Task processing method and device, storage medium and electronic equipment

Abstract

The invention discloses a task processing method and device, a storage medium and electronic equipment, and relates to the technical field of data processing.The task processing method comprises the steps that in response to a calling instruction for an intelligent assistant in a smart watch, candidate proxy equipment is determined, and the calling instruction is used for calling the intelligent assistant to execute a target task; performing consistency verification on the sensor feature data of the smart watch and the candidate proxy device; determining a target proxy device from the candidate proxy devices according to cost data and value data of data interaction between the smart watch and the candidate proxy devices under the condition that the consistency verification is passed; and sending an execution request of the target task to the target proxy device to request an intelligent assistant of the target proxy device to execute the target task and send an execution result of the target task to the intelligent watch. According to the method, the target task is executed through the target agent device and the result is returned, so that resource consumption generated by data acquisition of the smart watch and cloud interaction can be reduced, and the endurance pressure of the smart watch is relieved.

Description

Technical Field

[0001] This application relates to the field of data processing technology, and in particular to a task processing method, apparatus, storage medium and electronic device. Background Technology

[0002] Currently, when a user invokes the AI ​​assistant on their smartwatch to perform a task, the smartwatch uses its own sensors to collect data, which is then transmitted to a cloud service interface associated with the watch account. The smartwatch can then process the data based on the execution information returned from the cloud, and finally obtain the processing result from the AI ​​assistant and display it.

[0003] However, using this method, due to the limitations of the smartwatch's computing power and battery life, will result in significant resource consumption during data collection and cloud interaction, thus affecting the watch's battery life. Summary of the Invention

[0004] In view of this, this application provides a task processing method, apparatus, storage medium, and electronic device, the main purpose of which is to improve the technical problem that the existing technology, due to the limitations of the computing power and battery life of smartwatches, will generate large resource consumption during data collection and cloud interaction, thus affecting the battery life of the watch.

[0005] Firstly, this application provides a task processing method for use in a smartwatch, comprising: In response to a call command to the smart assistant in the smartwatch, a candidate agent device corresponding to the smartwatch is determined, wherein the call command is used to call the smart assistant to perform the target task; Perform consistency verification on the sensor feature data of the smartwatch and the candidate agent device; If the consistency check passes, the target agent device is determined from the candidate agent devices based on the cost data and value data of the data interaction between the smartwatch and the candidate agent devices. Send an execution request for the target task to the target agent device to request the smart assistant of the target agent device to execute the target task and send the execution result of the target task to the smartwatch.

[0006] Optionally, determining the target agent device from the candidate agent devices based on the cost and value data of data interaction between the smartwatch and the candidate agent devices includes: Determine the cost and value data of data interaction between the smartwatch and the candidate agent device; If the cost data is less than the value data, the device with the highest value data among the candidate agent devices is determined as the target agent device.

[0007] Optionally, determining the cost and value data for data interaction between the smartwatch and the candidate agent device includes: Determine the load data and power data corresponding to the smartwatch and the candidate agent device respectively, and generate the cost data based on the load data and power data; The multi-dimensional transmission data of the smartwatch and the candidate agent device are determined, and the value data is generated based on the multi-dimensional transmission data.

[0008] Optionally, after determining the cost and value data of the data interaction between the smartwatch and the candidate agent device, the method further includes: If the cost data is greater than the value data, an execution request for the target task is sent to the server to request the server to execute the target task and send the execution result of the target task to the smartwatch. In response to receiving the execution result, the execution result is displayed on the smartwatch.

[0009] Optionally, the consistency verification of the sensor feature data between the smartwatch and the candidate agent device includes: Determine the first target sensor data corresponding to the smartwatch; Receive the second target sensor data sent by the candidate agent device; Based on the first target sensor data and the second target sensor data, a consistency verification is performed on the sensor feature data of the smartwatch and the candidate agent device.

[0010] Optionally, determining the first target sensor data corresponding to the smartwatch includes: Determine the multi-dimensional sensor data corresponding to the smartwatch; Feature extraction is performed on the multi-dimensional sensor data to obtain the first target sensor data corresponding to the smartwatch.

[0011] Optionally, before receiving the second target sensor data sent by the candidate agent device, the method further includes: Generate a monitoring notification to listen to data from the candidate agent device; The monitoring notification is broadcast to the candidate agent device, which responds to the monitoring notification, determines the second target sensor data, and sends the second target sensor data to the smartwatch.

[0012] Optionally, the step of verifying the consistency of sensor feature data between the smartwatch and the candidate proxy device based on the first target sensor data and the second target sensor data includes: Based on the trend consistency analysis of the first target sensor data and the second target sensor data, trend consistency data is obtained. If the trend consistency data is greater than the trend consistency threshold, it is determined that the smartwatch and the candidate agent device have passed the consistency check.

[0013] Optionally, after sending the execution request for the target task to the target agent device, the method further includes: Receive the execution result of the target task sent by the target agent device; The execution result is adapted and transformed based on the trend consistency data to obtain the target execution result corresponding to the target task; The results of the target execution will be displayed on the smartwatch.

[0014] Secondly, this application provides a task processing method applied to a target agent device, comprising: In response to receiving an execution request for a target task sent by a smartwatch, the system acquires the target sensor data corresponding to the target task. Based on the target sensor data and the target task, a smart assistant execution request corresponding to the target agent device is generated. The smart assistant execution request is used to request the server to execute the target task. The system receives the execution result of the target task sent by the server and sends the execution result to the smartwatch.

[0015] Thirdly, this application provides a task processing system, including a smartwatch and a target agent device, wherein the smartwatch is configured to implement the task processing method described in the first aspect, and the target agent device is configured to implement the task processing method described in the second aspect.

[0016] Fourthly, this application provides a task processing apparatus, comprising: The determination module is configured to determine a candidate agent device corresponding to the smartwatch in response to a call command to the smart assistant in the smartwatch, wherein the call command is used to call the smart assistant to perform the target task; The verification module is configured to perform consistency verification on the sensor feature data of the smartwatch and the candidate agent device. The determination module is also configured to, if the consistency check passes, determine the target agent device from the candidate agent devices based on the cost data and value data of the data interaction between the smartwatch and the candidate agent devices; The sending module is configured to send an execution request for the target task to the target agent device, so as to request the smart assistant of the target agent device to execute the target task and send the execution result of the target task to the smartwatch.

