Ionization type multi-mode touch sensor based on polyelectrolyte interface regulation and control and preparation method and decoupling method of ionization type multi-mode touch sensor

Abstract

The invention provides an ionization type multi-mode tactile sensor based on polyelectrolyte interface regulation and a preparation method and a decoupling method thereof, the sensor is composed of an upper electrode, an electrolyte layer and a lower electrode in sequence, the electrolyte layer is a thin film made of polyelectrolyte materials, double-layer capacitance can be formed between the electrolyte layer and the electrodes, and the lower electrode is a thin film made of polyelectrolyte materials. The surface of the electrode is formed by arranging micro-structure areas and micro-structure-free areas at intervals, and a micro-nano structure array with the micro-structure areas can change the contact area between the electrode and the electrolyte layer when being pressed; the ionizing multi-mode tactile sensor based on polyelectrolyte interface regulation and control can accurately respond to temperature and pressure in a complex environment, overcome the problem that temperature change interferes with a force measurement result, realize bimodal accurate measurement of temperature and force, is suitable for the complex environment with large temperature fluctuation, and has a wide application prospect. And the sensor can stably work in a complex environment in which both the temperature and the pressure are changed, and high sensitivity and stability are kept.

Description

Technical Field

[0001] The present invention relates to the field of sensor technology, and in particular to an ionized multimodal tactile sensor based on polyelectrolyte interface regulation, and a preparation method and a decoupling method thereof. Background Art

[0002] With the continuous advancement of technology, tactile sensors are increasingly being used in fields such as virtual reality, robotics, smart wearable devices, and health monitoring. The main function of a tactile sensor is to sense changes in external force or pressure and provide real-time feedback. In these applications, high-precision tactile perception requires not only high sensitivity to changes in external force but also stable performance in environments with large temperature fluctuations. However, traditional capacitive tactile sensors are susceptible to temperature fluctuations, resulting in a significant decrease in force measurement accuracy. Specifically, the operating principle of double-layer capacitive tactile sensors is based on changes in the double-layer capacitance at the interface between the electrode and the electrolyte. The ion mobility and double-layer structure of the electrolyte material are highly sensitive to temperature. Temperature fluctuations can alter the ion mobility characteristics of the electrolyte and the double-layer capacitance, causing capacitance measurement deviations and ultimately affecting the accuracy of the force measurement. This temperature interference limits the reliable application of these sensors in complex environments.

[0003] To solve the above problems, researchers have developed temperature-pressure dual-modal sensors, aiming to achieve simultaneous measurement of temperature and pressure. However, traditional temperature-pressure dual-modal sensors still have significant shortcomings in practical applications, which limits their promotion in key areas. These shortcomings mainly include complex preparation processes, mismatched mechanical properties, insufficient mechanical robustness, and limited decoupling capabilities of temperature and pressure signals. Specifically, traditional temperature-pressure dual-modal sensors often adopt a vertical stacking or in-plane integration design, which makes the manufacturing process cumbersome and costly. At the same time, due to the differences in the mechanical properties of materials in different functional layers, it is easy to cause overall mechanical properties to be inconsistent, affecting the stability of the sensor. In addition, its complex structure is prone to deformation or damage in flexible applications, making it difficult to meet the conformal bonding requirements of flexible electronic devices. More importantly, existing temperature-pressure dual-modal sensors perform poorly in high-resolution separation and detection of temperature and pressure signals, and temperature fluctuations will still interfere with force measurement, resulting in insufficient reliability of measurement data.

[0004] To overcome these technical challenges, the present invention proposes an ionized multimodal tactile sensor based on polyelectrolyte interface regulation. By optimizing the interfacial properties of polyelectrolyte materials and combining them with microstructural design, this sensor achieves high-resolution decoupled detection of pressure and temperature signals, eliminating temperature interference on force measurement. This approach also simplifies the fabrication process, improves mechanical property matching and robustness, and meets the conformal performance requirements of flexible electronic devices, providing a new solution for the practical application of tactile sensors in a variety of complex application scenarios. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an ionized multimodal tactile sensor based on polyelectrolyte interface regulation.

[0006] Another technical problem to be solved by the present invention is to provide a method for preparing the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation.

[0007] Another technical problem to be solved by the present invention is to provide a decoupling method for the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation.

[0008] In order to solve the above technical problems, the technical solution of the present invention is:

[0009] An ionized multimodal tactile sensor based on polyelectrolyte interface regulation includes multiple temperature-pressure dual-mode sensing units. The temperature-pressure dual-mode sensing unit refers to a basic sensing structure that can simultaneously sense temperature and pressure. Each temperature-pressure dual-mode sensing unit is composed of an upper electrode, an electrolyte layer and a lower electrode in sequence. The material of the electrolyte layer film adopts a polyelectrolyte material that can form a double-layer capacitor between the electrolyte layer and the electrode. Its interface shape is composed of microstructured areas and non-microstructured areas arranged at intervals.

[0010] Preferably, the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation has a microstructure in the microstructure area formed by a micro-nanostructure array that can change the contact area between the electrode and the electrolyte layer when under pressure, and a non-microstructure area is used to respond to temperature changes for temperature measurement. Its capacitance response value is almost unaffected by external force, and the response to force is much smaller than the response to temperature changes. It can effectively compensate for the interference of temperature on force measurement to correct the force measurement value.

[0011] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the structure of the micro-nanostructure array is cone, pyramid, prism or hemisphere.

[0012] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the structure of the micro-nanostructure array adopts a hemispherical structure to enhance the response to external force, and is particularly suitable for precision force measurement.

[0013] Preferably, in the above-mentioned ion-type multimodal tactile sensor based on polyelectrolyte interface regulation, after comprehensively considering the demolding effect, force measurement range and sensitivity, the size of a single microstructure in the microstructure array is 50 μm in length and width and 25 μm in height, and the interval between two adjacent microstructures is 5 μm.

[0014] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the material of the electrolyte layer film is a polyelectrolyte material that can form a double-layer capacitor between the electrolyte layer and the electrode.

[0015] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the polyelectrolyte material that can form a double-layer capacitor between the electrolyte layer and the electrode is PVA (polyvinyl alcohol) gel with added phosphoric acid, chemically modified PVA gel containing ionic groups, polyelectrolyte polymer or ion gel.

[0016] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the polyelectrolyte material capable of forming a double-layer capacitor between the electrolyte layer and the electrode is PVA (polyvinyl alcohol) gel added with phosphoric acid.

[0017] Preferably, the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the polyelectrolyte material that can form a double-layer capacitor between the electrolyte layer and the electrode is mixed with 1g of PVA particles, 5-10g of water and 700-900μl of phosphoric acid. By adding phosphoric acid, the film forms an ionic electrolyte layer, which significantly improves the sensitivity of the capacitive response.

[0018] Preferably, in the above-mentioned ionotropic multimodal tactile sensor based on polyelectrolyte interface regulation, the polyelectrolyte material capable of forming a double-layer capacitor between the electrolyte layer and the electrode is prepared by mixing 1g of PVA particles, 9g of water and 825μl of phosphoric acid.

[0019] Preferably, in the above-mentioned ionotropic multimodal tactile sensor based on polyelectrolyte interface regulation, the chemically modified PVA gel containing ionic groups is prepared by copolymerizing PVA with monomers containing ionic groups (such as acrylic acid, maleic acid, etc.) or introducing ionic groups through chemical reactions (such as esterification, etherification, etc.); the ion gel is prepared by mixing a polymer gel with an electrolyte salt (such as NaCl, LiCl, etc.) or an ionic liquid (such as 1-butyl-3-methylimidazolium hexafluorophosphate); the polyelectrolyte polymer is selected from water-soluble polyelectrolytes such as sodium polyacrylate, sodium polystyrene sulfonate, and chitosan.

[0020] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the electrode array structure of the upper electrode and the lower electrode is the same, both consisting of m×n electrodes (preferably square electrodes), the upper electrode is equipped with m unidirectional lead wires, and the lower electrode is equipped with n unidirectional lead wires, and the directions of the upper lead wires and the lower lead wires are perpendicular to each other, with an angle of 90°. This design effectively reduces signal crosstalk, thereby improving the tactile perception performance of the sensor. The electrodes and lead wires are processed using screen printing technology. The electrolyte layer is a thin film made of polyelectrolyte material, and the interface of the film has m×n sensing areas, some of which are provided with microstructures, and the remaining areas are flat areas without microstructures, and the areas with microstructures and the areas without microstructures are arranged alternately. The m×n sensing areas on the electrolyte layer film correspond one-to-one in position to the m×n electrodes of the upper electrode and the lower electrode. During the packaging process, the m×n electrodes of the upper electrode and the lower electrode are precisely aligned with the m×n sensing areas on the electrolyte layer film to form m×n double-layer capacitors.

