Task processing method and device, electronic equipment and intelligent robot

Abstract

The embodiment of the invention discloses a task processing method and device, electronic equipment and an intelligent robot, and the method is applied to the intelligent robot, and comprises the steps: obtaining task information of a to-be-processed task, inputting the task information into a target task processing model obtained through pre-training, and outputting a task emergency identifier of the to-be-processed task; in response to an event including the same task emergency identifier in the to-be-executed tasks, task information of the to-be-executed tasks with the same task emergency identifier is reprocessed, so that task execution attributes of the to-be-executed tasks are updated based on a reprocessing result; wherein the to-be-executed task is a task which has confirmed a task emergency identifier and is not executed yet; and executing the to-be-executed task based on the updated task execution attribute. According to the technical scheme, the effect of accurately determining the task emergency level of the to-be-executed task according to the neural network model based on the robot is achieved, and the task execution efficiency is effectively improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the fields of artificial intelligence and robotics, and in particular to a task processing method, device, electronic device and intelligent robot. Background Art

[0002] With the development of the robotics industry and artificial intelligence technology, intelligent robots are gradually entering users' life scenarios. For example, intelligent robots are applied in home service scenarios to enable them to perform home-related tasks.

[0003] In the related art, before applying the robot to the user's life scenario, it is necessary to configure relevant execution logic for the robot so that the robot can perform corresponding tasks based on the configured execution logic when applied.

[0004] However, the logic configured in the robot is pre-configured, and may have a low degree of compatibility with the user's actual application needs. As a result, there may be a problem that the robot cannot effectively provide corresponding services to the user. Summary of the invention

[0005] The embodiments of the present application provide a task processing method, device, electronic device and intelligent robot to accurately determine the task urgency level of the task to be executed and effectively improve the task execution efficiency.

[0006] In a first aspect, an embodiment of the present application provides a task processing method, which is applied to an intelligent robot and includes:

[0007] Acquire task information of the task to be processed, wherein the task information includes at least one of audio information, video information and text information corresponding to the task to be processed;

[0008] Inputting the task information into a pre-trained target task processing model, and outputting a task urgency mark of the task to be processed;

[0009] In response to an event that the tasks to be executed include the same task urgency identifier, reprocessing the task information of the tasks to be executed with the same task urgency identifier, so as to update the task execution attribute of the tasks to be executed based on the reprocessing result; wherein the tasks to be executed are tasks with confirmed task urgency identifiers and have not yet been executed;

[0010] The to-be-executed task is executed based on the updated task execution attribute.

[0011] Furthermore, the obtaining of task information of the task to be processed includes at least one of the following methods:

[0012] Collecting video information including a target object and an environment to which the target object belongs based on a video acquisition device in the intelligent robot;

[0013] Collecting audio information emitted by the target object based on the audio collection device in the intelligent robot;

[0014] In response to the editing operation of the target object on the intelligent robot, the text information edited by the target object is obtained.

[0015] Furthermore, the method further comprises:

[0016] After acquiring the task information of the task to be processed, re-arranging the task information based on at least one preset target field to determine the target content under the at least one target field;

[0017] The task information of the task to be processed is updated based on the target content.

[0018] Furthermore, the at least one target field includes at least one of an event type field, an event occurrence time field, an event duration field, a user feedback field, an environmental parameter field of the environment to which the intelligent robot belongs, a user motion parameter field, a user type field, a user tone feature field, a user facial feature field, a festival field, and a date field.

[0019] Furthermore, the reprocessing of the task information of the pending tasks with the same task urgency identifier includes:

[0020] For pending tasks with the same task urgency mark, obtaining the task information of the pending tasks, and determining whether there are mergeable tasks and the task priority of the pending tasks according to the data content of the user type field, the user voice tone feature field, the user facial feature field and the event type field in the task information; wherein the mergeable tasks are tasks in the pending tasks;

[0021] In response to an event that there are mergeable tasks, merge the mergeable tasks to update the to-be-executed tasks;

[0022] Determine the task priority of the updated pending tasks;

[0023] Correspondingly, updating the task execution attribute of the to-be-executed task based on the reprocessing result includes:

[0024] Based on the updated task priority of the task to be executed, updating the task attribute of the task to be executed;

[0025] The task attributes include the execution order of the tasks to be executed.

[0026] Furthermore, the method further comprises:

[0027] Get the pending task information of historical tasks;

[0028] Marking the task information to be processed based on the constraint conditions corresponding to the preset emergency flags of each task to determine the historical emergency flags of the historical tasks;

[0029] Based on the to-be-processed task information of the historical tasks and the corresponding historical emergency identifiers, sample data participating in training the target task processing model is determined.

[0030] Further, the marking of the task information to be processed based on the constraint conditions corresponding to the preset task urgency flags to determine the historical urgency flags of the historical tasks includes:

[0031] For each historical task, determine the constraint conditions satisfied by the task information to be processed of the historical task according to the traversal order corresponding to all pre-set task emergency identifiers;

[0032] The task emergency flag corresponding to when the constraint condition is met is used as the historical emergency flag of the historical task.

[0033] Furthermore, the historical emergency identification includes a first emergency identification, a second emergency identification, a third emergency identification, a fourth emergency identification and a fifth emergency identification;

[0034] The first constraint condition corresponding to the first emergency indicator includes that the field content of the environmental parameter field is the first content or the field content of the event type field is at least one of the preset types; the second constraint condition corresponding to the second emergency indicator includes the condition of a preset reminder information; the third constraint condition corresponding to the third emergency indicator is that the pre-entered facial features meet the condition of a preset event or a preset first task; the fourth constraint condition corresponding to the fourth emergency indicator is a condition including a preset second task; the fifth constraint condition corresponding to the fifth emergency indicator is a condition that at least one target field does not have preset content.

[0035] Furthermore, the method further comprises:

[0036] Training the sample data based on at least one algorithm deployed in the model to be trained to obtain a loss value corresponding to the at least one algorithm;

[0037] Based on the loss value corresponding to the at least one algorithm, the model parameters in the model to be trained are modified, and the convergence of the loss function in the model to be trained is taken as the training goal to obtain the model to be verified;

[0038] The model to be verified corresponding to the algorithm is verified based on the test sample, and the model to be verified whose verification result meets the preset conditions is used as the target task processing model.

[0039] Furthermore, the method further comprises:

[0040] The target task processing model is deployed in the intelligent robot, and the model parameters of the target task processing model are updated based on the task information, execution result and response result of the task to be processed.

[0041] In a second aspect, an embodiment of the present application further provides a task processing device, which is configured in an intelligent robot and includes:

[0042] A task information acquisition module, used to acquire task information of a task to be processed, wherein the task information includes at least one of audio information, video information and text information corresponding to the task to be processed;

[0043] A task identification determination module, used to input the task information into a pre-trained target task processing model, and output a task emergency identification of the task to be processed;

[0044] A task attribute updating module, for responding to an event that the to-be-executed tasks include the same task urgency identifier, reprocessing the task information of the to-be-executed tasks with the same task urgency identifier, so as to update the task execution attribute of the to-be-executed tasks based on the reprocessing result; wherein the to-be-executed tasks are tasks with confirmed task urgency identifiers and have not yet been executed;

[0045] The task execution module is used to execute the to-be-executed task based on the updated task execution attribute.

[0046] In a third aspect, an embodiment of the present application provides an electronic device, the electronic device comprising:

[0047] one or more processors;

[0048] A memory for storing one or more programs;

[0049] When the one or more programs are executed by the one or more processors, the one or more processors implement the task processing method provided in any embodiment of the present application.

[0050] In a fourth aspect, an embodiment of the present application provides an intelligent robot, the intelligent robot comprising:

[0051] A display input device, a video acquisition device, an audio acquisition device, and an electronic device provided in an embodiment of the present application; the electronic device is communicatively connected with the display input device, the video acquisition device, and the audio acquisition device, respectively.

[0052] In a fifth aspect, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored, and when the program is executed by a processor, a task processing method provided in any embodiment of the present application is implemented.

