Haptic feedback device and data format used therein

Abstract

The present invention discloses a tactile feedback device and a data format used therein. The tactile feedback device is configured to receive a target static force applied by a user at time T2, transmit the target static force to a cloud server, and receive a dynamic drive signal corresponding to the target static force returned by the cloud server. The dynamic drive signal is determined based on the target static force at time T2, a tactile sampling signal at time T1, and a preset model. The preset model is determined by machine learning based on a sample set of multiple sets of tactile sampling signals and dynamic drive signals constructed in advance. The tactile feedback device is further configured to perform an action based on the dynamic drive signal. Time T1 and time T2 are different times, thereby achieving asynchronous feedback of the tactile feedback device. The present invention achieves asynchronous tactile feedback through an asynchronous communication mechanism, making the tactile feedback reproduction more realistic and reliable.

Description

Technical Field

[0001] This invention is a divisional application of the original invention entitled "Asynchronous tactile feedback system and model training method thereof, computer device and medium", the application number of the original application is 202210139565.2, and the filing date of the original application is February 16, 2022. The present invention relates to the field of tactile feedback technology, and in particular to a tactile feedback device and the data format used therein. Background Art

[0002] With the rapid development of Internet technology, remote human-computer interaction systems based on network technology have developed rapidly in game development, smart cities, medical diagnosis and rescue and other fields.

[0003] Remote human-computer interaction systems typically rely on video and voice interaction, while the development and application of tactile feedback is still in its infancy. Existing remote tactile feedback typically uses electrical drive signals to drive motors or hydraulic devices, driving mechanical linkages to achieve tactile simulation, or uses electromechanical units to achieve tactile simulation.

[0004] Regardless of the principle, existing tactile feedback typically involves the actuating end first sending a touch signal, which is then transmitted to the receiving end. The receiving end then generates a feedback signal based on this signal and transmits it to the actuating end. The actuating end then generates a corresponding pressure signal based on the feedback signal, which the user perceives. To ensure the authenticity of the entire touch response, this process must be completed very quickly. That is, the correspondence between the touch signal and the feedback signal must be determined almost synchronously with the touch action. Otherwise, a significant delay will occur, significantly reducing authenticity. However, this is almost impossible to achieve. This is because not only do communications, sensors, and processors have time delays, but the actuator also needs to act on the remote touched object based on the touch signal, which is difficult to achieve without delay and can actually lead to inaccurate communication results. Furthermore, since each touch signal is unlikely to be completely consistent or have specific patterns, it is impossible to study the regularity of the feedback signal. In some cases, real-time communication is not necessary, and standardized and regular research is more important. For example, remote research on the properties of a touched object does not need to be carried out in real time; asynchronous transmission according to a protocol-specified format is sufficient, ensuring data standardization and compliance.

[0005] In addition, in the existing technology, most of the signals sent by the acquisition end (or receiving end) are used to directly drive the actuator to output action. However, due to the large differences in the properties of the actuators, even if the dynamic tactile sampling signals actually collected by the acquisition end are used for driving, the same tactile feeling cannot be obtained, which greatly affects the feedback accuracy and reproducibility of the feedback system.

[0006] In the prior art, the signal sent from the acquisition end to the reproduction end usually only includes the dynamic vibration signal, that is, the dynamic signal and the static signal are not standardized and packaged together and uploaded to the server. This cannot be used as a basis for subsequent data analysis and is not convenient for research. Summary of the Invention

[0007] The present invention provides a tactile feedback device and a data format used therein, so as to realize asynchronous transmission of tactile feedback signals and improve reproduction accuracy.

[0008] According to one aspect of the present invention, a tactile feedback device is provided for receiving a target static force applied by a user at time T2, sending the target static force to a cloud server, and receiving a dynamic drive signal corresponding to the target static force returned by the cloud server, wherein the dynamic drive signal is determined based on the target static force at time T2, the tactile sampling signal at time T1, and a preset model, and the preset model is determined by machine learning based on a sample set of multiple sets of tactile sampling signals and dynamic drive signals constructed in advance; the tactile feedback device is further configured to perform an action based on the dynamic drive signal, wherein time T1 and time T2 are different times, thereby achieving asynchronous feedback of the tactile feedback device.

[0009] According to another aspect of the present invention, a data format for a tactile feedback device is provided, including at least one of the following: each set of data includes a static tactile sampling signal A S and a dynamic tactile sampling signal A D (t), the static tactile sampling signal A S and the dynamic tactile sampling signal A D (t) There is a mapping relationship; the dynamic tactile sampling signal A D (t) The signal duration is at least 5 seconds; the static tactile sampling signal A S Meet the following conditions: 0 S <40KPa; the dynamic tactile sampling signal A D (t) satisfies the following conditions: 0 <max(A D (t))<50KPa.

