CN117219079A - Control method and device of intelligent home robot - Google Patents

Abstract

The application provides a control method and a device of an intelligent home robot, wherein after a server obtains a voice signal input by a user, the intelligent home robot associated with the user is determined, at least one project label service expected to be executed by the user is determined by utilizing a large language model based on the voice signal and service description information of a plurality of services in a configured service list, and the intelligent home robot is instructed to execute the at least one project label service. The scheme of the application can promote the control flexibility of controlling the intelligent home robot.

Description

Control method and device of intelligent home robot

Technical Field

The application relates to the technical field of intelligent control, in particular to a control method and device of an intelligent home robot.

Background

The intelligent home robot can execute corresponding actions or control various intelligent devices in the home to work according to voice instructions input by a user.

However, the intelligent home robot can only respond to the pre-configured voice operation command. If the content of the voice command input by the user is partially different from the configured voice operation command, the intelligent home robot cannot accurately and even cannot understand the intention of the user, and cannot respond to the voice command of the user, so that the control flexibility of the intelligent home robot is poor.

Disclosure of Invention

The application provides a control method and a control device for an intelligent home robot, which can improve the control flexibility of the intelligent home robot.

In one aspect, the application provides a control method of an intelligent home robot, which is applied to a server, and the method comprises the following steps:

obtaining a voice signal input by a user;

determining the intelligent home robot associated with the user;

determining a service list configured for the intelligent home robot, wherein the service list comprises service description information of a plurality of services which can be provided by the intelligent home robot;

determining at least one target service expected to be executed by the user by utilizing a large language model based on the voice signal and service description information of the plurality of services, wherein the target service belongs to the plurality of services;

and instructing the intelligent home robot to execute the at least one project label service.

In one possible implementation, the service list further includes: at least one service parameter associated with the service and at least one parameter value corresponding to each service parameter;

the determining, based on the speech signal and the service description information of the plurality of services, at least one target service that the user desires to execute by using a large language model includes:

Based on the voice signals, determining at least one project label service expected to be executed by the user and target parameter values of target service parameters required to be referenced for executing the target service by utilizing a large language model according to service description information of the plurality of services and at least one parameter value corresponding to each service parameter associated with the service;

the instructing the intelligent home robot to perform the at least one landmark service includes:

and the intelligent home robot is instructed to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service.

In yet another possible implementation manner, at least some of the services in the service list are provided with a mandatory service parameter, and the mandatory service parameter of the service belongs to at least one service parameter associated with the service;

before the intelligent home robot is instructed to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service, the intelligent home robot further comprises:

for each target service, if the target service is provided with a necessary service parameter and the necessary service parameter of the target service is not included in the target service parameters required to be referred to by the target service determined by the large language model, generating a parameter reminding statement by using the large language model based on the necessary service parameter of the target service, wherein the parameter reminding statement is used for prompting a user to provide a parameter value of the necessary service parameter of the target service;

And returning a parameter input prompt matched with the parameter prompt statement to target terminal equipment so as to output the parameter input prompt through the target terminal, wherein the parameter input prompt is used for prompting a user to provide a parameter value of a necessary service parameter of the target service, and the target terminal equipment is an intelligent terminal or the intelligent home robot for sending the voice signal.

In yet another possible implementation manner, after the determining, by using the large language model, at least one landmark service that the user desires to perform, the method further includes:

for each target service, a service processing module corresponding to the target service is called, and a service execution instruction corresponding to the target service is generated through the service processing module, wherein the service execution instruction is used for indicating the intelligent home robot to execute the target service;

the instructing the intelligent home robot to perform the at least one landmark service includes:

and sending a service execution instruction corresponding to the at least one project label service to the intelligent home robot so that the intelligent home robot executes the at least one project label service.