[0053] In a sixth aspect, an embodiment of the present application further provides a computer program product, comprising a computer program, characterized in that when the computer program is executed by a processor, the computer program implements the task processing method as described in any embodiment of the present application.

[0054] The technical solution provided by the embodiment of the present application obtains the task information of the task to be processed, and further, inputs the task information into the target task processing model obtained by pre-training, and outputs the task urgency identifier of the task to be processed; further, in response to the event that the task to be executed includes the same task urgency identifier, the task information of the task to be executed with the same task urgency identifier is reprocessed to update the task execution attribute of the task to be executed based on the reprocessing result; further, the task to be executed is executed based on the updated task execution attribute, which solves the problem that the adaptability between the robot execution logic and the actual application needs of the user in the related technology is low, and thus, the robot cannot effectively provide corresponding services to the user, and achieves the effect of accurately determining the task urgency level of the task to be executed based on the robot according to the neural network model, improves the real-time and accuracy of task classification, and achieves the effect of dynamically adapting to environmental changes or special scene requirements during task classification, and by updating the task execution attributes of the task to be executed based on the tasks with the same task urgency level, the task execution efficiency is effectively improved, the task execution decision output by the robot is made more intelligent, the service efficiency of the robot is improved, and the intelligence level of the robot is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the exemplary embodiments of the present application, the following briefly introduces the drawings required to describe the embodiments. Obviously, the drawings introduced are only drawings of a part of the embodiments to be described in the present application, rather than all the drawings. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0056] Figure 1 A flowchart of a task processing method provided in an embodiment of the present application;

[0057] Figure 2A flowchart of a task processing method provided in an embodiment of the present application;

[0058] Figure 3 A flowchart of a task processing method provided in an embodiment of the present application;

[0059] Figure 4 A schematic diagram of the architecture of a task processing solution provided in an embodiment of the present application;

[0060] Figure 5 A schematic diagram of a task processing flow provided in an embodiment of the present application;

[0061] Figure 6 A schematic diagram of the structure of a task processing device provided in an embodiment of the present application;

[0062] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0063] Figure 8 A schematic diagram of the structure of an intelligent robot provided in an embodiment of the present application. DETAILED DESCRIPTION

[0064] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only the parts related to the present application, rather than all structures, are shown in the accompanying drawings.

[0065] The present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for ease of description, only the parts related to the present application, rather than all structures, are shown in the accompanying drawings.

[0066] Before introducing the technical solution provided by the embodiment of the present application, an exemplary description of the application scenario can be given first. For example, in a scenario where an intelligent robot (such as a household service robot) is used to perform tasks. Among them, an intelligent robot is a robot that can simulate object behavior and has learning and adaptability through built-in artificial intelligence technology. The intelligent robot can make corresponding reactions and decisions according to changes in the environment and external inputs, thereby performing various complex tasks. The intelligent robot can be controlled by voice, touch screen or other means. When the intelligent robot receives a task to be performed, it can perform task classification according to the urgency of the task to be performed to determine the task priority of each task to be performed. Furthermore, the corresponding task to be performed can be performed according to the task priority of each task to be performed.

[0067] Figure 1 A flowchart of a task processing method provided in an embodiment of the present application. This embodiment can be applied to any scenario in which task classification is performed based on an intelligent robot and corresponding tasks are executed based on the task classification results. The task processing method provided in this embodiment is applied to an intelligent robot and can be executed by the intelligent robot or by the cooperation of the intelligent robot and the server. The task processing device integrated in the intelligent robot and / or the server can be implemented by software and / or hardware and integrated in an electronic device, which can be a mobile terminal or a PC, etc.

[0068] like Figure 1 As shown, the method specifically comprises the following steps:

[0069] S110: Obtain task information of tasks to be processed.

[0070] In this embodiment, the intelligent robot can be a machine device that can perform task functions, and optionally, it can be a household service robot, that is, a robot that can perform any task in the home, such as fire warning, intrusion detection, daily cleaning, entertainment and care, etc.

[0071] Among them, the pending task may be a task that has not been processed. Not processed may include an undetermined task emergency mark and / or unexecuted. Optionally, the pending task may be a fire warning task, an intrusion detection task, a daily cleaning task, or an entertainment and accompanying task. Task information may be information associated with the pending task for describing the pending task. Task information may include any type of information associated with the pending task, and optionally, includes at least one of audio information, video information, and text information corresponding to the pending task. Audio information may include audio emitted by a target object associated with the pending task or ambient audio of an environment to which the target object belongs. For example, audio information includes conversation information between a user and an intelligent robot, a doorbell or a knock on the door, etc. Video information may include ambient video of an environment to which the target object associated with the pending task belongs or a video including the target object, etc. In this embodiment, the video information may include only video screen content; or, it may include audio information and video screen content associated with the audio information. Text information may be information describing the task content of the pending task in text form. For example, text information may include important schedule reminders, holiday reminders, or important event reminders, etc.

[0072] In this embodiment, there are multiple ways to obtain task information, for example, automatically detecting through a sensor device deployed in the intelligent robot; and / or determining based on an information editing operation (for example, an audio upload operation, a video upload operation, or a text upload operation, etc.) input by a user to the intelligent robot. The following are the various ways to obtain task information:

[0073] The first method is to use a video acquisition device in an intelligent robot to acquire video information including a target object and the environment to which the target object belongs.

[0074] Among them, the video acquisition device can be a device deployed in the intelligent robot, which can be used to collect video data in a preset detection area corresponding to the intelligent robot. The video acquisition device can be any device that can realize the video acquisition function, and optionally, it is a camera device (such as a camera, etc.). The target object can be an object associated with the task to be processed. The target object can be an object pre-input into the intelligent robot. Optionally, the target object can be a person, a pet, or a household item, etc. It should be noted that the target object can be an object associated with a task trigger instruction input by the user; or, it can also be an object with abnormal conditions during the daily detection process of the robot, etc.

[0075] In this embodiment, when the intelligent robot moves along a preset path in a preset area, the preset detection area can be detected by a detection device deployed in the intelligent robot. Further, when a target object is detected in the preset detection area, video information including the target object and the environment to which the target object belongs can be collected by a video acquisition device, and the collected video information is used as task information of the task to be processed.

[0076] As an optional implementation of this embodiment, at least one object can be predetermined and the determined object can be input into the intelligent robot. Furthermore, in the process of the intelligent robot moving along a preset path in a preset area, the preset detection area can be detected by a detection device deployed in the intelligent robot. Furthermore, in the case where an abnormality is detected in the preset detection area, the object with the abnormality can be determined and the object can be used as the target object. Furthermore, the target object and the environment to which the target object belongs can be video captured by a video capture device deployed in the intelligent robot to obtain video information including the target object and the environment to which the target object belongs.

[0077] The second method is to collect audio information emitted by the target object based on the audio collection device in the intelligent robot.

[0078] The audio collection device may be a device deployed in an intelligent robot, and the device may be used to collect audio information. The audio collection device may be any device capable of realizing the audio collection function, and may optionally be a microphone or the like.

[0079] In this embodiment, when the intelligent robot receives a task execution trigger instruction, the audio information emitted by the target object associated with the task to be processed can be collected by an audio collection device deployed in the intelligent robot to obtain audio information corresponding to the target object, and the collected audio information can be used as the task information of the task to be processed.

[0080] The third method: in response to the editing operation of the target object on the intelligent robot, the text information edited by the target object is obtained.

[0081] Among them, the editing operation may include any form of text editing operation for the intelligent robot input, optionally including text input operation, text upload operation, text modification operation and text deletion operation, etc.

[0082] In this embodiment, a text editing control can be set in advance on the intelligent robot. Further, when a trigger operation of the target object on the text editing control is detected, a text editing interface can be displayed on the display screen of the intelligent robot based on the trigger operation of the control. Further, the target object can input an editing operation in the text editing interface, and after triggering the confirmation control, the text information edited by the target object in the text editing interface can be obtained, and the obtained text information can be used as the task information of the task to be processed.

[0083] It should be noted that the benefits of obtaining task information of pending tasks based on the above three methods are: enriching the diversity of methods for obtaining task information, enriching the information sources of task information, and thus laying a data foundation for subsequent model-based information processing.