[0010] According to another aspect of the present invention, there is provided a collection device for generating, transmitting and / or storing the above-mentioned data for the tactile feedback device.

[0011] According to another aspect of the present invention, a data set for a tactile feedback device is provided, comprising: at least three sets of the above data.

[0012] According to another aspect of the present invention, a data set for a tactile feedback device is provided, comprising: a static tactile sampling signal A S ​and dynamic tactile sampling signal A D (t), the static tactile sampling signal A S and the dynamic tactile sampling signal A D (t) There is a mapping relationship; the static tactile sampling signal A S Including first static tactile sampling signal data Second static tactile sampling signal data and the third static tactile sampling signal data in, and The following conditions must be met: as well as

[0013] The beneficial effects of the present invention are:

[0014] 1. By setting up an asynchronous communication mechanism between the acquisition end and the reproduction end, the entire system's dependence on the communication channel bandwidth and stability is reduced.

[0015] 2. By setting a standardized asynchronous communication data format, the tactile feedback reproduction is made more realistic, reliable, and statistically analyzable for research purposes.

[0016] 3. By establishing multiple relationships between multiple signals at the acquisition end and the reproduction end, especially by establishing a dedicated preset model, it is avoided to directly use the signal from the acquisition end (or its simple deformation) to drive the tactile reproduction end, so that the action ultimately applied to the user by the tactile reproduction end is more accurate and realistic.

[0017] 4. Through pre-set model training, the team creatively proposed placing the acquisition device directly on the surface of the tactile feedback device. By actively applying different dynamic drive signals to the tactile feedback device, the tactile sampling signals generated by the tactile feedback device are collected. The actively applied dynamic drive signals and the tactile sampling signals generated by the tactile feedback device are used as learning samples. This avoids the problem of constantly trying dynamic drive signals and the difficulty in simulating the tactile sensation of the target object. This simplifies the sample acquisition method, improves the accuracy and efficiency of sample acquisition, and ensures the accuracy of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a structural diagram of a tactile asynchronous feedback system provided in Example 1 of the present invention.

[0019] Figure 2 This is a structural diagram of a tactile asynchronous feedback system provided in Example 2 of the present invention.

[0020] Figure 3 This is a structural diagram of a tactile asynchronous feedback system provided in Example 3 of the present invention.

[0021] Figure 4 This is a flowchart of a model training method for a tactile asynchronous feedback system provided in Example 4 of the present invention.

[0022] Figure 5 This is a structural diagram of a computer device provided in Example 5 of the present invention. DETAILED DESCRIPTION

[0023] The present invention will be further described in detail below with reference to the accompanying drawings and examples. It will be understood that the specific embodiments described herein are intended only to illustrate the present invention and are not intended to limit the present invention. It should also be noted that, for ease of description, the accompanying drawings only illustrate portions relevant to the present invention, not all structures.

[0024] Figure 1 This is a schematic diagram of the structure of a tactile asynchronous feedback system provided by the first embodiment of the present invention. This embodiment is applicable to application scenarios where remote tactile reproduction is achieved through wireless communication technology. In particular, the sampling end and the reproduction end of the tactile asynchronous feedback system adopt asynchronous communication.

[0025] like Figure 1 As shown, the tactile asynchronous feedback system 00 includes: a cloud server 03, and a collection device 01 and a tactile feedback device 02 that are communicatively connected to the cloud server 03.

[0026] The acquisition device 01, located at the acquisition end, is used to collect tactile sampling signals generated by the target object at time T1 when different preset static forces N are applied to the target object. The acquisition device 01 normalizes the multiple sets of tactile sampling signals to form tactile sampling signal data, which is then transmitted to the cloud server 03 for the first communication. The tactile sampling signals include static tactile sampling signals and dynamic tactile sampling signals.

[0027] The tactile feedback device 02 is at the reproduction end and is used to receive the target static force P applied to the tactile feedback device 02 by the user at time T2 and send it to the cloud server 03. At the same time, it receives the dynamic drive signal corresponding to the target static force P returned by the cloud server 03, and takes action according to the dynamic drive signal to achieve the second communication.

