Touch screen virtual key feedback method, device and equipment

Abstract

The invention relates to the technical field of touch screens, and discloses a touch screen virtual key feedback method, device and equipment, and the method comprises the steps: carrying out the normalization processing of touch data collected by a touch screen, obtaining a standardized position matrix and a standardized pressure matrix, and constructing a touch response matrix; performing singular value decomposition and reconstruction operation on the touch response matrix to obtain a reconstruction matrix; classifying the touch points according to the reconstruction matrix to obtain a touch track, and performing region area calculation and pressure distribution center calculation to obtain a touch feature descriptor; establishing a target touch response equation according to the touch feature descriptors; performing signal linearization processing and interference compensation to obtain a compensated touch signal; according to the method, feedback parameters are calculated, stability constraint calculation is conducted on the feedback parameters, virtual key feedback signals are output, accurate control over the virtual key feedback intensity is achieved, and the sense of reality and operation experience of tactile feedback are improved.

Description

Technical Field

[0001] The present invention relates to the field of touch screen technology, and in particular to a touch screen virtual key feedback method, device and equipment. Background Art

[0002] With the widespread application of touch screen technology, virtual buttons have attracted much attention as an important human-computer interaction method. However, traditional touch screen virtual buttons generally have problems such as lack of tactile feedback and poor operating experience, which seriously affect the user's operating accuracy and usage experience.

[0003] The current touch screen virtual button feedback solutions on the market mainly rely on simple vibration feedback, which cannot provide accurate tactile response based on the user's touch characteristics. At the same time, due to problems such as signal interference and data loss during touch data collection, the touch response is unstable, affecting the tactile feedback effect of the virtual button. Summary of the invention

[0004] The present invention provides a touch screen virtual key feedback method, device and equipment, which realizes accurate control of virtual key feedback intensity and improves the realism of tactile feedback and operating experience.

[0005] In a first aspect, the present invention provides a touch screen virtual key feedback method, the touch screen virtual key feedback method comprising: Normalizing the touch data collected by the touch screen to obtain a standardized position matrix and a standardized pressure matrix, and constructing a touch response matrix; Performing singular value decomposition and reconstruction operations on the touch response matrix to obtain a reconstructed matrix; Classifying the touch points according to the reconstruction matrix to obtain a touch trajectory, and performing area calculation and pressure distribution center calculation on the touch trajectory to obtain a touch feature descriptor; Establishing a target touch response equation according to the touch feature descriptor; Performing signal linearization processing and interference compensation based on the target touch response equation to obtain a compensated touch signal; A feedback parameter is calculated based on the compensated touch signal, and a stability constraint calculation is performed on the feedback parameter to output a virtual key feedback signal.

[0006] In a second aspect, the present invention provides a touch screen virtual key feedback device, the touch screen virtual key feedback device comprising: A normalization processing module is used to perform normalization processing on the touch data collected by the touch screen to obtain a standardized position matrix and a standardized pressure matrix, and to construct a touch response matrix; A reconstruction operation module, used for performing singular value decomposition and reconstruction operation on the touch response matrix to obtain a reconstructed matrix; A calculation module, used for classifying the touch points according to the reconstruction matrix to obtain a touch track, and performing area calculation and pressure distribution center calculation on the touch track to obtain a touch feature descriptor; An establishing module, used for establishing a target touch response equation according to the touch feature descriptor; An interference compensation module, used for performing signal linearization processing and interference compensation based on the target touch response equation to obtain a compensated touch signal; The output module is used to calculate feedback parameters based on the compensated touch signal, perform stability constraint calculation on the feedback parameters, and output a virtual key feedback signal.

[0007] The third aspect of the present invention provides a touch screen virtual key feedback device, comprising: a memory and at least one processor, wherein the memory stores instructions; the at least one processor calls the instructions in the memory so that the touch screen virtual key feedback device executes the above-mentioned touch screen virtual key feedback method.

[0008] A fourth aspect of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the above-mentioned touch screen virtual key feedback method.

[0009] In the technical solution provided by the present invention, the problem of missing data in the touch data acquisition process is effectively solved by normalizing the touch data and reconstructing the touch data through singular value decomposition, thereby improving the integrity and reliability of the touch data; the touch trajectory multi-point recognition and area calculation method are adopted to accurately extract the touch feature parameters, and the accurate recognition of the user's touch intention is realized; a response dynamic equation based on the touch feature component is established, and the optimal parameter set is obtained by combining the optimization algorithm to ensure the accuracy of the virtual key feedback; the influence of external interference factors such as electromagnetic interference, temperature drift and mechanical vibration is effectively suppressed through signal linearization processing and interference compensation model; the feedback control strategy based on stability constraint is adopted to ensure the stability and reliability of the virtual key feedback; the amplitude modulation and frequency modulation of the tactile waveform are used to achieve accurate control of the virtual key feedback intensity, thereby improving the realism and operation experience of the tactile feedback.

[0010] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0011] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic diagram of an embodiment of a touch screen virtual key feedback method in an embodiment of the present invention; Figure 2 A schematic diagram of an embodiment of a touch screen virtual key feedback device in an embodiment of the present invention; Figure 3 Schematic diagram of an embodiment of a touch screen virtual key feedback device in an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0014] The terms "including" and "having" and any variations thereof mentioned in the embodiments of the present invention are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device end including a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or device ends.

[0015] To facilitate understanding of this embodiment, a touch screen virtual key feedback method disclosed in an embodiment of the present invention is first described in detail. Figure 1 As shown, the method comprises the following steps: 101. Normalize the touch data collected by the touch screen to obtain a standardized position matrix and a standardized pressure matrix, and construct a touch response matrix; It is understandable that the execution subject of the present invention may be a touch screen virtual key feedback device, or a terminal or a server, which is not limited here. The embodiment of the present invention is described by taking a server as the execution subject as an example.

[0016] Specifically, the touch data of the touch screen is collected. The touch data includes touch position coordinate data and touch pressure value data, which are from the sensor on the touch screen or the corresponding detection device. The touch position coordinate data represents the specific position of the touch point on the screen in the form of two-dimensional coordinates, while the touch pressure value data represents the pressure applied to the touch point, which is a scalar value. The maximum and minimum values ​​of the touch position coordinate data are extracted to obtain the value range of the touch data, that is, to find the farthest point and the closest point of the touch position coordinate. By extracting the extreme value data of the touch position coordinate, the area range where the touch operation occurs is obtained. The touch position coordinate data is interval mapped using the position coordinate extreme value data. The values ​​of the touch position coordinates are mapped to a unified standardized range, usually the [0,1] interval, to obtain a standardized position matrix. The maximum and minimum values ​​of the touch pressure value data are extracted to determine the maximum pressure and the minimum pressure applied on the touch screen, so as to standardize the pressure value. The touch pressure value is interval mapped using the extreme value data of the pressure value, and it is standardized to the [0,1] interval, so as to ensure that the pressure data on different touch points are compared on the same scale, and a standardized pressure matrix is ​​obtained. Perform matrix concatenation on the standardized position matrix and the standardized pressure matrix to obtain a concatenated response matrix, which contains both the position information and pressure information of the touch points. Perform transposition and data normalization on the concatenated response matrix. The purpose of the matrix transposition operation is to exchange the rows and columns of the matrix so that each touch sampling moment corresponds to the row vector of the matrix, and the column vector of the matrix corresponds to different characteristic parameters of the touch. The data normalization operation is to scale all the data in the matrix to a uniform range, usually [0,1], to obtain a touch response matrix. The row vector of the touch response matrix represents the moment of touch sampling, while the column vector corresponds to the characteristic parameters of the touch, such as touch position, touch pressure, etc.

