Control signal output method and device, intelligent wearable equipment and storage medium

Abstract

The invention provides a control signal output method and device, intelligent wearable equipment and a computer readable storage medium. The control signal output method comprises the following steps: acquiring target sensing data of interaction control equipment of intelligent wearable equipment; when a jitter signal generated by key operation of the interaction control equipment is detected, starting a filter to filter the target sensing data to obtain filtered sensing data; and outputting a control signal of a controlled object of the intelligent wearable device according to the filtered sensing data. According to the method and the device, the influence of vibration or jitter signals of the interaction control equipment on interaction control can be reduced, the control error of the controlled object is reduced, the control accuracy of the interaction control equipment on the controlled object is improved, and the interaction operation fluency and stability of the interaction control equipment are further improved.

Description

Technical Field

[0001] The present application relates to the field of interactive control technologies, and particularly relates to a control signal output method, apparatus, intelligent wearable device, and computer-readable storage medium. Background Art

[0002] With the development of intelligent wearable technologies, augmented reality devices, virtual reality devices, etc. have entered users' lives. Currently, there are various main interaction methods for intelligent wearable devices such as augmented reality devices and virtual reality devices, such as eye movement tracking interaction, gesture recognition interaction, voice command interaction, interaction with an interaction control device (such as a ring, etc.).

[0003] Among them, when interacting with an interaction control device (such as a ring, etc.), interaction control signals are output based on the motion data of the interaction control device. However, the inventors of the present application found during actual research and development that: in some cases, the interaction control device will generate a certain degree of vibration or jitter. For example, when a user presses a button on the interaction control device, the interaction control device usually vibrates or jitters to a certain degree, and this vibration or jitter will be captured by the sensors of the interaction control device, resulting in control errors of the controlled object (such as a cursor), affecting the fluency and stability of the interaction operation; for example, when a user presses the ring button, the vibration of the ring will be manifested as irregular jitter of the cursor, and the cursor jitter will affect the fluency and stability of the interaction operation. Summary of the Invention

[0004] The present application provides a control signal output method, apparatus, intelligent wearable device, and computer-readable storage medium, which can reduce the control error of the controlled object, improve the control accuracy of the interaction control device over the controlled object, and further improve the fluency and stability of the interaction operation of the interaction control device.

[0005] In a first aspect, the present application provides a control signal output method, and the method includes:

[0006] Obtain target sensing data of an interaction control device of an intelligent wearable device;

[0007] When a jitter signal generated by a button operation of the interaction control device is detected, perform filtering processing on the target sensing data through a filter to obtain filtered sensing data;

[0008] Output a control signal of a controlled object of the intelligent wearable device according to the filtered sensing data.

[0009] In a second aspect, the present application provides a control signal output apparatus, and the control signal output apparatus includes:

[0010] An obtaining unit, configured to obtain target sensing data of an interaction control device of an intelligent wearable device;

[0011] A filtering unit, configured to filter the target sensing data through a filter to obtain filtered sensing data when detecting a jitter signal generated by a key operation of the interaction control device;

[0012] A control unit, configured to output a control signal of a controlled object of the intelligent wearable device according to the filtered sensing data.

[0013] In a third aspect, the present application further provides an intelligent wearable device, including a processor and a memory. A computer program is stored in the memory, and when the processor calls the computer program in the memory, it executes any one of the control signal output methods provided by the present application.

[0014] In a fourth aspect, the present application further provides a computer-readable storage medium, on which a computer program is stored, and the computer program is loaded by a processor to execute the control signal output method.

[0015] In the present application, when detecting a jitter signal generated by a key operation of an interaction control device, the target sensing data of the interaction control device is filtered through a filter to obtain filtered sensing data; then, the filtered sensing data is used to output a control signal of a controlled object of the intelligent wearable device. In this way, the filtered sensing data can filter out the vibration or jitter signal of the interaction control device, reduce the influence of the vibration or jitter signal of the interaction control device on the interaction control, reduce the control error of the controlled object, improve the control accuracy of the interaction control device for the controlled object (such as a cursor, display content), and further improve the interaction operation fluency and stability of the interaction control device. Description of the Drawings

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present application. For those skilled in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0017] Figure 1 is a structural schematic block diagram of an intelligent wearable device provided by an embodiment of the present application;

[0018] Figure 2 is a schematic flowchart of a control signal output method provided by an embodiment of the present application;

[0019] Figure 3 is a schematic flowchart of an embodiment for determining a learned filtering coefficient of a filter provided by an embodiment of the present application;

[0020] Figure 4 It is a schematic diagram of an embodiment of step 303 in an embodiment of the present application;

[0021] Figure 5 It is a schematic diagram of another embodiment of step 303 in an embodiment of the present application;

[0022] Figure 6 It is a schematic structural diagram of an embodiment of a control signal output device provided in an embodiment of the present application. Detailed implementation manners

[0023] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative efforts fall within the protection scope of the present application.

[0024] The flowcharts shown in the accompanying drawings are only illustrative, and do not necessarily include all the contents and operations / steps, nor do they necessarily need to be executed in the described order. For example, some operations / steps can also be decomposed, combined, or partially merged, so the actual execution order may change according to the actual situation.

[0025] In the description of the embodiments of the present application, it should be understood that the terms "first" and "second" are only used for descriptive purposes, and cannot be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the described features. In the description of the embodiments of the present application, "a plurality of" means two or more, unless otherwise specifically defined.

[0026] In order for any person skilled in the art to implement and use the present application, the following description is given. In the following description, details are set forth for the purpose of explanation. It should be understood that those skilled in the art can realize that the present application can be implemented without using these specific details. In other instances, well-known processes will not be elaborated in detail to avoid unnecessary details from obscuring the description of the embodiments of the present application. Therefore, the present application is not intended to be limited to the embodiments shown, but is consistent with the widest scope that conforms to the principles and features disclosed in the embodiments of the present application.

[0027] Embodiments of the present application provide a control signal output method, apparatus, smart wearable device, and computer-readable storage medium. Among them, the control signal output apparatus can be integrated in the smart wearable device. Among them, the smart wearable device can be a smart glasses, a smart helmet, etc. The smart glasses can be AR (augmented reality) glasses, VR (Virtual Reality) glasses, MR (Mixed Reality) glasses, XR (eXtended Reality) glasses, etc. The smart helmet can be an AR helmet, etc.

[0028] The execution subject of the control signal output method in the embodiments of the present application can be the control signal output apparatus provided in the embodiments of the present application, or a smart wearable device integrated with the control signal output apparatus. Among them, the control signal output apparatus can be implemented in a hardware or software manner.

[0029] The following will describe in detail some embodiments of the present application with reference to the accompanying drawings. Without conflict, the following embodiments and the features in the embodiments can be combined with each other.

[0030] Figure 1 It is a structural schematic block diagram of a smart wearable device provided by an embodiment of the present application.

[0031] As Figure 1 shown, the smart wearable device 100 includes a processor 101 and a memory 102. The processor 101 and the memory 102 are connected through a bus 103, and this bus is, for example, an I2C (Inter-integrated Circuit) bus.