[0028] The cloud server 03 stores the preset model and the tactile sampling signal data sent by the acquisition device, and is used to use the preset model and the tactile sampling signal data sent by the acquisition device, and obtain a dynamic driving signal corresponding to the target static force P according to the target static force P applied by the user sent by the tactile feedback device 02, and send it to the tactile feedback device 02.

[0029] The asynchronous communication process is as follows:

[0030] (1) Time T1: The acquisition device 01 acquires tactile sampling signals generated by the target object when different preset static forces N are applied to the target object. The acquisition device 01 normalizes the data of these multiple sets of tactile sampling signals and transmits them to the cloud server 03, completing the first communication feedback process. The tactile sampling signals include static tactile sampling signals and corresponding dynamic tactile sampling signals.

[0031] (2) Time T2: The tactile feedback device receives the target static force P applied by the user to the tactile feedback device at time T2, and sends the target static force P to the cloud server 03. The cloud server 03 obtains a dynamic drive signal corresponding to the target static force P applied by the user based on the preset model stored therein and the standard data corresponding to the multiple sets of static tactile sampling signals and dynamic tactile sampling signals sent by the above-mentioned acquisition device, and returns the dynamic drive signal to the tactile feedback device 02. The tactile feedback device 02 then acts according to the dynamic drive signal, thereby achieving the second communication feedback.

[0032] Here, T1 and T2 represent different time points or time periods. This means that the first and second communications are completed at different locations and times, without requiring an immediate response. This allows for asynchronous feedback. This enables asynchronous communication between the acquisition device and the tactile feedback device. It should be understood that the asynchrony described here does not mean that the acquisition and reproduction ends are out of sync due to communication delays, but rather that the acquisition and reproduction feedback actions are completed at different times specified by the system.

[0033] Because the asynchronous communication process means that the actions of the acquisition end and the reproduction end do not need to occur at the same time, sufficient data can be collected during the acquisition process without being affected by the reproduction end, resulting in more data combinations. Without being constrained by time, the acquisition end can fully utilize its role (avoiding the incomplete collection caused by the need to respond immediately to the reproduction end's actions in existing technologies), thereby enabling comprehensive collection of the target object's properties, facilitating subsequent research. This reduces the entire system's dependence on the bandwidth and stability of the communication channel.

[0034] By setting up an asynchronous communication process, the reproduction end does not need to wait too long for the action time of the acquisition end. Instead, the driving signal is sent directly by the cloud server, which makes the response faster.

[0035] During the first communication process, data normalization is performed on the static tactile sampling signal and the corresponding dynamic tactile sampling signal to obtain normalized tactile sampling signal data.

[0036] In one embodiment, the tactile sampling signal data complies with preset data format requirements; the preset data format requirements include at least one of the following: a mapping relationship between static tactile sampling signals and dynamic tactile sampling signals, a numerical range of the static tactile sampling signals and the dynamic tactile sampling signals, and a duration of the dynamic tactile sampling signals.

[0037] Exemplarily, the preset data format requirements for the tactile sampling signal data include:

[0038] (1) Each set of data includes an A S Signal and an A D (t) signal, there is a mapping relationship between the two.

[0039] (2)0 S <40KPa, 0 <max(A D (t))<50KPa.

[0040] (3)A D (t) The signal duration is at least 5 seconds.

[0041] Among them A S is the static tactile sampling signal, A D (t) is the dynamic tactile sampling signal.

[0042] The above data format can be optimized based on a large number of experiments, which can ensure that subsequent reproduction is true and accurate with a small algorithm burden.

[0043] The present invention sets a standardized asynchronous communication data format to make tactile feedback reproduction more realistic, reliable, and statistically analyzable, thereby expanding the research and application fields of tactile asynchronous feedback.

[0044] like Figure 1 ​As shown, the tactile feedback device 02 includes a microcontroller 210, a vibration simulation unit 220, and a static force sampling unit 230. The static force sampling unit 230 is used to collect the static force applied to the vibration simulation unit 220 and upload the static force to the cloud server 03. The cloud server 03 determines a dynamic drive signal based on the static force and a preset regression model, and sends the dynamic drive signal to the microcontroller 210. The microcontroller 210 is used to control the vibration simulation unit 220 to output a tactile feedback signal based on the dynamic drive signal, thereby realizing the asynchronous transmission of the tactile feedback signal in the second stage of asynchronous tactile feedback. This solves the problem of delayed response and lack of effective reproduction data foundation of the existing tactile feedback algorithm, improves the accuracy of tactile feedback, and is conducive to expanding the application scenarios of tactile feedback, with strong practicality. In addition, since the collection of model training data and tactile reproduction do not need to be performed synchronously, the inaccuracy of the data due to time delay is avoided. And due to the standardization of the data format, it is possible to study the regularity of the touched object. Before the system is built, the preset model is pre-trained based on the tactile sampling signal uploaded by the acquisition device 01 and its corresponding dynamic drive signal, and the trained preset model is stored in the cloud server 03. It can be understood that in the process of training to obtain samples, the acquisition device does not collect the target object, but is used as a signal acquisition device for the tactile feedback device. That is, in this process, the acquisition device and the tactile feedback device are in the same space, which is different from the actual use process. In other words, the acquisition device has two uses, one is to obtain samples before building the system, and the other is to collect the target object after the system is built. This setting method of multiple uses of one device has not been proposed in the prior art. There is no need to set up a separate sensing unit, which simplifies the system structure and makes it more convenient to obtain samples.