[0084] In this embodiment, the acquired task information of the task to be processed may include at least one form of information, and the acquired task information may include information with a low correlation with the task to be processed. Furthermore, in view of the above situation, in order to reduce the amount of information to be processed and improve the efficiency of information processing and task execution, after acquiring the task information of the task to be processed, the method further includes: re-arranging the task information based on at least one preset target field, determining the target content under at least one target field; and updating the task information of the task to be processed based on the target content.

[0085] Among them, the target field can be a key field associated with the task. At least one target field can include any key field associated with the task, optionally including at least one of an event type field, an event occurrence time field, an event duration field, a user feedback field, an environmental parameter field of the environment to which the intelligent robot belongs, a user motion parameter field, a user type field, a user tone feature field, a user facial feature field, a festival field, and a date field. The event type field can be used to characterize the type of event that triggers the generation of the task to be processed. The target content under the event type field can be a specific event type that characterizes the event that triggers the generation of the task to be processed. Optionally, the target content under the event type field can be a fire warning, a medical warning, an intrusion warning, a schedule reminder, daily cleaning, or entertainment care, etc. The event occurrence time field can be used to characterize the occurrence time of the event that triggers the generation of the task to be processed. The event duration field can be used to characterize the duration of the event that triggers the generation of the task to be processed. The user feedback field can be used to characterize the user's feedback on the execution of the intelligent robot task. The target content under the user feedback field can be the user's score for the execution of the task. The environmental parameter field of the environment to which the intelligent robot belongs can be used to characterize the environmental conditions of the environment to which the intelligent robot belongs. Optionally, the environmental parameter field of the environment to which the intelligent robot belongs may include the ambient temperature, ambient humidity, ambient smog conditions, etc. The user motion parameter field may be used to characterize the user's motion behavior pattern. The user type field may be used to characterize the category to which the user belongs. The user pitch feature field may be used to characterize the pitch of the audio information emitted by the user. The user facial feature resource may be used to characterize the user's facial contour. The holiday field may be used to characterize whether the current date is a holiday and the specific holiday corresponding to the current date if the current date is a holiday. The date field may be used to characterize whether the current date is a date corresponding to an important schedule.

[0086] As an optional implementation of this embodiment, at least one target field may be pre-set and input into the intelligent robot. Further, after obtaining the task information of the task to be processed, the task information may be extracted based on the at least one pre-set target field to determine the target content under the at least one target field based on the task information. Further, the task information of the task to be processed may be updated based on the target content under the at least one target field, so that the task information of the task to be processed is represented by the target content under each target field.

[0087] S120: Input the task information into the pre-trained target task processing model, and output the task urgency mark of the task to be processed.

[0088] Among them, the target task processing model can be a neural network model that takes task information as an input object to determine the urgency of the task based on the task information. The target task processing model can be obtained by training a pre-built neural network model based on the task information and historical emergency identification of historical tasks. The target task processing model can be a neural network model that deploys any algorithm. Optionally, the algorithm deployed in the target task processing model can be a decision tree algorithm, a support vector machine algorithm, a random forest tree algorithm, a gradient boosting tree algorithm, a K-nearest neighbor algorithm, etc. The task emergency identification can be used to characterize the urgency of the corresponding task. The task emergency identification can be used to identify the task priority corresponding to the task to be processed. The task emergency identification can be any form of information that can characterize the priority and urgency of the task, and optionally, it is at least one of a number, a text, and an icon. In this embodiment, multiple task priorities can be determined based on the scope of influence, urgency, and / or potential risk of the task. Further, a corresponding task emergency identification can be set for each task priority to determine the execution priority of the corresponding task based on the task emergency identification.

[0089] In practical applications, when the intelligent robot has at least two tasks to be performed, the execution order of the tasks is usually determined according to the execution priority of the tasks. In general, the execution priority of the tasks is implemented using a preset logical structure according to the grading rules pre-deployed in the intelligent robot. This task grading method requires pre-setting fixed grading rules to be implemented. Once these rules are defined, it is difficult to adapt to changes in the environment or the needs of special scenarios. In addition, these rules are usually based on logical conditional judgments, which may have low flexibility and cannot effectively process fuzzy or continuous data.

[0090] In view of the above situation, in this embodiment, when determining the tasks to be processed, the task information of the tasks to be processed can be obtained. Further, the obtained task information can be input into the pre-trained target task processing model to process the information to be processed based on the target task processing model. And the task priority of the task to be processed is determined based on the information to be processed. Furthermore, based on the mapping relationship between the execution priority and the emergency identifier, the task emergency identifier of the task to be processed can be determined and output.

[0091] It should be noted that the benefit of updating the task information based on the target content under the field associated with the user and determining the task urgency mark of the pending task based on the updated task information is that different users have different needs and expectations for the urgency and handling methods of the same event. Furthermore, the user-related features are introduced when determining the task urgency mark, which achieves the effect of determining the task urgency of the pending task based on the user's preferences and behavior patterns, meets the personalized needs of the user, makes the behavior pattern of the intelligent robot more compatible with the behavior pattern of the relevant user, and improves the intelligence of the intelligent robot.

[0092] It should also be noted that the advantage of determining the task urgency mark of the task to be processed based on the target task processing model according to the task information is that it can dynamically adapt to environmental changes or the needs of special scenarios, improve the real-time and accuracy of task classification, and make the task execution decision output by the intelligent robot more intelligent, thereby improving the intelligence of the intelligent robot. In addition, the response level and strategy are automatically adjusted in combination with environmental perception and real-time user portrait data, realizing a more flexible and efficient event processing mechanism.

[0093] S130 , in response to an event that the tasks to be executed include the same task urgency identifier, reprocess the task information of the tasks to be executed with the same task urgency identifier, so as to update the task execution attributes of the tasks to be executed based on the reprocessing results.

[0094] Among them, the task to be executed is a task whose task emergency mark has been confirmed and has not been executed. The task to be executed can be any task in the task execution list of the intelligent robot. In this embodiment, after determining the task emergency mark of the task to be processed, the task emergency mark can be associated with the task to be processed and stored in the task execution list to wait for execution in the task execution list, and the task that contains the task emergency mark and has not been executed in the task execution list can be used as the task to be executed. It should be noted that after determining the task emergency mark of the task to be processed, the task to be processed is also updated to the task to be executed, that is, the task to be executed with the same task emergency mark includes the task to be processed with the determined task emergency mark in the task to be executed. The event that includes the same task emergency mark in the task to be executed can be understood as the event that determines the task to be executed from the task to be executed with the same task emergency mark as the task to be processed. It should be noted that the event that includes the same task emergency mark in the task to be executed can be automatically triggered when the task emergency mark of the task to be processed is obtained; or, it can also be triggered when the event execution operation input by the user for the intelligent robot is received, and this embodiment does not make specific restrictions on this. The task execution attribute can be understood as information characterizing the task execution status. Optionally, the task execution attribute includes the execution order of the tasks to be executed.

[0095] In this embodiment, when the pending tasks that have not been executed by the intelligent robot include at least two cases, after determining the task urgency identifier of the pending tasks, there may be a case where the pending tasks that have been determined and have not been executed have the same task urgency identifier as the pending tasks. The same task urgency identifier can indicate that the task priorities of these tasks are the same. Furthermore, in order to improve the efficiency of task execution so that the pending tasks with the same task priority can be merged for execution, after determining the task urgency identifier of the pending tasks, the pending tasks can be traversed according to the task urgency identifier to determine the pending tasks with the same task urgency identifier from the pending tasks. Furthermore, for the pending tasks with the same task urgency identifier, the task information of the pending tasks can be obtained. Thereafter, the task information of the pending tasks that have been obtained can be reprocessed. Furthermore, the task execution attributes of the pending tasks and the pending tasks can be updated according to the reprocessing results.