[0017] 102. Perform singular value decomposition and reconstruction operations on the touch response matrix to obtain a reconstructed matrix; Specifically, the touch response matrix input is subjected to matrix decomposition operation by means of a singular value decomposition algorithm. Singular value decomposition is a method of decomposing a matrix into three relatively simple matrices, which can help extract the main characteristic information of the matrix. The singular value decomposition algorithm decomposes the touch response matrix into three parts: a left singular matrix, a singular value diagonal matrix, and a right singular matrix. The left singular matrix represents the basis vectors of the data in the original space, while the right singular matrix represents the basis vectors of the data in the target space, and the diagonal elements in the singular value diagonal matrix represent the weight or importance of the original data on these basis vectors. Through decomposition, each component of the touch response matrix can be accurately expressed as a linear combination of these basis vectors. The singular value diagonal matrix is ​​processed. All diagonal elements in the singular value diagonal matrix are extracted, and these elements are singular values, which represent the importance of each feature in the matrix. The singular values ​​are arranged in descending order to obtain a sorted singular value sequence. The contribution ratio of each singular value in the total data is analyzed by calculating the cumulative contribution rate, and a cumulative contribution rate curve is obtained. Based on the cumulative contribution rate curve, a preset contribution rate threshold is set, and the truncation position of the singular value is determined according to the threshold. The sorted singular value sequence is screened, and the singular values ​​less than the truncation threshold are set to zero to obtain the screened singular value diagonal matrix. The screened singular value diagonal matrix is ​​multiplied with the left singular matrix by matrix multiplication to obtain the left reconstruction matrix. The left reconstruction matrix is ​​multiplied with the right singular matrix to obtain the reconstructed response matrix. The reconstructed response matrix is ​​compared with the original touch response matrix, and the mean square error is used to measure the difference between them. The purpose of the mean square error calculation is to evaluate the similarity between the reconstructed matrix and the original matrix. The smaller the error, the better the reconstruction effect and the less data loss. If the reconstruction error is large, it means that important feature information is lost during the reconstruction process. The quality of the reconstruction is judged by calculating the mean square error value. According to the preset error threshold, the validity of the calculated reconstruction error value is judged. If the reconstruction error is lower than the set threshold, it means that the reconstructed matrix can effectively restore the characteristics of the original matrix, and the reconstruction operation is considered to be effective. If the reconstruction error exceeds the threshold, it means that the reconstruction quality does not meet the standard and the truncation threshold or reconstruction strategy needs to be readjusted. Based on this judgment result, a reconstruction validity mark is assigned to the reconstructed response matrix. If the reconstruction validity mark is valid, the reconstructed response matrix is ​​data corrected to improve the reconstruction accuracy and finally obtain the reconstructed matrix. The data correction process includes correcting the outliers in the matrix or fine-tuning the values ​​in the matrix to ensure that the reconstructed matrix reflects the characteristics of the original touch response matrix as accurately as possible.

[0018] 103. Classify the touch points according to the reconstruction matrix to obtain a touch trajectory, and calculate the area and pressure distribution center of the touch trajectory to obtain a touch feature descriptor; Specifically, the touch point coordinates in the reconstructed matrix are analyzed for time continuity. By studying the temporal change relationship of the touch points, the touch point data adjacent in time are found to reflect the dynamic process of the touch operation. By calculating the spatial distance of the adjacent touch point data, the distance matrix of the touch points is obtained to reflect the distribution of the touch points in space, thereby helping to understand the overall shape and change trend of the touch trajectory. The touch points are clustered according to the preset distance threshold, and the distance threshold defines the maximum allowable distance between adjacent touch points. Through clustering analysis, the touch points are divided into several different groups, and each group of touch points represents a continuous touch trajectory. The touch point clustering results are subjected to trajectory marking operation to identify the starting point, the end point and the change direction of each trajectory to obtain the touch trajectory. The convex hull calculation is performed on the touch points in the touch trajectory to find the minimum area enclosed by the touch trajectory and obtain the contour point set of the touch area. The polygon area calculation is performed using the contour point set to obtain the area data of the touch area and quantify the size of the touch area. The pressure value data in the touch track is spatially mapped, and the pressure value on the touch point is matched with its spatial position on the screen to generate a pressure distribution matrix, which reflects the pressure distribution of the touch point at different positions. The centroid of the pressure distribution matrix is ​​calculated. The center position of the pressure distribution in the touch track, that is, the "center of gravity" of the pressure distribution, is found. This position reflects the concentrated area of ​​the touch operation and is an important indicator for judging the user's touch strength and position. The coordinates of the center of the pressure distribution are obtained by calculating the centroid of the pressure distribution matrix. According to the touch area area data and the coordinates of the center of the pressure distribution, a touch feature vector is constructed to obtain an initial feature vector. The initial feature vector is normalized. Different eigenvalues ​​are converted to a unified scale range to eliminate the difference in dimension or value size to obtain a standardized feature vector. The standardized feature vector is weighted according to the preset feature weight. The influence of each feature on the virtual key feedback system is evaluated. Through the weighted operation, the features that are more important to the system response are emphasized, while other minor features are suppressed. The weighted feature vector is more in line with the needs of practical applications and can better reflect the user's touch behavior and needs. The weighted feature vector is input into the feature encoder for feature encoding operation to obtain feature encoding data. A touch feature code is generated based on the feature coding data, and the touch feature code is combined with the touch trajectory parameters to obtain a touch feature descriptor. The touch feature descriptor contains multi-dimensional information such as the spatial position, area size, pressure distribution, and time characteristics of the touch.

[0019] 104. Establish a target touch response equation according to the touch feature descriptor; Specifically, the touch feature descriptor is subjected to feature decomposition operation to obtain multiple component sets of touch features, including a touch area feature matrix , pressure distribution characteristic matrix , touch timing feature vector And the touch trajectory feature vector Each feature component reflects different aspects of touch behavior, for example, the touch area feature matrix Describes the distribution of touch on the plane, pressure distribution feature matrix reflects the pressure distribution during touch, and the touch timing feature vector Represents the time variation trend of touch behavior, touch trajectory feature vector Then the trajectory information of the touch point is captured. The initial touch response equation is established based on the touch feature component set The initial touch response equation combines the contribution of each characteristic component to the touch response by weighted integration. The form of the equation is: ; in, is the feature weight coefficient, which indicates the relative importance of each feature component in the touch response; and are Gaussian weighted functions for touch area and pressure distribution, and is the spatial scale parameter, controlling the width of the Gaussian function; Indicates the timing characteristics of touch, is the time decay factor, is the time attenuation coefficient, which controls the attenuation speed of the touch signal; is the spatial gradient norm of the touch trajectory, reflecting the rate of change of the trajectory; It is an environmental noise function that describes the error and noise caused by the external environment or equipment. In order to build a more accurate touch response equation, the noise data sequence of the touch screen is collected to build the environmental noise function By analyzing these noise data, we can obtain the noise characteristics and substitute them into the initial touch response equation, calculate the noise response data, and obtain a touch response model that includes the influence of environmental noise. This noise model helps to improve the robustness of the touch response and enable it to better adapt to complex practical application scenarios. In order to optimize the initial touch response equation, we construct an optimization objective function The objective function is to minimize the touch response equation and the ideal response function The error between the two while maintaining the smoothness and stability of the touch response. The objective function is as follows: ; in, Represents the L2 norm, which measures the error between the touch response equation and the ideal response equation. represents the L1 norm, which measures the smoothness of the gradient and second-order gradient of the touch response equation. and is the regularization coefficient, which is used to control the smoothness of the gradient and the second-order gradient. The optimization objective function comprehensively considers the accuracy and smoothness of the touch response to ensure that the system can avoid overfitting and noise while ensuring accuracy. After calculating the gradient of the optimization objective function, an optimization algorithm (such as gradient descent or other numerical optimization methods) is used to solve the optimal parameter set. ,in is the optimized feature weight coefficient, including and ,These weight coefficients are determined through optimization and can more accurately represent the contribution of each feature to the touch response; is the optimized spatial scale parameter, including and ,Optimizing these parameters helps adjust the width of the Gaussian function, making the touch response more consistent with the actual touch behavior; The optimized time attenuation coefficient can adjust the attenuation speed of the touch signal to better match the timing characteristics of the touch. Substitute the optimal parameter set into the initial touch response equation to obtain the target touch response equation , which is of the form: ; in, is the noise suppression coefficient, and , which is used to suppress the impact of noise on the final response and ensure that the system can maintain stability and accuracy when dealing with noise. The target touch response equation is used to simulate and feedback the user's touch operation.