[0045] In this embodiment, a sample set consisting of tactile sampling signals and dynamic drive signals can be obtained through a big data platform or in an experimental environment. For example, when establishing training samples, the acquisition device 01 can be placed on the surface of the tactile feedback device 02 and multiple preset static forces N can be applied to the acquisition device 01. Alternatively, the acquisition device 01 can actively apply multiple preset static forces N to the tactile feedback device. While the tactile feedback device receives different preset static forces N, multiple dynamic drive signals are applied to the tactile feedback device 02 to obtain the tactile sampling signals collected and output by the acquisition device 01. A sample set is established based on multiple sets of dynamic drive signals and tactile sampling signals corresponding to the multiple different preset static forces N. This sample set is used to perform offline training on the basic regression model. During model training, the tactile sampling signals can be used as input parameters of the model, and the dynamic drive signals can be used as output parameters of the model. A set of tactile sampling signals corresponds one-to-one to a set of dynamic drive signals. After deep machine learning of the data in the sample set, the final preset model is obtained. Furthermore, during the tactile feedback process, the tactile sampling signals in the preset model may be compared according to the target static force P applied by the user to the tactile feedback device 02 , and the dynamic driving signal may be determined according to the tactile sampling signals corresponding to the target static force P.

[0046] In one embodiment, the tactile sampling signal includes a static tactile sampling signal and a dynamic tactile sampling signal, wherein the static tactile sampling signal is a static pressure sampling signal applied to the surface of the touched target object. The static tactile sampling signal can be converted from a static pressure setting value, or detected by a pressure sensor, without limitation. The sampling device 01 can directly apply the pressure or pressure value to the surface of the target object according to a pre-set pressure or pressure value in a program, and the acquisition device 01 can acquire the signal. The sampling device 01 can also directly apply the pressure or pressure value to the surface of the target object according to a pre-set pressure or pressure value in a program, and the acquisition device can directly read the pressure from the preset program. The sampling device 01 can also apply the pressure or pressure value to the surface of the target object according to a pre-set pressure or pressure value in a program, and the acquisition device can directly read the pressure from the preset program. The acquisition device can also apply the pressure or pressure value to the surface of the target object by the user at the acquisition end and acquire the signal. The dynamic tactile sampling signal is a dynamic force signal emitted by the touched target object itself under a specific static pressure.

[0047] In this embodiment, the dynamic force signal may be a dynamic pressure signal or a dynamic pressure signal that varies with time; a set of tactile sampling signals includes a set of static tactile sampling signals and dynamic tactile sampling signals that correspond one-to-one to the static pressure. During the model training phase, multiple dynamic driving signals with different values ​​may be applied under the same static pressure, and the dynamic force signal generated by each dynamic driving signal may be collected. Then, the corresponding relationship between the dynamic driving signal and the tactile sampling signal may be established through model training.

[0048] In one embodiment, the dynamic driving signal corresponds one-to-one to the target static force P applied by the user to the tactile feedback device 02 at time T2.

[0049] The dynamic driving signal is a driving control signal for driving the component to vibrate. For example, when the vibration simulation unit 220 is a motor matrix, the dynamic driving signal may be a position signal and a velocity signal, or a position signal and an acceleration signal.

[0050] In the cloud server 03, according to the received target static force P applied by the user at time T2, the tactile sampling signal data A (including the static tactile sampling signal A) corresponding to the target static force P stored in the server 03 during the previous communication process is searched. S and dynamic tactile sampling signal A D , and the mapping relationship between them). This search can usually be done by comparing the target static force P with a certain static tactile sampling signal A of the tactile sampling signal data. S Correspondingly, the target static force P is correlated with the static tactile sampling signal A. S A corresponding dynamic tactile sampling signal A D At the same time, the dynamic tactile sampling signal A D Input, according to the preset model, a model output dynamic driving signal D can be obtained, which is output to the tactile feedback device 02 to drive the tactile feedback device 02 to operate, so that the tactile feedback device produces the same tactile effect as the dynamic tactile sampling signal AD.