[0032] Specifically, the processor 101 is used to provide computing and control capabilities to support the operation of the entire smart wearable device 100. The processor 101 can be a central processing unit (CPU). The processor 101 can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0033] Specifically, the memory 102 can be a Flash chip, a read-only memory (ROM), a magnetic disk, an optical disc, a USB flash drive, a portable hard drive, or the like.

[0034] Those skilled in the art can understand that Figure 1 the structure shown in is only a block diagram of some structures related to the solution of the embodiment of the present application, and does not constitute a limitation on the intelligent wearable device to which the solution of the embodiment of the present application is applied. The specific intelligent wearable device may include more or fewer components than those shown in the figure, or combine some components, or have a different component layout.

[0035] The processor 101 is configured to run a computer program stored in the memory 102, and when executing the computer program, implement any one of the control signal output methods provided by the embodiments of the present application. For example, the processor 101 is configured to run a computer program stored in the memory 102, and when executing the computer program, the following steps may be implemented:

[0036] Obtain target sensing data of an interaction control device of the intelligent wearable device;

[0037] When a jitter signal generated by a key operation of the interaction control device is detected, filter the target sensing data through a filter to obtain filtered sensing data;

[0038] Output a control signal of a controlled object of the intelligent wearable device according to the filtered sensing data.

[0039] In some embodiments, the processor 101 is configured to run a computer program stored in the memory 102, and when executing the computer program, the following steps may be implemented:

[0040] Obtain sample sensing data of the interaction control device and expected output data of the sample sensing data;

[0041] Filter the sample sensing data through the filter according to the initialized filtering coefficient to obtain the actual output data of the sample sensing data;

[0042] Adjust the initialized filtering coefficient of the filter according to the error between the actual output data and the expected output data to obtain the learned filtering coefficient of the filter:

[0043] Filter the target sensing data through the filter according to the learned filtering coefficient to obtain filtered sensing data.

[0044] In some embodiments, the processor 101 is configured to run a computer program stored in the memory 102, and when executing the computer program, the following steps may be implemented:

[0045] Obtain a target error between the actual output data and the expected output data;

[0046] Obtain a target learning rate of the filter;

[0047] With the goal of minimizing the sum of squares of the target error, adjust the initialized filter coefficients according to the target error and the target learning rate to obtain the learned filter coefficients of the filter.

[0048] In some embodiments, the target learning rate includes a first learning rate and a second learning rate, the first learning rate is greater than the second learning rate, the sample sensing data includes first sample sensing data in the key-pressing stage and second sample sensing data in the first target stage, the jitter intensity of the interaction control device in the first target stage is less than the jitter intensity in the key-pressing stage, the target error includes a first error between the actual output data of the first sample sensing data and the expected output data of the first sample sensing data, and a second error between the actual output data of the second sample sensing data and the expected output data of the second sample sensing data. The processor 101 is configured to run a computer program stored in the memory 102, and when executing the computer program, the following steps may be implemented:

[0049] With the goal of minimizing the sum of squares of the first error, adjust the initialized filter coefficients according to the first error and the first learning rate to obtain the learned filter coefficients in the key-pressing stage;

[0050] With the goal of minimizing the sum of squares of the second error, adjust the initialized filter coefficients according to the second error and the second learning rate to obtain the learned filter coefficients in the first target stage;

[0051] Filter the target sensing data through the filter according to the learned filter coefficients in the key-pressing stage to obtain the filtered sensing data corresponding to the key-pressing stage;

[0052] And / or, filter the target sensing data through the filter according to the learned filter coefficients in the first target stage to obtain the filtered sensing data corresponding to the first target stage.

[0053] In some embodiments, the target learning rate includes a third learning rate and a fourth learning rate, the third learning rate is less than the fourth learning rate, the sample sensing data includes third sample sensing data in the key release stage and fourth sample sensing data in the second target stage, the jitter intensity of the interaction control device in the second target stage is greater than that in the key release stage, the target error includes a third error between the actual output data and the expected output data of the third sample sensing data, and a fourth error between the actual output data and the expected output data of the fourth sample sensing data. The processor 101 is configured to run a computer program stored in the memory 102 and, when executing the computer program, may implement the following steps:

[0054] Taking the minimization of the sum of squares of the third error as the goal, adjusting the initial filter coefficients according to the third error and the third learning rate to obtain the learned filter coefficients in the key release stage;

[0055] Taking the minimization of the sum of squares of the fourth error as the goal, adjusting the initial filter coefficients according to the fourth error and the fourth learning rate to obtain the learned filter coefficients in the second target stage;

[0056] Filtering the target sensing data by a filter according to the learned filter coefficients in the key release stage to obtain the filtered sensing data corresponding to the key release stage;

[0057] And / or, filtering the target sensing data by a filter according to the learned filter coefficients in the second target stage to obtain the filtered sensing data corresponding to the second target stage.

[0058] In some embodiments, the processor 101 is configured to run a computer program stored in the memory 102 and, when executing the computer program, may implement the following steps:

[0059] Obtain the target error between the actual output data and the expected output data;

[0060] Obtain the covariance matrix of the sample sensing data;

[0061] Taking the minimization of the sum of squares of the target error as the goal, adjusting the initial filter coefficients according to the target error and the covariance matrix to obtain the learned filter coefficients of the filter.

[0062] In some embodiments, the controlled object is a cursor or display content, and the processor 101 is configured to run a computer program stored in the memory 102 and, when executing the computer program, may implement the following steps:

[0063] Output a control signal for the cursor of the smart wearable device according to the filtered sensing data to control the display position of the cursor on the smart wearable device;

[0064] Alternatively, output a control signal for the display content of the smart wearable device according to the filtered sensing data to control the display position of the display content on the smart wearable device.

[0065] It should be noted that those skilled in the art can clearly understand that, for the sake of convenience and brevity of description, the specific working process of the smart wearable device described above can refer to the corresponding process in the embodiment of the following control signal output method, and will not be elaborated here.

[0066] Hereinafter, Figure 1 taking the smart wearable device shown in

[0067] as the execution subject of this control signal output method as an example, the control signal output method provided by the embodiments of the present application will be introduced in detail. For the sake of simplification and convenience of description, the execution subject will be omitted in the subsequent method embodiments. Figure 2 , Figure 2 is a schematic flowchart of a control signal output method provided by an embodiment of the present application. The control signal output method includes steps 201 to 203, wherein:

[0068] 201. Obtain target sensing data of an interaction control device of the smart wearable device.

[0069] Among them, the interaction control device is a device for controlling the controlled object of the smart wearable device. For example, the interaction control device may be a ring, a watch, etc. In some embodiments, the controlled object is a cursor, and the interaction control device may be a device for controlling the cursor of the smart wearable device. In some embodiments, the controlled object is display content, and the interaction control device may be a device for controlling the display content displayed at the screen position of the smart wearable device.