[0020] 105. Perform signal linearization processing and interference compensation based on the target touch response equation to obtain a compensated touch signal; Specifically, the integral term in the target touch response equation F*(x, y, t) is discretized and sampled. The calculation of the touch area response matrix and the pressure distribution response matrix requires the spatial information of the touch area and the pressure distribution to be discretized so that it can be processed numerically. In order to discretize the integral term, a sampling method is used, such as dividing the spatial area of ​​the touch area and the pressure distribution into small grid units, and then calculating the response value on each grid unit to obtain a discrete touch area response matrix and a pressure distribution response matrix, which reflect the response characteristics of the touch area and the pressure distribution under specific touch conditions. The exponential decay term in the target touch response equation is piecewise linear interpolated to obtain a time response vector. Since the decay term in the touch response is time-dependent, an exponential decay function is used to describe the process of the touch signal decaying over time. The piecewise linear interpolation is used to make the decay function more accurately approximate the actual situation, and a time response vector is obtained, which represents the change of the touch response at different time points. The gradient term in the target touch response equation is linearly approximated to obtain a trajectory response vector. The gradient term represents the rate of change of the touch trajectory. The gradient of the touch trajectory is calculated by the linear approximation method to obtain the trajectory response vector that reflects the change of the touch trajectory, effectively capturing the spatial change characteristics of the touch trajectory. The touch area response matrix, pressure distribution response matrix, time response vector and trajectory response vector are weighted and combined according to the optimized weight coefficient to obtain the linear state equation. The interference observation matrix is ​​constructed based on the linear state equation. The interference observation matrix captures the impact of external interference caused by factors such as electromagnetic interference, temperature drift and mechanical vibration on the touch signal. Collect electromagnetic interference data, temperature drift data and mechanical vibration data, and input these data into the interference observation matrix for state estimation operation to obtain the interference state vector. The interference state vector can reflect the degree of influence of various interference factors in the current system. Perform frequency domain transformation operation on the interference state vector to obtain the interference spectrum characteristics. Frequency domain analysis can convert the interference signal from the time domain to the frequency domain, thereby more clearly revealing the spectrum characteristics of the interference signal. Through frequency domain transformation, the interference characteristics of different frequency components are identified. According to the interference spectrum characteristics, the corresponding compensation filter group is designed to filter out the unnecessary interference frequency components. By designing a compensation filter group, the interference compensation coefficient is obtained, and the interference compensation coefficient is used to construct a compensation matrix with the linear state equation to obtain an interference compensation model. The interference compensation model removes the influence of electromagnetic interference, temperature drift and mechanical vibration from the touch signal. The electromagnetic interference data, temperature drift data and mechanical vibration data are input into the interference compensation model for response compensation operation to obtain the compensated touch signal.

[0021] 106. Calculate feedback parameters based on the compensated touch signal, perform stability constraint calculation on the feedback parameters, and output a virtual key feedback signal.

[0022] Specifically, the compensated touch signal is input into the wavelet transformer for time-frequency domain decomposition operation. Wavelet transform is a tool that can simultaneously analyze the performance of a signal in the time domain and frequency domain. Through operation, the characteristics of the touch signal in different frequency bands and time scales are extracted to obtain the time-frequency decomposition characteristics. According to the time-frequency decomposition characteristics, the intensity of the touch signal is quantified and calculated to obtain the touch intensity feedback parameter, which reflects the strength of the touch signal. The touch intensity feedback parameter is subjected to state space modeling operation. By inputting the touch intensity feedback parameter into the state space modeling algorithm, a feedback state vector describing the dynamic behavior of the feedback process is constructed. The feedback state vector contains the state information of the feedback system within a given time. Based on the feedback state vector, a tactile response dynamic equation is constructed to obtain the state equation of the feedback system. The state equation of the feedback system is subjected to characteristic root analysis operation. Through characteristic root analysis, the system characteristic equation is obtained, which reveals the inherent characteristics of the feedback system. The result of the characteristic root analysis is used to extract the stability criterion of the system. Stability is an important indicator of whether the touch feedback system can work normally, and the system characteristic equation provides a mathematical basis for analyzing the stability. Based on the stability criterion extracted from the system characteristic equation, the stability constraint boundary conditions are determined to ensure that the system operates stably within a certain parameter range. Based on the stability constraint boundary conditions, the feedback state vector is subjected to pole configuration operation. Pole configuration is a method in control theory. By adjusting the pole position of the system, the response of the system is more in line with expectations, especially in terms of stability and response speed. The pole configuration operation can help determine the appropriate pole position of the system, and calculate the feedback gain based on these results to obtain the tactile feedback gain parameter. The feedback gain parameter controls the intensity and effect of the feedback signal, and directly affects the perceived quality of the tactile feedback. The tactile feedback gain parameter is input into the controller for parameter optimization operation. Through the optimization algorithm, the optimal control parameter is obtained according to the performance requirements of the system. A tactile waveform template is generated according to the target control parameter to obtain a tactile waveform sequence containing the ideal tactile feedback signal characteristics. An amplitude modulation operation is performed on the tactile waveform sequence. According to the intensity requirement of the touch feedback, the amplitude of the tactile waveform is adjusted so that the final tactile signal can better simulate the actual touch feedback effect. The modulated tactile signal will make corresponding adjustments in amplitude to adapt to different tactile feedback intensity requirements. Frequency modulation calculation is performed based on the modulated tactile signal. Frequency modulation is to adjust the frequency component of the tactile waveform in time to achieve different tactile feedback effects. Through frequency modulation, the perception effect of the tactile signal is optimized to make it more in line with user needs. The modulated tactile signal will have certain frequency characteristics and can produce different tactile perception effects on the physical device. The modulated tactile drive signal undergoes power amplification operation to amplify the tactile signal to a sufficient voltage or current level to drive the actuator to produce actual physical feedback.After the amplification operation is completed, the drive circuit is matched to ensure that the amplified signal can match the drive requirements of the actuator. Through circuit matching, the effect of the drive signal is maximized to ensure that the actuator can accurately respond to the tactile signal. Based on the matched drive signal, the tactile actuator is effectively driven and controlled to output a virtual key feedback signal. This signal is converted into actual tactile feedback through the actuator, and the user can perceive the tactile effect related to the virtual key through the touch screen, thereby enhancing the interactive experience of the virtual key.

[0023] In the embodiment of the present invention, by normalizing and reconstructing the touch data through singular value decomposition, the problem of missing data in the touch data acquisition process is effectively solved, and the integrity and reliability of the touch data are improved; by adopting the touch trajectory multi-point recognition and area calculation method, the touch feature parameters are accurately extracted, and the accurate recognition of the user's touch intention is realized; a response dynamic equation based on the touch feature component is established, and the optimal parameter set is obtained by combining the optimization algorithm, thereby ensuring the accuracy of the virtual key feedback; through signal linearization processing and interference compensation model, the influence of external interference factors such as electromagnetic interference, temperature drift and mechanical vibration is effectively suppressed; by adopting a feedback control strategy based on stability constraints, the stability and reliability of the virtual key feedback are ensured; through the amplitude modulation and frequency modulation of the tactile waveform, the precise control of the virtual key feedback intensity is realized, and the realism and operation experience of the tactile feedback are improved.