[0051] Therefore, the format requirement for the data group consisting of multiple tactile sampling signal data stored in the server is:

[0052] (1) Include no less than three standard tactile sampling signal data;

[0053] (2) Satisfaction as well as

[0054] By preferentially limiting the above-mentioned format, data availability is improved, and tactile feedback can be realistically simulated. In the prior art, most of the dynamic tactile sampling signals are directly used to drive the actuator to output action. However, due to the large differences in the properties of the actuator, even if the dynamic tactile sampling signals actually collected by the acquisition end are used for driving, the same tactile sensation cannot be obtained, which greatly affects the feedback accuracy and reproducibility of the feedback system. The present invention solves this problem by adopting the above-mentioned multiple corresponding relationships, which can restore the tactile sensation of the acquisition end to the greatest extent, so that users in different spaces and at different times can obtain the same tactile sensation as the acquisition end by touching the feedback device of the reproduction end. Moreover, all signals are digitized, and the relationships have been determined, so further accurate and in-depth research on the properties of the target object at the acquisition end can be carried out at different spaces and at different times.

[0055] Optionally, Figure 2This is a structural diagram of a tactile asynchronous feedback system provided in Example 2 of the present invention.

[0056] like Figure 2 As shown, the acquisition device 01 includes a detection unit 110 and an analog-to-digital conversion unit 120; the detection unit 110 is used to obtain a dynamic tactile sampling signal of the surface of the touched target under a specific static force; the analog-to-digital conversion unit 120 is used to convert the dynamic tactile sampling signal A D The analog-to-digital conversion is performed to obtain a dynamic coding signal, which is a binary coding sequence. The acquisition device may also include a static force application unit, which applies a preset static force to the target object under the control of a set program, thereby acquiring a dynamic tactile sampling signal.

[0057] The polarity and amplitude of the electrical signal output by the detection unit 110 are related to the direction and speed of the dynamic force signal of the touched object.

[0058] In this embodiment, the detection unit 110 can be a detector, which includes a shell and an inertial body. A permanent magnet is fixed on the shell, and the inertial body is a coil. When relative motion occurs between the shell and the inertial body, the inertial body cuts the magnetic field to generate an alternating voltage, i.e., an electrical vibration signal.

[0059] Specifically, during the application of the tactile asynchronous feedback system 00, the detection unit 110 is placed on the surface of the touched target object. When the touched target object outputs a dynamic force signal, the outer shell of the detection unit 110 vibrates accordingly. Since the inertial body does not vibrate or does not completely vibrate with the outer shell, a relative motion is generated between the inertial body and the outer shell. The inertial body cuts the magnetic field to generate an alternating voltage, i.e., an electrical vibration signal. This electrical vibration signal is the tactile sampling signal. Furthermore, the analog-to-digital conversion unit 120 performs a quantization conversion on the tactile sampling signal from an analog signal to a digital signal to form a binary code sequence, and uses wireless communication technology to transmit the binary code sequence to the cloud server 03. After receiving the binary code sequence, the cloud server 03 decodes the binary code sequence to obtain the corresponding dynamic tactile sampling signal A. D The detector can effectively detect weak dynamic force signals, and the output signal of the detector can be converted into a coded signal suitable for remote transmission by combining with the analog-to-digital conversion unit to realize remote human-computer interaction.

[0060] In one embodiment, the detection unit 110 may further include a pressure sensor, a pressure intensity sensor, or other detection equipment capable of detecting pressure or pressure intensity signals.

[0061] Optionally, Figure 3 This is a structural diagram of a tactile asynchronous feedback system provided in Example 3 of the present invention.

[0062] like Figure 3As shown, the acquisition device 01 further includes a filtering unit 130 and an amplifying unit 140, which are arranged between the detection unit 110 and the analog-to-digital conversion unit 120; the filtering unit 130 is used to sample the dynamic tactile sampling signal A D Perform filtering and convert the filtered dynamic tactile sampling signal dynamic tactile sampling signal A D＇ The amplifying unit 140 is used to transmit the dynamic tactile sampling signal A to the amplifying unit 140; D＇ Amplify the dynamic tactile sampling signal A after amplification D ” is transmitted to the analog-to-digital conversion unit 120.