[0021] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, both m and n are even numbers.

[0022] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the temperature and pressure dual-mode sensing unit is a basic sensing unit in the overall sensor, which is composed of an upper electrode, a lower electrode and a corresponding electrolyte layer part; the upper electrode and the lower electrode each contain p×q electrodes, and the electrolyte layer has p×q sensing areas corresponding to the electrodes, wherein some sensing areas are provided with microstructures, and the remaining sensing areas are planar areas without microstructures, and the sensing areas with microstructures are arranged at intervals from the sensing areas without microstructures; during the packaging process, the p×q electrodes of the upper electrode and the lower electrode are precisely aligned with the p×q sensing areas on the electrolyte layer to form p×q double-layer capacitors.

[0023] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, in the temperature-pressure dual-mode sensing unit, p=q=2, and among the 2×2 sensing areas of the electrolyte layer, the two sensing areas on the diagonal are provided with microstructures, and the other two sensing areas are planar areas without microstructures.

[0024] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, each basic temperature-pressure dual-mode sensing unit consists of an upper electrode and a lower electrode, each containing 2×2 square electrodes, and a corresponding electrolyte layer portion. The electrolyte layer has 2×2 square areas in the sensing unit, of which the square areas on the two diagonals are provided with microstructures, and the other two square areas are flat areas without microstructures. During the packaging process, the 2×2 square electrodes of the upper and lower electrodes are precisely aligned with the 2×2 square areas on the electrolyte layer, forming four double-layer capacitors.

[0025] Preferably, in the above-mentioned ion-type multimodal tactile sensor based on polyelectrolyte interface regulation, in one of the basic temperature-pressure dual-mode sensing units, the pressure sensing unit is a region with a microstructure, which utilizes the change in the contact area of ​​the double-layer capacitor when under pressure to cause a change in the capacitance response; the temperature sensing unit is a planar region without a microstructure, which utilizes the change in the ion mobility of the polyelectrolyte material in the electrolyte layer caused by temperature change to cause a change in the capacitance response, and the temperature sensing unit has an extremely low response to pressure; the decoupling process is: first detect the temperature, and then use the capacitance response generated by the temperature sensor to temperature compensate the capacitance response of the pressure sensor, and decouple the capacitance response difference caused by temperature and pressure through a force decoupling algorithm based on temperature compensation to ensure the stability and accuracy of the sensor in a complex environment and achieve accurate decoupling of temperature and pressure data.

[0026] Preferably, the above-mentioned ion-type multimodal tactile sensor based on polyelectrolyte interface regulation uses a single-chip microcomputer (STM32F407) as the core controller and combines it with a capacitance acquisition chip (FDC2214) to efficiently collect, convert and process the signals of the m×n sensor array to obtain the capacitance value of each double-layer capacitor; the capacitance response difference between temperature and force is decoupled through a force decoupling algorithm based on temperature compensation, and the stability and accuracy of the sensor in complex environments are ensured.

[0027] In the aforementioned polyelectrolyte interface-controlled ionized multimodal tactile sensor, assuming the lead wires of the upper electrode are oriented longitudinally and the lead wires of the lower electrode are oriented transversely, the capacitance of the double-layer capacitor in the upper left corner can be determined by capturing and processing the electrical signals from the leftmost lead wire of the upper electrode and the topmost lead wire of the lower electrode. Similarly, by processing both longitudinal and transverse electrical signals, the capacitance of each capacitor in the sensor can be determined.

[0028] The above-mentioned ionized multimodal tactile sensor uses a temperature-compensated force decoupling algorithm as a temperature and pressure signal decoupling algorithm, aiming to accurately separate the effects of temperature and pressure, thereby ensuring high-precision pressure measurement. The electrolyte layer film of the ionized multimodal tactile sensor is composed of microstructured regions and non-microstructured regions arranged alternately. The microstructured regions are mainly used to respond to external forces (pressure), while the non-microstructured regions are used to respond to temperature changes. Because temperature also affects the capacitive response of the microstructured regions, this algorithm effectively eliminates temperature interference on pressure measurement through multi-step signal processing, regression modeling, and data fusion methods.

[0029] Specifically, by utilizing the characteristic that the non-microstructure area is more sensitive to temperature than to force, the above-mentioned interface regulation is used to achieve temperature and pressure dual-mode sensing. The ionized multimodal tactile sensor realizes temperature sensing based on the non-microstructure area. The non-microstructure area is insensitive to pressure. The temperature is first detected and then the capacitive response generated by the temperature sensor is used to compensate the capacitive response of the pressure sensor. The difference in the capacitive response of temperature and force is decoupled through a force decoupling algorithm based on temperature compensation, and the stability and accuracy of the sensor in complex environments are ensured, thereby achieving accurate decoupling of temperature and pressure data.

[0030] The force decoupling algorithm based on temperature compensation is a compensation algorithm based on multi-layer perceptron (MLP). 0_avg (average capacitance signal of the part without microstructure) and C m_avg (the average capacitance signal of the microstructure part) is used as the input feature, and the predicted pressure P is directly output through the neural network, effectively decoupling the pressure value under the influence of temperature and achieving high-precision pressure measurement.

[0031] Preferably, in the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the microstructured area on the electrolyte layer film responds to both pressure and temperature, while the non-microstructured area only responds to temperature and is insensitive to pressure. Based on this characteristic, temperature sensing is achieved through the non-microstructured area, and its capacitive response is used to temperature compensate the pressure sensing capacitive response of the microstructured area. A force decoupling algorithm based on temperature compensation is used to decouple the capacitance response difference of temperature and pressure, ensure the stability and accuracy of the sensor in complex environments, and achieve accurate decoupling of temperature and pressure data. The specific decoupling method steps of the force decoupling algorithm based on temperature compensation are as follows:

[0032] (1) Calculate the average capacitance and

[0033] (2) After preprocessing, it is input into the trained neural network model.

[0034] Preferably, in the above-mentioned ion-type multimodal tactile sensor based on polyelectrolyte interface regulation, the training method of the neural network model is as follows:

[0035] 1. Use multi-layer perceptron (MLP) neural network for regression modeling. The model architecture includes input layer, hidden layer and output layer;

[0036] 2. Collect and preprocess input data and calculate the average value and Among them, C 01 、C 02 is the capacitance without microstructure, C m1 、C m2 is a capacitor with microstructure; for C 0_avg and C m_avg Normalization is performed separately, and the formula is as follows: Where C represents the original capacitance value, C min and C max are the corresponding minimum and maximum values ​​in the data set respectively;

[0037] 3. Through the training strategy of loss function, optimizer, batch training, learning rate scheduling and early stopping method, the model can effectively learn the mapping relationship between capacitance signal and pressure;

[0038] 4. Use grid search to optimize key hyperparameters. The optimization parameters include:

[0039] (1) Learning rate: candidate values ​​{0.0001, 0.001, 0.01};

[0040] (2) Number of neurons in the first hidden layer: candidate values ​​{32, 64, 128};

[0041] (3) Number of neurons in the second hidden layer: candidate values ​​{16, 32, 64};

[0042] (4) Batch size: candidate values ​​{16, 32, 64};

[0043] 5. Validate the model through cross-validation. Use the same model architecture and training strategy for each cross-validation, and record the loss value and prediction deviation of each fold.