[0017] Fourthly, this application provides a task processing apparatus, comprising: The acquisition module is configured to acquire target sensor data corresponding to the target task in response to receiving an execution request for the target task sent by the smartwatch; The generation module is configured to generate a smart assistant execution request corresponding to the target agent device based on the target sensor data and the target task. The smart assistant execution request is used to request the server to execute the target task. The sending module is configured to receive the execution result of the target task sent by the server and send the execution result to the smartwatch.

[0018] Fifthly, this application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the task processing methods described in the first and second aspects.

[0019] In a sixth aspect, this application provides an electronic device, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the computer program to implement the task processing methods described in the first and second aspects.

[0020] By employing the above technical solutions, this application provides a task processing method, apparatus, storage medium, and electronic device. This application determines candidate proxy devices in response to a smart assistant's invocation command, providing potential device selection for the proxy execution of the target task; ensures the adaptability of the proxy device to execute the target task by performing consistency verification on the sensor feature data of the smartwatch and the candidate proxy devices; ensures the rationality of the proxy execution by determining the target proxy device from the candidate proxy devices based on cost and value data after the consistency verification passes; and reduces the resource consumption caused by the smartwatch's own data collection and cloud interaction by sending an execution request for the target task to the target proxy device, requesting the smart assistant to execute the task and return the result, thus alleviating the battery life pressure on the smartwatch. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating a task processing method provided in an embodiment of this application is shown; Figure 2 A flowchart illustrating a task processing method provided in an embodiment of this application is shown; Figure 3 A flowchart illustrating a task processing method provided in an embodiment of this application is shown; Figure 4 The illustration shows a flowchart of a smartwatch AI assistant collaborative agent example based on trend consistency difference provided in an embodiment of this application; Figure 5 This paper shows a schematic diagram of the structure of a task processing device provided in an embodiment of this application; Figure 6 This paper shows a schematic diagram of the structure of a task processing device provided in an embodiment of this application; Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of this application is shown. Detailed Implementation

[0024] The embodiments of this application will now be described in more detail with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features described herein can be combined with each other.

[0025] To address the technical issue of limited computing power and battery life in smartwatches, which leads to significant resource consumption and impacts battery life during data collection and cloud interaction, this embodiment provides a task processing method, such as... Figure 1 As shown, the method includes: Step 101: In response to the call command of the smart assistant in the smartwatch, determine the candidate agent device corresponding to the smartwatch.

[0026] The invocation command is used to invoke the smart assistant to execute the target task.

[0027] In this embodiment, the invocation command can be a command triggered by the user through the smartwatch's operating interface, or it can be a command automatically generated by the smartwatch. The invocation command can be used to trigger the smart assistant to execute a specific target task. For example, the invocation command in this embodiment may specifically include a voice wake-up command, a touch operation command, a timed trigger command, etc., and the target task may specifically include a health data monitoring task, a location query task, an information interaction task, etc.

[0028] In this embodiment, the candidate proxy device can be a device located within the same central point range as the current smartwatch, capable of data interaction, and possessing the ability to perform target tasks on behalf of the smartwatch. The candidate proxy device can be used to perform target tasks on behalf of the smart assistant of the current smartwatch, thereby reducing the resource consumption of the current smartwatch. For example, the candidate proxy device in this embodiment may specifically include children's smartwatches within the same system, smartphones that support connection to the current smartwatch system, and other devices from the same source.

[0029] In this embodiment, a compatible device can be a device capable of data interaction with the current smartwatch and meeting sensor compatibility requirements. Compatible devices can be identified and filtered based on broadcast interaction. For example, compatible devices in this embodiment may specifically include children's smartwatches running the same operating system, or smartphones with interactive applications adapted to the current smartwatch.

[0030] In this embodiment of the application, the smartwatch CW can obtain surrounding source devices through broadcast scanning and identify qualified source devices as candidate proxy devices. Specifically, the smartwatch CW can first send a source device check broadcast message (PM), which may contain the current device identifier ID and the list of sensors (LSE) required to perform the target task. After receiving the PM, the surrounding device can read its own available sensor list (LSS) (excluding sensors that are currently occupied and have not been released). If the LSE is a subset of the LSS, it indicates that the surrounding device has proxy capabilities. The surrounding device can respond with a fixed message (ML), which may contain its own device identifier IDS, current status information (remaining battery power P, computing load CL), and message synchronization time delay (tl). The smartwatch CW can summarize all responding surrounding devices to form a surrounding source device list (SR), and the devices in the surrounding source device list SR can be candidate proxy devices.

[0031] Step 102: Perform consistency verification on the sensor feature data of the smartwatch and the candidate agent device.

[0032] In this embodiment, sensor feature data can be a dataset extracted from sensor data collected by a smartwatch or candidate agent device that reflects data trends and core features. Sensor feature data can be used to determine the similarity and compatibility of data collected by different devices.

[0033] In this embodiment, consistency verification can be a verification process based on the trend similarity of sensor feature data. Consistency verification can be used to determine whether the data collected by the candidate agent device has trend consistency with the data collected by the current smartwatch, thereby determining whether the candidate agent device is suitable for performing the target task. For example, in this embodiment, consistency verification can be specifically implemented by calculating the trend consistency degree and comparing it with a threshold.

[0034] In this embodiment, consistency verification of sensor feature data between the smartwatch and candidate agent devices can be performed by the smartwatch CW sending a sensor monitoring notification M to all devices in the candidate agent device list SR. The sensor monitoring notification M can include a message type identifier SLS and a sensor monitoring initiation time ts (ts is calculated from the current time tn and the maximum synchronization delay of all candidate devices to ensure that all devices start data collection synchronously). The candidate agent device can start data collection of its own sensor list LSS at time ts to obtain raw data DT. After feature extraction and compression, it generates feature compressed data DZT and sends it to the smartwatch CW. At the same time, the smartwatch CW can also start data collection of its own sensor LSE to obtain raw data DS. After feature extraction and compression in the same way, feature compressed data DZS can be generated. A trend comparison can be performed based on DZS and DZT to complete the consistency verification.