[0070] Exemplarily, the interaction control device may be built with an IMU sensor for collecting sensing data of the interaction control device. The sensing data of the interaction control device (hereinafter referred to as "device sensing data" for simplicity of expression) may include, but is not limited to: motion data such as acceleration and angular velocity of the interaction control device. By capturing the sensing data of the interaction control device such as acceleration and angular velocity and other motion data, it can be used to control the display of the controlled object on the smart wearable device.

[0071] Among them, the target sensing data is device sensing data used to generate a control signal for a controlled object.

[0072] In some embodiments, the controlled object is a cursor. During the process of controlling the cursor on the smart wearable device using an interactive control device, motion data such as acceleration and angular velocity of the interactive control device can be collected by the built-in IMU sensor of the interactive control device as the target sensing data.

[0073] In some embodiments, the controlled object is display content. During the process of controlling the display content on the smart wearable device using an interactive control device, motion data such as acceleration and angular velocity of the interactive control device can be collected by the built-in IMU sensor of the interactive control device as the target sensing data.

[0074] 202. When a jitter signal generated by a key operation of the interactive control device is detected, the target sensing data is filtered by a filter to obtain filtered sensing data.

[0075] In some embodiments, it can be determined whether there is jitter in the interactive control device by detecting whether there is a key operation on the interactive control device. For example, when the interactive control device detects a key press operation, it generates a key press signal and sends the key press signal to the smart wearable device. Thus, when the smart wearable device receives the key press signal sent by the interactive control device, the smart wearable device can determine that a jitter signal generated by the key operation of the interactive control device is detected. Another example is that when the interactive control device detects a key release operation, it generates a key release signal and sends the key release signal to the smart wearable device. Thus, when the smart wearable device receives the key release signal sent by the interactive control device, the smart wearable device can determine that a jitter signal generated by the key operation of the interactive control device is detected. In this way, the anti-jitter mechanism can be started when there is jitter in the interactive control device, such as during the key press stage and the key release stage, to identify and compensate for the vibration signals caused by the key press stage, the key release stage, etc., and reduce the effective posture changes of these vibration signals being recognized as the control of the controlled device by the interactive control device.

[0076] There are multiple implementation manners for step 202. Exemplarily, they include but are not limited to the following manners (1), (2), (3), and (4). Step 202 can perform filtering processing on the target sensing data through any one of manners (1), (2), (3), and (4) to obtain the filtered sensing data; or, according to the requirements of the actual business scenario, step 202 can also simultaneously combine at least two of the following manners (1), (2), (3), and (4) to perform filtering processing on the target sensing data to obtain the filtered sensing data. For example, simultaneously combining manners (1) and (2), or simultaneously combining manners (1) and (3), or simultaneously combining manners (2) and (4):

[0077] (1) In some embodiments, the filter performs filtering processing on the target sensing data according to the learned filtering coefficients in the key-press stage. At this time, step 202 includes: when the intelligent wearable device receives a key-press signal sent by the interactive control device, the intelligent wearable device can determine that a jitter signal generated by the key operation of the interactive control device is detected, and the intelligent wearable device enables the filter to perform filtering processing on the target sensing data according to the learned filtering coefficients in the key-press stage to obtain the filtered sensing data. The determination of the learned filtering coefficients in the key-press stage can refer to the relevant descriptions in the later steps 301 to 303, which will not be elaborated here. In this way, the filtered sensing data can filter out the jitter signal caused by the interactive control device (such as a ring) in the key-press stage, improve the control accuracy of the interactive control device (such as a ring) in the key-press stage for the controlled object (such as a cursor, display content). For example, improve the position control accuracy of the cursor and reduce the cursor jitter problem caused by the jitter of the interactive control device (such as a ring) in the key-press stage; or, improve the dragging control accuracy of the display content and reduce the display content jitter problem caused by the jitter of the interactive control device (such as a ring) in the key-press stage.

[0078] (2) In some embodiments, the filter filters the target sensing data according to the learned post - release filtering coefficients of the key. At this time, step 202 includes: when the intelligent wearable device receives the key - release signal sent by the interactive control device, the intelligent wearable device can determine the jitter signal generated by the key operation of the interactive control device, and the intelligent wearable device enables the filter to filter the target sensing data according to the learned post - release filtering coefficients of the key to obtain the filtered sensing data. The determination of the learned post - release filtering coefficients of the key can refer to the relevant descriptions in steps 301 - 303 below and will not be elaborated here. In this way, the filtered sensing data can filter out the jitter signal caused by the interactive control device (such as a ring) during the key - release stage, improving the control accuracy of the interactive control device (such as a ring) over the controlled object (such as a cursor, display content) during the key - release stage. For example, it improves the position control accuracy of the cursor and reduces the cursor jitter problem caused by the jitter of the interactive control device (such as a ring) during the key - release stage; another example is to improve the drag - control accuracy of the display content and reduce the display - content jitter problem caused by the jitter of the interactive control device (such as a ring) during the key - release stage.

[0079] (3) In some embodiments, the filter filters the target sensing data according to the learned post - filtering coefficients of the first target stage. The jitter intensity of the interactive control device in the first target stage is less than that in the key - press stage. For example, when the jitter intensity in the key - press stage > the jitter intensity in the key - release stage > the jitter intensity in other stages (other stages refer to the time period when the interactive control device jitters after the key - release stage), the first target stage includes the key - release stage and other stages; another example is when the jitter intensity in other stages > the jitter intensity in the key - press stage > the jitter intensity in the key - release stage, the first target stage is the key - release stage. At this time, step 202 includes: when the intelligent wearable device detects the jitter signal of the interactive control device in the first target stage, the intelligent wearable device enables the filter to filter the target sensing data according to the learned post - filtering coefficients of the first target stage to obtain the filtered sensing data. The determination of the learned post - filtering coefficients of the first target stage can refer to the relevant descriptions in steps 301 - 303 below and will not be elaborated here. In this way, the filtered sensing data can filter out the jitter signal caused by the interactive control device (such as a ring) in the first target stage, improving the control accuracy of the interactive control device (such as a ring) over the controlled object (such as a cursor, display content) in the first target stage. For example, it improves the position control accuracy of the cursor and reduces the cursor jitter problem caused by the jitter of the interactive control device (such as a ring) in the first target stage; another example is to improve the drag - control accuracy of the display content and reduce the display - content jitter problem caused by the jitter of the interactive control device (such as a ring) in the first target stage.

