application CN114378805B proposes to fix the same flexible tensile strain sensor on the limbs of a quadruped crawling pneumatic soft robot to detect the bending state of each foot separately. This solution increases the detection range by using multiple flexible sensors, but the method of obtaining global information by splicing is affected by the consistency of the performance of each sensor, which is not conducive to accurately interpreting deformation signals at different positions.

[0005] In summary, currently, the self-sensing soft machines only detect the bending angle and local single tactile load, and cannot detect the deformation state of the whole body in a large range like human skin, or perceive the stimulation of the external environment in a non-structural environment. Therefore, it is necessary to develop a self-sensing soft machine with large-area sensing function, which is of great value to improve the accuracy of soft machine motion control. Summary of the invention

[0006] In view of the above problems, the purpose of the present invention is to realize a self-sensing software machine with large-area sensing function. The device can realize the sensing function of deformation state and environment in a large range.

[0007] The device has a large-area sensing function, can detect the state of the soft machine during deformation, can detect the loading signal from the external environment, and adjust the parameters of the soft machine motion control module according to the detection results of the large-area sensing. The present invention is implemented based on the following technical solutions.

[0008] A self-sensing software machine with large-area sensing function, comprising a flexible sensing and driving module, a signal detection module, and a closed-loop control module, characterized in that:

[0009] The flexible sensing and driving module integrates flexible sensing function and software driving function, including:

[0010] The elastically deformable main body is made of a superelastic material, the elastic modulus is lower than 1Mpa, the elastically deformable main body is bendable and stretchable, and the maximum stretching amount is greater than 100%;

[0011] Large-area sensing part, made of strain-sensitive elastic material, can cover a large range, and the detectable area is not less than 100cm 2 The shape of the sensing area is not restricted, and the sensing signal is obtained from the detection electrodes arranged at the boundary of the sensing area; the large-area sensing part can bend and stretch with the elastic deformable main part without changing the movement and dynamic characteristics of the deformable main part, and does not restrict the deformation movement of the soft machine, and can truly reflect the deformation state of the soft machine;

[0012] Preferably, the signal detection module comprises a boundary current driving unit and a boundary signal collecting unit, characterized in that:

[0013] The boundary current driving unit is capable of inputting a driving current to the boundary electrode of the large-area sensing unit;

[0014] The boundary signal acquisition unit can acquire voltage signals on boundary electrodes of the large-area sensing unit.

[0015] Preferably, the closed-loop control module includes a regional information reconstruction unit and a feedback control unit, characterized in that:

[0016] The regional information reconstruction unit can obtain deformation information in the sensing area according to the voltage signal of the signal acquisition module through the regional information reconstruction process, reflecting the state of the software machine itself or the state of the environment in which it is located;

[0017] The feedback control unit can determine the deformation state or environmental state of the software device based on the reconstructed image of the region information reconstruction unit, and output a control signal of the software device.

[0018] Preferably, the region information reconstruction process includes the following steps:

[0019] Step 1: Establish a sensing model according to the shape of the sensing area in the large-area sensing part and determine the boundary of the sensing area; the sensing model is characterized in that the interior of the sensing area is continuous, and the shape of the sensing area is not restricted and can be regular or irregular; arrange detection electrodes at the boundary of the sensing model and determine the coordinates of each boundary electrode.

[0020] Step 2: Estimate the electrical property distribution s(x,y) in the sensing area based on the strain-sensitive functional materials constituting the sensing area and their preparation methods. For the uniform deposition preparation method, the electrical property distribution of the strain-sensitive functional materials in the sensing area is uniform, and the electrical parameter distribution of the surface conductivity and surface resistivity can be determined by the four-point method.

[0021] Step 3: Solve the distribution of theoretical boundary voltage. According to the estimated electrical property distribution, calculate the theoretical potential distribution of the sensing area, and after selecting the position of the grounding signal, obtain the theoretical boundary voltage distribution F(s) at the boundary electrode; further, according to the electrical property characteristics of the strain-sensitive functional material, determine the electrical performance parameters before and after deformation, so as to determine the electrical property distribution of the sensing area of ​​the soft machine in the previous state and the current state, which are recorded as s1 and s2 respectively; further, calculate the boundary voltage signals in the previous state and the current state, and obtain the difference between the theoretical boundary voltage signals in the two states, which is recorded as ΔF=F(s2)-F(s1).