In yet another possible implementation manner, before generating, by the service processing module, a service execution instruction corresponding to the target service, the method further includes:

Determining a target equipment type of intelligent home equipment which is required to be controlled by the intelligent home robot to execute the target service through the service processing module;

determining a target intelligent home device associated by the user and belonging to the target device type;

the service execution instruction is used for instructing the intelligent home robot to control the target intelligent home device to execute the target service.

In yet another possible implementation manner, the determining, using a large language model, at least one landmark service that the user desires to perform includes:

determining at least one project label service expected to be executed by the user and the execution sequence of the at least one project label service by utilizing a large language model;

the instructing the intelligent home robot to perform the at least one landmark service includes:

and sequentially indicating the intelligent home robot to execute the at least one project label service according to the execution sequence of the at least one project label service.

In yet another possible implementation manner, the method further includes:

if the target service expected to be executed by the user cannot be determined by utilizing the large language model, outputting an abnormal reminding statement through the large language model, wherein the abnormal reminding statement is used for prompting that the service indicated by the voice signal cannot be identified;

And returning an abnormal prompt matched with the abnormal prompt statement to target terminal equipment so as to output the abnormal prompt through the target terminal equipment, wherein the abnormal prompt is used for prompting that the user cannot recognize the service indicated by the voice signal, and the target terminal equipment is an intelligent terminal or an intelligent home robot for sending the voice signal.

In still another aspect, the present application further provides a control device for an intelligent home robot, applied to a server, the device comprising:

a signal obtaining unit for obtaining a voice signal input by a user;

a robot determining unit, configured to determine an intelligent home robot associated with the user;

a list determining unit configured to determine a service list configured for the intelligent home robot, the service list including service description information of a plurality of services that the intelligent home robot can provide;

a service determining unit, configured to determine, based on the voice signal and service description information of the plurality of services, at least one target service that the user desires to execute, using a large language model, the target service belonging to the plurality of services;

and the service indicating unit is used for indicating the intelligent home robot to execute the at least one project label service.

In one possible implementation manner, the service list determined by the list determining unit further includes: at least one service parameter associated with the service and at least one parameter value corresponding to each service parameter;

the service determination unit includes:

a service determining subunit, configured to determine, based on the voice signal, service description information of the multiple services and at least one parameter value corresponding to each service parameter associated with the service, and determine, by using a large language model, at least one target service that the user desires to execute and a target parameter value of a target service parameter to which the target service needs to be referred;

the service instruction unit includes:

and the first service indicating subunit is used for indicating the intelligent home robot to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service.

In yet another possible implementation manner, at least some of the services in the service list determined by the list determining unit are provided with a mandatory service parameter, where the mandatory service parameter of the service belongs to at least one service parameter associated with the service;

the apparatus further comprises:

A first sentence generating unit, configured to generate, for each target server, a parameter alert sentence by using the large language model based on the candidate service parameter of the target service if the candidate service parameter of the target service is set in the target service and the candidate service parameter of the target service to be referred to by the large language model is not included in the target service parameter of the target service determined by the large language model before the first service instruction subunit instructs the intelligent home robot to execute the at least one target service, where the parameter alert sentence is used to prompt a user to provide a parameter value of the candidate service parameter of the target service;

the first reminding feedback unit is used for returning a parameter input reminder matched with the parameter reminding statement to target terminal equipment so as to output the parameter input reminder through the target terminal, wherein the parameter input reminder is used for prompting a user to provide a parameter value of a necessary service parameter of the target service, and the target terminal equipment is an intelligent terminal or an intelligent home robot for sending the voice signal.