[0080] (4) In some embodiments, the filter filters the target sensing data according to the learned post-filtering coefficients in the second target stage. The jitter intensity of the interaction control device in the second target stage is greater than that in the key-release stage. For example, when the jitter intensity in the key-press stage > the jitter intensity in the key-release stage > the jitter intensity in other stages, the second target stage is the key-press stage; or when the jitter intensity in other stages > the jitter intensity in the key-release stage > the jitter intensity in the key-press stage, the second target stage includes other stages. At this time, step 202 includes: when the smart wearable device detects the jitter signal of the interaction control device in the second target stage, the smart wearable device enables the filter to filter the target sensing data according to the learned post-filtering coefficients in the second target stage to obtain the filtered sensing data. The determination of the learned post-filtering coefficients in the second target stage can refer to the relevant descriptions in parts of steps 301-303 later, which will not be elaborated here. In this way, the filtered sensing data can filter out the jitter signal caused by the interaction control device (such as a ring) in the second target stage, improving the control accuracy of the interaction control device (such as a ring) on the controlled object (such as a cursor, display content) in the second target stage. For example, it improves the position control accuracy of the cursor and reduces the cursor jitter problem caused by the jitter of the interaction control device (such as a ring) in the second target stage; or it improves the dragging control accuracy of the display content and reduces the display content jitter problem caused by the jitter of the interaction control device (such as a ring) in the second target stage.

[0081] For example, taking "the output process of the control signal includes two stages: the key-press operation stage and the key-release operation. During the key-press stage, filtering is performed according to the learned post-filtering coefficient corresponding to the key-press stage, and during the key-release stage, filtering is performed according to the learned post-filtering coefficient corresponding to the key-release stage" as an example, from the key-press operation to the key-release operation, the specific process of steps 202 and 203 combined to filter out the influence of the jitter signal is as follows: In step 202, when detecting the jitter signal generated by the key-press operation of the interactive control device, the target sensing data is filtered by the filter according to the learned post-filtering coefficient of the key-press stage to obtain the filtered sensing data; In step 203, during the key-press stage, using the filtered sensing data corresponding to the key-press stage (that is, the filtered sensing data obtained by filtering the target sensing data according to the learned post-filtering coefficient of the key-press stage), the control signal of the controlled object of the intelligent wearable device is output; In step 202, when detecting the jitter signal generated by the key-release operation of the interactive control device, the target sensing data is filtered by the filter according to the learned post-filtering coefficient of the key-release stage to obtain the filtered sensing data; In step 203, during the key-release stage, using the filtered sensing data corresponding to the key-release stage (that is, the filtered sensing data obtained by filtering the target sensing data according to the learned post-filtering coefficient of the key-release stage), the control signal of the controlled object of the intelligent wearable device is output. In this way, according to the jitter signals of different intensities generated by the key operation (such as the jitter intensity 1 in the key-press stage and the jitter intensity 2 in the key-press stage), different filtering coefficients can be used to filter the target sensing data, improving the filtering accuracy of the jitter signal.

[0082] For another example, taking "the output process of the control signal includes two stages: the key-pressing operation stage and the first target stage. In the key-pressing stage, filtering is performed according to the learned post-filtering coefficient corresponding to the key-pressing stage, and in the first target stage, filtering is performed according to the learned post-filtering coefficient corresponding to the first target stage" as an example, the specific process of combining steps 202 and 203 to filter out the influence of the jitter signal is as follows: In step 202, when the jitter signal generated by the key-pressing operation of the interactive control device is detected, the target sensing data is filtered by the filter according to the learned post-filtering coefficient of the key-pressing stage to obtain the filtered sensing data corresponding to the key-pressing stage; In step 203, in the key-pressing stage, using the filtered sensing data corresponding to the key-pressing stage (that is, the filtered sensing data obtained by filtering the target sensing data according to the learned post-filtering coefficient of the key-pressing stage), the control signal of the controlled object of the intelligent wearable device is output; In step 202, when the jitter signal generated in the first target stage of the interactive control device is detected, the target sensing data is filtered by the filter according to the learned post-filtering coefficient of the first target stage to obtain the filtered sensing data corresponding to the first target stage; In step 203, in the first target stage, using the filtered sensing data corresponding to the first target stage (that is, the filtered sensing data obtained by filtering the target sensing data according to the learned post-filtering coefficient of the key-release stage), the control signal of the controlled object of the intelligent wearable device is output.

[0083] For another example, taking "the output process of the control signal includes two stages: the key release operation stage and the second target stage 2. During the key release stage, filtering is performed on the target sensing data according to the learned post-filtering coefficient corresponding to the key press stage. During the second target stage, filtering is performed on the target sensing data according to the learned post-filtering coefficient corresponding to the second target stage" as an example, the specific process of steps 202 and 203 combined to filter out the influence of the jitter signal is as follows: In step 202, when the jitter signal generated by the key release operation of the interactive control device is detected, the target sensing data is filtered by the filter according to the learned post-filtering coefficient of the key release stage to obtain the filtered sensing data corresponding to the key release stage; In step 203, during the key release stage, the filtered sensing data corresponding to the key release stage (that is, the filtered sensing data obtained by filtering the target sensing data according to the learned post-filtering coefficient of the key release stage) is used to output the control signal of the controlled object of the intelligent wearable device; In step 202, when the jitter signal generated by the second target stage of the interactive control device is detected, the target sensing data is filtered by the filter according to the learned post-filtering coefficient of the second target stage to obtain the filtered sensing data corresponding to the second target stage; In step 203, during the second target stage, the filtered sensing data corresponding to the second target stage (that is, the filtered sensing data obtained by filtering the target sensing data according to the learned post-filtering coefficient of the second target stage) is used to output the control signal of the controlled object of the intelligent wearable device.

[0084] Exemplarily, as Figure 3 shown, the learned post-filtering coefficient of the filter is determined by the following steps 301 to 303:

[0085] 301. Obtain the sample sensing data of the interactive control device and the expected output data of the sample sensing data.

[0086] Among them, the sample sensing data is the device sensing data used for the filtering coefficient learning of the filter.

[0087] Among them, the expected output data refers to the output value expected after the filter filters the sample sensing data, and is used to indicate the performance value of the sample sensing data after removing the jitter interference of the interactive control device.

[0088] 302. Filter the sample sensing data by the filter according to the initial filtering coefficient to obtain the actual output data of the sample sensing data.

[0089] Among them, the actual output data is the device sensing data obtained by filtering the sample sensing data through a filter according to the initialized filter coefficients. The initialized filter coefficients are preset values of the filter coefficients, and the specific values of the initialized filter coefficients can be set according to the requirements of the actual business scenario. The specific values of the initialized filter coefficients are not limited here.

[0090] Exemplarily, as shown in Formula 1 below, taking "the device sensing data after filtering at the k-th acquisition moment is equal to the weighted sum of the device sensing data at the previous M + 1 acquisition moments" as an example, the specific implementation of step 302 can be as follows: Assume that the initialized coefficients of the filter are {w0, w1…, w M}, the sample sensing data is the device sensing data x(k) at the k-th acquisition moment, and the sample sensing data x(k) is used as an input sensing data of the filter. For each new input sensing data x(k), through the sliding window mechanism, a convolution or inner product operation is performed with the initialized coefficients to obtain the output sensing data y(k), which is used as the actual output data of the sample sensing data x(k).

[0091]

[0092] In Formula 1, x(k) represents the input sensing data (i.e., the sample sensing data, specifically representing the device sensing data at the k-th acquisition moment), y(k) represents the actual output data (i.e., the actual output data of the sample sensing data x(k), specifically representing the actual value of the device sensing data after filtering at the k-th acquisition moment), w i represents the i-th initialized coefficient of the filter, and x(k - i) represents the device sensing data at the (k - i)-th acquisition moment.