[0035] Step 103: If the consistency verification passes, determine the target agent device from the candidate agent devices based on the cost data and value data of data interaction between the smartwatch and the candidate agent devices.

[0036] In this embodiment, the cost data can be a quantified value of resource consumption generated when the smartwatch CW and the candidate agent device interact collaboratively. The cost data can be calculated based on the power status and computing load of both parties.

[0037] In the embodiments of this application, the value data can be the quantified value of the revenue brought by the agent's execution of the target task. The value data can be obtained by considering multiple dimensions such as data transmission efficiency, real-time interaction, and device adaptability.

[0038] In the embodiments of this application, the target proxy device can be the device that satisfies the condition that the value data is greater than the cost data and has the largest value data among the candidate devices that have passed the consistency check. The target proxy device can be used to reduce the resource consumption of the smartwatch CW through proxy execution.

[0039] For the embodiments of this application, the criterion for passing the consistency verification is that the trend consistency degree TRL is greater than or equal to the preset trend synchronization coefficient TRS (TRS is a preset empirical value or a dynamically adjusted value of the system). Passing the consistency verification indicates that the smartwatch CW and the candidate device are in similar scenarios and have consistent data trends, and can have the basis for proxying. The collaborative interaction cost data CE (based on the difference in power consumption and computing power load between the two parties) and the proxy value VP (based on parameters such as data transmission length, transmission speed, synchronization delay, device security area area, and sensor accuracy) can be calculated using cost data and value data. If VP is greater than CE, it means that the proxy benefit is higher than the consumption, and the device with the largest VP can be selected from the candidate devices as the target proxy device. If VP is less than or equal to CE, the proxy can be abandoned, and the smartwatch CW can directly interact with the server.

[0040] Step 104: Send an execution request for the target task to the target agent device to request the smart assistant of the target agent device to execute the target task and send the execution result of the target task to the smartwatch.

[0041] In this embodiment, the execution request can be a structured message containing core information about the target task. For example, the execution request in this embodiment specifically includes the call destination fu, the call time t, and the caller device identifier ID. The execution request can be used by the target proxy device to identify the task requirements and the source of the request.

[0042] In this embodiment, the smartwatch CW can send an execution request RA to the target agent device PT. After receiving the execution request RA, the target agent device PT can start its own sensors to collect the required data D according to the calling purpose fu, generate an AI intelligent assistant execution request RE (which may contain fu and D), and send it to the AI ​​intelligent assistant execution center under the corresponding cloud disk account. The AI ​​intelligent assistant execution center deploys a multimodal hybrid artificial intelligence large model, supports vectorized input of sensor data, and can perform model inference based on fu and D in RE to obtain the inference result RM. The AI ​​intelligent assistant execution center can feed back RM to the target agent device PT, and the target agent device PT can forward RM to the smartwatch CW and other source devices that requested the proxy.

[0043] Compared with existing technologies, the embodiments of this application determine candidate proxy devices by responding to the call instructions of the smart assistant, providing potential device selection for the proxy execution of the target task; ensure the adaptability of the proxy device to execute the target task by performing consistency verification on the sensor feature data of the smartwatch and the candidate proxy devices; ensure the rationality of proxy execution by determining the target proxy device from the candidate proxy devices based on cost data and value data when the consistency verification passes; and reduce the resource consumption caused by the smartwatch's own data collection and cloud interaction by sending the execution request of the target task to the target proxy device, requesting the smart assistant to execute the task and return the result, thus alleviating the battery life pressure of the smartwatch.

[0044] Optionally, when performing the task of "determining the target agent device from the candidate agent devices based on the cost and value data of data interaction between the smartwatch and the candidate agent devices," the following methods may be used, but are not limited to them: Figure 2 As shown, it includes: Step 201: Determine the cost and value data for data interaction between the smartwatch and the candidate agent device.

[0045] In the embodiments of this application, the determination of cost data can be based on the quantification of resource consumption in collaborative interaction, and the determination of value data can be based on the quantification of actual benefits of agent execution. Cost data and value data can be calculated based on device status parameters and transmission data characteristics.

[0046] Step 202: If the cost data is less than the value data, the device with the highest value data among the candidate agent devices shall be identified as the target agent device.

[0047] If the value data VP is greater than the cost data CE, the candidate agent device with the largest VP can be selected as the target agent device. If there are multiple candidate agent devices with the same maximum value of VP, the size of CE can be further compared (select the device with the smaller CE), or the device can be determined by random selection. If the VP of all candidate devices is less than or equal to CE, no agent is needed, and the smartwatch CW can directly interact with the server to execute tasks.

[0048] Optionally, when performing the "determining the cost data and value data of data interaction between the smartwatch and the candidate agent device", the following methods can be used, but are not limited to: determining the load data and power data corresponding to the smartwatch and the candidate agent device respectively, and generating cost data based on the load data and power data; determining the multi-dimensional transmission data of data interaction between the smartwatch and the candidate agent device, and generating value data based on the multi-dimensional transmission data.

[0049] In this embodiment of the application, the load data may be the computing load CL (processor utilization, task queue length, etc.).

[0050] In this embodiment of the application, the power data can be the remaining power percentage P.

[0051] For embodiments of this application, multi-dimensional data transmission may include characteristic data transmission length tdl, transmission speed tsp, and message synchronization time delay tl.

[0052] For embodiments of this application, device parameters may include the safe area as set by the smartwatch CW, the sensor accuracy pr, and the number of supported sports modes nss.

[0053] For the embodiments of this application, the calculation formula for cost data CE can be as shown in Formula 1, where PC can represent the remaining battery power of the smartwatch CW, CLC can represent the remaining battery power of the candidate agent device ST, CL can represent the computing power load of the smartwatch CW, and CLS can represent the computing power load of the candidate agent device ST; Formula 1 can compare the differences in battery power and computing power load between the smartwatch and the candidate agent device, and take the maximum value of the difference as the cost data. If PC A larger CLC indicates that the smartwatch's CW battery is under more strain, allowing the distributor to conserve more power; if the CL... A larger CLS indicates that the smartwatch faces greater computational pressure from the CW (Computational Written Power) and that a proxy can alleviate more of the computational burden.