[0044] The preparation method of the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation has the following specific steps:

[0045] (1) Preparation of film material: an ion gel film is prepared using a polyelectrolyte material (preferably a PVA ion gel prepared by mixing 1 g of PVA particles, 5-10 g of water, and 700-900 μl of phosphoric acid solution);

[0046] (2) Patterned design and manufacturing of electrolyte layer film: The manufacturing process of electrolyte layer film includes mold design and processing, intermediate mold preparation and film forming; the mold is manufactured by two-photon printing, 3D printing, photolithography, soft lithography, laser direct writing or inkjet printing technology, and the mold surface is designed with a pattern of microstructure areas and non-microstructure areas arranged at intervals. The shape of the microstructure area can be selected from micro-nano structures such as cone, pyramid, prism or hemisphere; through the polydimethylsiloxane (PDMS) demolding process, the mold is used to prepare a PDMS thick film with a surface of microstructure areas and non-microstructure areas arranged at intervals as an intermediate mold. The polyelectrolyte material is cast on the PDMS intermediate mold and demolded after curing to prepare an electrolyte layer film. The surface of the film directly forms a pattern of microstructure areas and non-microstructure areas arranged at intervals corresponding to the mold;

[0047] (3) Assembly and packaging of electrode layers: Assemble and package the electrode layer and electrolyte layer;

[0048] (4) Implementation of temperature and force decoupling: The capacitance response value of the sensor is collected through a capacitance acquisition chip (FDC2214) and the data is processed using a force decoupling algorithm based on temperature compensation.

[0049] Preferably, in the method for preparing the above-mentioned ionized multimodal tactile sensor based on polyelectrolyte interface regulation, the steps of the two-photon printing technology are as follows:

[0050] Ⅰ. Drawing mold model:

[0051] A mold model is drawn using SolidWorks software. The model is a pattern in which microstructured areas and non-microstructured areas are alternately arranged. The model is then exported as an STL file to be imported into a printer for processing.

[0052] Ⅱ. Silanization treatment of wafers:

[0053] ① Prepare diluted glacial acetic acid solution: Mix water and glacial acetic acid in a ratio of 1:10 to prepare 1.5 ml of diluted glacial acetic acid solution (136 μl water and 1360 μl glacial acetic acid), and let it stand for at least 5 minutes;

[0054] ② Prepare silanization solution: Mix 50 ml of ethanol with 250 μl of 3-(trimethoxysilyl)propyl methacrylate reagent (CAS No. 2530-85-0), then add the diluted glacial acetic acid solution prepared in step ① and shake to mix well;

[0055] ③ Plasma treatment of the wafer: Place the wafer with the printed surface facing up and perform plasma treatment to remove surface impurities and improve adhesion;

[0056] ④ Soak the wafer: Place the plasma-treated wafer in the solution prepared in step ② and soak for 7 minutes to fully perform silanization;

[0057] ⑤ Rinse and dry: After taking out the wafer, rinse it with isopropyl alcohol and then blow dry the wafer surface with nitrogen;

[0058] III. Two-photon printing of microstructures:

[0059] Import the exported STL model into the slicing software of the two-photon printer. After the software processing is completed, operate the printer according to the processed file to print the microstructure;

[0060] IV. Development treatment:

[0061] After printing is completed, the wafer is developed: soak the printed wafer in developer for 10 minutes, and then soak it in isopropyl alcohol for 5 minutes to remove uncured photoresist;

[0062] V. PDMS demoulding treatment:

[0063] (a) Before demolding, trichlorosilane is used as a release agent and attached to the wafer surface by evaporation;

[0064] (b) Pour the prepared PDMS onto the wafer structure and let it sit overnight;

[0065] (c) Heating at 60-80° C. for 4 hours to solidify the PDMS; after heating, removing the wafer and demolding it to complete the production of the PDMS intermediate mold.

[0066] VI. Electrolyte layer ion film demoulding treatment:

[0067] The polyelectrolyte material is cast on the PDMS middle mold and demoulded after curing to prepare the electrolyte layer film.

[0068] The above-mentioned decoupling method of the ionized multimodal tactile sensor based on polyelectrolyte interface regulation adopts a force decoupling algorithm based on temperature compensation. It measures the temperature by measuring the capacitive response of the non-microstructure area (temperature sensing unit), and uses the temperature or directly uses the capacitive response of the non-microstructure area to compensate for the capacitive response of the microstructure area (pressure sensing unit), thereby achieving decoupling of temperature and pressure to ensure the stability and accuracy of the sensor in complex environments.

[0069] Preferably, the decoupling method of the above-mentioned ion-type multimodal tactile sensor based on polyelectrolyte interface regulation uses two 70HC4051 multiplexing chips to receive the upper and lower electrode plate signals of the sensor array respectively. The multiplexing chip switches each sensor unit in turn by adopting a polling mechanism, and outputs the upper and lower plate signals of each sensor unit to the FDC2214 capacitance acquisition chip in turn. The FDC2214 capacitance acquisition chip processes the signal to calculate the capacitance value and outputs it to the stm32 microcontroller in the form of a digital signal.

[0070] Beneficial effects:

[0071] The aforementioned ion-type multimodal tactile sensor achieves dual-mode sensing of temperature and pressure by fabricating microstructured and non-microstructured areas on the same film. Traditional dual-mode sensors, because temperature and pressure sensing cannot be performed using the same mechanism, typically require vertical stacking or in-plane integration of the two different sensing mechanisms. This approach often results in complex fabrication processes, mismatched mechanical properties, and insufficient mechanical robustness, making it difficult to meet the conformal requirements of flexible electronic devices. The present application cleverly adopts a design of alternating arrangement of microstructures and non-microstructures on a single film, and can achieve simultaneous sensing of temperature and pressure only through surface interface regulation, thereby realizing high-resolution decoupling detection of pressure and temperature signals, effectively overcoming the above-mentioned defects in traditional technologies; the mold is manufactured using two-photon printing technology, which can achieve high precision of 1μm in the XY direction and 0.6um in the Z direction, and can achieve precise manufacturing at the micron level. The surface shape of the microstructured area of ​​the electrolyte layer film is preferably a hemispherical structure. Compared with the cone, prism, and pyramid structures, the contact area between the electrolyte layer and the electrode changes more when subjected to force, thereby achieving a more significant capacitive response; by regulating the ionized multimodal tactile sensor based on the polyelectrolyte interface, it can accurately respond to temperature and pressure in complex environments, overcome the problem of temperature changes interfering with force measurement results, realize dual-modal precise measurement of temperature and force, provide reliable perception data, and is suitable for stable operation in complex environments with high and low temperatures and large temperature and pressure fluctuations, maintaining high sensitivity and stability. It has broad application prospects in the fields of robotic tactile, virtual reality, smart wearable devices, and health monitoring. The sensor can provide accurate temperature and force measurement data, and achieves dual-modal precise measurement of temperature and force while simultaneously responding to external force and temperature changes. The preparation method is simple, low-cost, suitable for large-scale batch production, and has strong potential for industrial application. Specifically:

[0072] (1) Temperature and pressure dual-mode measurement:

[0073] The ionized multimodal tactile sensor, based on polyelectrolyte interface regulation, utilizes a thin electrolyte layer design with alternating microstructured and non-microstructured regions. This effectively reduces the impact of temperature on sensor measurements, enabling dual-modal sensing of external temperature and force. The microstructured regions primarily respond to force changes, while the non-microstructured regions primarily respond to and compensate for temperature changes. This decoupling of temperature and force successfully decouples temperature and force measurements, enabling the sensor to accurately measure both temperature and force simultaneously in complex environments with large temperature fluctuations. This effectively prevents temperature interference with force measurements, maintains stable force measurement accuracy, and achieves high-precision dual-modal measurement.

[0074] (2) Application of ionic electrolyte layer:

[0075] By adding phosphoric acid to the hydrogel to form an ionic electrolyte layer, the electrical properties and response sensitivity of the film are significantly improved, the response sensitivity of the sensor to external force and temperature changes and the measurement accuracy are improved, and the amplitude of capacitance change is increased, making the sensor more sensitive to changes in external pressure and temperature, solving the dual-modal measurement problem of temperature and force, and enabling it to maintain stable working performance under different environmental conditions; the ionization properties of the film further enhance the response capability and stability of the sensor, adapting to more complex working environments.

[0076] (3) Application of two-photon printing technology:

[0077] The mold is manufactured using two-photon printing technology. The mold surface is patterned with alternating microstructured and non-microstructured areas. The mold's processing accuracy in the XY direction (horizontal plane) reaches 1μm, and the Z direction (vertical direction) processing accuracy reaches 0.6μm. This technology enables high-precision manufacturing of complex microstructures, ensuring the accuracy of the mold pattern, and ultimately demolding to produce an electrolyte layer film with corresponding microstructured and non-microstructured areas on the surface. This manufacturing method ensures the accuracy of the film's surface microstructure and provides feasibility for mass production of sensors.