[0024] In a specific embodiment, the process of executing step 101 may specifically include the following steps: Collecting touch data of the touch screen, the touch data including touch position coordinate data and touch pressure value data; Extracting the maximum and minimum values ​​of the touch position coordinate data to obtain position coordinate extreme value data, and performing interval mapping operation on the touch position coordinate data according to the position coordinate extreme value data to obtain a standardized position matrix; Extracting the maximum and minimum values ​​of the touch pressure value data to obtain extreme pressure value data, and performing interval mapping operation on the touch pressure value data according to the extreme pressure value data to obtain a standardized pressure matrix; A matrix splicing operation is performed on the standardized position matrix and the standardized pressure matrix to obtain a splicing response matrix, and a matrix transposition operation and a data normalization operation are performed on the splicing response matrix to obtain a touch response matrix, wherein the row vectors of the touch response matrix correspond to the touch sampling moments, and the column vectors correspond to the touch feature parameters.

[0025] Specifically, the touch data of the touch screen is collected, and the touch data includes touch position coordinate data and touch pressure value data. These data are collected in real time by the sensor of the touch screen and recorded when a touch event occurs. The touch position coordinate data is represented by two-dimensional coordinates, usually the X and Y coordinates on the screen; the touch pressure value data reflects the strength or pressure of the touch, which is obtained through the pressure sensing layer of the sensor. The maximum and minimum values ​​of the touch position coordinate data are extracted. Assume that the touch position coordinate data is a series of two-dimensional coordinate points , where each pair of coordinates Indicates the position of a touch point. For the X and Y coordinates, calculate their maximum and minimum values ​​respectively: ; ; The maximum and minimum values ​​are the extreme value data of the position coordinates. and . Perform interval mapping operation on the touch position coordinate data according to the extreme value data, and map the original data to a standardized range, usually the [0,1] interval, for better subsequent processing. For the X coordinate and Y coordinate of each touch point, the standardized calculation formula is: ; ; Through the standardization process, the standardized position matrix is ​​obtained and , eliminating the differences in position coordinate scales between different devices and different touch events. Extract the maximum and minimum values ​​of the touch pressure value data. The touch pressure data is , represents the pressure value measured at each touch point. Calculate the maximum and minimum pressure values: ; Then, the pressure values ​​are mapped to intervals based on these extreme value data. The standardized calculation formula for pressure data is: ; Get the standardized pressure matrix . Perform matrix concatenation operation on the standardized touch position matrix and the standardized pressure matrix. Integrate the position and pressure information into a unified response matrix. Concatenate the standardized position matrix and the standardized pressure matrix by column to obtain a concatenated response matrix , which is of the form: ; The row vectors of the matrix represent different touch sampling moments, and the column vectors correspond to different characteristic parameters of touch - position and pressure. For each touch point, its response information includes the position coordinates and pressure value. Each row of the spliced ​​response matrix is ​​a complete data set of touch sampling. Since these data will change over time, the matrix Perform a transposition operation to obtain a column vector, which represents the matrix of each touch feature parameter. The transposed matrix is ​​expressed as: ; in, represents the transpose of the normalized position matrix, Represents the transpose of the standardized pressure matrix. Through matrix transposition and data normalization operations, all touch feature parameters are represented on a unified scale. is regarded as the final touch response matrix. In this matrix, the row vector corresponds to the touch sampling moment, and the column vector corresponds to different touch feature parameters.

[0026] In a specific embodiment, the process of executing step 102 may specifically include the following steps: Performing matrix decomposition operation on the touch response matrix input by using a singular value decomposition algorithm to obtain a left singular matrix, a singular value diagonal matrix, and a right singular matrix; Perform a diagonal element extraction operation on the singular value diagonal matrix to obtain a singular value sequence, and perform a descending sorting operation on the singular value sequence to obtain a sorted singular value sequence; The cumulative contribution rate is calculated based on the sorted singular value sequence to obtain a cumulative contribution rate curve, and the truncation position of the cumulative contribution rate curve is determined according to a preset contribution rate threshold to obtain a singular value truncation threshold; The singular values ​​in the sorted singular value sequence that are smaller than the singular value cutoff threshold are set to zero to obtain a screened singular value diagonal matrix; Perform matrix multiplication operation on the screened singular value diagonal matrix and the left singular matrix to obtain the left reconstruction matrix, and perform matrix multiplication operation on the left reconstruction matrix and the right singular matrix to obtain the reconstructed response matrix; Calculating the mean square error between the reconstructed response matrix and the touch response matrix to obtain a reconstruction error value, and judging the validity of the reconstruction error value according to a preset error threshold to obtain a reconstruction validity mark; A data correction operation is performed on the reconstructed response matrix according to the reconstruction validity mark to obtain a reconstructed matrix.

[0027] Specifically, the touch response matrix input is decomposed by a singular value decomposition algorithm. Given a touch response matrix (Assuming a The matrix of Represents the touch sampling moment, represents the number of touch features), and decomposes it into three matrices through the singular value decomposition algorithm: the left singular matrix , singular value diagonal matrix and the right singular matrix ,Right now: ; in, is a An orthogonal matrix containing the touch response matrix The left singular vector of ; is a A diagonal matrix of , where the elements on the diagonal are singular values; is a An orthogonal matrix containing the touch response matrix The right singular vectors of . Singular values Arranged by size, used to represent matrices The main components or important features of the singular value diagonal matrix are extracted to obtain the singular value sequence. In the singular value decomposition process, the diagonal elements of the singular value diagonal matrix Each element in represents the weight of a different component of the matrix. By extracting these singular values , and get a sequence of singular values ​​sorted by size: ; The singular values ​​are arranged in descending order, namely: ; The cumulative contribution rate is calculated based on the sorted singular value sequence. The cumulative contribution rate indicates the contribution of each singular value to the total amount of information. The cumulative contribution rate is calculated using the following formula: ; in, Before The cumulative contribution rate of singular values ​​is It is singular values. According to the preset contribution rate threshold , determine a cutoff position , that is, when the cumulative contribution rate curve reaches the threshold, stop considering subsequent singular values. , only keep the front singular values, which are considered to contribute most to the explanatory power of the matrix. Based on the cutoff position, the sorted singular value sequence is processed and the singular values ​​less than the cutoff threshold are set to zero. Assume is the determined cutoff position, for each , its singular value Set to zero to get the filtered singular value diagonal matrix : ; The selected singular value diagonal matrix With left singular matrix Perform matrix multiplication to obtain the left reconstructed matrix : ; Reconstruct the matrix on the left With right singular matrix Perform matrix multiplication to obtain the reconstructed response matrix : ; Reconstructing the response matrix It is a low-rank approximation matrix after singular value screening, which approximates the original touch response matrix by retaining the most important singular values. In order to evaluate the effectiveness of the reconstructed matrix, the reconstruction error is calculated. The reconstruction error is measured using the mean square error. and the reconstructed response matrix , the mean square error is calculated by the following formula: ; in, is a matrix Middle Line The elements of the column, is a matrix The smaller the mean square error, the closer the reconstruction result is to the original matrix, indicating that the reconstruction effect is better. According to the calculated mean square error value, it is compared with the preset error threshold to determine whether the reconstruction is valid. If the reconstruction error is less than the preset error threshold, the reconstruction is marked as valid; if the reconstruction error is greater than the error threshold, it is marked as invalid and further data correction is required. If the reconstruction is valid, use the reconstructed response matrix As the final touch response matrix. In the case of invalid reconstruction, data correction operation is performed to improve the accuracy of reconstruction.

[0028] In a specific embodiment, the process of executing step 103 may specifically include the following steps: Performing time continuity analysis on the touch point coordinates in the reconstructed matrix to obtain time-adjacent touch point data, and performing spatial distance calculation based on the time-adjacent touch point data to obtain a touch point distance matrix; Performing cluster analysis on the touch point distance matrix according to a preset distance threshold to obtain a touch point clustering result, and performing a trajectory marking operation on the touch point clustering result to obtain a touch trajectory; Calculate the convex hull of the touch points in the touch trajectory to obtain a touch area contour point set, and calculate the polygon area based on the touch area contour point set to obtain touch area area data; Perform spatial distribution mapping on the pressure value data in the touch trajectory to obtain a pressure distribution matrix, and calculate the centroid of the pressure distribution matrix to obtain the center coordinates of the pressure distribution; Construct a feature vector based on the touch area data and the center coordinates of the pressure distribution to obtain an initial feature vector; Normalizing the initial feature vector to obtain a standardized feature vector, and performing a weighted operation on the standardized feature vector according to a preset feature weight to obtain a weighted feature vector; The weighted feature vector is input into a feature encoder for feature encoding operation to obtain feature encoding data, and a touch feature code is generated according to the feature encoding data. The touch feature code is combined with a touch trajectory parameter to obtain a touch feature descriptor.