[0022] Step 4: Test the boundary voltage signal of the large-area sensor of the software machine in the previous state and the current state through the signal detection module. The boundary voltage signal in the previous state is recorded as V1, and the boundary voltage signal in the current state is recorded as V2. The difference between the boundary voltage signals in the previous state and the current state is obtained, which is recorded as ΔV=V2-V1.

[0023] Step 5. Determine the difference between the test signal and the theoretical signal. According to the boundary signal obtained by the test and the boundary signal obtained by theoretical calculation, calculate the absolute value of the difference between the two (|ΔV-ΔF|), and judge whether the estimated electrical performance distribution meets the actual situation. Set the error amount 1% x|ΔF| as the judgment threshold. If |ΔV-ΔF| is lower than the threshold, the change in the estimated electrical performance distribution can reflect the electrical parameter distribution of the sensing area in the previous state and the current state; if |ΔV-ΔF| is not lower than the threshold, the estimated electrical parameter distribution does not meet the requirements. Repeat step 3, update the electrical performance distribution s1 and s2, and recalculate the difference ΔF of the theoretical boundary voltage signal until |ΔV-ΔF| meets the limit of the set threshold.

[0024] Step 6: Mapping imaging: According to the model of the sensing area, the distribution of the electrical property changes is mapped to the sensing area to obtain a reconstructed image of the sensing area.

[0025] Preferably, the strain-sensitive functional material of the large-area sensing part has a low elastic modulus, and its resistance or capacitance parameters change with strain. The embedded position of the large-area sensing part is located in the stretchable deformable layer of the elastically deformable main body of the software machine, which neither affects the deformation of the main body nor can deform with the main body or withstand external loading.

[0026] Preferably, the regional information reconstruction unit is capable of obtaining an image of the electrical signal changes reflecting the strain distribution in the sensing area, characterized in that global information detection of the deformation state of the software machine and the external environment loading is achieved based only on the electrical signals at the edge of the sensing area, and the sensing area has various shapes to meet the needs of global self-perception of various types of software machines.

[0027] Preferably, the detection wire is made of carbon-based fiber with a resistivity lower than 0.001Ω / cm, the carbon-based fiber is bendable and non-stretchable, the conductivity does not change when bent, and does not interfere with the signal of the sensing area. The detection wire is connected to the sensing area through the same strain-sensitive flexible sensing material.

[0028] Preferably, the boundary electrodes are located at the edge of the sensing area, the number and position of the electrodes are adjustable, and the boundary electrodes are used for both signal detection and current driving.

[0029] Compared with the prior art, the present invention has the following beneficial effects:

[0030] (1) The present invention realizes information perception in irregular and large-area sensing areas, and only uses electrodes arranged at the boundaries of the sensing area for signal detection, which greatly reduces the complexity of the wires and ensures the detection continuity within the sensing area. It breaks through the traditional method of splicing multiple isolated sensing units to detect information in a larger range and avoids the problem of blind spots in array sensing. The above-mentioned large-area sensing method not only realizes the ability of the software machine to perceive the deformation state of the whole body and the global environmental state, but also facilitates the simplicity of the self-perception software machine system integration;

[0031] (2) The present invention establishes a corresponding large-area sensing model according to different shapes of the soft machine, detects the boundary voltage signals of the previous state and the current state, iteratively reduces the difference between the test signals of the current state and the previous state and the theoretical calculation, and finally obtains an image reflecting the change of electrical properties of the soft machine in the deformation state or environmental loading state. The image can not only quantitatively reflect the deformation state, but also obtain the loading information in the large-area sensing area of ​​the soft machine;

[0032] (3) The regional information reconstruction method of the present invention uses the difference between the boundary voltage signal of the current state and the previous state to predict the deformation movement or external load of the current state of the soft machine. The image reconstruction method focuses on the change amount between the two states, is not affected by the model error, and has the advantages of high precision and high reliability.

[0033] (4) The large-area sensing method of the present invention can realize information detection in a continuous sensing area of ​​no less than 100 cm2, and can meet the large-area self-sensing needs of soft machines of various shapes such as single-joint, multi-joint, quadruped, and bionic animals. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a module schematic diagram of a self-sensing software machine with large-area sensing function of the present invention. Explanation of the numbers in the figure: 1. Flexible sensing and driving module 1; 2. Signal detection module; 3. Closed-loop control module; 4. Large-area sensing unit; 5. Elastic deformable body unit; 6. Boundary current driving unit; 7. Boundary signal acquisition unit; 8. Regional information reconstruction unit; 9. Feedback control unit.