[0078] (4) Capacitive decoupling of temperature and force response:

[0079] By using a temperature-compensated force decoupling algorithm to decouple the capacitive responses to temperature and force, the sensor can precisely separate and independently process temperature and pressure signals, eliminating temperature interference on the pressure signal. This allows the sensor to accurately distinguish the effects of both temperature and force changes when faced with simultaneous changes, and achieves precise compensation and correction through data processing, thereby achieving high-precision force measurement. This algorithm provides precise temperature and pressure decoupling technology for applications such as dexterous hands and robotic tactile perception, enabling stable operation in dynamic environments and ensuring high-performance tactile perception and force measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 This is the overall structure diagram of a 6×6 array of ionized multimodal tactile sensors based on polyelectrolyte interface regulation.

[0081] Figure 2 This is a diagram of the temperature-pressure dual-mode sensing unit structure, taking a 2×2 array of ionized multimodal tactile sensors based on polyelectrolyte interface regulation as an example;

[0082] Figure 3 This is a diagram of the structure of an ion film with an electrolyte layer arranged at intervals for multimodal sensing;

[0083] Figure 4 This is a data graph showing the response of the sensor capacitance to temperature changes when there is no microstructured film as the electrolyte layer;

[0084] Figure 5 This is a data graph showing the response of the sensor capacitance to pressure changes when there is no microstructured film as the electrolyte layer;

[0085] Figure 6 This is a comparison data chart of the contact area changes between cone, pyramid, prism, and hemispherical microstructures under pressure and the pressure surface obtained through COMSO simulation;

[0086] Figure 7 This is a graph showing the relationship between capacitance and pressure response in the microstructured part.

[0087] In the figure: 1: a basic temperature-pressure dual-mode sensing unit; 2: upper electrode; 3: electrolyte layer; 4: lower electrode; 5: area without microstructure; 6: area with microstructure. DETAILED DESCRIPTION

[0088] A basic temperature-pressure dual-mode sensing unit of the ionization multi-modal tactile sensor of the present invention will be described below with reference to the embodiments and drawings.

[0089] Example 1

[0090] The ionization multimodal tactile sensor includes a plurality of temperature and pressure dual-mode sensing units, such as Figure 1 As shown, each temperature and pressure dual-mode sensing unit serves as a basic sensing unit in the overall sensor, as shown in Figure 2 、 Figure 3 As shown, it consists of an upper electrode 2 and a lower electrode 4, each containing 2×2 square electrodes, and a corresponding intermediate electrolyte layer 3.

[0091] (1) Electrode layer:

[0092] The upper electrode 2 and the lower electrode 4 serve as the core electrode components of the double-layer capacitive sensor, playing the role of charge collection and signal transmission. The two form a double-layer capacitor structure through interfacial contact with the electrolyte layer 3, which is used to detect changes in capacitance caused by changes in the external environment. Specifically, changes in external temperature or pressure will cause changes in capacitance, and the electrode transmits this capacitance change signal to an external measurement system through a lead. To ensure the stability of signal transmission, the upper electrode 2 and the lower electrode 4 are arranged in a criss-cross pattern, and the lead angle is designed to be 90 ° to effectively reduce the electrical signal interference between the upper and lower electrode leads.

[0093] (2) Electrolyte layer:

[0094] The electrolyte layer 3 is located between the upper electrode 2 and the lower electrode 4, employing a 2×2 interface design that precisely mates with the 2×2 surfaces of the two electrodes. The two diagonal square regions are microstructured. The microstructure arrays are hemispherical (or pyramidal, prism-shaped, or conical structures that can change the contact area between the electrode and the electrolyte layer under pressure). The other two square regions are flat, without microstructures. During the packaging process, the 2×2 square electrodes of the upper and lower electrodes are precisely aligned with the 2×2 square regions of the electrolyte layer. The electrolyte layer is an ionic thin film made of PVA gel with added phosphoric acid (or a polyelectrolyte material such as chemically modified PVA gel containing ionic groups, a polyelectrolyte polymer, or an ion gel that forms a double-layer capacitor between the electrolyte layer and the electrodes). Four double-layer capacitors are formed at the interface with the upper electrode 2 and the lower electrode 4. This double-layer structure is the basis for the sensor to sense external changes. Through its ion migration and dielectric properties, the electrolyte layer 3 establishes a stable electric field between the electrodes, supporting the formation and change of capacitance. During the packaging process, the upper and lower electrodes 2 and 4 form a complete capacitor unit through close physical contact with the electrolyte layer 3. The electrolyte layer 3 not only provides the necessary electrochemical environment but also directly influences the capacitor's sensitivity to temperature and pressure through its material properties, thereby enabling high-precision measurement of environmental parameters.

[0095] Figure 3The electrolyte layer structure of the basic temperature and pressure dual-mode sensing unit is demonstrated, which includes a microstructured area 6 and a non-microstructured area 5. The microstructured area 6 and the non-microstructured area 5 are arranged at intervals. The specific design is intended to achieve dual-modal perception of external temperature and force. The electrolyte layer is an ion film made of PVA gel with added phosphoric acid. The ratio is 1g PVA particles, 9g water and 825μl phosphoric acid. The addition of phosphoric acid forms an ionic electrolyte layer, which significantly improves the response sensitivity of the sensor. Due to the special properties of the hydrogel material, it can sense temperature changes well and generate a capacitive response, which plays a role in enhancing the capacitive response. The capacitance change of the sensor is measured by the lead electrical signals of the upper and lower electrodes, thereby realizing the perception of temperature and pressure changes.

[0096] Microstructured Area 6: The ion membrane in this area features a microstructured design, employing an array of hemispherical structures. Each microstructure measures 50μm in length, width, and height, and is 25μm high, with 5μm spacing between adjacent microstructures. This design aims to enhance the sensor's sensitivity to external forces while ensuring effective demolding. When applied force, the capacitance changes due to changes in the contact area of ​​the double-layer capacitor. This capacitance response is measured using electrical signals from the leads of the upper and lower electrodes.

[0097] The experimental steps for the change of contact area with the pressure surface under pressure obtained by COMSO simulation are as follows:

[0098] 1. Use SolidWorks to draw microstructure models such as cones, pyramids, prisms, and hemispheres;

[0099] 2. Use SolidWorks to draw a cuboid and assemble it with the microstructure drawn in the first step. The cuboid is located above the microstructure and serves as the surface for applying pressure and calculating contact area during simulation (simulating the sensor's electrodes).

[0100] 3. Select the mechanical physics field and solid mechanics module, set the material's elastic modulus, density, Poisson's ratio and other physical parameters, and import each model into COMSOL;

[0101] 4. Select the upper surface of the force-applying cuboid and set the prescribed displacement, select the lower surface of the microstructure and set it as a fixed constraint, select the microstructure part and set it as a hyperelastic material, mesh the force-applying cuboid and microstructure separately, set the calculation of contact area and pressure in the results, and click Calculate in the study;

[0102] 4. Derive the data of contact area changing with displacement and pressure changing with displacement. After processing the data, obtain the data of contact area changing with pressure between each microstructure and the pressure surface. Use origin to draw the curve of contact area changing with pressure for each microstructure.

[0103] like Figure 6 As shown in the figure, the simulation results show that when the hemispherical structure is under pressure, the change in the contact area with the pressure surface is greater than that of the conical, pyramidal, and prism-shaped micro-nano structures, thereby enhancing the capacitance response, making the sensor more sensitive to changes in external force and improving measurement accuracy.

[0104] Non-microstructured region 5: This region is specifically used for measuring temperature changes. The capacitance response of the non-microstructured region is almost entirely caused by temperature changes. Since the contact area between the electrolyte layer and the electrode hardly changes when the ion membrane in the non-microstructured region is under pressure, there is almost no response to external force.