[0029] Specifically, the touch point coordinates in the reconstructed matrix are analyzed for time continuity. The touch response matrix is ​​continuous in time, and each row represents the touch position coordinate data at a touch sampling moment. The touch point data adjacent in time are extracted, and the spatial distance is calculated based on the touch points adjacent in time. Assume that at time and Two touch point locations are collected and , its spatial distance Calculated by Euclidean distance: ; By calculating the spatial distance of all time-adjacent touch points in the reconstructed matrix, the distance matrix D of the touch points is obtained. Each item in the matrix Indicates time and The spatial distance between two touch points. Based on the touch point distance matrix, cluster analysis is performed to identify the correlation between touch points. For example, suppose there is a preset distance threshold , cluster all adjacent touch points using a clustering algorithm (such as DBSCAN or K-means algorithm). If the distance between two points is less than , then these two points are considered to be of the same class. Through this clustering method, the clustering results of the touch points are obtained, that is, the clustering labels corresponding to each touch point. The touch point clustering results are marked with trajectories. In the clustering results, the trajectory to which each touch point belongs is marked according to the time sequence to obtain the touch trajectory data. The touch trajectory is a path composed of a series of continuous touch points, and the marked trajectory is used for subsequent feature extraction. The convex hull of the touch points in the touch trajectory is calculated to find the minimum convex hull of all touch points in the touch trajectory to form a closed area around all touch points. Assume that there are Touch points , use the convex hull algorithm to calculate the boundary of the convex hull. The boundary point set of the convex hull is The contour of the touch area is formed. The polygon area is calculated based on the contour point set of the touch area. For the calculation of the polygon area, the area formula of the polygon (Shoelace formula) is used to calculate the area of ​​the touch area. : ; in, is the number of points of the touch area outline, is each point of the convex hull. The area value reflects the size of the touch area. At the same time, the pressure value data in the touch track is spatially mapped to obtain the pressure distribution matrix , whose elements represent the pressure values ​​of the touch points at different positions. By calculating the centroid of the pressure distribution matrix, the coordinates of the pressure distribution center are obtained. , and its calculation formula is: ; in, It is The pressure value of each touch point, is the corresponding position coordinate. By calculation, the pressure distribution center of the touch area is determined. Using the touch area data and the center coordinates of the pressure distribution Constructing the initial feature vector , which contains the characteristic information about the touch area and the center of pressure distribution. The initial characteristic vector is expressed as: ; The initial eigenvectors are normalized to ensure that different eigenvalues ​​are compared on the same scale. The normalization formula is: ; in, and are the minimum and maximum values ​​of each component in the eigenvector, respectively. The standardized eigenvector eliminates the influence of different feature dimensions, so that they are processed on the same scale. The standardized eigenvector is weighted according to the preset feature weights to obtain the weighted eigenvector Assume that the feature weight is , then the calculation formula of the weighted eigenvector is: ; The weighted feature vector reflects the importance of each feature and improves the accuracy of the touch feature descriptor. The weighted feature vector is input into the feature encoder for feature encoding operation to obtain feature encoding data. The feature encoder uses different encoding methods, such as Huffman coding or other adaptive coding methods. Based on the feature encoding data e, a touch feature code is generated. , and combine the touch feature code with the touch trajectory parameters (such as the number of touch points, touch time, etc.) to obtain the touch feature descriptor : ; Through this step, effective touch features are extracted from the reconstructed touch response matrix, and touch feature descriptors are generated using these features for touch analysis or virtual key feedback generation.

[0030] In a specific embodiment, the process of executing step 104 may specifically include the following steps: Performing feature decomposition operation on the touch feature descriptor to obtain a touch feature component set, the touch feature component set includes a touch area feature matrix A(x, y), a pressure distribution feature matrix P(x, y), a touch timing feature vector T(t) and a touch trajectory feature vector M(x, y, t); The initial touch response equation F(x, y, t) is established based on the touch feature component set. The initial touch response equation F(x, y, t) is: ; Where: x, y are the coordinate values ​​of the rectangular coordinate system of the touch plane; t is the touch sampling time point; is the touch center coordinate; is the feature weight coefficient; G is the Gaussian kernel function, is the spatial scale parameter; λ is the time attenuation coefficient; represents the spatial gradient norm of the touch trajectory; ε(x, y, t) is the environmental noise function; Collect the touch screen noise data sequence to construct the environmental noise function ε(x, y, t), substitute the environmental noise function into the initial touch response equation to obtain the noise response data; Construct the optimization objective function: ; in: is the ideal response function; represents the L2 norm, represents the L1 norm; is the regularization coefficient, J(w,σ,λ) represents the optimization objective function, w represents the feature weight, and σ represents the spatial scale; Perform gradient calculation and parameter optimization on the optimization objective function to obtain the optimal parameter set ,in, Expressed as the optimized feature weight coefficient, include , is the optimized spatial scale parameter, include , is the optimized time decay coefficient; Substitute the optimal parameter set into the initial touch response equation to obtain the target touch response equation , target touch response equation for: ; Where k is the noise suppression coefficient, and 0 <k<1。

[0031] Specifically, the touch feature descriptor is decomposed and the initial touch response equation is established based on the touch feature component set. The touch screen response is optimized through mathematical modeling to ensure the accuracy and stability of the feedback. The touch feature descriptor includes four main feature components: touch area feature matrix , pressure distribution characteristic matrix , touch timing feature vector and touch trajectory feature vector , where each feature represents an important aspect of touch behavior. In order to extract these features, the touch data collected by the touch screen is decomposed to obtain these feature components. Touch area feature matrix Indicates the distribution of the touch area on the plane, which is represented by the distribution of touch points at different locations on the touch screen. Pressure distribution feature matrix It describes the pressure value distribution at different positions in the touch area, reflecting the pressure intensity applied by the touch point. Touch timing feature vector Indicates the change of touch in time, which is related to the start duration and end time of the touch event. Touch trajectory feature vector The motion trajectory of the touch point on the touch screen is captured. The initial touch response equation F(x, y, t) is established based on the touch feature component set to describe the relationship between the touch response and the touch feature. The initial touch response equation is in the form of: ; in, and Represents the coordinates of the touch plane in the rectangular coordinate system, Indicates the time point of touch sampling, are the coordinates of the touch center, is the feature weight coefficient, which indicates the importance of each feature in the response. and is a Gaussian kernel function, which is used to smooth the touch area and pressure distribution data, and is a parameter that controls the spatial scale. is the time decay coefficient, which controls the decay of the effect of time, represents the spatial gradient norm of the touch trajectory, capturing the rate of change of the trajectory, is the environmental noise function, which describes the impact of environmental factors on touch response. In order to optimize the response equation, the noise data sequence of the touch screen is collected to construct the environmental noise function By analyzing the actual use environment of the touch screen, these noise data sequences are substituted into the initial touch response equation to obtain noise response data, thereby modeling the noise of the system so that subsequent optimization can take environmental interference into account. Construct an optimization objective function. The optimization objective function aims to minimize the error. By adjusting the parameters Minimize the difference between the touch response and the ideal response. The form of the optimization objective function is: ; in, is the ideal response function, which represents the expected touch response; and Represent the L2 norm and L1 norm, respectively, used to measure error and gradient; and is the regularization coefficient, which controls the weight of the gradient and second-order derivative terms. Through this objective function, the difference between the touch response and the ideal response is minimized, and the smoothness and continuity of the response are constrained. In order to solve the optimization problem, the objective function is gradient calculated and parameter optimized to obtain the optimal parameter set .in, Represents the optimized feature weight coefficient, including ,and Represents the optimized spatial scale parameters, including and , is the optimized time attenuation coefficient. Substitute the optimal parameters into the initial touch response equation to obtain the target touch response equation : ; in, is the noise suppression coefficient, and , used to reduce the impact of environmental noise on touch response.