[0035] Figure 2 The structure diagram of the flexible sensing and driving module with large-area sensing function of the present invention is shown in the figure. The numbers in the figure are as follows: 10, large-area sensing skin; 11, boundary electrode; 12, elastic deformable cavity; 13, bending deformation layer; 14, stretchable deformation layer.

[0036] Figure 3 The invention discloses a method for reconstructing regional information of a self-sensing software machine with a large-area sensing function.

[0037] Figure 4 It is a schematic diagram of the perception of the self-sensing software machine with large-area sensing function in the embodiment of the present invention.

[0038] Figure 5 It is a schematic diagram of the perception of the external environment by a self-sensing software machine with large-area sensing function in an embodiment of the present invention. DETAILED DESCRIPTION

[0039] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments.

[0040] Figure 1It is a module schematic diagram of a self-sensing software machine with a large-area sensing function. The self-sensing software machine with a large-area sensing function includes a flexible sensing and driving module 1, a signal detection module 2, and a closed-loop control module 3. Among them, the flexible sensing and driving module 1 includes a large-area sensing part 4 and an elastic deformable body part 5. The large-area sensing part 4 collects the self-state information and environmental information of the software machine; the elastic deformable body part 5 can realize the elastic deformation of its own structure and the movement of the software through external drive, and the external drive can be in the form of fluid drive, voltage drive, electrochemical drive, etc. The signal detection module 2 includes a boundary current drive part 6 and a boundary signal acquisition part 7. The boundary current drive part 6 applies an excitation current to the large-area sensing part 4, and the magnitude of the current is determined according to the range of the signal acquisition device and the range of change of the electrical signal of the large-area sensor; the boundary signal acquisition part 7 is responsible for collecting the corresponding boundary signals, and this information provides basic data for subsequent signal processing and control. The closed-loop control module 3 includes a regional information reconstruction part 8 and a feedback control part 9. The regional information reconstruction unit 8 obtains the reconstruction information of the sensing area of ​​the software machine through calculation based on the boundary signal collected by the signal detection module 2; the feedback control unit 9 adjusts the driving state of the elastically deformable main body 5 in real time according to the reconstructed regional information, thereby realizing the control of the self-sensing software machine with large-area sensing function. The advantages of the present invention are: first, through boundary signal detection and regional information reconstruction, the large-area self-sensing function of the software machine to the external environment and its own state is realized, which solves the problem that the current self-sensing software machine cannot realize the information perception of large areas and continuous areas; second, only the detection electrodes arranged at the boundary of the sensing area are used, which greatly reduces the difficulty of self-sensing software machine system integration; third, the large-area self-sensing function is combined with the closed-loop control module, and the whole system has modular design characteristics, which can adapt to the collaborative self-sensing and control of multiple software machines.

[0041] Figure 21 is a schematic diagram of the structure of the flexible sensing and driving module with large-area sensing function of the present invention. Taking a self-sensing soft machine with large-area sensing function in the form of a quadruped pneumatic as an example, it includes a large-area sensing skin 10, a boundary electrode 11, an elastic deformable cavity 12, a bending deformation layer 13 and a stretchable deformable layer 14. The elastic deformable cavity 12 can produce expansion deformation in the inflated state, and the deformation of the stretchable deformable layer 14 is greater than the deformation of the bending deformation layer 13. The large-area sensing skin 10 is located in the stretchable deformable layer 14, which is used to sense external pressure or deformation information and transmit the signal to the closed-loop control module. This embedded position can obtain strain sensitive signals with high signal-to-noise ratio and improve the sensing performance; at the same time, the stretchable deformable layer 14 covers the large-area sensing skin 10 to form a comprehensive package, which can prevent the flexible sensing and driving module from peeling off when inflated and deformed. The boundary electrode 11 is set at the boundary of the sensing area of ​​the soft machine, which is used to collect the boundary electrical signal of the large-area sensing skin 10 and transmit it to the closed-loop control module. The elastic deformable cavity 12 serves as the core driving component of the soft machine, and realizes various deformations and movements of the soft machine by driving changes in air pressure. The bending deformation layer 13 constrains the deformation mode of the soft machine by limiting the output force or displacement of the soft machine, and provides additional structural strength and external protection. The soft machine combines the elastic deformation principle and the signal acquisition function of the large-area sensing skin to achieve the adjustment of the motion state of the soft machine and the perception of the external environment. Preferably, the four feet and the central area of ​​the quadruped soft machine are independently air-pressure controlled, which can achieve a variety of different deformation postures, and is more conducive to demonstrating the large-area sensing function of the present invention.