[0105] The method for measuring the response data of the sensor capacitance value to temperature changes is as follows:

[0106] 1. Evenly coat the prepared hydrogel on the surface of the glass slide;

[0107] 2. Let the glass slide coated with hydrogel stand overnight to allow the hydrogel to fully solidify;

[0108] 3. After the hydrogel solidifies, demold the hydrogel on the glass slide;

[0109] 4. Cut part of the hydrogel to serve as a thin film of electrolyte layer without microstructure;

[0110] 5. Sandwich the electrolyte layer film between the upper and lower screen-printed electrodes and attach a glass slide under the lower electrode for encapsulation;

[0111] 6. Place the packaged sensor on the heating plate, connect the capacitance acquisition device to the lead wires of the upper and lower electrodes, and measure the capacitance value;

[0112] 7. Start the heating plate and set its temperature to 70°C, so that the temperature of the sensor rises from room temperature (20°C) to 70°C at one time. During this process, the capacitance acquisition device will record the change of capacitance value and corresponding time in real time;

[0113] 8. Based on Newton's law of cooling, establish a functional relationship between sensor temperature and time, and then convert the data of capacitance change over time recorded by the capacitance acquisition device into data of capacitance change over temperature;

[0114] 9. Plot the data of capacitance versus temperature.

[0115] The method for measuring the response data of the sensor capacitance value to pressure changes is as follows:

[0116] 1. Fix the sensor on the pressure test platform and connect the capacitance acquisition device to the lead wires of the upper and lower electrodes to measure the capacitance value;

[0117] 2. Control the pressure test platform to gradually apply pressure from 0 to 8N, and record the changes in pressure and capacitance values ​​during the process;

[0118] 3. Since the pressure acts on an area of ​​16 square centimeters, the applied pressure of 0 to 8N can be converted to a pressure of 0 to 500 kPa. Plot the relationship between the change in capacitance and pressure.

[0119] Experiments have shown that the capacitive response of thin films without microstructures to force changes (e.g. Figure 5 is much smaller than the capacitance response to temperature changes (as shown in Figure 4 As shown in Figure 2 ). Therefore, the capacitive response of the non-microstructured film portion can be considered to be entirely due to temperature changes. This makes the non-microstructured film ideal for precise temperature monitoring in environments with large temperature fluctuations, while accurately compensating for the interference caused by temperature changes on the force measurement.

[0120] The preparation method of the above-mentioned ionization type multimodal tactile sensor has the following specific steps:

[0121] (1) The prepared PVA particles, phosphoric acid, and water were mixed in a ratio of 1 g PVA particles, 9 g water, and 825 μl phosphoric acid to form a hydrogel. The hydrogel has good ionization properties and high capacitance response sensitivity.

[0122] (2) The mold is manufactured using two-photon printing technology, and the mold surface is designed to have a pattern with alternating microstructure areas and non-microstructure areas. The processing accuracy of two-photon printing technology in the XY direction (horizontal plane) reaches 1μm, and the processing accuracy in the Z direction (vertical direction) reaches 0.6μm, ensuring the precise molding of the microstructure on the mold. The microstructure area adopts a micro-nanostructure array design such as cone, pyramid, prism or hemisphere, which can change the contact area between the electrode and the electrolyte layer when under pressure. Subsequently, polydimethylsiloxane (PDMS) is used to prepare a PDMS intermediate mold based on the above mold through a demolding process. The polyelectrolyte material prepared in step (1) is cast on the PDMS intermediate mold, and demolded after curing to prepare the electrolyte layer 3. The resulting film surface forms a pattern with alternating microstructure areas 6 and non-microstructure areas 5 corresponding to the mold, meeting the design requirements of the sensor. This step ensures that the surface morphology of the electrolyte layer film is consistent with the design requirements through precise mold manufacturing and reasonable process flow, providing a reliable guarantee for the performance of the sensor. The specific steps of the two-photon printing technology are as follows:

[0123] Ⅰ. Drawing mold model:

[0124] A mold model is drawn using SolidWorks software. The model is a pattern in which microstructured areas and non-microstructured areas are alternately arranged. The model is then exported as an STL file to be imported into a printer for processing.

[0125] Ⅱ. Silanization treatment of wafers:

[0126] ① Prepare diluted glacial acetic acid solution: Mix water and glacial acetic acid in a ratio of 1:10 to prepare 1.5 ml of diluted glacial acetic acid solution (136 μl water and 1360 μl glacial acetic acid), and let it stand for at least 5 minutes;

[0127] ② Prepare silanization solution: Mix 50 ml of ethanol with 250 μl of 3-(trimethoxysilyl)propyl methacrylate reagent (commercially available, CAS No. 2530-85-0), then add the diluted glacial acetic acid solution prepared in step ① and shake to mix well;

[0128] ③ Plasma treatment of the wafer: Place the wafer with the printed surface facing up and perform plasma treatment to remove surface impurities and improve adhesion;

[0129] ④ Soak the wafer: Place the plasma-treated wafer in the solution prepared in step ② and soak for 7 minutes to fully perform silanization;

[0130] ⑤ Rinse and dry: After taking out the wafer, rinse it with isopropyl alcohol and then blow dry the wafer surface with nitrogen;

[0131] III. Two-photon printing of microstructures:

[0132] Import the exported STL model into the slicing software of the two-photon printer. After the software processing is completed, operate the printer according to the processed file to print the microstructure;

[0133] IV. Development treatment:

[0134] After printing is completed, the wafer is developed: soak the printed wafer in developer for 10 minutes, and then soak it in isopropyl alcohol for 5 minutes to remove uncured photoresist;

[0135] V. PDMS demoulding treatment:

[0136] (a) Before demolding, trichlorosilane is used as a release agent and attached to the wafer surface by evaporation;

[0137] (b) Pour the prepared PDMS onto the wafer structure and let it sit overnight;

[0138] (c) Heating at 60-80° C. for 4 hours to solidify the PDMS; after heating, removing the wafer and demolding it to complete the production of the PDMS intermediate mold.

[0139] VI. Electrolyte layer ion film demoulding treatment:

[0140] The polyelectrolyte material is cast on the PDMS middle mold and demoulded after curing to prepare the electrolyte layer film.

[0141] (3) Combine the upper electrode 2 and the lower electrode 4 with the electrolyte layer 3 and connect them through electrode leads to complete the assembly of the sensor.

[0142] (4) The assembled sensor is fixed in shape using PI tape or other packaging processes to ensure the stability and reliability of the sensor.

[0143] (5) The capacitance response of the sensor is collected in real time through the STM32 control system and the FDC2214 capacitance acquisition chip, and the force decoupling algorithm based on temperature compensation is used for real-time decoupling to ensure that the sensor can work stably under different environmental conditions.

[0144] This technology allows precise control of the size and shape of the microstructures, ensuring that the microstructured regions 6 are spaced apart from the non-microstructured regions 5, meeting the requirements for precise force and temperature measurement. Two-photon printing precisely controls the size and spacing of the microstructures at the micron level, ensuring precise spacing between the microstructured and non-microstructured regions on the electrolyte layer, meeting the high-precision requirements of the present invention.

[0145] This design allows the sensor to accurately compensate for temperature effects on force measurements while simultaneously measuring changes in external temperature and force. The specific operating principle is as follows: the microstructured region 6 responds to both force and temperature changes, with changes in capacitance transmitted through electrical signals in the electrode leads. In contrast, the non-microstructured region 5 responds only to temperature changes, with the capacitance response being almost entirely due to temperature fluctuations. By decoupling the capacitance response, the sensor can accurately measure changes in external temperature and force, ensuring stable measurement results.

[0146] This embodiment uses the STM32F407 microcontroller as the core controller, combined with the FDC2214 capacitance acquisition chip to measure the capacitance response value, efficiently collects, converts and processes the signals of the 2×2 capacitive sensor array, and uses a force decoupling algorithm based on temperature compensation to decouple the capacitance response of temperature and force. Specifically,

[0147] 1. System Architecture and Hardware Connection

[0148] The signal processing hardware design includes the following core modules, which work together through specific connections to achieve accurate capture and transmission of tactile signals:

[0149] 1. Capacitive sensor array

[0150] (1) Structure: A two-dimensional array consisting of 4 capacitive sensor units (2×2).

[0151] (2) Working principle: Each sensor unit contains two upper and lower electrode plates. External pressure causes the contact area between the electrolyte layer microstructure and the electrodes to change, and temperature causes the dielectric constant and ion migration speed to change, thereby changing the capacitance value.

[0152] (3) Signal output: Each sensor unit outputs two capacitance signals through the upper and lower electrode plates for subsequent circuit processing.

[0153] 2. Multiplexing circuit

[0154] (1) Composition: Two 70HC4051 multiplexing chips are used to process the upper and lower electrode plate signals respectively.