[0096] As an optional implementation of the present embodiment, after determining the task emergency identifier of the pending task, an event that includes the same task emergency identifier in the pending tasks can be triggered. Further, in response to the event, a pending task with the same task emergency identifier as the task is determined from the pending tasks. Afterwards, for the pending tasks with the same task emergency identifier, the task information of the pending tasks is obtained. The task information of the pending tasks is information updated based on the target content under at least one target field, and then, the field content comparison can be performed on the acquired task information of the pending tasks to determine whether there are mergeable tasks in the pending tasks based on the content under the field, and when mergeable tasks are determined, the mergeable tasks are merged. Then, the pending tasks are updated based on the task merging results, and the task execution attributes of the updated pending tasks are determined, and the task execution attributes of the pending tasks are updated based on the task execution attributes.

[0097] S140: Execute the task to be executed based on the updated task execution attribute.

[0098] In this embodiment, after the task execution attribute of the to-be-executed task is updated, the corresponding to-be-executed task may be executed based on the updated task execution attribute.

[0099] As an optional implementation of this embodiment, when the task execution attribute includes the task execution order, after the task execution order of the tasks to be executed is updated, the tasks to be executed may be executed sequentially based on the updated task execution order.

[0100] The technical solution provided by the embodiment of the present application obtains the task information of the task to be processed, and further, inputs the task information into the target task processing model obtained by pre-training, and outputs the task urgency identifier of the task to be processed; further, in response to the event that the task to be executed includes the same task urgency identifier, the task information of the task to be executed with the same task urgency identifier is reprocessed to update the task execution attribute of the task to be executed based on the reprocessing result; further, the task to be executed is executed based on the updated task execution attribute, which solves the problem that the adaptability between the robot execution logic and the actual application needs of the user in the related technology is low, and thus, the robot cannot effectively provide corresponding services to the user, and achieves the effect of accurately determining the task urgency level of the task to be executed based on the robot according to the neural network model, improves the real-time and accuracy of task classification, and achieves the effect of dynamically adapting to environmental changes or special scene requirements during task classification, and by updating the task execution attributes of the task to be executed based on the tasks with the same task urgency level, the task execution efficiency is effectively improved, the task execution decision output by the robot is made more intelligent, the service efficiency of the robot is improved, and the intelligence level of the robot is improved.

[0101] Figure 2 A flowchart of a task processing method provided for an embodiment of the present application, based on the above-mentioned embodiment, for the pending tasks with the same task emergency mark, it is determined whether there are mergeable tasks, and when it is determined that there are mergeable tasks, the mergeable tasks are merged and processed to update the pending tasks, and the task priority of the updated pending tasks is determined. Furthermore, the pending tasks can be executed based on the task priority of the updated pending tasks. The specific implementation method can be referred to the detailed description of this embodiment, wherein the technical terms that are the same as or corresponding to the above-mentioned embodiments are not repeated in this embodiment.

[0102] As shown in 2, the method includes:

[0103] S210: Obtain task information of tasks to be processed.

[0104] S220: Input the task information into the pre-trained target task processing model, and output the task urgency mark of the task to be processed.

[0105] S230. In response to an event that includes the same task urgency identifier in tasks to be executed, for tasks to be executed with the same task urgency identifier, obtain task information of the tasks to be executed, and determine whether there are tasks that can be merged and the task priority of the tasks to be executed based on the data content of the user type field, user tone feature field, user facial feature field and event type field in the task information.

[0106] Among them, the mergeable tasks can be tasks that can be merged. The mergeable tasks are tasks in the tasks to be executed. Task priority can be used to characterize the urgency or urgency level of the task execution order. Optionally, the task priority may include emergency level, high priority, medium priority, low priority, entertainment and social level, etc. Exemplarily, for intelligent robots, tasks with an emergency priority level may include tasks that require immediate response, such as fire warnings and medical warnings; tasks with a high priority level may include more urgent tasks such as intrusion warnings and important schedule reminders; tasks with a medium priority level may include tasks such as family members coming home and specific housework tasks; tasks with a low priority level may include tasks such as daily cleaning and daily maintenance; tasks with an entertainment and social level may include tasks such as companionship chatting and entertainment activities. Among them, tasks with an emergency priority level must be executed first; tasks with an entertainment and social level priority level can be executed last.

[0107] In this embodiment, after determining the task emergency identifier of the task to be processed, the task to be processed can be updated to the task to be executed. Further, in response to an event including the same task emergency identifier in the task to be executed, a task with the same task emergency identifier as the task to be processed can be determined from the tasks to be executed. Further, for the tasks to be executed with the same task emergency identifier, task information of the tasks to be executed can be obtained, and the acquired task information is represented based on the target content under at least one target field. Further, the data content under the user type field, the user tone feature field, the user facial feature field and the event type field can be obtained from the task information, and the obtained data content can be compared with the field content, and the to-be-executed tasks with the same task emergency identifier corresponding to the same user can be determined based on the data content under the user type field, the user tone feature field and the user facial feature field, and the to-be-executed tasks with the same task emergency identifier corresponding to the same event can be determined based on the data content under the event type field. Further, it can be determined whether there are mergeable tasks in the tasks to be executed with the same task emergency identifier based on the field content comparison result, and the task priority of the tasks to be executed can be determined.

[0108] S240 . In response to an event that there are mergeable tasks, merge the mergeable tasks to update the tasks to be executed.

[0109] Among them, the event of the existence of mergeable tasks can be used to indicate that there are mergeable tasks in the pending tasks with the same task emergency mark. In this embodiment, the event execution triggering method of the existence of mergeable tasks can include at least one of the following: automatically triggering execution when the existence of mergeable tasks is detected; inputting a trigger operation for a preset event execution control; receiving instructions (including audio information and / or action information) associated with the event execution trigger operation, etc.

[0110] In this embodiment, when it is detected that there are mergeable tasks among the pending tasks with the same task urgency mark, an event of executing the mergeable tasks is triggered. Further, in response to the event, the mergeable tasks are obtained and merged to serve as new pending tasks. Further, the pending tasks that have been determined to have task urgency marks and have not yet been executed can be updated according to the task merging result to reduce the number of pending tasks.

[0111] S250: Determine the task priority of the updated task to be executed, and update the task execution attribute of the task to be executed based on the updated task priority of the task to be executed.

[0112] The task execution attributes include the execution order of the tasks to be executed.

[0113] In this embodiment, after merging the mergable tasks with the same task urgency identifier and updating the tasks to be executed based on the task merging result, the number of tasks to be executed will decrease, and the execution order of the tasks to be executed may also change. Then, the task urgency identifier of the updated task to be executed can be obtained, and the task priority of the updated task to be executed can be determined based on the mapping relationship between the task urgency identifier and the task priority. Furthermore, the task execution attribute of the task to be executed is updated according to the updated task priority of the task to be executed.

[0114] S260: Execute the task to be executed based on the updated task execution attribute.

[0115] The advantage of the above method is that it can determine whether there are mergeable tasks among the tasks to be executed with the same task urgency level, and merge the mergeable tasks when it is determined that there are mergeable tasks to update the task execution properties of the tasks to be executed, thereby effectively improving the task execution efficiency, making the task execution decisions output by the intelligent robot more intelligent, and improving the intelligence level of the intelligent robot.

[0116] Figure 3A flowchart of a task processing method provided in an embodiment of the present application, based on the above-mentioned embodiment, before applying the target task processing model, the sample data involved in training the target task processing model can be determined first, so as to obtain the target task processing model based on the sample data training. The specific implementation method can refer to the detailed description of this embodiment, wherein the technical terms that are the same as or corresponding to the above-mentioned embodiment are not repeated in this embodiment.

[0117] As shown in 3, the method includes:

[0118] S310: Obtain pending task information of historical tasks.

[0119] Among them, the historical tasks may be tasks that the intelligent robot has performed within a historical time period. Optionally, the historical tasks may include fire warning, medical warning, intrusion detection, important schedule reminders, daily cleaning, entertainment and care, etc. The pending task information may be information associated with the historical tasks for describing the specific task content of the historical tasks. The pending task information may include data content under at least one preset target field. Optionally, at least one target field includes at least one of an event type field, an event occurrence time field, an event duration field, a user feedback field, an environmental parameter field of the environment to which the intelligent robot belongs, a user motion parameter field, a user type field, a user tone feature field, a user facial feature field, a holiday field, and a date field.