As can be seen from the above, in the embodiment of the present application, after the server of the intelligent home robot acquires the voice signal input by the user, the server may determine the target service that the user desires to execute by using the large language model based on the voice signal and the description information of the plurality of services in the configured service list. By means of powerful language understanding capability of the large language model and combining with service description information of various services provided by the configured intelligent home robot, as long as the voice signals input by the user contain content of related services or the semantics of the voice signals are consistent with the service description information of the services, target services expected to be executed by the user can be determined through the large language model, so that voice instructions corresponding to different services are not required to be configured for the intelligent home robot in advance, and the user is not required to accurately input corresponding voice instructions each time, the intelligent home robot can be controlled to execute the related services, and the control flexibility of the intelligent home robot is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only embodiments of the present application, and other drawings may be obtained according to the provided drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a control method of an intelligent home robot provided by the application;

fig. 2 shows a schematic flow chart of a control method of the intelligent home robot provided by the application;

FIG. 3 is a schematic diagram of a scenario architecture to which the present approach is applicable;

fig. 4 is a schematic diagram showing a composition structure of a control device for an intelligent home robot according to an embodiment of the present application.

Detailed Description

The scheme of the application can promote the control flexibility of the intelligent home robot, does not need a user to accurately know the voice instructions of various services provided by the intelligent home robot, and can also realize the instruction of the home robot to execute related services.

In the present application, the type of intelligent home robot may be a plurality of possibilities.

For example, the intelligent home robot may be an intelligent home robot with a walking function, so that the intelligent home robot can assist a human in drying home or providing other related services while simultaneously performing intelligent home devices in a home. For example, the smart home robot may be a robot having a plurality of moving wheels and capable of moving in different directions; or, the intelligent household robot can dynamically balance the quadruped robot, so that the intelligent robot can walk more flexibly and can climb stairs and the like.

Of course, other forms of intelligent home robot are equally applicable to this embodiment without limitation.

The following description of the embodiments of the present application will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without any inventive effort, are intended to be within the scope of the application.

Fig. 1 shows a schematic flow chart of a control method of a smart home robot according to an embodiment of the present application, where the method of the embodiment may be applied to a server, and the server may be a server for controlling the smart home robot. For example, the server may be a cloud server for controlling the intelligent home robot, which is not limited.

The method of the embodiment can comprise the following steps:

s101, obtaining a voice signal input by a user.

For example, in one possible scenario, a user may input a voice signal to the smart home robot, in which case the smart home robot may send the voice signal input by the user to the server.

In yet another possible scenario, if the user is out or otherwise not around the smart home robot, but wishes to provide a corresponding service through the smart home robot, the user may also input a voice signal to a smart terminal such as a cell phone or computer in which the robot control application is installed, and the smart terminal may send the voice signal to the server.

Wherein the voice signal may contain information about a user's desire for the intelligent home robot to perform an operation.

S102, determining the intelligent home robot associated with the user.

The intelligent home robot associated with the user is the intelligent home robot which establishes a binding relation with the user in advance.

For example, the server side may store the smart home robots associated with the user identifications of different users in advance, and correspondingly, when obtaining the voice signal input by the user, the server side may obtain the user identification of the user sent by the sending end of the voice signal (such as the aforementioned smart terminal or the smart home robot), and determine the smart home robot associated with the user based on the user identification.

S103, determining a service list configured for the intelligent home robot.

Wherein the service list includes service description information of a plurality of services that the intelligent home robot can provide.

The service that the intelligent home robot can provide can be a function or a completed operation that the intelligent home robot can provide. Depending on the type of intelligent home robot, the specific configuration, etc., the services that the intelligent home robot can provide will vary.

For example, the services that the intelligent home robot can provide may include, but are not limited to, some or all of safety alarm services, baby care services, pet care services, elderly care services, clothes washing services, home cleaning services, and intelligent light control services.

The service description information of each service is used to describe service contents of the service and information related to the service such as roles. For example, the service description information of the service may include: the service can realize specific functions and roles, and one or more of operation steps related to the service, intelligent home equipment required to be used for executing the service, service modes related to the service and the like.