[0054] (Formula 1) In this embodiment, the calculation of the value data VP can be achieved by first determining the feature data transmission length tdl. The calculation formula for the feature data transmission length tdl is shown in Formula 2, where DZS can represent the smartwatch CW feature compression data and DZT can represent the candidate device feature compression data. The calculation formula for the value data VP is shown in Formula 3, where tsp can represent the historical transmission rate statistics of both parties, tl can represent the synchronization delay fed back by the candidate device, as can represent the geographical security range set by the smartwatch CW, pr can represent the measurement accuracy level of the smartwatch CW sensor, and nss can represent the total number of motion monitoring modes supported by the watch.

[0055] (Formula 2) (Formula 3) As an optional approach, after performing the "determination of cost data and value data for data interaction between the smartwatch and candidate agent devices", the following method can be used, but is not limited to: if the cost data is greater than the value data, sending an execution request for the target task to the server to request the server to execute the target task and send the execution result of the target task to the smartwatch; in response to receiving the execution result, displaying the execution result on the smartwatch.

[0056] In the embodiments of this application, when the cost data is greater than the value data, it can be concluded that the resource consumption of the proxy execution is higher than the actual benefit, and it is more reasonable for the smartwatch CW to execute tasks by directly connecting to the server. The execution request sent by the smartwatch CW to the server can include information such as the target task type, the raw data or feature data collected by its own sensors, and the device identifier. The server starts the corresponding AI intelligent assistant model, performs inference processing based on the received data, and can generate the execution result RM and provide feedback. The smartwatch CW can display the result to the user through screen display (such as numerical values ​​and charts), voice broadcast, etc., and can also directly save the RM to the family data of the corresponding cloud disk account for easy subsequent query and management.

[0057] As an optional approach, when performing the "consistency verification of sensor feature data between the smartwatch and the candidate agent device", the following method can be used, but is not limited to: determining the first target sensor data corresponding to the smartwatch; receiving the second target sensor data sent by the candidate agent device; and performing consistency verification of sensor feature data between the smartwatch and the candidate agent device based on the first target sensor data and the second target sensor data.

[0058] In this embodiment, the first target sensor data can be the feature compressed data DZS of a smartwatch CW. The smartwatch CW can activate the sensor list LSE to collect multi-dimensional raw data DS, and then perform preprocessing (denoising, outlier removal, standardization), feature extraction (time domain features such as mean and variance, frequency domain features such as Fourier transform results) and compression processing on the DS to obtain DZS.

[0059] In this embodiment, the second target sensor data can be the feature-compressed data (DZT) of the candidate agent device. The smartwatch CW can broadcast a sensor monitoring notification M to the candidate agent device, triggering the candidate agent device to broadcast synchronous sensor data acquisition and generate DZT.

[0060] In this embodiment of the application, the consistency verification of the sensor feature data of the smartwatch and the candidate agent device based on the first target sensor data and the second target sensor data can be performed by determining whether the changing trends of DZS and DZT are consistent through trend consistency analysis. If the changing trends of DZS and DZT meet the consistency conditions, the consistency verification can be passed.

[0061] As an optional approach, when performing the "determining the first target sensor data corresponding to the smartwatch", the following method can be used, but is not limited to: determining the multi-dimensional sensor data corresponding to the smartwatch; and extracting features from the multi-dimensional sensor data to obtain the first target sensor data corresponding to the smartwatch.

[0062] In this embodiment, the multi-dimensional sensor data can be data collected by the smartwatch CW through the sensor list LSE according to the target task requirements. The collection of multi-dimensional sensor data can be carried out according to the sampling frequency and can be aligned by timestamp to form a structured dataset. The structured dataset formed by multi-dimensional sensor data can ensure the temporal consistency of the data.

[0063] In this embodiment, feature extraction of multi-dimensional sensor data to obtain the first target sensor data corresponding to the smartwatch can be achieved by first dividing the original data into segments of a fixed length sl, with each segment denoted as pd_m (m being the segment number), and the total number of segments being... sd / sl (sd is the total size of the original data); perform Fast Fourier Transform (F) on each segment pd_m to obtain frequency domain data; extract the part of the frequency domain data that is higher than fh (couh) and lower than f_l (coul) according to the high frequency component threshold fh and the low frequency component threshold f_l of the system; perform Fast Fourier Transform (F^(-1)) on coul and coul respectively, and then concatenate the transformation results to obtain the compressed result of a single segment of data; the concatenation of the compressed results of all segments is the feature compressed data DZS (first target sensor data). The calculation formula of the first target sensor data can be shown in Formula 4, where pdz_m can represent the segmentation of the original data DZ according to the length sl. In the m-th segment after the initial data DT, sdz can represent the total size of data DZ; the calculation formula for the second target sensor data can be shown in Formula 5, where pd_m can represent the m-th segment after segmenting the original data DT by length sl, sd can represent the total size of data DT, and the z function can represent the compression of the data segments; the calculation formula for z(pdm) can be shown in Formula 6, where f_h can represent the high-frequency component threshold, f_l can represent the low-frequency component threshold, couh can represent the part of the frequency domain data that is higher than the high-frequency component threshold f_h, coul can represent the part of the frequency domain data that is lower than the high-frequency component threshold f_l, and pd_m can represent each segment of the first target sensor data (m is the segment number). sd / sl It can represent the number of segments in the first target sensor data (sd can represent the total size of the original data).

[0064] (Formula 4) (Formula 5) (Formula 6) As an alternative approach, before performing the "receiving second target sensor data sent by the candidate agent device", the following method may be used, but not limited to: generating a monitoring notification to monitor the data of the candidate agent device; broadcasting the monitoring notification to the candidate agent device, which responds to the monitoring notification, determines the second target sensor data, and sends the second target sensor data to the smartwatch.

[0065] In this embodiment, the monitoring notification M can be used by the candidate agent device to identify it as a sensor monitoring instruction. The format of the monitoring notification M can be M=(SLS,ts), where SLS can represent a fixed message identifier, and ts can represent the sensor monitoring initiation time. The calculation formula for ts can be as shown in Formula 7, where tn can represent the current system time of the smartwatch CW. The 'nss' parameter can represent the message synchronization delay from the devices in the candidate agent device list (SR), and nss can represent the total number of devices in the SR.