[0093] 303. Adjust the initialized filter coefficients of the filter according to the error between the actual output data and the expected output data to obtain the learned filter coefficients of the filter.

[0094] There are multiple implementation manners for step 303. Exemplarily, they include the following manners ① and ②:

[0095] ① In some embodiments, the least mean square error algorithm is used for learning the filter coefficients. At this time, as Figure 4 shown, step 303 can specifically include the following steps 3031A to 3033A:

[0096] 3031A. Obtain the target error between the actual output data and the expected output data.

[0097] For example, as shown in Formula 2 below, the difference between the expected output data and the actual output data can be calculated as the error between the actual output data and the expected output data.

[0098] e(k) = d(k) - y(k) Formula 2

[0099] In Formula 2, e(k) represents the error between the expected output data d(k) and the actual output data y(k). y(k) represents the actual output data of the sample sensing data (i.e., the actual value of the device sensing data filtered at the k-th acquisition moment), and d(k) represents the expected output data of the sample sensing data (i.e., the expected value of the device sensing data filtered at the k-th acquisition moment).

[0100] 3032A. Obtain the target learning rate of the filter.

[0101] Among them, the target learning rate is a parameter used to control the amplitude of the filter coefficient adjustment each time, so as to control the update speed of the filter.

[0102] There are various ways to obtain the target learning rate in step 3032A. Exemplarily, they include:

[0103] Method 1: In some embodiments, the same learning rate is used in all stages, such as the button press stage, the button release stage, etc. For example, the target learning rate is set to 10 -6 .

[0104] Method 2: In some embodiments, the interaction control device generates jitter signals of different intensities due to button operations. The jitter intensities of the interaction control device in different stages are as follows: the jitter intensity in the button press stage > the jitter intensity in the first target stage. The target learning rate is divided into the first learning rate in the button press stage and the second learning rate in the first target stage, and the first learning rate is greater than the second learning rate. Considering that the vibration of the interaction control device is relatively intense in the button press stage, in order to quickly adapt to the change of vibration, a relatively large learning rate can be set for the button press stage and a relatively small learning rate can be set for the first target stage; in this way, the learning rate of the filter can be dynamically adjusted, so that when the vibration is strong, the learning rate of the filter increases to quickly adapt to the vibration change; when the vibration weakens, the learning rate decreases to avoid excessive smoothing. At this time, the target learning rate includes the first learning rate and the second learning rate, and step 3032A includes: obtaining the first learning rate of the sample sensing data in the button press stage and obtaining the second learning rate of the sample sensing data in the first target stage.

[0105] Mode 3: In some embodiments, the interaction control device generates jitter signals of different intensities due to key operations. The jitter intensities of the interaction control device at different stages are as follows: the jitter intensity in the key release stage < the jitter intensity in the second target stage. The target learning rate is divided into a third learning rate in the key release stage and a fourth learning rate in the second target stage, and the third learning rate is less than the fourth learning rate. Considering that the vibration of the interaction control device in the key release stage is relatively weak, in order to quickly adapt to the change of vibration, a relatively small learning rate can be set for the key release stage and a relatively large learning rate can be set for the second target stage; in this way, the learning rate of the filter can be dynamically adjusted, so that when the vibration is strong, the learning rate of the filter increases to quickly adapt to the change of vibration; when the vibration weakens, the learning rate decreases to avoid over-smoothing. At this time, the target learning rate includes the third learning rate and the fourth learning rate, and step 3032A includes: obtaining the third learning rate of the sample sensing data in the key release stage and obtaining the fourth learning rate of the sample sensing data in the second target stage.

[0106] Mode 4: In some embodiments, the interaction control device generates three jitter signals of different intensities due to key operations. The three stages of the jitter signal generated by the interaction control device due to key operations include: the key press stage, the key release stage, and other stages. The jitter intensities of the interaction control device in the three stages are as follows: the jitter intensity in the key press stage > the jitter intensity in the key release stage > the jitter intensity in other stages. The target learning rate is divided into a first learning rate in the key press stage, a third learning rate in the key release stage, and a fifth learning rate in other stages. The first learning rate is greater than the third learning rate, and the third learning rate is greater than the fifth learning rate. First, considering that the vibration of the interaction control device in the key press stage is the most intense, in order to quickly adapt to the change of vibration, a relatively large learning rate can be set for the key press stage; second, considering that the vibration of the interaction control device in the key release stage is relatively weak, in order to quickly adapt to the change of vibration, a relatively small learning rate can be set for the key release stage; third, considering that the vibration of the interaction control device in other stages is the weakest, in order to quickly adapt to the change of vibration, the smallest learning rate can be set for other stages; in this way, the learning rate of the filter can be dynamically adjusted, so that when the vibration is strong, the learning rate of the filter increases to quickly adapt to the change of vibration; when the vibration weakens, the learning rate decreases to avoid over-smoothing. At this time, the target learning rate includes the first learning rate, the third learning rate, and the fifth learning rate, and step 3032A includes: obtaining the first learning rate of the sample sensing data in the key press stage, obtaining the third learning rate of the sample sensing data in the key release stage, and obtaining the fifth learning rate of the sample sensing data in other stages.

[0107] 3033A. With the goal of minimizing the sum of squares of the target error, adjust the initialized filter coefficients according to the target error and the target learning rate to obtain the learned filter coefficients of the filter.

[0108] Corresponding to step 3032A, there are multiple implementation manners for step 3033A. Exemplarily, they include:

[0109] Manner 1: In some embodiments, the same learning rate is used in all stages, such as the button press stage, the button release stage, etc. For example, the target learning rate is set to 10 -6 .

[0110] Exemplarily, the filter coefficient update manner can be as shown in Formula 3 below:

[0111] w (k+1) = w k + μ * e(k) * x(k) Formula 3

[0112] In Formula 3, w (k+1) represents the learned filter coefficient at the (k + 1)-th acquisition moment, μ represents the target learning rate (used to control the update speed of the filter), x(k) represents the input sensing data (i.e., the sample sensing data, specifically representing the device sensing data at the k-th acquisition moment), e(k) represents the target error (specifically representing the error between the expected output data d(k) and the actual output data y(k)), and w k represents the learned filter coefficient at the k-th acquisition moment.

[0113] At this time, in step 3033A, with the goal of minimizing the sum of squares of the target error e(k), the target learning rate u (such as u = 10 -6 ), the target error e(k), the sample sensing data x(k), and the learned filter coefficient w k at the k-th acquisition moment can be substituted into a preset filter coefficient update formula (such as Formula 3) to calculate the learned filter coefficient w (k+1) at the (k + 1)-th acquisition moment.