[0042] It should be noted that the large-area sensing skin in the above-mentioned embodiment is only a preferred form of the four-legged pneumatic soft machine. For different forms of soft machines, sensing skins of corresponding shapes can be selected to form a large-area sensing part of a self-sensing soft machine with large-area sensing function.

[0043] Figure 3 The invention is a method for reconstructing regional information of a self-sensing software machine with large-area sensing function. The method mainly includes sensing area modeling, electrical performance distribution estimation, theoretical boundary voltage calculation, mapping imaging, etc. The processing process connects the signal detection module and the closed-loop control module, and can perform on-chip real-time information processing, which is conducive to realizing closed-loop feedback control of integrated sensing and driving.

[0044] It should be noted that the amount of voltage data obtained from the signal detection module is related to the driving mode of the current driving unit and the number of boundary electrodes. When the number of boundary electrodes N≤128, the quality of regional information reconstruction is positively correlated with the number of boundary electrodes. When the number of boundary electrodes N>128, the reconstruction quality of regional information cannot be improved by increasing the number of electrodes. Instead, it will cause a huge amount of data, greatly reduce the speed of signal processing, and affect the real-time performance of the closed-loop control module.

[0045] The following is an explanation of the implementation methods of the main steps:

[0046] Step 1: Establish a model of the sensing area. According to the shape of the sensing area, determine the boundary of the sensing area in a two-dimensional coordinate system; arrange detection electrodes at the boundary of the sensing model, and determine the coordinates of each boundary electrode.

[0047] The characteristic of this sensing model is that the interior of the sensing area is continuous, and the shape of the sensing area is not restricted and can be regular or irregular. It can reflect the actual shape of the large-area sensing part and meet the large-area perception needs of software machines of various shapes.

[0048] Step 2: Estimate the distribution of electrical properties. According to the strain-sensitive functional materials constituting the sensing area and their preparation methods, estimate the distribution of electrical properties in the sensing area, denoted as s(x, y); wherein x and y are the coordinate values ​​of each point in the sensing area. The distribution of electrical properties is related to the formula and preparation process of the flexible sensing material used, and can be uniform or unevenly distributed. Preferably, for the preparation method of uniform deposition, the method for estimating electrical properties is that the electrical properties of the strain-sensitive functional materials in the sensing area are uniformly distributed, and the four-point method is used to measure the surface conductivity and surface resistivity, thereby obtaining the uniformly distributed electrical parameters of the sensing area.

[0049] Step three, solve the theoretical boundary voltage distribution. According to the estimated electrical property distribution, calculate the theoretical potential distribution of the sensing area. The potential distribution includes the potential values ​​of each position inside the sensing area and the potential values ​​of each position on the boundary of the sensing area. After the grounding position is selected, the theoretical boundary voltage distribution F(s) at the boundary electrode is obtained; wherein F(·) is the forward operator for calculating the theoretical voltage distribution according to the electrical property distribution s. Further, according to the electrical property characteristics of the strain-sensitive functional material, the electrical performance parameters before and after deformation are determined, thereby determining the electrical performance distribution of the sensing area of ​​the software machine in the previous state and the current state, which are recorded as s1 and s2, respectively. Further, the boundary voltage signals in the previous state and the current state are calculated to obtain the difference between the theoretical boundary voltage signals in the two states, recorded as ΔF=F(s2)-F(s1).

[0050] Step 4: Test the boundary voltage signals of the previous state and the current state. Through the signal detection module, test the boundary voltage signals of the large-area sensor of the software machine in the previous state and the current state. The boundary voltage signal in the previous state is recorded as V1, and the boundary voltage signal in the current state is recorded as V2.

[0051] Furthermore, the difference between the boundary voltage signal in the previous state and the current state is calculated and recorded as ΔV=V2-V1, and the difference is the test signal.

[0052] Step 5: Determine the difference between the test signal and the theoretical signal. Based on the boundary signal obtained by the test and the boundary signal obtained by theoretical calculation, calculate the absolute value of the difference between the two (|ΔV-ΔF|). The specific calculation method is:

[0053]

[0054] Where M is the number of the boundary data, i is an integer ranging from 1 to M, V1 and V2 are the test values ​​of the boundary signals of the previous state and the current state respectively, and F(·) is a forward operator for calculating the boundary voltage according to the electrical performance distribution.