[0066] (Formula 7) In this embodiment of the application, the smartwatch CW can broadcast a notification M via short-range communication methods such as Bluetooth and Wi-Fi; the candidate agent device can parse the SLS and ts in the broadcast notification M, and at time ts, it can start its own available sensor list LSS to collect raw data DT, generate second target sensor data DZT according to the same feature extraction and compression process as the smartwatch CW, and then send DZT to the smartwatch CW through the same communication method; if the candidate device sensor is occupied or cannot meet the collection requirements, it can choose not to respond or provide an unavailable prompt.

[0067] As an optional approach, when performing the "consistency verification of sensor feature data between the smartwatch and the candidate agent device based on the first target sensor data and the second target sensor data", the following method can be used, but is not limited to: performing trend consistency analysis based on the first target sensor data and the second target sensor data to obtain trend consistency data; and determining that the smartwatch and the candidate agent device have passed the consistency verification if the trend consistency data is greater than the trend consistency threshold.

[0068] In this embodiment of the application, trend consistency analysis can be performed by calculating trend consistency data TRL, and the trend consistency threshold can be the trend synchronization coefficient TRS preset by the system (e.g., 0.8, which can be dynamically adjusted according to the task type).

[0069] For the embodiments of this application, the calculation of trend consistency data TRL can first calculate the minimum non-zero length cl of the feature compressed data. The calculation formula of cl can be as shown in Formula 8, where len(DZS) 0) can represent the length of non-zero terms in DZS, len(DZT) 0) can represent the length of non-zero terms in DZT; the formula for calculating the trend consistency data TRL can be shown in Formula 9, where cl can represent the minimum non-zero length of the feature compression data, DZT[m] can represent the m-th non-zero term in DZT, DZS[m] can represent the m-th non-zero term in DZS, and avg can represent the average proportion of non-zero components. It can represent the sensor noise ratio; if TRL is greater than or equal to TRS, it can indicate that the changing trends of DZS and DZT are highly consistent, and it can indicate that the smartwatch CW is in a similar scenario to the candidate device, thus passing the consistency check; if TRL is less than TRS, it fails the consistency check.

[0070] (Formula 8) (Formula Nine) As an optional approach, after executing "sending the execution request of the target task to the target agent device", the following methods can be used, but are not limited to: receiving the execution result of the target task sent by the target agent device; adapting and converting the execution result according to trend consistency data to obtain the target execution result corresponding to the target task; and displaying the target execution result on the smartwatch.

[0071] In this embodiment of the application, the execution result received by the smartwatch CW can be the cloud service inference result RM forwarded by the proxy device PT. Since the smartwatch CW and the proxy device may have different locations, resulting in different data intensity, the RM can be adapted and transformed using Trend Consistency Ratio (TRL) to obtain the target execution result FR that conforms to the actual scenario of the smartwatch CW. The formula for the adaptation and transformation can be as shown in Formula 10, where... This can represent a loop expansion and addition operation. Specifically, the loop expansion and addition operation can include: loop expansion of the TRL to the same length as the RM, and XOR operation between the expanded TRL and the RM. This can ensure that the FR after TRL adaptation and conversion is consistent with the execution result of the data collected by the smartwatch CW itself.

[0072] (Formula 10) In the embodiments of this application, the smartwatch CW can display the target execution result FR to the user through screen display, voice broadcast, etc.; at the same time, the smartwatch CW can upload TRL to the family cloud space of the corresponding cloud disk account for storage and establish a connection with RM; the agent device PT will also store RM to the logged-in family cloud disk.

[0073] Compared with existing technologies, this application's embodiments generate cost data by determining the load and power data of the smartwatch and candidate proxy devices, and generate value data by determining multi-dimensional transmission data, thus achieving precise quantification of cost and value. By sending an execution request to the server when the cost data exceeds the value data, receiving and displaying the results, it ensures that the task can still be executed normally when a proxy is unsuitable. It provides data support for proxy device selection by determining the first target sensor data of the smartwatch, receiving the second target sensor data of the candidate proxy devices, and performing consistency verification based on both. It simplifies the data source processing for consistency verification by determining the multi-dimensional sensor data of the smartwatch and extracting features to obtain the first target sensor data. It ensures data acquisition for consistency verification by generating a listening notification and broadcasting it to the candidate proxy devices, prompting them to determine and send the second target sensor data. It obtains trend consistency data by analyzing the trend consistency of the first and second target sensor data, and determines that the verification passes when the trend consistency data exceeds a threshold, thus clarifying the judgment criteria for consistency verification. Finally, it receives the execution results from the target proxy devices, adapts and converts them based on the trend consistency data to obtain the target execution results, and displays them, ensuring that the execution results meet the usage requirements of the smartwatch.

[0074] This embodiment also provides a task processing method, such as Figure 3 As shown, it includes: Step 301: In response to receiving the execution request for the target task sent by the smartwatch, obtain the target sensor data corresponding to the target task.

[0075] In this embodiment, in response to receiving an execution request for a target task sent by a smartwatch, the target agent device can first parse the calling destination fu in the request to determine the sensor type required by the target task. The target agent device can then activate its own available corresponding sensors to collect data according to the sampling frequency and collection duration, thereby obtaining the target sensor data D. During the collection process, if the required sensor is occupied by other applications, the target agent device can wait for the sensor to be released before collecting data, or send a prompt message to the smartwatch that sent the request indicating that the sensor is unavailable, to ensure the effectiveness of data collection.

[0076] Step 302: Based on the target sensor data and the target task, generate a smart assistant execution request corresponding to the target agent device. The smart assistant execution request is used to request the server to execute the target task.

[0077] In this embodiment of the application, the target agent device can encapsulate the parsed call destination fu and the collected target sensor data D to generate a smart assistant execution request RE, the format of which can be RE=(fu,D,ID).

[0078] In this embodiment, the server can be the AI ​​intelligent assistant execution center corresponding to the cloud disk account logged in by the target agent device. The multimodal hybrid artificial intelligence large model deployed in the AI ​​intelligent assistant execution center supports the input of various sensor data after vectorization, and can adapt to different types of target task inference needs.

[0079] Step 303: Receive the execution result of the target task sent by the server and send the execution result to the smartwatch.