[0063] In one embodiment, the filtering unit 130 may be a line filter. The amplification unit 140 includes a preamplifier and a main amplifier. The preamplifier is provided before the line filter and is used to perform a first-stage amplification of the weak signal and clutter signal output by the detection unit 110. The main amplifier is used to perform a second-stage amplification of the tactile sampling signal according to the voltage amplitude requirements of the analog-to-digital conversion unit 120.

[0064] Specifically, after the detection unit 110 detects the dynamic signal emitted by the target object, the preamplifier receives the dynamic tactile sampling signal A output by the detection unit 110. D The main amplifier performs a second stage of amplification on the dynamic tactile sampling signal to meet the voltage amplitude requirements of the analog-to-digital conversion unit 120. This filtering and amplification eliminates interference from interference signals on the tactile sampling signal, preventing the impact of clutter on tactile feedback. This helps improve the accuracy of model training data and the matching effect of tactile feedback drive signals.

[0065] Optionally, the preset model is a one-dimensional or multi-dimensional, one-dimensional or multiple-dimensional polynomial curve model between the tactile sampling signal and the dynamic driving signal.

[0066] Among them, the polynomial curve model is the dynamic tactile sampling signal A D A polynomial function is obtained by performing polynomial curve fitting between the tactile sampling signal and the dynamic driving signal D. The input parameter of the polynomial function can be the tactile sampling signal, and the output parameter of the polynomial function can be the dynamic driving signal.

[0067] Optionally, the mathematical expression of the polynomial curve model is as follows:

[0068]

[0069] Among them, y is the output parameter, x is the input parameter, W is the polynomial coefficient, M is the maximum power of the polynomial (i.e. the order of the polynomial), x j is x raised to the power of j, w j is x j The specific value of the polynomial coefficient W can be determined through regression training.

[0070] It should be noted that those skilled in the art can set a specific mathematical expression of the polynomial curve model according to the number of characteristic parameters in the sample set, and there is no limitation to this.

[0071] Specifically, the cloud server 03 receives the target static force P uploaded by the tactile feedback device 02, and matches the corresponding dynamic tactile sampling signal A according to the target static force P. D , and the dynamic tactile sampling signal A D Substituting the pre-learned polynomial curve model into the predicted dynamic drive signal D, the corresponding dynamic drive signal D can be a position signal and a velocity signal, or a position signal and an acceleration signal. The microcontroller 210 of the tactile feedback device 02 controls the vibration simulation unit 220 to output a tactile feedback signal based on the dynamic drive signal, achieving asynchronous tactile feedback. This solves the problem of lag response and lack of effective replication data in existing tactile feedback algorithms, improves tactile feedback accuracy, and facilitates the expansion of tactile feedback application scenarios, with strong practicality.

[0072] Based on the above embodiments, a fourth embodiment of the present invention provides a model training method for a tactile asynchronous feedback system, which is used in the above tactile asynchronous feedback system. The tactile asynchronous feedback system includes an acquisition device and a tactile feedback device.

[0073] Figure 4 This is a flowchart of a model training method for a trigger feedback system provided in Example 4 of the present invention.

[0074] like Figure 4 As shown, the model training method includes the following steps:

[0075] Step S1: placing a collection device on the surface of a tactile feedback device, and applying at least one static force to the tactile feedback device.

[0076] In this embodiment, the method of applying static force includes at least one of the following: the acquisition device itself actively applies static force to the tactile feedback device; or, an external force is used to press the acquisition device against the tactile feedback device. The specific method is not limited as long as the tactile feedback device receives a specific static force.

[0077] In one embodiment, the acquisition device includes a detection unit and an analog-to-digital conversion unit; the detection unit acquires tactile signals and converts the tactile signals into tactile sampling signals; the analog-to-digital conversion unit performs analog-to-digital conversion on the tactile sampling signals to obtain dynamic coding signals, which are binary coding sequences.

[0078] In one embodiment, the acquisition device further includes a filtering unit and an amplifying unit, which are arranged between the detection unit and the analog-to-digital conversion unit; the filtering unit is used to filter the tactile sampling signal and transmit the filtered tactile sampling signal to the amplifying unit; the amplifying unit is used to amplify the tactile sampling signal and transmit the amplified tactile sampling signal to the analog-to-digital conversion unit.

[0079] Step S2: applying multiple dynamic drive signals to the tactile feedback device to obtain tactile sampling signals output by the acquisition device.