[0055] According to the calculation results, it is judged whether the estimated electrical performance distribution meets the actual situation, and the error amount 1% x|ΔF| is set as the judgment threshold.

[0056] If |ΔV-ΔF|<threshold, the estimated change in the electrical property distribution satisfies the change in the electrical property distribution of the actual sensing area, that is, s2-s1.

[0057] If |ΔV-ΔF|≥threshold, the change in the estimated electrical performance distribution cannot meet the change in the electrical performance distribution in the actual sensing area, and step three needs to be repeated to update the electrical performance distributions s1 and s2, recalculate the difference ΔF of the theoretical boundary voltage signal, and again determine the relationship between |ΔV-ΔF| and the threshold until the set threshold limit is met.

[0058] It should be noted that the size of the threshold affects the accuracy of the reconstructed information; however, if the threshold is too small, the regional information reconstruction process will not converge. Therefore, the threshold should be selected reasonably to balance the accuracy and accessibility of the data.

[0059] Step 6: Mapping imaging: After determining the distribution of the electrical property changes in the sensing area in the previous state and the current state, the distribution of the electrical property changes is mapped to the sensing area according to the model of the sensing area to obtain a reconstructed image of the area information.

[0060] Figure 4It is a schematic diagram of the perception of the self-sensing software machine with large-area sensing function in the embodiment of the present invention. On the basis of the previous state, the software machine injects different amounts of gas into the two left feet and the middle area, causing the soft machine to bend and deform. If the inflation amount of foot 1 is less than the inflation amount of foot 2, the bending angle of foot 1 is less than the bending angle of foot 2, that is, θ1<θ2. The surface strain value of the low modulus layer of foot 1 is less than the surface strain of the low modulus layer of foot 2. The boundary signals of the previous state and the current state after bending deformation are respectively collected by the signal acquisition module, and after regional information reconstruction, an image reflecting the changes in electrical properties in the sensing area of ​​the bending deformation state is obtained. This image corresponds to the strain distribution of the soft machine in the deformed state, and can judge the self-state of the soft machine.

[0061] It should be noted that the software machine in the previous state may be in an uninflated state or a partially inflated state. Preferably, in the embodiment, the inflation states of the various parts of the software machine are independent.

[0062] Figure 5 This is a schematic diagram of the perception of the external environment by a self-sensing software machine with a large-area sensing function in an embodiment of the present invention. When a certain deformation state of the software machine is selected as the previous state, the external environment applies loading stimuli at two different positions to the software machine, causing corresponding strains on the surface of the software machine, causing a signal response of the strain-sensitive large-area sensing skin. The signal detection module collects the boundary signals of the previous state and the environment after loading, and reconstructs the regional information to obtain an image of the changes in electrical properties in the sensing area reflecting the external loading state. The image corresponds to the loading state of the external environment, and can realize the perception of the environmental state by the software machine.

[0063] It should be noted that the mapping image obtained by the regional information reconstruction unit reflects the difference in strain state between the previous state and the current state. For a series of continuous deformation processes, the state at a fixed moment can be selected as the previous state, or a certain moment can be dynamically selected as the previous state.

[0064] It should be emphasized that the embodiments of the present invention are illustrative. For those skilled in the art, it is easy to think of applying the self-sensing software machine with large-area sensing function disclosed in the present invention to various software machines, not limited to the four-legged software machine described in the specific embodiments of the present invention. Therefore, the above embodiments are only preferred and not restrictive.