[0155] (2) Connection method:

[0156] (a) The four input pins of the upper electrode plate multiplexing chip are connected to the upper electrode plate of each sensor unit in the array, and the output pin is connected to the IN0 channel of FDC2214.

[0157] (b) The four input pins of the lower electrode plate multiplexing chip are connected to the lower electrode plate of each sensor unit, and the output pin is connected to the IN1 channel of FDC2214.

[0158] (3) Function: The STM32F407 outputs a 2-bit binary control signal (e.g., 00 to 11) through the GPIO pins, selects a specific sensor unit, and transmits its capacitance signal to the FDC2214. The multiplexing circuit uses a polling mechanism to sequentially switch the four sensor units to ensure comprehensive signal acquisition.

[0159] 3. Capacitance-to-digital conversion circuit

[0160] (1) Core components: FDC2214 capacitance acquisition chip, combined with external LC oscillator circuit.

[0161] (2) Connection relationship:

[0162] (a) The IN0 and IN1 channels of the FDC2214 receive the upper and lower electrode plate signals, respectively, from the multiplexing circuit. The excitation frequencies for the IN0 and IN1 channels are set to 1 MHz and 1.2 MHz, respectively, to avoid crosstalk.

[0163] (b) Each channel is connected to an LC oscillating circuit consisting of a fixed inductor L (e.g., 10 μH) and a capacitor C provided by the sensor unit.

[0164] (3) Working principle:

[0165] (a) The FDC2214 applies an excitation signal from 10kHz to 10MHz to the LC circuit, generating an oscillation frequency.

[0166] (b) When the sensor unit capacitance C changes, the oscillation frequency f changes accordingly. The FDC2214 measures this frequency using an internal high-precision counter and converts it into a 28-bit digital signal, which indirectly reflects the capacitance value.

[0167] 4. STM32F407 microcontroller

[0168] (1) Function: As a microcontroller unit (MCU), it is responsible for system coordination, data processing and communication.

[0169] (2) Connection method:

[0170] (a) Communicate with FDC2214 through the I2C interface and receive digital signals.

[0171] (b) The multiplexing circuit is controlled by GPIO pins.

[0172] (c) Connect to the host computer through the RS-485 interface to transmit the processing results.

[0173] (3) Function:

[0174] (a) Configure the operating parameters of FDC2214 (such as excitation frequency and sampling time).

[0175] (b) Control the polling sequence and cycle of the multiplexing circuit.

[0176] (c) Process the digital signal, calculate the double-layer capacitance value, and transmit it to the host computer via RS-485 for subsequent temperature and force decoupling.

[0177] 5. RS-485 communication circuit

[0178] (1) Function: To realize long-distance, anti-interference data transmission between STM32F407 and host computer.

[0179] (2) Connection: Convert the signal to RS-485 through the UART interface of STM32F407 and connect it to the RS-485 port of the host computer.

[0180] 2. Signal Processing Flow

[0181] The complete workflow of the signal processing section is as follows, ensuring efficiency and accuracy from signal acquisition to data output:

[0182] 1. Signal Acquisition

[0183] (1) The STM32F407 outputs a 2-bit address signal (00 to 11), which is used to select four double-layer capacitors in a 2×2 array in sequence through a multiplexing circuit.

[0184] (2) The upper and lower electrode plate signals of the selected sensor unit are transmitted to the IN0 and IN1 channels of FDC2214 respectively.

[0185] 2. Capacitance-to-digital conversion

[0186] (1) FDC2214 provides excitation signal for the LC oscillator circuit of each channel, with a frequency range of 10kHz to 10MHz.

[0187] (2) Capacitance changes cause oscillation frequency changes. FDC2214 uses the formula Calculate the capacitance value and output a 28-bit digital signal, which is transmitted to the STM32F407 through the I2C interface.

[0188] 3. Data Processing

[0189] STM32F407 optimizes the received digital signal and uses a sliding average filter algorithm to eliminate noise and improve data stability;

[0190] 4. Data Transmission

[0191] The STM32F407 transmits the processed data (including sensor unit number, timestamp, capacitance value, etc.) to the host computer through the RS-485 interface at a rate of 9600bps.

[0192] The FDC2214 capacitive acquisition chip mentioned above only provides two active channels, while a 2×2 array needs to process signals from four sensor units. The multiplexing circuit achieves channel expansion through time-sharing acquisition, avoiding hardware redundancy. This results in an acquisition time of approximately 10ms per sensor unit and a total polling period of 40ms, ensuring real-time performance.

[0193] The temperature-compensated force decoupling algorithm accurately distinguishes capacitance responses caused by temperature changes from those caused by external force during real-time measurement. It then compensates for the temperature effect on force measurements through real-time compensation. This allows the sensor to operate stably in complex environments and produce highly accurate temperature and force data.

[0194] The capacitor with microstructure (denoted as C with ) exhibits a complex nonlinear response under the combined effects of temperature (T) and pressure (P), the causes of which are as follows:

[0195] 1.1 Effect of temperature on capacitance of non-microstructured parts

[0196] Experimental data show that Figure 3, the capacitor C without microstructure no It has a linear relationship with temperature T. Based on this linear relationship, a mathematical model is established:

[0197] C no =a·T+b

[0198] in:

[0199] T is temperature in degrees Celsius (℃);

[0200] a is the linear rate of change of capacitance with temperature, in F / ℃;

[0201] b is the initial capacitance value at 0°C, in F.

[0202] The specific values ​​of a and b can be determined by experimental calibration. Figure 3 As an example of the non-microstructure capacitor used, at 20°C, C no =5.98992×10 -10 F; at 70℃, C no =3.97033×10 -8 F. can be calculated.

[0203] a=7.820862×10 -10 F / ℃

[0204] b=-1.5042732×10 -8 F

[0205] Therefore, the linear relationship can be approximated as:

[0206] C no =7.820862×10 -10 T-1.5042732×10 -8

[0207] 1.2 Solving Temperature Through Capacitance

[0208] In the above 2×2 sensor array, there are two capacitors without microstructures (denoted as C 01 and C 02 ), whose average capacitance is defined as Calculate C 0_avg After the value of , the temperature can be calculated by inverse formula:

[0209]

[0210] For example, if C is calculated 0_avg =2.0×10 -8 F, then:

[0211]

[0212] This method utilizes the linear response characteristics of the capacitance of the non-microstructured part to provide a reliable basis for temperature calibration.

[0213] 1.3 Effect of temperature on capacitance of microstructured parts

[0214] For the capacitor C with microstructure with The influence mechanism of temperature is similar to that of the part without microstructure, and still shows a linear relationship. with The response with temperature can be expressed as:

[0215] C with =a′·T+b′

[0216] in:

[0217] a′ is the rate of temperature change, in F / °C;

[0218] b′ is the initial capacitance value at 0°C, in F.

[0219] The specific values ​​of a′ and b′ are determined through experimental calibration, and although the microstructure may adjust the slope and intercept of the linear model, its overall response remains linear.

[0220] 1.3 Effect of pressure on the capacitance of microstructured parts

[0221] Pressure affects C by changing the contact area of ​​the double layer capacitor (i.e. the contact area A between the electrolyte layer and the electrode). with , and its relationship follows the classic capacitance formula:

[0222]

[0223] in:

[0224] ∈0 is the dielectric constant of vacuum;

[0225] ∈ r is the relative dielectric constant;

[0226] A(P) is the contact area that changes with pressure;

[0227] d is the distance between the electrodes (assumed to remain constant under pressure).

[0228] like Figure 7 As shown in the experimental data, the pressure has a great influence on the C with The effect is approximately linear.

[0229] 1.4 Effect of temperature and pressure coupling on the capacitance of microstructured parts

[0230] Although the effects of temperature and pressure on capacitance under their own action can be approximated to be linear, the coupling of the two makes C withIt exhibits a significant nonlinear response, which may be caused by:

[0231] (1) Temperature dependence of microstructure mechanical properties: The elastic modulus of PVA gel decreases with increasing temperature, and the deformation becomes larger at high temperature, resulting in an enhanced capacitance response to pressure;

[0232] (2) Interaction between thermal expansion and deformation: Temperature increase causes thermal expansion of the microstructure, which changes the initial contact area or electrode spacing and affects the pressure response;

[0233] (3) Dielectric constant interaction: Temperature and pressure synergistically change ∈ r , the regulating effect of pressure on dielectric constant is more significant at high temperature;

[0234] (4) Dynamic behavior of microstructure: Phase transition or irreversible deformation may occur under high temperature and high pressure, introducing path dependence or hysteresis effects.