[0032] In a specific embodiment, the process of executing step 105 may specifically include the following steps: Target touch response equation The integral term in is discretized and sampled to obtain the touch area response matrix and the pressure distribution response matrix; Perform piecewise linear interpolation operation on the exponential decay term in the target touch response equation to obtain a time response vector, and perform linear approximation operation on the gradient term in the target touch response equation to obtain a trajectory response vector; Performing weighted combination operation on the touch area response matrix, the pressure distribution response matrix, the time response vector and the trajectory response vector according to the optimized weight coefficient to obtain a linear state equation; An interference observation matrix is ​​constructed according to the linear state equation, and electromagnetic interference data, temperature drift data and mechanical vibration data are input into the interference observation matrix for state estimation operation to obtain an interference state vector; Perform frequency domain transformation on the interference state vector to obtain interference spectrum characteristics, and design a compensation filter group according to the interference spectrum characteristics to obtain interference compensation coefficients; The interference compensation coefficient and the linear state equation are used to perform compensation matrix construction operation to obtain an interference compensation model, and the electromagnetic interference data, temperature drift data and mechanical vibration data are input into the interference compensation model for response compensation operation to obtain a compensated touch signal.

[0033] Specifically, the target touch response equation The integral term in is discretized and sampled, and the integral term of the target touch response equation is expressed as a weighted sum of the touch area and the pressure distribution, in the following form: ; in, is the touch area matrix, is the pressure distribution matrix, and is a Gaussian kernel function, which is used to smooth the touch area and pressure distribution data. and Controls the spatial scale of the Gaussian kernel. For the discretization of these integral terms, within a given touch region and The coordinates are discretized into a finite number of points, thus converting the continuous integral into a discrete summation. For example, given the touch area matrix , discretized sampling is performed using the following formula: ; in, represents the discretized touch area response matrix, and are the discretized coordinates within the touch area, and They are and In a similar way, the pressure distribution response matrix is ​​obtained: , thus obtaining the touch area response matrix and the pressure distribution response matrix. Perform piecewise linear interpolation on the exponential decay term in the target touch response equation to obtain the time response vector. The exponential decay term in the target touch response equation is expressed as the influence of the time decay factor: ; in, is the touch timing feature vector, describing the change of touch behavior over time. represents exponential decay, is the attenuation coefficient. To calculate the time response of this term, the time is segmented and linearly interpolated. Assume that the time vector Discretized into a series of time points , then the response at each time point is calculated by piecewise linear interpolation: ; The resulting time response vector With the attenuation factor Multiply them to get the attenuated time response vector. Ensure that the response characteristics of the touch screen in different time periods can be accurately simulated. Perform linear approximation on the gradient term in the target touch response equation. The gradient term in the target touch response equation represents the rate of change of the touch trajectory, in the form of: ; in, is the touch trajectory feature vector, describing the trajectory of the touch point in space and time. To simplify the calculation, the gradient term is linearly approximated. Assume that the touch trajectory data is in discrete Coordinate points and time points It is known that the gradient is calculated by the finite difference method: ; in, are the sampling intervals of space and time respectively. Through the above approximation, the trajectory response vector is obtained and added to the target touch response equation. The touch area response matrix, pressure distribution response matrix, time response vector and trajectory response vector are weighted and combined according to the optimized weight coefficient to obtain the linear state equation: ; Based on the linear state equation, the interference observation matrix is ​​constructed , which is used to estimate the interference factors in the system. The interference comes from electromagnetic interference, temperature drift and mechanical vibration. These interference data are input into the interference observation matrix for state estimation. Assume that the electromagnetic interference data is , temperature drift data is , the mechanical vibration data is , then the state estimation is performed through Kalman filtering or other filtering algorithms: ; in, is the estimated interference state vector. Perform frequency domain transformation on the interference state vector to obtain the interference spectrum characteristics. Use fast Fourier transform or other spectrum analysis methods to extract the spectrum characteristics of the interference signal and then design the compensation filter group , get the interference compensation coefficient , the interference signal is processed by the filter bank: ; The interference compensation coefficient and the linear state equation are used to construct the compensation matrix to obtain the interference compensation model: ; Through this step, the influence of electromagnetic interference, temperature drift and mechanical vibration is eliminated. The interference compensated signal is input into the response compensation operation to obtain the compensated touch signal. The specific calculation method is: ; in, is the ambient noise function, and the final compensated touch signal That is the accurate touch response signal after interference compensation and noise processing.

[0034] In a specific embodiment, the process of executing step 106 may specifically include the following steps: The compensated touch signal is input into the wavelet transformer for time-frequency domain decomposition operation to obtain time-frequency decomposition features, and the touch signal strength is quantified and calculated according to the time-frequency decomposition features to obtain touch strength feedback parameters; Perform state space modeling operation on the touch intensity feedback parameter to obtain a feedback state vector, and construct a tactile response dynamic equation based on the feedback state vector to obtain a feedback system state equation; Perform characteristic root analysis on the state equation of the feedback system to obtain the system characteristic equation, and extract the stability criterion based on the system characteristic equation to obtain the stability constraint boundary condition; Performing pole configuration operation on the feedback state vector based on the stability constraint boundary condition to obtain a pole configuration result, and performing feedback gain calculation based on the pole configuration result to obtain a tactile feedback gain parameter; The tactile feedback gain parameter is input into the controller for parameter optimization operation to obtain the target control parameter, and a tactile waveform template is generated according to the target control parameter to obtain a tactile waveform sequence; Performing amplitude modulation operation on the tactile waveform sequence to obtain a modulated tactile signal, and performing frequency modulation calculation based on the modulated tactile signal to obtain a tactile driving signal; The tactile drive signal is amplified to obtain a power amplified signal, and the drive circuit is matched according to the power amplified signal to obtain a matched drive signal. The tactile actuator is driven and controlled based on the matched drive signal to obtain a virtual key feedback signal.