Claims

1. A self-sensing software machine with large-area sensing function, characterized in that: Including flexible sensing and driving module, signal detection module, closed-loop control module, The flexible sensing and driving module integrates flexible sensing function and software driving function, including: The elastically deformable main body is made of a superelastic material, the elastic modulus is lower than 1Mpa, the elastically deformable main body is bendable and stretchable, and the maximum stretching amount is greater than 100%; Large-area sensing part, made of elastic material, can detect an area of ​​no less than 100cm 2 The continuous sensing area, the sensing signal is obtained from the detection electrodes arranged at the boundary of the sensing area; the large-area sensing part follows the bending and stretching of the elastically deformable main body; The signal detection module comprises: A boundary current driving unit capable of inputting a driving current to a boundary electrode of the large-area sensing unit; A boundary signal acquisition unit capable of acquiring voltage signals on boundary electrodes of a large-area sensing unit; The closed-loop control module comprises: The regional information reconstruction unit can obtain deformation information in the sensing area according to the voltage signal of the signal acquisition module through the regional information reconstruction process, reflecting the state of the software machine itself or the state of the environment in which it is located; A feedback control unit capable of determining a deformation state or an environmental state of the software machine according to the reconstructed image of the region information reconstruction unit, and outputting a control signal of the software machine; The regional information reconstruction process includes the following steps: Step 1: Establish a sensing model according to the shape of the sensing area in the large-area sensing part, and determine the boundary of the sensing area; the interior of the sensing area of ​​the sensing model is continuous; arrange detection electrodes at the boundary of the sensing model, and determine the coordinates of each boundary electrode; Step 2: Estimate the electrical property distribution s(x,y) in the sensing area; for the uniform deposition preparation method, the electrical properties of the strain-sensitive functional material in the sensing area are uniformly distributed, and the electrical parameter distribution of the surface conductivity and the surface resistivity is determined by the four-point method; Step 3, solve the distribution of theoretical boundary voltage; according to the estimated electrical property distribution, calculate the theoretical potential distribution of the sensing area, and after selecting the position of the grounding signal, obtain the theoretical boundary voltage distribution F(s) at the boundary electrode; according to the electrical property characteristics of the strain-sensitive functional material, determine the electrical performance parameters before and after deformation, so as to determine the electrical property distribution of the sensing area of ​​the soft machine in the previous state and the current state, which are recorded as s1 and s2 respectively; further, calculate the boundary voltage signals in the previous state and the current state, and obtain the difference between the theoretical boundary voltage signals in the two states, which is recorded as ΔF=F(s2)-F(s1); Step 4: Test the boundary voltage signal of the large-area sensor of the software machine in the previous state and the current state through the signal detection module. The boundary voltage signal in the previous state is recorded as V1, and the boundary voltage signal in the current state is recorded as V2. The difference between the boundary voltage signals in the previous state and the current state is obtained, which is recorded as ΔV=V2-V1; Step 5, determine the difference between the test signal and the theoretical signal; calculate the absolute value of the difference between the boundary signal obtained by the test and the boundary signal obtained by theoretical calculation (|ΔV-ΔF|), and judge whether the estimated electrical performance distribution meets the actual situation; set the error amount 1% x|ΔF| as the judgment threshold, if |ΔV-ΔF| is lower than the threshold, the change of the estimated electrical performance distribution can reflect the electrical parameter distribution of the sensing area in the previous state and the current state; if |ΔV-ΔF| is not lower than the threshold, the estimated electrical parameter distribution does not meet the requirements, repeat step 3, update the electrical performance distribution s1 and s2, and recalculate the difference ΔF of the theoretical boundary voltage signal until |ΔV-ΔF| meets the limit of the set threshold; Step six: mapping imaging; according to the model of the sensing area, the distribution of the change in electrical properties is mapped to the sensing area to obtain a reconstructed image of the sensing area.

2. The self-sensing software machine with large-area sensing function according to claim 1, characterized in that: The strain-sensitive functional material constituting the large-area sensing part has a low elastic modulus, and its resistance or capacitance parameters change with strain. The embedded position of the large-area sensing part is located in the stretchable deformable layer of the elastically deformable main body of the soft machine, which neither affects the deformation of the main body nor can deform along with the main body or withstand external loading.

3. The self-sensing software machine with large-area sensing function according to claim 1, characterized in that: The regional information reconstruction unit can obtain an image of the electrical signal changes that reflects the strain distribution in the sensing area, and realize global information detection of the deformation state of the soft machine and the external environment loading based only on the electrical signal at the edge of the sensing area.

4. The self-sensing software machine with large-area sensing function according to claim 1, characterized in that: The detection wire is made of carbon-based fiber with a resistivity lower than 0.001Ω / cm. The carbon-based fiber is bendable but not stretchable. The conductivity does not change when bent, and does not interfere with the signal in the sensing area. The detection wire is connected to the sensing area through the same strain-sensitive flexible sensing material.

5. The self-sensing software machine with large-area sensing function according to claim 1, characterized in that: The boundary electrodes are located at the edge of the sensing area. The number and position of the electrodes are adjustable. The boundary electrodes are used for signal detection or current driving.

Images