[0080] In this embodiment, after receiving the execution request RE, the server-side AI intelligent assistant execution center can extract the calling destination fu and target sensor data D from the execution request RE, and vectorize D (converting it into a numerical format recognizable by the model). Based on the inference logic matched to fu, the vectorized D is input into the model to execute the inference task, obtaining the execution result RM. The server feeds back the RM to the target agent device. The target agent device receives the execution result RM and can identify the corresponding request-initiating device based on the smartwatch device identifier ID contained in the execution request RA. It then sends the RM to the smartwatch via short-range communication (such as Bluetooth). If multiple smartwatches send the same type of execution request to the target agent device, the target agent device can receive the RM from the server in batches and forward it one by one according to the device ID, achieving multi-device collaborative agent execution and improving overall resource utilization efficiency. Simultaneously, the target agent device can store the RM in its logged-in cloud storage account's home cloud drive for easy subsequent association queries and data management.

[0081] As an optional approach, this application also provides an example of a smartwatch AI assistant collaborative agent based on trend consistency differences. The flowchart of the smartwatch AI assistant collaborative agent example based on trend consistency differences is as follows: Figure 4 As shown, Figure 4 The specific steps may include: Step 1: After the children's smartwatch CW calls the AI ​​smart assistant, it first reads the list of nearby source devices SR (the children's smartwatch CW is the smartwatch in this application embodiment, the list of nearby source devices SR is the candidate proxy device list in this application embodiment, and the AI ​​smart assistant is the smart assistant in this application embodiment).

[0082] Step 2: After obtaining the list of nearby devices of the same origin (SR), the children's smartwatch sends a sensor monitoring notification (M) to the list of devices of the same origin (SR).

[0083] Step 3: The source device ST in the source device list SR (the source device ST is the candidate proxy device in this application embodiment) initiates time ts in the sensor listening notification M to start reading the data DT of its own sensor list LSS (the data DT is the original data of the second target sensor data in this application embodiment).

[0084] Step 4: After the children's smartwatch CW obtains the list of surrounding devices of the same origin SR and sends the sensor monitoring notification M, it also starts monitoring its own sensor LSE and reads the sensor data DS (the sensor data DS is the original data of the first target sensor data in this embodiment).

[0085] Step 5: The source device ST in the source device list SR performs feature extraction and compression on the read data DT to obtain feature compressed data DZT (feature compressed data DZT is the second target sensor data in this embodiment of the application), and sends it to the children's smartwatch CW.

[0086] Step 6: After receiving the feature compression data DZT sent by all the same source devices in the same source device list SR, the children's smartwatch CW uses feature extraction to compress the data DZ read by its own sensor to obtain its own feature compression data DZS (the own feature compression data DZS is the first target sensor data in the embodiment of this application).

[0087] Step 7: The child smartwatch CW compares its own feature compression data DZS with the feature compression data DZT of the same source device ST in the same source device list SR. When the trend consistency degree TRL of the self-feature compression data DZS and the feature compression data DZT of the same source device ST in the same source device list SR is greater than or equal to the trend synchronization coefficient TRS, it is considered that the trend is consistent, and the proxy value VP is calculated (the trend consistency degree TRL is the trend consistency data in this application embodiment, the trend synchronization coefficient TRS is the trend consistency threshold in this application embodiment, trend consistency means that the consistency verification in this application embodiment is passed, and the proxy value VP is the value data in this application embodiment).

[0088] Step 8: When the proxy value VP is greater than the collaborative interaction cost CE, use the same source device ST with the highest proxy value in the same source device list SR as the proxy same source device PT (the proxy same source device PT is the target proxy device in this application embodiment) to proxy the AI ​​assistant requests of the children's smartwatch CW and the same source devices in the same source device list SR with a proxy value greater than VP, excluding the same source device ST.

[0089] Step 9: The AI ​​assistant of the child smartwatch CW and the source device ST in the source device list SR requests the result MR through the proxy source device PT, performs trend consistency difference TRL transformation, and obtains the final result FR (the result MR is the execution result of the target task in this application embodiment, and the final result FR is the target execution result in this application embodiment).

[0090] Step 10: When the agent value VP is less than or equal to the collaborative interaction cost CE (the collaborative interaction cost CE is the cost data in the embodiment of this application), it indicates that the agent value cost is higher than the direct interaction. Therefore, the children's smartwatch CW interacts directly with the cloud disk server, obtains the result MR returned by the cloud disk server as the final result FR, and saves the result directly to the family data of the corresponding cloud disk account.

[0091] Compared with existing technologies, this application embodiment obtains target sensor data by responding to the target task execution request received from the smartwatch, providing necessary data support for the server to execute the target task; it generates a smart assistant execution request based on the target sensor data and the target task, thereby accurately initiating the request to the server to execute the target task; and it receives the target task execution result sent by the server and sends it to the smartwatch, completing the result feedback of the proxy execution process, reducing the resource consumption caused by the smartwatch directly interacting with the server, and alleviating the battery life pressure on the smartwatch.

[0092] Furthermore, as Figure 1 and Figure 2 To provide a specific implementation of the method shown, this embodiment offers a task processing device, such as... Figure 5 As shown, the device includes: a determination module 41, a verification module 42, and a transmission module 43.

[0093] The determination module 41 is configured to determine the candidate agent device corresponding to the smart watch in response to the invocation command of the smart assistant in the smart watch. The invocation command is used to invoke the smart assistant to perform the target task. Verification module 42 is configured to perform consistency verification on the sensor feature data of the smartwatch and the candidate agent device; The determination module 41 is also configured to determine the target agent device from the candidate agent devices based on the cost data and value data of data interaction between the smartwatch and the candidate agent devices, provided that the consistency check passes. The sending module 43 is configured to send an execution request for the target task to the target agent device, so as to request the smart assistant of the target agent device to execute the target task and send the execution result of the target task to the smartwatch.

[0094] In some examples of this embodiment, the determining module 41 is specifically configured to determine the cost data and value data of data interaction between the smartwatch and the candidate agent device; if the cost data is less than the value data, the device with the largest value data among the candidate agent devices is determined as the target agent device.

[0095] In some examples of this embodiment, the determining module 41 is further configured to determine the load data and power data corresponding to the smartwatch and the candidate agent device respectively, generate cost data based on the load data and power data, determine the multi-dimensional transmission data of the data interaction between the smartwatch and the candidate agent device, and generate value data based on the multi-dimensional transmission data.