[0080] In one embodiment, the dynamic driving signal corresponds one-to-one to the static force applied to the tactile feedback device; and the tactile sampling signal includes a static tactile sampling signal and a dynamic tactile sampling signal.

[0081] The dynamic driving signal may be a control signal applied to the tactile feedback device by the test system. For example, the dynamic driving signal may be a sine wave or a pulse signal.

[0082] Step S3: establishing a sample set according to the dynamic driving signals and tactile sampling signals corresponding to different static forces.

[0083] Step S4: Perform machine learning training on the preset model based on the sample set.

[0084] Specifically, using the acquisition device as a detector as an example, when establishing training samples, the acquisition device is placed on the surface of the tactile feedback device. A test system is used to apply a set of dynamic drive signals and static forces to the tactile feedback device, activating the tactile feedback device. The acquisition device's housing vibrates along with the tactile feedback device, generating tactile sampling signals due to electromagnetic induction. The tactile sampling signals correspond one-to-one with the dynamic drive signals. By switching between different dynamic drive signals and static force levels, the tactile sampling signals corresponding to each dynamic drive signal under different static forces are obtained. After accumulating a large amount of data, a sample set is established. Furthermore, the tactile sampling signals can be used as input parameters of the model, and the dynamic drive signals as output parameters of the model. Through deep machine learning of the data in the sample set, the final preset model is obtained. This establishes a correlation between the tactile sampling signals and the dynamic drive signals. For any tactile sampling signal, a corresponding dynamic drive signal can be output, ensuring that the tactile feedback device is driven according to the dynamic drive signal, and the sensory effect produced by the action is the same as that of the tactile sampling signal obtained by direct touch.

[0085] Optionally, the preset model is a one-dimensional or multi-dimensional, one-dimensional or multiple-dimensional polynomial curve model between the tactile sampling signal and the dynamic driving signal.

[0086] Based on the above embodiments, embodiment five of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and runnable on the processor. When the processor executes the program, the model training method for triggering the feedback system as described above is implemented, and machine learning is performed based on a sample set of tactile sampling signals and dynamic drive signals to determine a preset model.

[0087] Optionally, the model training method can be completed locally offline. After the preset model training is completed, the preset model can be deployed to a local controller or a cloud server for calculating the dynamic drive signal based on the real-time sampled tactile sampling signal.

[0088] Figure 5 This is a structural diagram of a computer device provided in Example 5 of the present invention. Figure 5 A block diagram of an exemplary computer device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 5 The computer device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0089] like Figure 5 As shown, computer device 12 is implemented as a general-purpose computing device. Components of computer device 12 may include, but are not limited to, one or more processors 16, system memory 28, and a bus 18 that connects various system components (including system memory 28 and processor 16).

[0090] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor, or a local bus using any of a variety of bus architectures. Examples of these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0091] The computer device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the computer device 12, including volatile and non-volatile media, removable and non-removable media.

[0092] System memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. Computer device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, storage system 34 may be configured to read and write non-removable, non-volatile magnetic media ( Figure 5 Not shown, often called a "hard drive"). Although Figure 5 Not shown, a magnetic disk drive for reading and writing to a removable non-volatile magnetic disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to bus 18 via one or more data media interfaces. Memory 28 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of various embodiments of the present invention.

[0093] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which, or some combination thereof, may include an implementation of a network environment. Program modules 42 generally implement the functions and / or methods of the embodiments described herein.

[0094] The computer device 12 can also communicate with one or more external devices 14 (e.g., a keyboard, pointing device, display 24, etc.), one or more devices that enable a user to interact with the computer device 12, and / or any device that enables the computer device 12 to communicate with one or more other computing devices (e.g., a network card, a modem, etc.). Such communication can occur via an input / output (I / O) interface 22. Furthermore, the computer device 12 can communicate with one or more networks (e.g., a local area network (LAN), a wide area network (WAN), and / or a public network such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with the other modules of the computer device 12 via a bus 18. It should be understood that, although not shown, other hardware and / or software modules can be used in conjunction with the computer device 12, including but not limited to microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0095] The processor 16 executes various functional applications and data processing by running programs stored in the system memory 28, such as implementing the model training method for the tactile asynchronous feedback system provided by the embodiment of the present invention.

[0096] Based on the above embodiments, embodiment 6 of the present invention provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the model training method as described above is implemented.

[0097] Based on the above embodiments, a seventh embodiment of the present invention provides a computer-readable storage medium, which stores the tactile sampling signal data collected by the collection device provided in any of the above embodiments.

[0098] The computer storage medium of the embodiment of the present invention may adopt any combination of one or more computer-readable media. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or component, or any combination thereof. More specific examples (a non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device.