[0035] Specifically, the compensated touch signal Input to the wavelet transformer for time-frequency domain decomposition. Wavelet transform can convert the signal from the time domain to the time-frequency domain while maintaining local information in time and frequency. The definition formula of wavelet transform is: ; in, is the mother wavelet function, is the scale factor, is the translation factor, is the signal to be processed. Through wavelet transform, the time-frequency decomposition characteristics of the signal are obtained , so that the frequency components and time distribution of the signal can be accurately analyzed. According to the time-frequency decomposition characteristics, the intensity of the touch signal is quantified and calculated to obtain the touch intensity feedback parameter , the formula is as follows: ; in, Represents the energy of the wavelet coefficients, reflecting the intensity of the touch signal in a specific time-frequency region. Through this formula, the feedback parameter of the intensity of the touch signal changing over time is obtained . Feedback parameters for touch intensity Perform state space modeling. Assume that the state vector of the system is , the dynamic equation of the system is expressed as: ; in, is the state transition matrix of the system, is the control input matrix, is the touch intensity feedback parameter. Based on the state space equation, the tactile response dynamic equation is constructed: ; The dynamic equation describes the change of the system state over time and uses the change of the touch signal strength as the control input. The stability and responsiveness of the system are optimized by the state equation of the feedback system. In order to ensure the stability of the feedback system, the state equation of the feedback system is subjected to characteristic root analysis. By solving the characteristic equation of the system state equation, the characteristic root of the system is obtained. , that is, solve the equation: ; in, is the identity matrix, is the characteristic root. Solving the equation to obtain the characteristic root helps the system determine the stability of the system. According to the real part of the characteristic root, if the real part of all the characteristic roots is negative, the system is stable. Otherwise, the system has unstable behavior. According to the characteristic root of the system, the stability criterion is extracted and the stability constraint boundary conditions are obtained, such as: ; Constraints ensure the stability of the feedback system. Based on the stability constraint boundary conditions, the feedback state vector is pole-placed. By adjusting the pole position of the system, the stability of the system is ensured and its dynamic response is optimized. Assume that the state equation of the system is: ; in, is the input signal. The appropriate feedback gain matrix is ​​selected by pole placement method , so that the closed-loop poles of the system are located at the desired position. Calculated as: ; in, is the desired pole position, place It is a pole placement algorithm that returns the corresponding feedback gain matrix. Based on the feedback gain matrix , calculate the tactile feedback gain parameter , and optimize the control effect of the system according to this gain parameter. The input controller performs parameter optimization operations so that the system achieves the expected dynamic performance during the response process and minimizes energy loss. The optimization process is performed using optimization algorithms such as the least squares method or the gradient descent method. The optimization objective function is: ; in, and is the weight matrix, is the system status, is the control input. Through the optimization process, the target control parameters are obtained Based on these control parameters, a tactile waveform template is generated , which is used to drive the tactile actuator to generate feedback signals. Generated tactile waveform template After amplitude modulation operation, the modulated tactile signal is obtained , whose expression is: ; in, is the amplitude, is the frequency, is time. Based on the modulated tactile signal, frequency modulation calculation is performed to obtain the tactile driving signal , the formula is: ; in, is the modulated frequency, is the phase shift. The drive signal is used to control the tactile actuator. Amplify to obtain power amplified signal , the signal is amplified to the required drive level through the power amplifier. The formula is as follows: ; in, is the amplification factor. The power amplified signal is matched to the driving circuit to obtain the matched driving signal , through impedance matching and other circuit designs, the maximum power efficiency of signal transmission is ensured. Based on the matched driving signal, the tactile actuator is driven and controlled to generate a feedback signal of the virtual key.

[0036] The above describes the touch screen virtual key feedback method in the embodiment of the present invention. The following describes the touch screen virtual key feedback device in the embodiment of the present invention. Figure 2 , an embodiment of a touch screen virtual key feedback device in an embodiment of the present invention includes: A normalization processing module 201 is used to perform normalization processing on the touch data collected by the touch screen to obtain a standardized position matrix and a standardized pressure matrix, and to construct a touch response matrix; A reconstruction operation module 202, used for performing singular value decomposition and reconstruction operation on the touch response matrix to obtain a reconstructed matrix; The calculation module 203 is used to classify the touch points according to the reconstruction matrix to obtain the touch track, and calculate the area and pressure distribution center of the touch track to obtain the touch feature descriptor; Establishing module 204, used to establish a target touch response equation according to the touch feature descriptor; An interference compensation module 205 is used to perform signal linearization processing and interference compensation based on a target touch response equation to obtain a compensated touch signal; The output module 206 is used to calculate feedback parameters based on the compensated touch signal, perform stability constraint calculation on the feedback parameters, and output a virtual key feedback signal.

[0037] Through the coordinated cooperation of the above-mentioned components, the problem of missing data in the touch data acquisition process is effectively solved by normalizing the touch data and reconstructing it through singular value decomposition, thereby improving the integrity and reliability of the touch data; the touch trajectory multi-point recognition and area calculation method are adopted to accurately extract the touch feature parameters and realize the accurate recognition of the user's touch intention; a response dynamic equation based on the touch feature component is established, and the optimal parameter set is obtained by combining the optimization algorithm to ensure the accuracy of the virtual key feedback; through signal linearization processing and interference compensation model, the influence of external interference factors such as electromagnetic interference, temperature drift and mechanical vibration is effectively suppressed; a feedback control strategy based on stability constraints is adopted to ensure the stability and reliability of virtual key feedback; through amplitude modulation and frequency modulation of the tactile waveform, precise control of the virtual key feedback intensity is achieved, thereby improving the realism and operation experience of the tactile feedback.

[0038] above Figure 2 The touch screen virtual key feedback device in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The touch screen virtual key feedback device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0039] Figure 33 is a schematic diagram of the structure of a touch screen virtual key feedback device provided by an embodiment of the present invention. The touch screen virtual key feedback device 300 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 310 (for example, one or more processors) and a memory 320, and one or more storage media 330 (for example, one or more mass storage device terminals) storing application programs 333 or data 332. Among them, the memory 320 and the storage medium 330 can be short-term storage or permanent storage. The program stored in the storage medium 330 may include one or more modules (not shown in the figure), and each module may include a series of instruction operations in the touch screen virtual key feedback device 300. Furthermore, the processor 310 may be configured to communicate with the storage medium 330, and execute a series of instruction operations in the storage medium 330 on the touch screen virtual key feedback device 300 to implement the steps of the above-mentioned touch screen virtual key feedback method.

[0040] The touch screen virtual key feedback device 300 may also include one or more power supplies 340, one or more wired or wireless network interfaces 350, one or more input and output interfaces 360, and / or one or more operating systems 331, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Figure 3 The illustrated touch screen virtual key feedback device structure does not constitute a limitation on the touch screen virtual key feedback device provided by the present invention, and may include more or fewer components than illustrated, or a combination of certain components, or a different arrangement of components.

[0041] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the touch screen virtual key feedback method.

[0042] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, systems and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0043] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or the whole or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program code.

[0044] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A touch screen virtual key feedback method, characterized in that: The method comprises: Normalizing the touch data collected by the touch screen to obtain a standardized position matrix and a standardized pressure matrix, and constructing a touch response matrix; Performing singular value decomposition and reconstruction operations on the touch response matrix to obtain a reconstructed matrix; Classifying the touch points according to the reconstruction matrix to obtain a touch trajectory, and performing area calculation and pressure distribution center calculation on the touch trajectory to obtain a touch feature descriptor; Establishing a target touch response equation according to the touch feature descriptor; Performing signal linearization processing and interference compensation based on the target touch response equation to obtain a compensated touch signal; A feedback parameter is calculated based on the compensated touch signal, and a stability constraint calculation is performed on the feedback parameter to output a virtual key feedback signal.

2. The touch screen virtual key feedback method according to claim 1, characterized in that: The touch data collected by the touch screen is normalized to obtain a standardized position matrix and a standardized pressure matrix, and a touch response matrix is ​​constructed, including: Collecting touch data of the touch screen, wherein the touch data includes touch position coordinate data and touch pressure value data; Extracting the maximum and minimum values ​​of the touch position coordinate data to obtain position coordinate extreme value data, and performing interval mapping operation on the touch position coordinate data according to the position coordinate extreme value data to obtain a standardized position matrix; Extracting the maximum and minimum values ​​of the touch pressure value data to obtain extreme pressure value data, and performing interval mapping operation on the touch pressure value data according to the extreme pressure value data to obtain a standardized pressure matrix; A matrix splicing operation is performed on the standardized position matrix and the standardized pressure matrix to obtain a splicing response matrix, and a matrix transposition operation and a data normalization operation are performed on the splicing response matrix to obtain a touch response matrix, wherein the row vectors of the touch response matrix correspond to the touch sampling moments, and the column vectors correspond to the touch feature parameters.

3. The touch screen virtual key feedback method according to claim 2, characterized in that: The step of performing singular value decomposition and reconstruction operations on the touch response matrix to obtain a reconstructed matrix includes: Performing a matrix decomposition operation on the touch response matrix input by a singular value decomposition algorithm to obtain a left singular matrix, a singular value diagonal matrix, and a right singular matrix; Performing a diagonal element extraction operation on the singular value diagonal matrix to obtain a singular value sequence, and performing a descending order arrangement operation on the singular value sequence to obtain a sorted singular value sequence; Calculating the cumulative contribution rate based on the sorted singular value sequence to obtain a cumulative contribution rate curve, and determining a truncation position of the cumulative contribution rate curve according to a preset contribution rate threshold to obtain a singular value truncation threshold; Performing a zeroing operation on singular values ​​in the sorted singular value sequence that are smaller than the singular value cutoff threshold to obtain a screened singular value diagonal matrix; Performing a matrix multiplication operation on the screened singular value diagonal matrix and the left singular matrix to obtain a left reconstruction matrix, and performing a matrix multiplication operation on the left reconstruction matrix and the right singular matrix to obtain a reconstructed response matrix; Performing mean square error calculation on the reconstructed response matrix and the touch response matrix to obtain a reconstruction error value, and performing validity judgment on the reconstruction error value according to a preset error threshold to obtain a reconstruction validity mark; A data correction operation is performed on the reconstruction response matrix according to the reconstruction validity mark to obtain a reconstruction matrix.