[0096] In some examples of this embodiment, the determining module 41 is further configured to send an execution request for the target task to the server when the cost data is greater than the value data, so as to request the server to execute the target task and send the execution result of the target task to the smartwatch; in response to receiving the execution result, the execution result is displayed on the smartwatch.

[0097] In some examples of this embodiment, the verification module 42 is specifically configured to: determine the first target sensor data corresponding to the smartwatch; receive the second target sensor data sent by the candidate proxy device; and perform consistency verification on the sensor feature data of the smartwatch and the candidate proxy device based on the first target sensor data and the second target sensor data.

[0098] In some examples of this embodiment, the verification module 42 is further configured to determine the multi-dimensional sensor data corresponding to the smartwatch; and to extract features from the multi-dimensional sensor data to obtain the first target sensor data corresponding to the smartwatch.

[0099] In some examples of this embodiment, the verification module 42 is further configured to generate a monitoring notification for monitoring data from the candidate agent device; broadcast the monitoring notification to the candidate agent device, which responds to the monitoring notification, determines the second target sensor data, and sends the second target sensor data to the smartwatch.

[0100] In some examples of this embodiment, the verification module 42 is further configured to perform trend consistency analysis based on the first target sensor data and the second target sensor data to obtain trend consistency data; if the trend consistency data is greater than the trend consistency threshold, it is determined that the smartwatch and the candidate agent device have passed the consistency verification.

[0101] In some examples of this embodiment, the verification module 42 is further configured to receive the execution result of the target task sent by the target agent device; adapt and convert the execution result according to the trend consistency data to obtain the target execution result corresponding to the target task; and display the target execution result on the smartwatch.

[0102] It should be noted that other corresponding descriptions of the functional units involved in the task processing device provided in this embodiment can be found in [reference]. Figure 1 and Figure 2 The corresponding description in [the document] will not be repeated here.

[0103] Furthermore, as Figure 3 To provide a specific implementation of the method shown, this embodiment offers a task processing device, such as... Figure 6 As shown, the device includes: an acquisition module 51, a generation module 52, and a sending module 53.

[0104] The acquisition module 51 is configured to acquire target sensor data corresponding to the target task in response to receiving an execution request for the target task sent by the smartwatch; The generation module 52 is configured to generate a smart assistant execution request corresponding to the target agent device based on the target sensor data and the target task. The smart assistant execution request is used to request the server to execute the target task. The sending module 53 is configured to receive the execution result of the target task sent by the server and send the execution result to the smartwatch.

[0105] It should be noted that other corresponding descriptions of the functional units involved in the task processing device provided in this embodiment can be found in [reference]. Figure 3 The corresponding description in [the document] will not be repeated here.

[0106] Based on the above, Figure 1 , Figure 2 and Figure 3 Accordingly, this embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the above-described method. Figure 1 , Figure 2 and Figure 3 The method shown.

[0107] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause a computer device (such as personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.

[0108] like Figure 7 The diagram shown is a hardware structure schematic of an electronic device according to the present invention, comprising: At least one processor 601; and, A memory 602 is communicatively connected to at least one processor 601; wherein, The memory 602 stores instructions that can be executed by at least one processor, such that the instructions are executed by at least one processor to enable the at least one processor to perform the task processing method as described above.

[0109] Figure 7 Take the 601 processor as an example.

[0110] The electronic device may also include an input device 604 and an output device 604.

[0111] The processor 601, memory 602, input device 603, and output device 604 can be connected via a bus or other means. Figure 7 Taking the example of a connection between China and Israel via a bus.

[0112] The memory 602, as a non-volatile computer-readable storage medium, can be used to store non-volatile software programs, non-volatile computer-executable programs, and modules, such as the program instructions / modules corresponding to the task processing method in the embodiments of this application, for example, Figure 1 , Figure 2 and Figure 3 The method flow is shown. The processor 601 executes various functional applications and data processing by running non-volatile software programs, instructions, and modules stored in the memory 602, thereby implementing the task processing method in the above embodiments.

[0113] Memory 602 may include a program storage area and a data storage area, wherein the program storage area may store the operating system and applications required for at least one function; the data storage area may store data created according to the use of the task processing method, etc. Furthermore, memory 602 may include high-speed random access memory and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other non-volatile solid-state storage device. In some embodiments, memory 602 may optionally include memory remotely located relative to processor 601, and these remote memories may be connected to the apparatus performing the task processing method via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0114] Input device 603 can receive user clicks and generate signal inputs related to user settings and function control for task processing methods. Output device 604 may include display devices such as a display screen.

[0115] One or more modules are stored in memory 602, and when run by one or more processors 601, they execute the task processing method in any of the above method embodiments.

[0116] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0117] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0118] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0119] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platforms, or it can be implemented by hardware. By applying the solution of this embodiment, compared with the existing technology, this application embodiment determines candidate proxy devices by responding to the call instructions of the smart assistant, providing potential device selection for the proxy execution of the target task; by performing consistency verification on the sensor feature data of the smartwatch and the candidate proxy devices, the adaptability of the proxy device to execute the target task is ensured; by determining the target proxy device from the candidate proxy devices based on cost data and value data when the consistency verification passes, the rationality of the proxy execution is ensured; by sending the execution request of the target task to the target proxy device, requesting the smart assistant to execute the task and return the result, the resource consumption generated by the smartwatch's own data collection and cloud interaction can be reduced, alleviating the battery life pressure of the smartwatch; by determining the load data and power data of the smartwatch and the candidate proxy device to generate cost data, and by determining the multi-dimensional transmission data to generate value data, the accurate quantification of cost and value is achieved; by sending the execution request to the server when the cost data is greater than the value data, receiving the result and displaying it, the task can still be executed normally when it is not suitable for proxy; by determining the smartwatch and the candidate proxy device to generate the target proxy device, the rationality of the proxy execution is ensured; by determining the smart assistant to generate the target proxy device, requesting the smart assistant to execute the task and return the result ... sending the execution request to the server when the cost data is greater than the value data, receiving the result and displaying it, the rationality of the proxy execution is ensured; by determining the smart assistant to generate the target proxy device, requesting the smart assistant to execute the task and return the result The system receives primary target sensor data from the smartwatch and secondary target sensor data from candidate proxy devices, performing consistency verification between the two to support proxy device selection. The primary target sensor data is obtained by identifying and extracting features from the smartwatch's multi-dimensional sensor data, simplifying data source processing for consistency verification. A monitoring notification is generated and broadcast to candidate proxy devices, prompting them to identify and send the secondary target sensor data, ensuring data acquisition for consistency verification. Trend consistency data is obtained by analyzing the trend consistency between the primary and secondary target sensor data; if the trend consistency exceeds a threshold, the verification is passed, clarifying the criteria for consistency verification. The execution results from the target proxy devices are received, adapted, and displayed based on the trend consistency data, ensuring the execution results meet the smartwatch's usage requirements. Finally, the execution results of the target task sent by the server are received and sent back to the smartwatch, completing the result feedback of the proxy execution process, reducing resource consumption caused by direct interaction between the smartwatch and the server, and alleviating battery life pressure on the smartwatch.