[0099] A computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0100] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0101] Computer program code for performing the operations of the present invention may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0102] In summary, the model training method, computer equipment and medium provided by the present invention are used for a tactile asynchronous feedback system. The system is provided with a communication-connected acquisition device and a tactile feedback device. The acquisition device converts the acquired tactile signal into a tactile sampling signal and sends the tactile sampling signal to the tactile feedback device. The microcontroller of the tactile feedback device substitutes the tactile sampling signal into a preset model to obtain a dynamic driving signal; the vibration simulation unit drives the fluid flow in the simulated tissue according to the dynamic driving signal and outputs a tactile feedback signal. The dynamic driving signal corresponding to the tactile signal is determined by the preset model, and the simulated tissue vibration is driven according to the dynamic driving signal. The vibration tactile signal of the target object can be accurately reproduced, which solves the problem that the existing tactile feedback system cannot accurately reproduce the vibration perception of the target object, is conducive to improving the accuracy of tactile feedback and expanding application scenarios.

[0103] Note that the above are only preferred embodiments of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and that various obvious changes, readjustments, and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments and may include many other equivalent embodiments without departing from the concept of the present invention. The scope of the present invention is determined by the scope of the appended claims.

Claims

1. A tactile feedback device, characterized in that: for receiving a target static force applied by a user at time T2, sending the target static force to a cloud server, and receiving a dynamic driving signal corresponding to the target static force returned by the cloud server, wherein the dynamic driving signal is determined based on the target static force at time T2, the tactile sampling signal at time T1, and a preset model, wherein the preset model is determined by machine learning based on a sample set of multiple sets of tactile sampling signals and dynamic driving signals constructed in advance, the tactile sampling signals including static tactile sampling signals and dynamic tactile sampling signals, and a mapping relationship exists between the static tactile sampling signals and the dynamic tactile sampling signals; The tactile feedback device is further configured to perform an action according to the dynamic drive signal, wherein time T1 and time T2 are different times, thereby achieving asynchronous feedback of the tactile feedback device.

2. The tactile feedback device according to claim 1, wherein: The tactile sampling signal data complies with the preset data format requirements; The preset data format requirement includes at least one of the following: a mapping relationship between the static tactile sampling signal and the dynamic tactile sampling signal, a value range of the static tactile sampling signal and the dynamic tactile sampling signal, and a duration of the dynamic tactile sampling signal.

3. The tactile feedback device according to claim 1, wherein: The dynamic driving signal corresponds one-to-one to the target static force applied by the user to the tactile feedback device at time T2.

4. The tactile feedback device according to any one of claims 1 to 3, characterized in that: The tactile feedback device includes a microcontroller, a vibration simulation unit and a static force sampling unit; The vibration simulation unit includes at least one actuator, and the microcontroller is used to adjust the action stroke and action frequency of the actuator according to the dynamic driving signal.

5. A tactile sampling signal data, characterized in that: The data format of the data includes at least one of the following: Each set of data includes a static tactile sampling signal A S and a dynamic tactile sampling signal A D (t), the static tactile sampling signal A S and the dynamic tactile sampling signal A D (t) There is a mapping relationship; The dynamic tactile sampling signal A D (t) The signal duration is at least 5 seconds; The static tactile sampling signal A S Meet the following conditions: 0 S <40KPa; the dynamic tactile sampling signal A D (t) satisfies the following conditions: 0 <max(A D (t))<50KPa;​ The tactile sampling signal data is used to train a preset model, and the preset model is used to determine a dynamic driving signal according to the target static force and the tactile sampling signal.

6. The tactile sampling signal data according to claim 5, characterized in that: The data is the data stored in the server at time T1.

7. A collection device, characterized in that: Used to generate, transmit and / or store the tactile sampling signal data as claimed in any one of claims 5 or 6.

8. A tactile sampling signal data set, characterized in that: include: At least three sets of tactile sampling signal data as claimed in claim 5.

9. A tactile sampling signal data set, characterized in that: include: Static tactile sampling signal A S and dynamic tactile sampling signal A D (t), the static tactile sampling signal A S and the dynamic tactile sampling signal A D (t) There is a mapping relationship; the static tactile sampling signal A S Including first static tactile sampling signal data Second static tactile sampling signal data and the third static tactile sampling signal data in, and The following conditions must be met: as well as 10. The tactile sampling signal data set according to any one of claims 8 or 9, characterized in that: The data set is intended to be stored in a server.

Images