[0114] Method 2: In some embodiments, the target learning rate is divided into a first learning rate in the button-pressing stage and a second learning rate in the first target stage. At this time, the sample sensing data includes first sample sensing data in the button-pressing stage and second sample sensing data in the first target stage. The jitter intensity of the interaction control device in the first target stage is less than that in the button-pressing stage. The target error includes a first error between the actual output data and the expected output data of the first sample sensing data, and a second error between the actual output data and the expected output data of the second sample sensing data. Step 3033A may specifically include: aiming to minimize the sum of squares of the first error, adjusting the initialized filtering coefficient according to the first error and the first learning rate to obtain the learned filtering coefficient in the button-pressing stage; aiming to minimize the sum of squares of the second error, adjusting the initialized filtering coefficient according to the second error and the second learning rate to obtain the learned filtering coefficient in the first target stage.

[0115] For example, in step 3033A, on the one hand, aiming to minimize the sum of squares of the first error e 1 (k), taking the first standard learning rate μ 1 (such as μ 1 = 10 -3 ), the first error e 1 (k), the first sample sensing data x 1 (k), and the learned filtering coefficient at the k-th acquisition moment and substituting them into the preset filtering coefficient update formula (refer to formula 3), calculating to obtain the learned filtering coefficient at the (k + 1)-th acquisition moment as the learned filtering coefficient in the button-pressing stage. On the other hand, aiming to minimize the sum of squares of the second error e 2 (k), taking the second standard learning rate μ 2 (such as μ 2 = 10 -5 ), the second error e 2 (k), the second sample sensing data x 2 (k), and the learned filtering coefficient at the k-th acquisition moment and substituting them into the preset filtering coefficient update formula (refer to formula 3), calculating to obtain the learned filtering coefficient at the (k + 1)-th acquisition moment as the learned filtering coefficient in the first target stage.

[0116] Mode 3: In some embodiments, the target learning rate is divided into a third learning rate in the key-release stage and a fourth learning rate in the second target stage. At this time, the sample sensing data includes third sample sensing data in the key-release stage and fourth sample sensing data in the second target stage. The jitter intensity of the interaction control device in the second target stage is greater than that in the key-release stage. The target error includes a third error between the actual output data and the expected output data of the third sample sensing data, and a fourth error between the actual output data and the expected output data of the fourth sample sensing data. Step 3033A may specifically include: aiming to minimize the sum of squares of the third error, adjusting the initialized filtering coefficient according to the third error and the third learning rate to obtain the learned filtering coefficient in the key-release stage; aiming to minimize the sum of squares of the fourth error, adjusting the initialized filtering coefficient according to the fourth error and the fourth learning rate to obtain the learned filtering coefficient in the second target stage.

[0117] For example, in step 3033A, on the one hand, aiming to minimize the sum of squares of the third error e 3 (k), substituting the third target learning rate μ 3 (such as μ 3 = 10 -6 ), the third error e 3 (k), the third sample sensing data x 3 (k) and the learned filtering coefficient at the k-th acquisition moment into the preset filtering coefficient update formula (reference formula 3), and calculating to obtain the learned filtering coefficient at the (k + 1)-th acquisition moment as the learned filtering coefficient in the key-release stage. On the other hand, aiming to minimize the sum of squares of the fourth error e 4 (k), substituting the fourth target learning rate μ 4 (such as μ 4 = 10 -5 ), the fourth error e 4 (k), the fourth sample sensing data x 4 (k) and the learned filtering coefficient at the k-th acquisition moment into the preset filtering coefficient update formula (reference formula 3), and calculating to obtain the learned filtering coefficient at the (k + 1)-th acquisition moment as the learned filtering coefficient in the second target stage.

[0118] Mode 4: In some embodiments, the target learning rate is divided into a first learning rate in the key-pressing stage, a third learning rate in the key-releasing stage, and a fifth learning rate in other stages. At this time, the sample sensing data includes first sample sensing data in the key-pressing stage, third sample sensing data in the key-releasing stage, and fifth sample sensing data in other stages. The target error includes a first error between the actual output data and the expected output data of the first sample sensing data, a third error between the actual output data and the expected output data of the third sample sensing data, and a fifth error between the actual output data and the expected output data of the fifth sample sensing data. Step 3033A may specifically include: aiming at minimizing the sum of squares of the first error, adjusting the initialized filtering coefficient according to the first error and the first learning rate to obtain the learned filtering coefficient in the key-pressing stage; aiming at minimizing the sum of squares of the third error, adjusting the initialized filtering coefficient according to the third error and the third learning rate to obtain the learned filtering coefficient in the key-releasing stage; aiming at minimizing the sum of squares of the fifth error, adjusting the initialized filtering coefficient according to the fifth error and the fifth learning rate to obtain the learned filtering coefficient in other stages.

[0119] For example, in the first aspect, aiming at minimizing the sum of squares of the first error e 1 (k), substituting the first standard learning rate μ 1 (such as μ 1 = 10 -3 ), the first error e 1 (k), the first sample sensing data x 1 (k), and the learned filtering coefficient at the k-th acquisition moment into the preset filtering coefficient update formula (refer to Formula 3), and calculating to obtain the learned filtering coefficient at the (k + 1)-th acquisition moment as the learned filtering coefficient in the key-pressing stage. In the second aspect, aiming at minimizing the sum of squares of the third error e 3 (k), substituting the third standard learning rate μ 3 (such as μ 3 = 10 -6 ), the third error e 3 (k), the third sample sensing data x 3 (k), and the learned filtering coefficient at the k-th acquisition moment into the preset filtering coefficient update formula (refer to Formula 3), and calculating to obtain the learned filtering coefficient at the (k + 1)-th acquisition moment as the learned filtering coefficient in the key-releasing stage. In the third aspect, aiming at minimizing the fifth error e5 Taking the sum of squares of (k) as the target, the fifth learning rate μ 5 (such as μ 5 = 10 -5 ), the fifth error e 5 (k), the fifth sample sensing data x 5 (k) and the learned filter coefficient at the k-th acquisition moment are substituted into the preset filter coefficient update formula (reference formula 3), and the learned filter coefficient at the (k + 1)-th acquisition moment is calculated as the learned filter coefficient for other stages.

[0120] ② In some embodiments, a recursive least squares algorithm is used to learn the filter coefficient. For example Figure 5 as shown, at this time, step 303 may specifically include the following steps 3031B to 3033B:

[0121] 3031B. Obtain the target error between the actual output data and the desired output data.

[0122] The implementation of step 3031B is similar to the implementation of step 3031A. For specific reference, please refer to the relevant description above. Details are not repeated here.

[0123] 3032B. Obtain the covariance matrix of the sample sensing data.

[0124] Exemplarily, the target error, a preset forgetting factor (used to control the weight of the device sensing data at historical acquisition moments), and the sample sensing data can be substituted into a preset covariance formula (for example, the preset covariance formula can be shown as formula 4 below) to calculate the covariance matrix.

[0125]

[0126] In formula 4, P (k+1) represents the covariance matrix at the (k + 1)-th acquisition moment, P k represents the covariance matrix at the k-th acquisition moment, λ represents the preset forgetting factor, x(k) represents the input sensing data (i.e., the sample sensing data, specifically representing the device sensing data at the k-th acquisition moment), e(k) represents the target error (specifically representing the error between the desired output data d(k) and the actual output data y(k)), and w k represents the learned filter coefficient at the k-th acquisition moment.