[0120] In this embodiment, before training the target task processing model, sample data for training the target task processing model can be constructed to train the model based on the sample data. For an intelligent robot, initial task information of a historical task can be obtained from at least one data source associated with the intelligent robot. Furthermore, the initial task information can be rearranged based on at least one preset target field to determine the data content under at least one target field. Furthermore, the initial task information can be updated based on the data content under at least one target field to obtain the pending task information of the historical task. Optionally, at least one data source can include historical event records, sensor detection sources, user historical behavior data, and external data sources, etc.

[0121] Exemplarily, for an intelligent robot, the event type, event occurrence time, event duration, user feedback, and event processing results of a historical task can be obtained from the intelligent robot's historical event records; the environmental parameters of the intelligent robot's environment at the historical moment corresponding to the historical task (for example, including ambient temperature, ambient humidity, ambient smoke concentration, etc.) and the user's motion parameters can be obtained from the sensor collection data set in the intelligent robot; the user's evaluation information on the task priority of the historical task can be obtained from the user's historical behavior data; the weather forecast and holiday information for the date of the historical task can be obtained from an external data source.

[0122] S320: Mark the task information to be processed based on the preset constraint conditions corresponding to the urgency flags of each task, so as to determine the historical urgency flags of the historical tasks.

[0123] Among them, the constraint condition may be a division standard for dividing the task urgency. According to the constraint condition corresponding to the task emergency identifier, the task urgency corresponding to the historical task may be determined. It should be noted that the task emergency identifier may be used to characterize the urgency of the task, and the constraint condition corresponding to the task emergency identifier may be associated with the task information feature corresponding to the corresponding task emergency identifier. Optionally, the task emergency identifier includes a first emergency identifier, a second emergency identifier, a third emergency identifier, a fourth emergency identifier, and a fifth emergency identifier. The task priority corresponding to the first emergency identifier is the highest; the task priority corresponding to the fifth emergency identifier is the lowest. The first constraint condition corresponding to the first emergency identifier includes that the field content of the environmental parameter field is the first content or the field content of the event type field is at least one of the preset types. The first content may be environmental information associated with the environment to which the intelligent robot belongs and reaches a preset urgency. Optionally, the first content may include at least one of the ambient temperature reaching a preset temperature threshold, the ambient humidity reaching a preset humidity threshold, and the ambient smoke concentration reaching a preset concentration threshold. The preset type may be any time type, and optionally, it may be a fire warning event or a medical warning event. The second constraint condition corresponding to the second emergency identifier includes a condition of a preset reminder information. The preset reminder information may include important event reminder information and / or important schedule reminder information pre-entered by the user. The third constraint condition corresponding to the third emergency identifier is that the pre-entered facial features meet the conditions of the preset event or the preset first task. The preset event may be any event associated with the user, and optionally, includes the user arriving at a preset location. The preset first task may be any task associated with the user, and optionally, includes a specific housework task. The fourth constraint condition corresponding to the fourth emergency identifier is a condition including a preset second task. The preset second task may include any task performed by the intelligent robot, and optionally, includes daily cleaning tasks, daily maintenance tasks, etc. The fifth constraint condition corresponding to the fifth emergency identifier is a condition that at least one target field does not have a preset content. The preset content may be a field content that is predetermined and input into the intelligent robot. The absence of preset content in at least one target field may indicate that the task information received by the intelligent robot is information randomly input by the user. Exemplarily, the absence of preset content in at least one target field may include chatting and / or entertainment activities between the user and the intelligent robot.

[0124] In this embodiment, at least one task emergency identifier can be pre-set, and the constraint conditions corresponding to each task emergency identifier can be determined. Further, after obtaining the pending task information of the historical tasks, the pending task information of the historical tasks can be processed according to the constraint conditions corresponding to each task emergency identifier for each historical task, so as to determine the constraint conditions satisfied by the pending task information. Furthermore, the task emergency identifier corresponding to the satisfied constraint condition can be used as the historical emergency identifier of the historical task, so as to complete the identification marking of the pending task information.

[0125] Optionally, the task information to be processed is marked based on the constraints corresponding to the pre-set emergency identifiers of each task to determine the historical emergency identifier of the historical task, including: for each historical task, determining the constraints satisfied by the task information to be processed of the historical task based on the traversal order corresponding to all pre-set task emergency identifiers; and using the task emergency identifier corresponding to when the constraints are satisfied as the historical emergency identifier of the historical task.

[0126] The traversal order may be determined in the order of the task priorities corresponding to the task emergency identifiers. Optionally, when the task emergency identifiers include the first emergency identifier, the second emergency identifier, the third emergency identifier, the fourth emergency identifier and the fifth emergency identifier, and the task priority corresponding to the first emergency identifier is the highest and the task priority corresponding to the fifth emergency identifier is the lowest, the traversal order corresponding to all task emergency identifiers may be the first emergency identifier, the second emergency identifier, the third emergency identifier, the fourth emergency identifier and the fifth emergency identifier.

[0127] As an optional implementation of the present embodiment, for each historical task, the constraint conditions corresponding to each task emergency identifier can be matched with the pending task information of the historical task according to the traversal order corresponding to the preset task emergency identifier. Further, when it is determined that the pending task information matches the constraint conditions corresponding to any task emergency identifier, the constraint conditions can be determined as the constraint conditions satisfied by the pending task information. Further, the task emergency identifier corresponding to the constraint conditions can be determined, and the task emergency identifier can be used as the historical emergency identifier of the historical task. The advantage of such a setting is that the effect of the historical emergency identifier of the historical task can be accurately determined based on the constraint conditions and the traversal order corresponding to the task emergency identifier, so that the diversity and complexity of the environment are fully considered when training the model, thereby improving the flexibility of the emergency identifier determination process.

[0128] Exemplarily, for each historical task, the constraint conditions corresponding to the first emergency identifier, the second emergency identifier, the third emergency identifier, the fourth emergency identifier, and the fifth emergency identifier can be matched with the pending task information of the historical task according to the traversal order corresponding to the preset task emergency identifier. Further, when it is determined that the field content of the environmental parameter field in the pending task information is not the first content, the field content of the event type field is not the preset type, and the preset reminder information is included, it can be determined that the pending task information satisfies the second constraint condition corresponding to the second emergency identifier. Further, the second emergency identifier can be used as the historical task identifier of the historical task.

[0129] S330: Based on the pending task information of historical tasks and the corresponding historical emergency identification, determine the sample data involved in training the target task processing model.

[0130] In this embodiment, when the pending task information and the corresponding historical emergency identification of the historical tasks are obtained, for each historical task, a training sample corresponding to the historical task can be constructed based on the pending task information and the historical emergency identification of the historical task. Furthermore, the training sample corresponding to each historical task can be used as sample data for participating in the training of the target task processing model.

[0131] In this embodiment, after obtaining the sample data, the model to be trained can be trained based on the sample data, and the trained model can be used as the target task processing model.

[0132] On the basis of the above technical solutions, the training process of the target task processing model includes: training and processing sample data based on at least one algorithm deployed in the model to be trained to obtain the loss value corresponding to at least one algorithm; based on the loss value corresponding to at least one algorithm, correcting the model parameters in the model to be trained, taking the convergence of the loss function in the model to be trained as the training target, and obtaining the model to be verified; verifying the model to be verified corresponding to the algorithm based on the test sample, and taking the model to be verified whose verification result meets the preset conditions as the target task processing model. The advantage of this setting is that at least one algorithm is deployed in the model, so that the target task processing model finally obtained can accurately and flexibly cover various scenarios that can be encountered during task processing, and realize accurate classification and processing of tasks.

[0133] Among them, the model to be trained can be a neural network model whose model parameters are default values ​​or initial values. The algorithm deployed in the model to be trained can be an algorithm that can perform task information processing and task emergency identification classification. Optionally, at least one algorithm includes at least one of a decision tree algorithm, a support vector machine algorithm, a random forest tree algorithm, a gradient boosting tree algorithm, a K-nearest neighbor algorithm, and a long short-term memory network algorithm. The loss value can be a numerical value that characterizes the degree of difference between the predicted output and the actual output. The model to be verified can be a model obtained after the loss function in the model to be trained converges after the model parameters of the model to be trained are corrected according to the loss value corresponding to the corresponding algorithm. The model to be verified can correspond to the algorithm. In other words, the number of models to be verified can be consistent with the number of algorithms deployed in the model to be trained. The test sample can be a part of the sample data, or it can be sample data re-determined based on the sample data determination process described in steps S310-S330. The preset conditions can include at least one of the accuracy reaching a preset accuracy threshold, the recall reaching a preset recall threshold, the mean square error reaching a first error threshold, and the root mean square error reaching a second error threshold.