Illustrating:

for the security alarm service, the service description information of the service may include specific service content provided by the security alarm service, specific operations that can be achieved by the service, and the achieved purpose. For example, the service description information of the security alert service may be: the function can capture video information in the home and report the video information to the server; when a person which cannot be identified appears in a family, outputting an alarm prompt and automatically sending alarm information to a terminal of an associated user; outputting an alarm reminding when glass or a set object in a household is damaged; and when the closing state of the kitchen gas stove is monitored and the safety risk is confirmed, outputting an alarm prompt to an intelligent terminal of a user and the like.

For a child caretaking service, the service description information may describe the content items that the child caretaking can be implemented in, the specific manner in which the caretaking is employed, and so on. For example, baby care services may include voice-controlled cradles, turning on baby lights, turning on or off story machines, providing remote baby video viewing, and the like.

For a pet care service, the service description information may include various operations and control contents of the pet care support, and the like. For example, according to the user indication or timing requirement, a feeding instruction is issued to the pet feeding machine so as to feed the pet; and controlling information introduction related to the service, such as food amount of food fed by the pet feeding machine, and the like, according to the feeding requirement of the user.

The method is similar to other services, and the characteristics, the realized functions, the operations and the like of the service can be known through the service description information of the service, so that the method can be used as a corpus basis for analyzing the relevance between the user voice and the service by using a large language model.

It can be appreciated that in practical application, if all the service items provided by each intelligent home robot under the control of the server are the same, only one general service list can be configured in the server. In practical application, different service lists can be configured for intelligent home robots of different models or types. Of course, if the services opened by the users in different families to the intelligent home robots of the same model are different, the application can also store the service list corresponding to the different intelligent home robots in the server, and the service list can comprise all the services which the intelligent home robots possess and belong to which are opened.

S104, determining at least one project label service expected to be executed by the user by utilizing a large language model based on the voice signals and the service description information of the multiple services.

The large language model (Large Language Model, LLM) refers to a natural language processing technology based on machine learning, and a model capable of understanding human language and automatically generating language is constructed through learning of a large-scale corpus. The large language model is capable of performing a variety of natural language processing tasks such as automatic question-answering, machine translation, speech recognition, text generation, and the like.

The types of the large language model can be various, and the specific model type of the large language model in the application can be not limited, for example, the large language model can be an encoder-based model or a transducer-based model, etc.

In the application, the server is provided with the service description information of a plurality of services, the voice signals input by the user can be understood for a large language model based on the service description information of the services, the relevance of the text of the voice signals input by the user and the service description information of the services in terms of text or semantics and the like can be accurately analyzed, and the service expected to be executed, which is expressed by the voice signals of the user, can be finally determined.

For convenience of distinction, a service that the large language model determines that the user desires to execute based on the voice signal is referred to as a target service. It will be appreciated that the target service belongs to a plurality of services configured in a service list.

It will be appreciated that in most scenarios, the input of a large language model is text, and therefore, in the present application, it is also possible to first convert a speech signal into a target text, which is the text form of the speech signal. On the basis, the target text can be input into the large language model, and the large language model can analyze the service intention expressed by the target text by combining the service description information of each service in the service list, so that the target service expected to be executed by the user can be determined from multiple services.

In the application, the voice signal input by the user can only express the desire of executing one service, and can also express the desire of the intelligent home robot to execute a plurality of services.

If the user wants to execute the service a and then execute the service B, the execution sequence is to provide the service a and then provide the service B.

S105, instructing the intelligent home robot to execute at least one project label service.

The specific manner in which the server instructs the intelligent home robot to perform each target service may be a number of possible, and the application is not limited in this regard.

For example, the server may generate an execution instruction of each target service, and send the execution instruction of the at least one target service to the intelligent home robot, so that the intelligent home robot executes the at least one target service.

As another example, in one possible implementation, in order to be able to manage different services, and also facilitate flexibly adding or deleting control logic of different services in a server, in the present application, the server may be configured with service processing modules of different services. The service processing module of each service is a pre-configured program module and is configured to generate execution instructions for the service, and may also be configured to process other processing operations associated with the execution of the service.