4. The touch screen virtual key feedback method according to claim 3, characterized in that: The step of classifying the touch points according to the reconstruction matrix to obtain a touch track, and calculating the area and the pressure distribution center of the touch track to obtain a touch feature descriptor includes: Performing time continuity analysis on the touch point coordinates in the reconstructed matrix to obtain time-adjacent touch point data, and performing spatial distance calculation based on the time-adjacent touch point data to obtain a touch point distance matrix; Performing cluster analysis on the touch point distance matrix according to a preset distance threshold to obtain a touch point clustering result, and performing a track marking operation on the touch point clustering result to obtain a touch track; Performing convex hull calculation on the touch points in the touch trajectory to obtain a touch area contour point set, and performing polygon area calculation based on the touch area contour point set to obtain touch area area data; Performing spatial distribution mapping on the pressure value data in the touch trajectory to obtain a pressure distribution matrix, and performing centroid calculation on the pressure distribution matrix to obtain the pressure distribution center coordinates; Constructing a feature vector according to the touch area data and the center coordinates of the pressure distribution to obtain an initial feature vector; Normalizing the initial feature vector to obtain a standardized feature vector, and performing a weighted operation on the standardized feature vector according to a preset feature weight to obtain a weighted feature vector; The weighted feature vector is input into a feature encoder for feature encoding operation to obtain feature encoding data, and a touch feature code is generated according to the feature encoding data. The touch feature code is combined with a touch trajectory parameter to obtain a touch feature descriptor.

5. The touch screen virtual key feedback method according to claim 4, characterized in that: The step of establishing a target touch response equation according to the touch feature descriptor comprises: Performing feature decomposition operation on the touch feature descriptor to obtain a touch feature component set, wherein the touch feature component set includes a touch area feature matrix A(x, y), a pressure distribution feature matrix P(x, y), a touch timing feature vector T(t), and a touch trajectory feature vector M(x, y, t); An initial touch response equation F(x, y, t) is established based on the touch feature component set. The initial touch response equation F(x, y, t) is: ； Where: x, y are the coordinate values ​​of the rectangular coordinate system of the touch plane; t is the touch sampling time point; is the touch center coordinate; is the feature weight coefficient; G is the Gaussian kernel function, is the spatial scale parameter; λ is the time attenuation coefficient; represents the spatial gradient norm of the touch trajectory; ε(x, y, t) is the environmental noise function; Collecting a touch screen noise data sequence to construct an environmental noise function ε(x, y, t), substituting the environmental noise function into the initial touch response equation to obtain noise response data; Construct the optimization objective function: ； in: is the ideal response function; represents the L2 norm, represents the L1 norm; is the regularization coefficient, J(w,σ,λ) represents the optimization objective function, w represents the feature weight, and σ represents the spatial scale; Perform gradient calculation and parameter optimization on the optimization objective function to obtain the optimal parameter set ,in, Expressed as the optimized feature weight coefficient, include , is the optimized spatial scale parameter, include and , is the optimized time decay coefficient; Substitute the optimal parameter set into the initial touch response equation to obtain the target touch response equation , the target touch response equation for: ； Where k is the noise suppression coefficient, and 0 <k<1。 6. The touch screen virtual key feedback method according to claim 5, characterized in that: The performing signal linearization processing and interference compensation based on the target touch response equation to obtain a compensated touch signal includes: Performing a discretized sampling operation on the integral term in the target touch response equation F*(x, y, t) to obtain a touch area response matrix and a pressure distribution response matrix; Performing a piecewise linear interpolation operation on the exponential decay term in the target touch response equation to obtain a time response vector, and performing a linear approximation operation on the gradient term in the target touch response equation to obtain a trajectory response vector; Performing a weighted combination operation on the touch area response matrix, the pressure distribution response matrix, the time response vector and the trajectory response vector according to an optimized weight coefficient to obtain a linear state equation; Constructing an interference observation matrix according to the linear state equation, and inputting electromagnetic interference data, temperature drift data and mechanical vibration data into the interference observation matrix for state estimation operation to obtain an interference state vector; Performing frequency domain transformation operation on the interference state vector to obtain interference spectrum characteristics, and designing a compensation filter group according to the interference spectrum characteristics to obtain interference compensation coefficients; The interference compensation coefficient and the linear state equation are subjected to a compensation matrix construction operation to obtain an interference compensation model, and the electromagnetic interference data, the temperature drift data and the mechanical vibration data are input into the interference compensation model to perform a response compensation operation to obtain a compensated touch signal.

7. The touch screen virtual key feedback method according to claim 6, characterized in that: The step of calculating feedback parameters based on the compensated touch signal, performing stability constraint calculation on the feedback parameters, and outputting a virtual key feedback signal includes: Inputting the compensated touch signal into a wavelet transformer for time-frequency domain decomposition operation to obtain time-frequency decomposition features, and performing quantitative calculation on the touch signal strength according to the time-frequency decomposition features to obtain a touch strength feedback parameter; Performing state space modeling operation on the touch intensity feedback parameter to obtain a feedback state vector, and constructing a tactile response dynamic equation according to the feedback state vector to obtain a feedback system state equation; Performing characteristic root analysis on the state equation of the feedback system to obtain a system characteristic equation, and extracting stability criteria based on the system characteristic equation to obtain stability constraint boundary conditions; Performing a pole configuration operation on the feedback state vector based on the stability constraint boundary condition to obtain a pole configuration result, and performing a feedback gain calculation according to the pole configuration result to obtain a tactile feedback gain parameter; Inputting the tactile feedback gain parameter into the controller for parameter optimization calculation to obtain a target control parameter, and generating a tactile waveform template according to the target control parameter to obtain a tactile waveform sequence; Performing amplitude modulation operation on the tactile waveform sequence to obtain a modulated tactile signal, and performing frequency modulation calculation based on the modulated tactile signal to obtain a tactile driving signal; The tactile drive signal is amplified to obtain a power amplified signal, and a drive circuit is matched according to the power amplified signal to obtain a matched drive signal. The tactile actuator is driven and controlled based on the matched drive signal to obtain a virtual key feedback signal.

8. A touch screen virtual key feedback device, characterized in that: The device for executing the touch screen virtual key feedback method according to any one of claims 1 to 7 comprises: A normalization processing module is used to perform normalization processing on the touch data collected by the touch screen to obtain a standardized position matrix and a standardized pressure matrix, and to construct a touch response matrix; A reconstruction operation module, used for performing singular value decomposition and reconstruction operation on the touch response matrix to obtain a reconstructed matrix; A calculation module, used for classifying the touch points according to the reconstruction matrix to obtain a touch track, and performing area calculation and pressure distribution center calculation on the touch track to obtain a touch feature descriptor; An establishing module, used for establishing a target touch response equation according to the touch feature descriptor; An interference compensation module, used for performing signal linearization processing and interference compensation based on the target touch response equation to obtain a compensated touch signal; The output module is used to calculate feedback parameters based on the compensated touch signal, perform stability constraint calculation on the feedback parameters, and output a virtual key feedback signal.

9. A touch screen virtual key feedback device, characterized in that: The touch screen virtual key feedback device comprises: a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instruction in the memory to enable the touch screen virtual key feedback device to execute the touch screen virtual key feedback method according to any one of claims 1 to 7.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instruction is executed by the processor, the touch screen virtual key feedback method according to any one of claims 1 to 7 is implemented.

Images