[0134] In this embodiment, after obtaining sample data for training the target task processing model, the task information to be processed of the historical tasks in the sample data can be input into the model to be trained, so as to process the task information to be processed based on at least one algorithm deployed in the model to be trained, and output the model prediction identifiers corresponding to various algorithms. Further, for various algorithms, the model prediction identifier corresponding to the algorithm can be compared with the historical emergency identifier of the historical tasks in the sample data to determine the loss value, and the loss value is used as the loss value corresponding to the algorithm. Further, the model parameters in the model to be trained can be corrected based on the loss value corresponding to the algorithm, and then the training error of the loss function in the model to be trained, that is, the loss parameter, can be used as a condition for detecting whether the current loss function has reached convergence, for example, whether the training error is less than the preset error or whether the error change trend tends to be stable, or whether the current model iteration number is equal to the preset number, etc. If the detection reaches the convergence condition, for example, the training error of the loss function is less than the preset error or the error change tends to be stable, it indicates that the training of the model to be trained is completed, and at this time, the iterative training can be stopped. If it is detected that the convergence condition is not met at present, other sample data can be further obtained to train the model to be trained until the training error of the loss function is within the preset range. When the training error of the loss function reaches convergence, the trained model to be trained can be used as the model to be verified corresponding to the algorithm. Furthermore, in order to evaluate the model performance of the model to be verified corresponding to at least one algorithm, a test sample can be obtained, and for the model to be verified corresponding to at least one algorithm, the model to be verified corresponding to the algorithm can be tested and verified based on the test sample, and the verification result corresponding to the model to be verified can be obtained. Furthermore, it can be determined whether the verification result of at least one model to be verified meets the preset conditions. Then, the model to be verified whose verification result meets the preset conditions can be used as the target task processing model.

[0135] In this embodiment, in order to improve the model performance of the target task processing model during the model application process, and then improve the decision accuracy of the intelligent robot. In the process of applying the target task processing model based on the intelligent robot, the target task processing model can also be optimized. Optionally, the method also includes: deploying the target task processing model in the intelligent robot, and updating the model parameters of the target task processing model based on the task information, execution results and response results of the task to be processed.

[0136] The execution result may be the output information or status fed back after the pending task is completed. The execution result may be used to represent the execution details of the pending task. The response result may be the user's feedback on the execution of the pending task.

[0137] In this embodiment, after obtaining the target task processing model, the target task processing model can be deployed in the intelligent robot. Furthermore, in the case where the intelligent robot obtains the task information of the task to be processed, the task information can be processed based on the target task processing model, the execution result of the task to be processed can be determined, and the user's response result to the task execution status of the task to be processed can be received. Further, the model parameters of the target task processing model can be modified according to the task information, execution result and response result of the task to be processed. Thus, the model performance can be continuously optimized in the model application process of the target task processing model.

[0138] S340: Obtain task information of tasks to be processed.

[0139] S350: Input the task information into the pre-trained target task processing model, and output the task urgency mark of the task to be processed.

[0140] S360: In response to an event that the tasks to be executed include the same task urgency identifier, reprocess the task information of the tasks to be executed with the same task urgency identifier, so as to update the task execution attributes of the tasks to be executed based on the reprocessing results.

[0141] S370: Execute the task to be executed based on the updated task execution attribute.

[0142] For example, Figure 4 The following is a schematic diagram of the architecture of a task processing solution provided in an embodiment of the present application. Figure 4As shown, first, a task processing model to be trained is constructed; the target task processing model to be trained is established based on business scenarios, scenario process annotations, interpretation of safety and compliance requirements, task levels, historical task records, sensor data, hardware resource status, home scenario data, user feedback, external data (weather, calendar), business personalized rules, etc. Among them, the algorithms deployed in the task processing model to be trained may include convolutional neural network algorithm, K-nearest neighbor algorithm, long short-term memory network algorithm, BERT algorithm, hidden Markov algorithm, linear discriminant analysis algorithm and natural language processing algorithm, etc. Further, the task processing model to be trained can be trained with the acquired historical data based on real-time policy interpretation information, personalized business scenarios, real-time information rule change maintenance, real-time supplement of basic rules, information extraction, knowledge fusion, knowledge processing, etc., so as to obtain the trained target task processing model. Further, the target task processing model can be deployed to the intelligent robot. Based on the intelligent robot, the task type, task occurrence time, task-related description information, etc. of the task to be executed are obtained. Then, based on the target task processing model, the acquired task type, task occurrence time and task related description information are processed, and the task urgency level and priority of the task to be executed are determined. Furthermore, based on the task urgency level and priority, the response result of the task to be executed is determined and output.

[0143] The advantage of the above method is that it achieves the effect of training a target task processing model that can accurately perform task classification and processing based on the constructed sample data, and by adopting sample data from multiple data sources, the accuracy and robustness of the model are effectively improved, so that the trained target task processing model can support more complex task classification processing, effectively improving the model performance.

[0144] Figure 5 This is a schematic diagram of a task processing flow provided in an embodiment of the present application. This embodiment is a preferred embodiment of the above embodiments. Figure 5 As shown, the method includes:

[0145] First, data collection and preprocessing are performed, and task information of historical tasks is obtained based on sensor data, user behavior data, historical task records and external data sources associated with the intelligent robot. Further, the task information is preprocessed to obtain preprocessed feature data of the historical tasks; further, the task level classification is determined, and the historical task level of the historical tasks is determined based on the preprocessed feature data, so as to identify and mark the preprocessed feature data based on the historical task level. Among them, the historical task level can include emergency level, high priority, medium priority, low priority and entertainment and social level; then, model training and optimization are performed, and the training model is trained and optimized based on the preprocessed feature data of the historical tasks and the historical task level labels to obtain the target task processing model that has been trained; further, model deployment and real-time application are performed, and real-time sensor data and user behavior data associated with the intelligent robot are obtained, and task information of the task to be processed is determined based on the real-time sensor data and user behavior data. Further, the task information is processed based on the target task processing model deployed in the intelligent robot, and the task emergency identification of the task to be processed is output. Then, the response action of the task to be processed is determined according to the task urgency mark; further, user feedback and model improvement are carried out to obtain user feedback and real-time task processing results, and the model parameters of the target task processing model deployed in the intelligent robot are updated based on the user feedback and real-time task processing results to obtain an improved model.

[0146] The technical solution provided in the embodiment of the present application solves the problem that the adaptability between the robot execution logic and the actual application needs of the user in the related technology is low, which in turn causes the robot to be unable to effectively provide corresponding services to the user. It achieves the effect of accurately determining the task urgency level of the task to be executed based on the robot according to the neural network model, improves the real-time and accuracy of task classification, and achieves the effect of dynamically adapting to environmental changes or special scene requirements during task classification. In addition, by updating the task execution attributes of the tasks to be executed based on tasks with the same task urgency level, the task execution efficiency is effectively improved, the task execution decision output by the robot is made more intelligent, the service efficiency of the robot is improved, and the intelligence level of the robot is improved.

[0147] The following is an embodiment of the task processing device provided in the embodiment of the present application. The device and the task processing methods of the above embodiments belong to the same inventive concept. For details not described in detail in the embodiment of the task processing device, please refer to the embodiment of the above task processing method.

[0148] Figure 6This is a structural diagram of a task processing device provided in an embodiment of the present application, the device specifically includes: a task information acquisition module 410, a task identification determination module 420, a task attribute update module 430 and a task execution module 440.