[0235] In summary, C with There is a complex nonlinear relationship between the response of the fluid and temperature and pressure. A single linear model is difficult to accurately describe its behavior, and advanced modeling methods need to be introduced to compensate for it.

[0236] Based on this, the force decoupling algorithm based on temperature compensation is used as the temperature and pressure signal decoupling algorithm, aiming to accurately separate the influence of temperature and pressure, thereby ensuring high precision of pressure measurement. The electrolyte layer film of the ionized multimodal tactile sensor is composed of microstructured areas and non-microstructured areas arranged at intervals, wherein the microstructured areas are mainly used to respond to external forces (pressure), and the non-microstructured areas are used to respond to temperature changes. Since temperature also affects the capacitance response of the microstructured areas, this algorithm effectively eliminates the interference of temperature on pressure measurement through multi-step signal processing, regression modeling and data fusion methods. The specific decoupling method steps of the force decoupling algorithm based on temperature compensation are as follows:

[0237] 1. Model Architecture

[0238] A multi-layer perceptron (MLP) neural network is used for regression modeling to accurately predict pressure P from the capacitance signal. The model architecture includes an input layer, a hidden layer, and an output layer. The design details are as follows:

[0239] (1) Input layer

[0240] Accepts the following two input features:

[0241] C 0\_avg : The average capacitance signal of the part without microstructure, calculated as It mainly reflects the linear response caused by temperature changes and provides a reference benchmark for temperature effects;

[0242] C m\_avg: The average capacitance signal of the microstructure part is calculated as (In the 2×2 sensor array designed by the present invention, there are two capacitors with microstructures, denoted as C m1 and C m2 ), which contains nonlinear response information under the combined effects of temperature and pressure.

[0243] (2) Hidden layer

[0244] A two-layer hidden structure is used to fully capture the nonlinear relationship between input features:

[0245] (a) The first hidden layer: contains 64 neurons, using the ReLU activation function f(x) = max(0, x) to introduce nonlinear transformation and enhance the C 0_avg with C m_avg Ability to model complex relationships.

[0246] (b) The second hidden layer: contains 32 neurons and also uses the ReLU activation function to further refine the deep features output by the first hidden layer, while compressing the features to reduce redundant information.

[0247] (3) Output layer

[0248] A single neuron is set to directly output the predicted pressure P. As a regression task, the output layer uses linear output without activation function to ensure the continuity between the predicted value and the physical quantity.

[0249] 2. Data Preparation

[0250] To support model training and validation, the system collects and preprocesses input data. The specific steps are as follows:

[0251] (1) Data collection scope

[0252] (a) Temperature range: 20°C to 70°C (1°C increment);

[0253] (b) Pressure range: 0 kPa to 500 kPa (10 kPa step).

[0254] At each combination of temperature and pressure, measurements of four capacitors in the 2×2 sensor array were collected: two capacitors without microstructures, C 01 、C 02 and two microstructured capacitors C m1 、C m2 . Then calculate the average and To obtain more stable and accurate input features and ensure that the data covers the typical operating range of the tactile sensor.

[0255] (2) Data preprocessing

[0256] C 0_avg and C m_avg Normalization is performed separately, and the formula is as follows:

[0257]

[0258] Where C represents the original capacitance value, C min and C max are the corresponding minimum and maximum values ​​in the dataset, respectively. After normalization, both are mapped to the interval [0, 1], effectively eliminating feature scale differences and improving the stability of model training. Furthermore, outliers are detected and removed, and interpolation and smoothing (such as moving average) are used to reduce the interference of random noise.

[0259] 3. Training Methods

[0260] In order to enable the model to effectively learn the mapping relationship between capacitance signals and pressure, the present invention adopts the following training strategy:

[0261] (1) Loss function

[0262] The mean square error (MSE) is used as the loss function, which is defined as:

[0263]

[0264] Among them, P pred，i To predict pressure, P true，i is the actual pressure, and N is the number of samples. The square penalty mechanism of the mean square error helps drive the model closer to the true value.

[0265] (2) Optimizer

[0266] The Adam optimizer is used, with an initial learning rate of 0.001. Adam combines the advantages of the momentum method and RMSProp, accelerating model convergence by adaptively adjusting the learning rate. The specific parameter configuration is as follows:

[0267] (a) First-order momentum decay factor β1 = 0.9;

[0268] (b) Second-order momentum decay factor β2 = 0.999;

[0269] (c) Numerical stability factor ∈ = 1 × 10 -8 .

[0270] (3) Batch training

[0271] The Mini-Batch gradient descent method is used, and 32 samples are randomly selected for training each time to balance computational efficiency and memory usage.

[0272] (4) Learning rate scheduling

[0273] Implementing an exponential decay strategy: η t =η0·e -λt

[0274] Where, the initial learning rate η0 = 0.001, λ is the decay rate (e.g., 0.05), and t is the number of training rounds. This strategy accelerates learning in the early stages and ensures smooth convergence in the later stages.

[0275] (5) Early stopping method

[0276] Monitor the validation set loss. If the validation loss does not decrease after 10 consecutive epochs, terminate the training early to prevent the model from overfitting.

[0277] 4. Hyperparameter Optimization

[0278] To further improve model performance, this paper uses grid search to optimize key hyperparameters. The optimization parameters include:

[0279] (1) Learning rate: candidate values ​​{0.0001, 0.001, 0.01};

[0280] (2) Number of neurons in the first hidden layer: candidate values ​​{32, 64, 128};

[0281] (3) Number of neurons in the second hidden layer: candidate values ​​{16, 32, 64};

[0282] (4) Batch size: candidate values ​​{16, 32, 64}.

[0283] A total of 81 configuration combinations were considered. After preliminary evaluation of the average MSE value on the validation set, the optimal configuration was determined to be: learning rate 0.001; 64 neurons in the first hidden layer; 32 neurons in the second hidden layer; and batch size 32.

[0284] 5. Model Validation

[0285] To ensure the generalization ability and stability of the model, the following verification methods are used:

[0286] (1) Cross-validation

[0287] A 5-fold cross-validation was implemented, randomly dividing the dataset into 5 subsets, selecting 4 subsets for training and 1 subset for validation each time. The root mean square error (RMSE) was used as the evaluation metric:

[0288]

[0289] The goal is to control the relative error on the validation set to within 5% and the RMSE to be lower than 10kPa.

[0290] (2) Verification process

[0291] Each cross-validation uses the same model architecture and training strategy, and records the loss value and prediction deviation of each fold to ensure that the model responds stably to changes in temperature and pressure.

[0292] 6. Real-time Application

[0293] In practical applications, the sensor collects signals from four capacitors in a 2×2 array in real time: C 01 、C 02 、C m1 、C m2 First calculate the average and After preprocessing, the data is fed into a trained neural network model. The model's forward propagation time is less than 1 millisecond, and the corresponding pressure P is output in real time. This compensation algorithm is suitable for high-precision pressure measurement under various temperature and pressure conditions, such as in wearable devices and industrial tactile systems.

[0294] The above-mentioned force decoupling algorithm based on temperature compensation has the following technical advantages:

[0295] (1) High-precision design: Through two hidden layers and ReLU activation function, the model can accurately capture C with It significantly outperforms traditional linear compensation methods by taking into account the complex nonlinear characteristics in the system.

[0296] (2) Robustness guarantee: By averaging two capacitors with microstructures and two capacitors without microstructures in a 2×2 array, noise and spatial variability are reduced. The algorithm is trained over a wide range of 20°C to 70°C and 0kPa to 500kPa, showing good generalization ability and adaptability, and can operate stably in complex environments.

[0297] (3) Flexible scalability: The model architecture and hyperparameters can be adjusted according to different sensor types or measurement requirements, and have good adaptability and scalability.