On the basis, for each target service, a service processing module corresponding to the target service can be called, and a service execution instruction corresponding to the target service is generated through the service processing module of the target service. The service execution instruction corresponding to the target service is used for indicating the intelligent home robot to instruct to execute the target service. Correspondingly, the server sends a service execution instruction corresponding to at least one project label service to the intelligent home robot, so that the intelligent home robot executes the at least one project label service.

It may be appreciated that, in the case that the large language model determines the execution sequence of the at least one target service, the server may sequentially instruct the intelligent home robot to execute each target service according to the execution sequence of the at least one target service.

If so, the server generates and sends respective service execution instructions for instructing the intelligent home robot to the intelligent home robot according to the execution sequence of the at least one target service, so that the intelligent home robot sequentially executes the at least one target service according to the sequentially issued at least one service execution instruction.

Taking the example that the server generates service execution instructions through the service processing modules of the target services, the server can sequentially send the service execution instructions corresponding to the at least one target service to the intelligent home robot according to the execution sequence of the at least one target service.

Particularly, the server can call the service processing modules of the target services in sequence according to the execution sequence of at least one target service, the service processing modules can add the service execution instruction of the target service to the message queue, and the server can send the service execution instruction in the message queue to the corresponding intelligent home robot in sequence, so that the intelligent home robot of the user can execute the at least one target service in sequence.

In an alternative manner, if the target service that the user desires to execute cannot be determined by using the large language model, an abnormal reminding statement is output through the large language model, the abnormal reminding statement being used for prompting that the service indicated by the voice signal cannot be recognized. For example, the abnormal alert sentence may be "do not understand your semantics, please re-input a voice instruction", etc., which is not limited.

Correspondingly, the server can return the abnormal reminding matched with the abnormal reminding statement to the target terminal equipment so as to output the abnormal reminding through the target terminal equipment. The target terminal device is an intelligent terminal or an intelligent home robot which sends the voice signal, for example, the voice signal sent by the user through the intelligent home robot, and then the server returns the abnormal prompt to the intelligent home robot.

The abnormal reminding is used for reminding the user of being unable to identify the service indicated by the voice signal. The abnormal reminding can be a voice type reminding, for example, an abnormal reminding statement is converted into voice type reminding information, and the abnormal reminding is obtained. Of course, the abnormal alert may be other types of alerts, without limitation.

As can be seen from the above, in the embodiment of the present application, after the server of the intelligent home robot acquires the voice signal input by the user, the server may determine the target service that the user desires to execute by using the large language model based on the voice signal and the description information of the plurality of services in the configured service list. By means of powerful language understanding capability of the large language model and combining with service description information of various services provided by the configured intelligent home robot, as long as the voice signals input by the user contain content of related services or the semantics of the voice signals are consistent with the service description information of the services, target services expected to be executed by the user can be determined through the large language model, so that voice instructions corresponding to different services are not required to be configured for the intelligent home robot in advance, and the user is not required to accurately input corresponding voice instructions each time, the intelligent home robot can be controlled to execute the related services, and the control flexibility of the intelligent home robot is improved.

It can be understood that in practical application, the intelligent home robot provides more kinds of services, and if a certain service is simpler, the intelligent home robot determines that the service needs to be executed, and then the service can be directly executed. Alternatively, some services may be configured with default service parameters, and the server may instruct the intelligent home robot to perform the service according to the default service parameters of the service.

However, some services may involve a plurality of alternative service parameters, and each service parameter may be associated with at least one alternative parameter value, and when the parameter values of the service parameter selected by the user and the service parameter set by the user are different, the specific execution process of the intelligent home robot to execute the corresponding service may also be different.

Wherein, the service parameters of the service are used for representing the functions, modes or execution modes which can be selected by executing the service, and the like. And the optional parameter values of the service parameters may be parameter values that are optionally set under the service parameters.