[0149] Among them, the task information acquisition module 410 is used to obtain the task information of the task to be processed, wherein the task information includes at least one of the audio information, video information and text information corresponding to the task to be processed; the task identifier determination module 420 is used to input the task information into the pre-trained target task processing model, and output the task urgency identifier of the task to be processed; the task attribute update module 430 is used to respond to the event that the task to be executed includes the same task urgency identifier, and reprocess the task information of the task to be executed with the same task urgency identifier, so as to update the task execution attribute of the task to be executed based on the reprocessing result; wherein the task to be executed is a task with a confirmed task urgency identifier and has not yet been executed; the task execution module 440 is used to execute the task to be executed based on the updated task execution attributes.

[0150] The technical solution provided by the embodiment of the present application obtains the task information of the task to be processed, and further, inputs the task information into the pre-trained target task processing model to output the task urgency identifier of the task to be processed; further, in response to the event that the task to be executed includes the same task urgency identifier, the task information of the task to be executed with the same task urgency identifier is reprocessed to update the task execution attribute of the task to be executed based on the reprocessing result; further, the task to be executed is executed based on the updated task execution attribute, which solves the problem that the adaptability between the robot execution logic and the actual application needs of the user in the prior art is low, and thus the robot cannot effectively provide corresponding services to the user, and achieves the effect of accurately determining the task urgency level of the task to be executed based on the robot according to the neural network model, improves the real-time and accuracy of task classification, and achieves the effect of dynamically adapting to environmental changes or special scene requirements during task classification, and by updating the task execution attributes of the task to be executed based on the tasks with the same task urgency level, the task execution efficiency is effectively improved, the task execution decision output by the robot is made more intelligent, the service efficiency of the robot is improved, and the intelligence level of the robot is improved.

[0151] Based on the above technical solution, the task information acquisition module 410 is used to perform at least one of the following operations: based on the video acquisition device in the intelligent robot, collect video information including the target object and the environment to which the target object belongs; based on the audio acquisition device in the intelligent robot, collect audio information emitted by the target object; in response to the editing operation of the target object on the intelligent robot, obtain the text information edited by the target object.

[0152] On the basis of the above technical solution, the device further comprises:

[0153] An information rearrangement module, configured to, after acquiring the task information of the task to be processed, rearrange the task information based on at least one preset target field, and determine the target content under the at least one target field;

[0154] An information updating module is used to update the task information of the task to be processed based on the target content.

[0155] Based on the above technical solutions, the at least one target field includes at least one of an event type field, an event occurrence time field, an event duration field, a user feedback field, an environmental parameter field of the environment to which the intelligent robot belongs, a user motion parameter field, a user type field, a user tone feature field, a user facial feature field, a festival field and a date field.

[0156] On the basis of the above-mentioned technical solutions, the task attribute updating module 430 includes: a mergeable task determining unit, which is used to obtain the task information of the to-be-executed tasks with the same task emergency mark, and determine whether there is a mergeable task and the task priority of the to-be-executed task according to the data content of the user type field, the user tone feature field, the user facial feature field and the event type field in the task information; wherein the mergeable task is a task in the to-be-executed tasks; a task merging unit, which is used to merge the mergeable tasks in response to the event that there is a mergeable task, so as to update the to-be-executed task; a priority determining unit, which is used to determine the task priority of the updated to-be-executed task;

[0157] Correspondingly, the task attribute updating module 430 is specifically configured to update the task attribute of the task to be executed based on the updated task priority of the task to be executed; wherein the task attribute includes the execution order of the task to be executed.

[0158] On the basis of the above-mentioned technical solutions, the device also includes: a historical task information acquisition module, used to obtain the pending task information of historical tasks; a task information marking module, used to mark the pending task information based on the constraint conditions corresponding to the pre-set emergency identification of each task, so as to determine the historical emergency identification of the historical task; a sample data determination module, used to determine the sample data involved in training the target task processing model based on the pending task information of the historical tasks and the corresponding historical emergency identification.

[0159] On the basis of the above-mentioned technical solutions, the task information marking module includes: a constraint determination unit, which is used to determine the constraint conditions satisfied by the pending task information of each historical task according to the traversal order corresponding to all pre-set task emergency identifiers for each historical task; and a task information marking unit, which is used to use the task emergency identifier corresponding to when the constraint conditions are satisfied as the historical emergency identifier of the historical task.

[0160] On the basis of the above technical solutions, the historical emergency identification includes a first emergency identification, a second emergency identification, a third emergency identification, a fourth emergency identification and a fifth emergency identification; the first constraint condition corresponding to the first emergency identification includes that the field content of the environmental parameter field is the first content or the field content of the event type field is at least one of the preset types; the second constraint condition corresponding to the second emergency identification includes the condition of a preset reminder information; the third constraint condition corresponding to the third emergency identification is that the pre-entered facial features meet the conditions of a preset event or a preset first task; the fourth constraint condition corresponding to the fourth emergency identification is a condition including a preset second task; the fifth constraint condition corresponding to the fifth emergency identification is that at least one target field does not have a preset content.

[0161] On the basis of the above-mentioned technical solutions, the device also includes: a sample data processing module, which is used to train and process the sample data based on at least one algorithm deployed in the model to be trained, and obtain the loss value corresponding to the at least one algorithm; a model parameter correction module, which is used to correct the model parameters in the model to be trained based on the loss value corresponding to the at least one algorithm, and take the convergence of the loss function in the model to be trained as the training goal to obtain the model to be verified; a model verification module, which is used to verify the model to be verified corresponding to the algorithm based on the test sample, and use the model to be verified whose verification result meets the preset conditions as the target task processing model.

[0162] Based on the above technical solutions, the device also includes: a model deployment module, which is used to deploy the target task processing model in the intelligent robot, and update the model parameters of the target task processing model based on the task information, execution results and response results of the task to be processed.

[0163] The task processing device provided in the embodiments of the present application can execute the task processing method provided in any embodiment of the present application, and has the corresponding functional modules and beneficial effects for executing the task processing method.

[0164] It is worth noting that in the embodiment of the above-mentioned task processing device, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application.

[0165] Figure 7 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Figure 7 A block diagram of an exemplary server 12 suitable for implementing embodiments of the present application is shown. Figure 7 The server 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0166] like Figure 7 As shown, the server 12 is in the form of a general purpose computing device. The components of the server 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0167] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0168] The server 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the server 12, including volatile and non-volatile media, removable and non-removable media.

[0169] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The server 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 7 not shown, usually called a "hard drive"). Although Figure 7Not shown, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, DVD-ROM, or other optical medium) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The system memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present application.

[0170] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in system memory 28, such program modules 42 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. Program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0171] The server 12 may also communicate with one or more external devices 14 (e.g., keyboards, pointing devices, displays 24, etc.), may communicate with one or more devices that enable a user to interact with the server 12, and / or may communicate with any device that enables the server 12 to communicate with one or more other computing devices (e.g., network cards, modems, etc.). Such communication may be performed via an input / output (I / O) interface 22. In addition, the server 12 may also communicate with one or more networks (e.g., local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via a network adapter 20. As shown, the network adapter 20 communicates with other modules of the server 12 via a bus 18. It should be understood that, although not shown in the figure, other hardware and / or software modules may be used in conjunction with the server 12, including, but not limited to, microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0172] The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, implementing the task processing method steps provided in the first embodiment of the present invention, the method comprising:

[0173] Acquire task information of the task to be processed, wherein the task information includes at least one of audio information, video information and text information corresponding to the task to be processed;

[0174] Inputting the task information into a pre-trained target task processing model, and outputting a task urgency mark of the task to be processed;

[0175] In response to an event that the tasks to be executed include the same task urgency identifier, reprocessing the task information of the tasks to be executed with the same task urgency identifier, so as to update the task execution attribute of the tasks to be executed based on the reprocessing result; wherein the tasks to be executed are tasks with confirmed task urgency identifiers and have not yet been executed;

[0176] The to-be-executed task is executed based on the updated task execution attribute.

[0177] Of course, those skilled in the art can understand that the processor can also implement the technical solution of the task processing method provided in any embodiment of the present application.