[0298] The working principle of the above-mentioned ion-type multimodal tactile sensor is as follows:

[0299] The sensor detects changes in external temperature and pressure through changes in capacitance response values. The electrolyte layer in the sensor is designed with alternating microstructured areas and non-microstructured areas. The microstructured areas primarily respond to changes in external force, while the non-microstructured areas primarily respond to changes in temperature and are used to decouple the temperature and correction force measurements. By decoupling the capacitance response, the sensor can achieve dual-modal sensing of temperature and force and provide stable measurement results in complex environments. Specifically:

[0300] When the sensor is subjected to external force or temperature changes, the capacitance response value of the microstructure area will change significantly. This change is mainly due to the change in contact area caused by the external force.

[0301] In regions without microstructures, since their contact area remains virtually unchanged under pressure, their response to external forces is minimal, and their capacitance response is almost entirely due to temperature changes. Temperature changes affect the dielectric properties of the electrolyte and ion migration, leading to changes in the double-layer capacitance. This capacitance change is measured via electrical signals transmitted through the leads between the upper and lower electrodes.

[0302] Through this structural design, the sensor can compensate for force measurements while measuring temperature changes, thereby achieving dual-modal perception and decoupling of temperature and force, and enhancing the accuracy and stability of the sensor in complex environments.

[0303] In summary, in the above-mentioned ionized multimodal tactile sensor, the upper and lower electrodes, together with the electrolyte layer, form a double-layer capacitor structure for measuring capacitance changes. Changes in capacitance response are transmitted through the electrode lead signal, reflecting changes in external force and temperature. The microstructured area responds significantly to external force, and its capacitance value changes are mainly caused by changes in contact area. The non-microstructured area is almost unaffected by external force and mainly responds to temperature changes. Its capacitance response is almost entirely caused by temperature.

[0304] The above-mentioned ion-type multimodal tactile sensor uses two-photon printing technology to manufacture a mold with alternating arrangements of microstructured areas and "non-microstructured areas." The pre-designed high-precision model is directly imported into the printer for mold making. The mold processing accuracy in the XY direction (horizontal plane) reaches 1μm, and the processing accuracy in the Z direction (vertical direction) reaches 0.6μm. This process ensures the precise molding of the microstructured areas and non-microstructured areas on the mold. The electrolyte layer film is then prepared through the demolding process, realizing the precise manufacturing of the complex structure of the film surface. The high-precision characteristics of the mold enable the size, spacing and shape of the microstructures in the microstructured areas of the electrolyte layer film to be precisely controlled to meet the needs of different applications.

[0305] By using a microstructure-free area to measure temperature, the ionized multimodal tactile sensor's capacitive response is entirely driven by temperature changes, accurately compensating for temperature interference on force measurements. This solves the problem of temperature fluctuations interfering with force measurements in existing technologies. This design enables the sensor to operate stably in complex environments with large temperature fluctuations, maintaining high sensitivity and measurement accuracy while simultaneously responding to external forces and temperature changes.

[0306] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. An ionotropic multimodal tactile sensor based on polyelectrolyte interface regulation, characterized by: It includes multiple temperature-pressure dual-mode sensing units, each of which is composed of an upper electrode, an electrolyte layer and a lower electrode in sequence. The electrolyte layer is a single-layer film made of polyelectrolyte material, and its surface is arranged with microstructured areas and non-microstructured areas at intervals.

2. The ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to claim 1, characterized in that: The microstructure of the microstructured area is formed by a micro-nanostructure array that can change the contact area between the electrode and the electrolyte layer when under pressure, and the material of the electrolyte layer film is a polyelectrolyte material that can form a double-layer capacitor between the electrolyte layer and the electrode.

3. The ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to claim 2, characterized in that: The structure of the micro-nano structure array is cone, pyramid, prism or hemisphere; the polyelectrolyte material capable of forming a double-layer capacitor between the electrolyte layer and the electrode is PVA gel added with phosphoric acid, chemically modified PVA gel containing ionic groups, polyelectrolyte polymer or ion gel.

4. The ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to claim 3, characterized in that: The PVA gel with added phosphoric acid is prepared from 1g of PVA particles, 5-10g of water and 700-900μl of phosphoric acid; the chemically modified PVA gel containing ionic groups is prepared by copolymerizing PVA with monomers containing ionic groups or introducing ionic groups through chemical reaction; the ion gel is prepared by mixing a polymer gel with an electrolyte salt or an ionic liquid; and the polyelectrolyte polymer is a polyelectrolyte.

5. The ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to claim 1, characterized in that: The electrode array structure of the upper electrode and the lower electrode is the same, and both are composed of m×n electrodes; the upper electrode is equipped with m lead wires in the same direction, and the lower electrode is equipped with n lead wires in the same direction, and the directions of the upper lead wires and the lower lead wires are perpendicular to each other, with an angle of 90°; the electrolyte layer is a thin film made of polyelectrolyte material, and there are m×n sensing areas on the interface of the film, some of which are provided with microstructures, and the remaining areas are planar areas without microstructures, and the areas with microstructures and the areas without microstructures are arranged at intervals. The m×n sensing areas on the electrolyte layer film correspond one-to-one with the m×n electrodes of the upper electrode and the lower electrode in position. During the packaging process, the m×n electrodes of the upper electrode and the lower electrode are precisely aligned with the m×n sensing areas on the electrolyte layer film to form m×n double-layer capacitors.

6. The ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to claim 5, characterized in that: The temperature-pressure dual-mode sensing unit is a basic sensing unit in the overall sensor, consisting of an upper electrode, a lower electrode and a corresponding electrolyte layer. The upper electrode and the lower electrode each contain p×q electrodes, and the electrolyte layer has p×q sensing areas corresponding to the electrodes, wherein some sensing areas are provided with microstructures, and the remaining sensing areas are planar areas without microstructures, and the sensing areas with microstructures are arranged alternately with the sensing areas without microstructures. During the packaging process, the p×q electrodes of the upper electrode and the lower electrode are precisely aligned with the p×q sensing areas on the electrolyte layer to form p×q double-layer capacitors.

7. The ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to claim 6, characterized in that: In the temperature-pressure dual-mode sensing unit, p=q=2. Among the 2×2 sensing areas of the electrolyte layer, two sensing areas on the diagonal line are provided with microstructures, and the other two sensing areas are planar areas without microstructures.

8. The ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to claim 6, characterized in that: In one of the basic temperature-pressure dual-mode sensing units, the pressure sensing unit is a microstructured area, which uses the change in the contact area of ​​the double-layer capacitor when under pressure to cause a change in the capacitance response; the temperature sensing unit is a planar area without microstructures, which uses the change in ion mobility of the polyelectrolyte material in the electrolyte layer caused by temperature changes to cause a change in the capacitance response. The temperature sensing unit has an extremely low response to pressure; the decoupling process is: first detect the temperature, then use the capacitance response generated by temperature sensing to temperature compensate the capacitance response of the pressure sensor, and decouple the capacitance response difference caused by temperature and pressure through a force decoupling algorithm based on temperature compensation to ensure the stability and accuracy of the sensor in complex environments and achieve accurate decoupling of temperature and pressure data.

9. The method for preparing an ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to any one of claims 1 to 8, characterized in that: The specific steps are as follows: (1) Preparation of film materials: Ion gel films are made from polyelectrolyte materials; (2) Patterned design and manufacturing of electrolyte layer thin film: using two-photon printing, 3D printing, photolithography, soft lithography, laser direct writing or inkjet printing technology to manufacture a mold, the mold surface has a pattern with microstructured areas and non-microstructured areas arranged alternately; preparing an intermediate mold by a PDMS demolding process, the intermediate mold surface has a pattern corresponding to the mold; casting a polyelectrolyte material on the intermediate mold, and demolding after curing to prepare an electrolyte layer thin film, the film surface has a pattern with microstructured areas and non-microstructured areas arranged alternately corresponding to the mold; (3) Assembly and packaging of electrode layers: Assemble and package the electrode layer and electrolyte layer; (4) Implementation of temperature and force decoupling: The capacitance response value is collected by a capacitance acquisition chip and the data is processed using a force decoupling algorithm based on temperature compensation.

10. The decoupling method of an ionized multimodal tactile sensor based on polyelectrolyte interface regulation according to any one of claims 1 to 8, characterized in that: A force decoupling algorithm based on temperature compensation is adopted to measure the temperature by measuring the capacitance response of the area without microstructures. The capacitance response of the area with microstructures is compensated by using the temperature or directly using the capacitance response of the area without microstructures, thereby achieving decoupling of temperature and pressure to ensure the stability and accuracy of the sensor in complex environments.

Images