[0127] 3033B. Taking the minimization of the sum of squares of the target error as the goal, adjust the initial filter coefficient according to the target error and the covariance matrix to obtain the learned filter coefficient of the filter.

[0128] The implementation of step 3033B is similar to that of step 3033A. For specific details, please refer to the relevant descriptions above and will not be elaborated here.

[0129] In this way, by using the error between the actual output data and the expected output data, the initial filtering coefficients of the filter are adjusted so that the output data of the filter is as close as possible to the expected output data, thereby improving the filtering accuracy of the filter for the jitter signal.

[0130] 203. Output a control signal for the controlled object of the smart wearable device according to the filtered sensing data.

[0131] In some embodiments, the controlled object is a cursor. At this time, in step 203, a control signal for the cursor of the smart wearable device can be output according to the filtered sensing data to control the display position of the cursor on the smart wearable device. In this way, since the filtered sensing data can filter out the jitter signal of the interaction control device (such as a finger ring), it is avoided that the target sensing data carries the jitter signal of the interaction control device (such as a finger ring) resulting in cursor control error, and further the problem of irregular jitter of the cursor is avoided, and the control accuracy of the cursor is provided.

[0132] In some embodiments, the controlled object is display content (such as sliding list content). At this time, in step 203, a control signal for the display content of the smart wearable device can be output according to the filtered sensing data to control the display position of the display content (such as sliding list content) on the smart wearable device. In this way, since the filtered sensing data can filter out the jitter signal of the interaction control device (such as a finger ring), it is avoided that the target sensing data carries the jitter signal of the interaction control device (such as a finger ring) resulting in display content (such as sliding list content) control error, and further the problem of irregular jitter of the display content (such as sliding list content) is avoided, and the control accuracy of the display content (such as sliding list content) is provided.

[0133] In addition, to better implement the control signal output method in the embodiments of the present application, based on the control signal output method, an embodiment of a control signal output device is further provided in the embodiments of the present application, as Figure 6 shown, which is a schematic structural diagram of an embodiment of the control signal output device provided by the embodiments of the present application. The control signal output device 600 includes:

[0134] An acquisition unit 601, configured to acquire target sensing data of an interaction control device of the smart wearable device;

[0135] A filtering unit 602, configured to, when detecting a jitter signal generated by a key operation of the interaction control device, perform filtering processing on the target sensing data through a filter to obtain filtered sensing data;

[0136] A control unit 603 for outputting a control signal of a controlled object of the smart wearable device according to the filtered sensed data.

[0137] In some embodiments, the control signal output device further includes a learning unit (not shown in the figure), and the learning unit is configured to:

[0138] Obtain the sample sensed data of the interactive control device and the expected output data of the sample sensed data;

[0139] Filter the sample sensed data through the filter according to the initial filtering coefficients to obtain the actual output data of the sample sensed data;

[0140] Adjust the initial filtering coefficients of the filter according to the error between the actual output data and the expected output data to obtain the learned filtering coefficients of the filter.

[0141] In some embodiments, the filtering unit 602 is configured to:

[0142] Filter the target sensed data through the filter according to the learned filtering coefficients to obtain the filtered sensed data.

[0143] In some embodiments, the learning unit is configured to:

[0144] Obtain the target error between the actual output data and the expected output data;

[0145] Obtain the target learning rate of the filter;

[0146] With the goal of minimizing the sum of squares of the target error, adjust the initial filtering coefficients according to the target error and the target learning rate to obtain the learned filtering coefficients of the filter.

[0147] In some embodiments, the target learning rate includes a first learning rate and a second learning rate, the first learning rate is greater than the second learning rate, the sample sensed data includes first sample sensed data in the key-pressing stage and second sample sensed data in the first target stage, the jitter intensity of the interactive control device in the first target stage is less than the jitter intensity in the key-pressing stage, and the target error includes a first error between the actual output data and the expected output data of the first sample sensed data and a second error between the actual output data and the expected output data of the second sample sensed data. The learning unit is configured to:

[0148] With the goal of minimizing the sum of squares of the first error, adjust the initialized filter coefficients according to the first error and the first learning rate to obtain the learned filter coefficients in the key-pressing stage;

[0149] With the goal of minimizing the sum of squares of the second error, adjust the initialized filter coefficients according to the second error and the second learning rate to obtain the learned filter coefficients in the first target stage.

[0150] In some embodiments, the filtering unit 602 is configured to:

[0151] Filter the target sensing data through a filter according to the learned filter coefficients in the key-pressing stage to obtain the filtered sensing data corresponding to the key-pressing stage;

[0152] And / or, filter the target sensing data through a filter according to the learned filter coefficients in the first target stage to obtain the filtered sensing data corresponding to the first target stage.

[0153] In some embodiments, the target learning rate includes a third learning rate and a fourth learning rate, the third learning rate is less than the fourth learning rate, the sample sensing data includes third sample sensing data in the key-releasing stage and fourth sample sensing data in the second target stage, the jitter intensity of the interaction control device in the second target stage is greater than the jitter intensity in the key-releasing stage, the target error includes a third error between the actual output data and the expected output data of the third sample sensing data, and a fourth error between the actual output data and the expected output data of the fourth sample sensing data. The learning unit is configured to:

[0154] With the goal of minimizing the sum of squares of the third error, adjust the initialized filter coefficients according to the third error and the third learning rate to obtain the learned filter coefficients in the key-releasing stage;

[0155] With the goal of minimizing the sum of squares of the fourth error, adjust the initialized filter coefficients according to the fourth error and the fourth learning rate to obtain the learned filter coefficients in the second target stage.

[0156] In some embodiments, the filtering unit 602 is configured to:

[0157] Filter the target sensing data through a filter according to the learned filter coefficients in the key-releasing stage to obtain the filtered sensing data corresponding to the key-releasing stage;

[0158] And / or, filter the target sensing data according to the learned post-filtering coefficients in the second target stage by a filter to obtain the filtered sensing data corresponding to the second target stage.

[0159] In some embodiments, the learning unit is configured to:

[0160] Obtain the target error between the actual output data and the expected output data;

[0161] Obtain the covariance matrix of the sample sensing data;

[0162] With the goal of minimizing the sum of squares of the target error, adjust the initialized filtering coefficients according to the target error and the covariance matrix to obtain the learned post-filtering coefficients of the filter.

[0163] In some embodiments, the controlled object is a cursor or display content, and the control unit 603 is configured to:

[0164] Output a control signal of the cursor of the intelligent wearable device according to the filtered sensing data to control the display position of the cursor on the intelligent wearable device;

[0165] Or, output a control signal of the display content of the intelligent wearable device according to the filtered sensing data to control the display position of the display content on the intelligent wearable device.