[0120] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

[0121] The above are merely specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to these embodiments, but is to be accorded the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A task processing method, characterized in that, Applied to smartwatches, including: In response to a call command to the smart assistant in the smartwatch, a candidate agent device corresponding to the smartwatch is determined, wherein the call command is used to call the smart assistant to perform the target task; Perform consistency verification on the sensor feature data of the smartwatch and the candidate agent device; If the consistency check passes, the target agent device is determined from the candidate agent devices based on the cost data and value data of the data interaction between the smartwatch and the candidate agent devices. Send an execution request for the target task to the target agent device to request the smart assistant of the target agent device to execute the target task and send the execution result of the target task to the smartwatch.

2. The method according to claim 1, characterized in that, The step of determining the target agent device from the candidate agent devices based on the cost and value data of data interaction between the smartwatch and the candidate agent devices includes: Determine the cost and value data of data interaction between the smartwatch and the candidate agent device; If the cost data is less than the value data, the device with the highest value data among the candidate agent devices is determined as the target agent device.

3. The method according to claim 2, characterized in that, The determination of the cost and value data for data interaction between the smartwatch and the candidate agent device includes: Determine the load data and power data corresponding to the smartwatch and the candidate agent device respectively, and generate the cost data based on the load data and power data; The multi-dimensional transmission data of the smartwatch and the candidate agent device are determined, and the value data is generated based on the multi-dimensional transmission data.

4. The method according to claim 2, characterized in that, After determining the cost and value data of data interaction between the smartwatch and the candidate agent device, the method further includes: If the cost data is greater than the value data, an execution request for the target task is sent to the server to request the server to execute the target task and send the execution result of the target task to the smartwatch. In response to receiving the execution result, the execution result is displayed on the smartwatch.

5. The method according to claim 1, characterized in that, The consistency verification of sensor feature data between the smartwatch and the candidate agent device includes: Determine the first target sensor data corresponding to the smartwatch; Receive the second target sensor data sent by the candidate agent device; Based on the first target sensor data and the second target sensor data, a consistency verification is performed on the sensor feature data of the smartwatch and the candidate agent device.

6. The method according to claim 5, characterized in that, The step of determining the first target sensor data corresponding to the smartwatch includes: Determine the multi-dimensional sensor data corresponding to the smartwatch; Feature extraction is performed on the multi-dimensional sensor data to obtain the first target sensor data corresponding to the smartwatch.

7. The method according to claim 5, characterized in that, Before receiving the second target sensor data sent by the candidate proxy device, the method further includes: Generate a monitoring notification to listen to data from the candidate agent device; The monitoring notification is broadcast to the candidate agent device, which responds to the monitoring notification, determines the second target sensor data, and sends the second target sensor data to the smartwatch.

8. The method according to claim 5, characterized in that, The step of verifying the consistency of sensor feature data between the smartwatch and the candidate proxy device based on the first target sensor data and the second target sensor data includes: Based on the trend consistency analysis of the first target sensor data and the second target sensor data, trend consistency data is obtained. If the trend consistency data is greater than the trend consistency threshold, it is determined that the smartwatch and the candidate agent device have passed the consistency check.

9. The method according to claim 8, characterized in that, After sending the execution request for the target task to the target agent device, the method further includes: Receive the execution result of the target task sent by the target agent device; The execution result is adapted and transformed based on the trend consistency data to obtain the target execution result corresponding to the target task; The results of the target execution will be displayed on the smartwatch.

10. A task processing method, characterized in that, Applied to target agent devices, including: In response to receiving an execution request for a target task sent by a smartwatch, the system acquires the target sensor data corresponding to the target task. Based on the target sensor data and the target task, a smart assistant execution request corresponding to the target agent device is generated. The smart assistant execution request is used to request the server to execute the target task. The system receives the execution result of the target task sent by the server and sends the execution result to the smartwatch.

11. A task processing system, characterized in that, The device includes a smartwatch and a target agent device, wherein the target agent device is configured to implement the task processing method of any one of claims 1 to 9, and the target agent device is configured to implement the task processing method of claim 10.

12. A task processing device, characterized in that, include: The determination module is configured to determine a candidate agent device corresponding to the smartwatch in response to a call command to the smart assistant in the smartwatch, wherein the call command is used to call the smart assistant to perform the target task; The verification module is configured to perform consistency verification on the sensor feature data of the smartwatch and the candidate agent device. The determination module is also configured to, if the consistency check passes, determine the target agent device from the candidate agent devices based on the cost data and value data of the data interaction between the smartwatch and the candidate agent devices; The sending module is configured to send an execution request for the target task to the target agent device, so as to request the smart assistant of the target agent device to execute the target task and send the execution result of the target task to the smartwatch.

13. A task processing device, characterized in that, include: The acquisition module is configured to acquire target sensor data corresponding to the target task in response to receiving an execution request for the target task sent by the smartwatch; The generation module is configured to generate a smart assistant execution request corresponding to the target agent device based on the target sensor data and the target task. The smart assistant execution request is used to request the server to execute the target task. The sending module is configured to receive the execution result of the target task sent by the server and send the execution result to the smartwatch.

14. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 10.

15. An electronic device comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 10.

Images