[0178] Figure 8 This is a schematic diagram of the structure of an intelligent robot provided in an embodiment of the present application. Figure 8 As shown, the embodiment of the present application provides an intelligent robot 80, which includes a display input device 801, a video acquisition device 802, an audio acquisition device 803 and the above Figure 7 The electronic device 70 shown. The electronic device 70 can be respectively connected to a display input device 801 (for example, a display screen device with touch input and display), a video acquisition device 802 (for example, a camera, etc.) and an audio acquisition device 803 (for example, a microphone, etc.) to achieve related input and output functions. For example, the electronic device 70 can be a robot computer with a central processing unit.

[0179] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the steps of the task processing method provided in the above-mentioned embodiment of the present application are implemented. The method includes:

[0180] Acquire task information of the task to be processed, wherein the task information includes at least one of audio information, video information and text information corresponding to the task to be processed;

[0181] Inputting the task information into a pre-trained target task processing model, and outputting a task urgency mark of the task to be processed;

[0182] In response to an event that the tasks to be executed include the same task urgency identifier, reprocessing the task information of the tasks to be executed with the same task urgency identifier, so as to update the task execution attribute of the tasks to be executed based on the reprocessing result; wherein the tasks to be executed are tasks with confirmed task urgency identifiers and have not yet been executed;

[0183] The to-be-executed task is executed based on the updated task execution attribute.

[0184] The computer storage medium of the embodiment of the present application can adopt any combination of one or more computer-readable media. Computer-readable media can be computer-readable signal media or computer-readable storage media. Computer-readable storage media can be, for example, - but not limited to - electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or devices, or any combination of the above. More specific examples (non-exhaustive lists) of computer-readable storage media include: electrical connections with one or more wires, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the above. In this document, computer-readable storage media can be any tangible medium containing or storing a program, which can be used by an instruction execution system, device or device or used in combination with it.

[0185] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0186] Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0187] The computer program code for performing the operation of the embodiment of the present application can be written in one or more programming languages ​​or a combination thereof, and the programming language includes an object-oriented programming language, such as Java, Smalltalk, C++, and also includes a conventional procedural programming language, such as "C" language or similar programming language. The program code can be executed completely on the user's computer, partially on the user's computer, as an independent software package, partially on the user's computer, partially on the remote computer, or completely on the remote computer or server. In the case of a remote computer, the remote computer can be connected to the user's computer by any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (e.g., using an Internet service provider to connect to the Internet).

[0188] Note that the above are only preferred embodiments of the present application and the technical principles used. Those skilled in the art will understand that the present application is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present application. Therefore, although the present application is described in more detail through the above embodiments, the present application is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A task processing method, characterized in that: Applied in an intelligent robot, the method comprises: Acquire task information of the task to be processed; wherein the task information includes at least one of audio information, video information and text information corresponding to the task to be processed; Inputting the task information into a pre-trained target task processing model, and outputting a task emergency identifier of the task to be processed; In response to an event that the tasks to be executed include the same task urgency identifier, reprocessing the task information of the tasks to be executed with the same task urgency identifier, so as to update the task execution attribute of the tasks to be executed based on the reprocessing result; wherein the tasks to be executed are tasks with confirmed task urgency identifiers and have not yet been executed; The to-be-executed task is executed based on the updated task execution attribute.

2. The method according to claim 1, characterized in that The task information of the task to be processed may be obtained by at least one of the following methods: Collecting video information including a target object and an environment to which the target object belongs based on a video acquisition device in the intelligent robot; Collecting audio information emitted by the target object based on the audio collection device in the intelligent robot; In response to the editing operation of the target object on the intelligent robot, the text information edited by the target object is obtained.

3. The method according to claim 1, characterized in that After obtaining the task information of the task to be processed, the method further includes: Rearrange the task information based on at least one preset target field to determine target content under the at least one target field; The task information of the task to be processed is updated based on the target content.

4. The method according to claim 3, characterized in that The at least one target field includes at least one of an event type field, an event occurrence time field, an event duration field, a user feedback field, an environmental parameter field of the environment to which the intelligent robot belongs, a user motion parameter field, a user type field, a user tone feature field, a user facial feature field, a festival field, and a date field.

5. The method according to claim 1, characterized in that: The reprocessing of the task information of the pending tasks with the same task urgency identifier includes: For pending tasks with the same task urgency mark, obtaining the task information of the pending tasks, and determining whether there are mergeable tasks and the task priority of the pending tasks according to the data content of the user type field, the user voice tone feature field, the user facial feature field and the event type field in the task information; wherein the mergeable tasks are tasks in the pending tasks; In response to an event that there are mergeable tasks, merge the mergeable tasks to update the to-be-executed tasks; Determine the task priority of the updated pending tasks; Correspondingly, updating the task execution attribute of the to-be-executed task based on the reprocessing result includes: Based on the updated task priority of the task to be executed, updating the task attribute of the task to be executed; The task attributes include the execution order of the tasks to be executed.

6. The method according to claim 1, characterized in that Also includes: Get the pending task information of historical tasks; Marking the task information to be processed based on the constraint conditions corresponding to the preset emergency identifiers of each task to determine the historical emergency identifiers of the historical tasks; Based on the to-be-processed task information of the historical tasks and the corresponding historical emergency identifiers, sample data participating in training the target task processing model is determined.

7. The method according to claim 6, characterized in that The marking of the task information to be processed based on the constraint conditions corresponding to the preset task urgency flags of each task to determine the historical urgency flags of the historical tasks includes: For each historical task, determine the constraint conditions satisfied by the task information to be processed of the historical task according to the traversal order corresponding to all pre-set task emergency identifiers; The task emergency flag corresponding to when the constraint condition is met is used as the historical emergency flag of the historical task.

8. The method according to claim 6 or 7, characterized in that: The historical emergency identification includes a first emergency identification, a second emergency identification, a third emergency identification, a fourth emergency identification and a fifth emergency identification; The first constraint condition corresponding to the first emergency indicator includes that the field content of the environmental parameter field is the first content or the field content of the event type field is at least one of the preset types; the second constraint condition corresponding to the second emergency indicator includes the condition of a preset reminder information; the third constraint condition corresponding to the third emergency indicator is that the pre-entered facial features meet the condition of a preset event or a preset first task; the fourth constraint condition corresponding to the fourth emergency indicator is a condition including a preset second task; the fifth constraint condition corresponding to the fifth emergency indicator is a condition that at least one target field does not have preset content.

9. The method according to claim 7, characterized in that: The method further comprises: Training the sample data based on at least one algorithm deployed in the model to be trained to obtain a loss value corresponding to the at least one algorithm; Based on the loss value corresponding to the at least one algorithm, the model parameters in the model to be trained are modified, and the convergence of the loss function in the model to be trained is taken as the training goal to obtain the model to be verified; The model to be verified corresponding to the algorithm is verified based on the test sample, and the model to be verified whose verification result meets the preset conditions is used as the target task processing model.

10. The method according to claim 9, characterized in that Also includes: The target task processing model is deployed in the intelligent robot, and the model parameters of the target task processing model are updated based on the task information, execution result and response result of the task to be processed.

11. A task processing device, characterized in that: Configured in the intelligent robot, including: A task information acquisition module, used to acquire task information of a task to be processed, wherein the task information includes at least one of audio information, video information and text information corresponding to the task to be processed; A task identification determination module, used to input the task information into a pre-trained target task processing model, and output a task emergency identification of the task to be processed; A task attribute updating module, for responding to an event that the to-be-executed tasks include the same task urgency identifier, reprocessing the task information of the to-be-executed tasks with the same task urgency identifier, so as to update the task execution attribute of the to-be-executed tasks based on the reprocessing result; wherein the to-be-executed tasks are tasks with confirmed task urgency identifiers and have not yet been executed; The task execution module is used to execute the to-be-executed task based on the updated task execution attribute.

12. An electronic device, characterized in that: The electronic device comprises: one or more processors; A memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the task processing method as described in any one of claims 1-10.

13. An intelligent robot, characterized in that: The intelligent robot comprises a display input device, a video acquisition device, an audio acquisition device and the electronic device according to claim 12; the electronic device is communicatively connected with the display input device, the video acquisition device and the audio acquisition device respectively.

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

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the computer program implements the task processing method according to any one of claims 1 to 10.

Images