[0166] In specific implementation, the above-mentioned each unit can be implemented as an independent entity, or can be arbitrarily combined and implemented as the same or several entities. For the specific implementation of the above-mentioned each unit, reference can be made to the embodiments of the control signal output method described above, which will not be elaborated here.

[0167] Those of ordinary skill in the art can understand that all or part of the steps in the above control signal output method can be completed by instructions, or by controlling relevant hardware through instructions. These instructions can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0168] For this reason, an embodiment of the present application provides a computer-readable storage medium, in which multiple computer programs are stored, and these computer programs can be loaded by a processor to execute any control signal output method provided by the embodiments of the present application.

[0169] Among them, the computer-readable storage medium may include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disc, etc.

[0170] In the above embodiments of the control signal output device, computer-readable storage medium, and smart wearable device, the descriptions of the respective embodiments have their own focuses. For parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments. Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes and beneficial effects of the above-described control signal output device, computer-readable storage medium, smart wearable device, and their corresponding units can refer to the description of the control signal output method in the above embodiments, and will not be elaborated herein specifically.

[0171] The above has introduced in detail a control signal output method, device, smart wearable device, and computer-readable storage medium provided by the embodiments of the present application. Specific examples are used herein to elaborate on the principle and implementation manner of the present application. The description of the above embodiments is only used to help understand the method and its core idea of the present application; at the same time, for those skilled in the art, according to the idea of the present application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present application.

Claims

1. A method for outputting a control signal, characterized in that, The method includes: Obtaining target sensing data of an interaction control device of an intelligent wearable device; When a jitter signal generated by a key operation of the interaction control device is detected, filtering the target sensing data through a filter to obtain filtered sensing data; Outputting a control signal of a controlled object of the intelligent wearable device according to the filtered sensing data.

2. The control signal output method according to claim 1, wherein The filter learns filtering coefficients in the following manner: Obtaining sample sensing data of the interaction control device and expected output data of the sample sensing data; Filtering the sample sensing data through the filter according to initialized filtering coefficients to obtain actual output data of the sample sensing data; Adjusting the initialized filtering coefficients of the filter according to an error between the actual output data and the expected output data to obtain learned filtering coefficients of the filter; The filtering the target sensing data through the filter to obtain filtered sensing data includes: Filtering the target sensing data through the filter according to the learned filtering coefficients to obtain filtered sensing data.

3. The control signal output method according to claim 2, characterized in that The adjusting the initialized filtering coefficients of the filter according to an error between the actual output data and the expected output data to obtain learned filtering coefficients of the filter includes: Obtaining a target error between the actual output data and the expected output data; Obtaining a target learning rate of the filter; Adjusting the initialized filtering coefficients according to the target error and the target learning rate with the goal of minimizing the sum of squares of the target error to obtain learned filtering coefficients of the filter.

4. The control signal output method according to claim 3, wherein The target learning rate includes a first learning rate and a second learning rate, the first learning rate is greater than the second learning rate, the sample sensing data includes first sample sensing data in a key-pressing stage and second sample sensing data in a first target stage, the jitter intensity of the interaction control device in the first target stage is less than the jitter intensity in the key-pressing stage, and the target error includes a first error between the actual output data of the first sample sensing data and the expected output data of the first sample sensing data, and a second error between the actual output data of the second sample sensing data and the expected output data of the second sample sensing data; The adjusting the initialized filtering coefficients according to the target error and the target learning rate with the goal of minimizing the sum of squares of the target error to obtain learned filtering coefficients of the filter includes: Adjusting the initialized filtering coefficients according to the first error and the first learning rate with the goal of minimizing the sum of squares of the first error to obtain learned filtering coefficients in the key-pressing stage; Adjusting the initialized filtering coefficients according to the second error and the second learning rate with the goal of minimizing the sum of squares of the second error to obtain learned filtering coefficients in the first target stage; Filtering the target sensing data according to the learned filter coefficients to obtain filtered sensing data, including: Filtering the target sensing data according to the learned filter coefficients in the key-pressing stage to obtain the filtered sensing data corresponding to the key-pressing stage; And / or, filtering the target sensing data according to the learned filter coefficients in the first target stage to obtain the filtered sensing data corresponding to the first target stage.

5. The control signal output method according to claim 3, wherein The target learning rate includes a third learning rate and a fourth learning rate, the third learning rate is less than the fourth learning rate, the sample sensing data includes third sample sensing data in the key-releasing stage and fourth sample sensing data in the second target stage, the jitter intensity of the interaction control device in the second target stage is greater than the jitter intensity in the key-releasing stage, and the target error includes a third error between the actual output data and the expected output data of the third sample sensing data, and a fourth error between the actual output data and the expected output data of the fourth sample sensing data; Taking minimizing the sum of squares of the target error as the goal, and adjusting the initial filter coefficients according to the target error and the target learning rate to obtain the learned filter coefficients of the filter, including: Taking minimizing the sum of squares of the third error as the goal, and adjusting the initial filter coefficients according to the third error and the third learning rate to obtain the learned filter coefficients in the key-releasing stage; Taking minimizing the sum of squares of the fourth error as the goal, and adjusting the initial filter coefficients according to the fourth error and the fourth learning rate to obtain the learned filter coefficients in the second target stage; Filtering the target sensing data according to the learned filter coefficients to obtain filtered sensing data, including: Filtering the target sensing data according to the learned filter coefficients in the key-releasing stage to obtain the filtered sensing data corresponding to the key-releasing stage; And / or, filtering the target sensing data according to the learned filter coefficients in the second target stage to obtain the filtered sensing data corresponding to the second target stage.

6. The control signal output method according to claim 2, characterized in that Adjusting the initial filter coefficients of the filter according to the error between the actual output data and the expected output data to obtain the learned filter coefficients of the filter, including: Obtaining the target error between the actual output data and the expected output data; Obtaining the covariance matrix of the sample sensing data; Taking minimizing the sum of squares of the target error as the goal, and adjusting the initial filter coefficients according to the target error and the covariance matrix to obtain the learned filter coefficients of the filter.

7. The control signal output method according to claim 1, wherein The controlled object is a cursor or display content, and outputting a control signal for the controlled object of the smart wearable device according to the filtered sensing data, including: Output a control signal for the cursor of the smart wearable device according to the filtered sensing data to control the display position of the cursor on the smart wearable device; Alternatively, output a control signal for the display content of the smart wearable device according to the filtered sensing data to control the display position of the display content on the smart wearable device.

8. A control signal output device, characterized in that, The control signal output device includes: An acquisition unit, configured to acquire target sensing data of an interaction control device of the smart wearable device; A filtering unit, configured to perform filtering processing on the target sensing data through a filter when detecting a jitter signal generated by a key operation of the interaction control device to obtain filtered sensing data; A control unit, configured to output a control signal for a controlled object of the smart wearable device according to the filtered sensing data.

9. An intelligent wearable device, characterized in that, It includes a processor and a memory, and a computer program is stored in the memory. When the processor calls the computer program in the memory, it executes the control signal output method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored thereon, and the computer program is loaded by a processor to execute the control signal output method according to any one of claims 1 to 7.

Images