For ease of understanding, several services are illustrated:

taking the clothes washing service provided by the intelligent home robot as an example, the clothes can be washed in a plurality of different washing modes such as soft washing, energy saving or strong washing modes, and the clothes can be washed in different water temperature selections, so that the washing mode and the water temperature belong to service parameters which can be selected by the clothes washing service. The parameter values of the washing mode can be specific values such as soft washing, energy saving, strength and the like as parameter values, and the corresponding water temperature can be cold water, warm water or high temperature, and can also be specific selectable values such as water temperature values and the like.

For another example, taking a switch light service as an example, the service parameters pertaining to the service are turned on and off, and the specific lights that are turned on and off, such as a living room light, a main sleeping light, a secondary sleeping light 1, a secondary sleeping light 2, etc., all pertain to the parameter values that are selectable for turning on or off the relevant service parameters.

Of course, the above is exemplified by two cases, and in practical application, the service parameters of the service and the selectable parameter values under the service parameters can be set according to the service provided by the intelligent home robot and the self requirements. For example, the lamp may be turned on and off as different services, and the service parameter may be a lamp position, and the parameter value of the lamp position may be parameters such as a lamp in a living room and a lamp in a bedroom, which is not limited.

Based on the above, in the service list configured by the application, besides the service description information of each service, at least one service parameter and at least one parameter value corresponding to each service parameter can be associated with at least part of the services in the plurality of services in the service list.

Correspondingly, based on the voice signal input by the user, service description information of a plurality of services in a service list and at least one parameter value corresponding to each service parameter associated with the service, determining at least one target service expected to be executed by the user and a target parameter value of a target service parameter required to be referenced for executing each target service by using a large language model.

Based on this, the server may instruct the intelligent home robot to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service.

It will be appreciated that in some cases, the service parameter of a certain service provided by the smart home robot is a parameter that must be referred to for executing the service, in which case if the user does not select a specific parameter value of the service parameter, the smart home robot may not normally execute the service. Based on this, the server in the present application can also generate a sentence for prompting the user to input necessary parameters by means of a large language model.

For ease of understanding, the following description is provided in connection with fig. 2. Fig. 2 is a schematic flow chart of a control method of an intelligent home robot according to an embodiment of the present application, where, for convenience of description, a user inputs a voice signal through a terminal device such as a mobile phone, and the method of the present embodiment is applied to a server. The method of the embodiment can comprise the following steps:

s201, obtaining a voice signal sent by a user through the intelligent terminal.

The embodiment is described by taking the case that a user inputs a voice signal through an intelligent terminal.

It will be appreciated that the user may also input a voice signal through the smart home robot and send it to the server, and for ease of understanding, reference may be made to fig. 3, which shows that the solution of the present application is applicable to a scene frame diagram.

As can be seen from fig. 3, the user can not only interact with the server through the intelligent terminal such as a mobile phone or a computer, for example, send a voice signal to the server, or receive voice information fed back by the server, etc. Voice interaction with the server can also be performed through the intelligent home robot.

Of course, the user can also interact with the intelligent home robot and the server through other forms of information besides voice through the intelligent terminal, and the method is not limited.

S202, determining the intelligent home robot associated with the user.

S203, determining a service list configured for the intelligent home robot.

Wherein the service list comprises: the intelligent home robot can provide service description information of a plurality of services, at least one service parameter related to the services and at least one parameter value corresponding to each service parameter.

In the present application, at least some of the services in the service list may be provided with a mandatory service parameter, which is of at least one service parameter associated with the service. The optional service parameters of the service may be set in the service list or may be configured separately, which is not limited.

S204, determining at least one project mark service expected to be executed by a user and a target parameter value of a target service parameter to be referred by executing the target service by using a large language model based on the voice signal, service description information of a plurality of services and at least one parameter value corresponding to each service parameter associated with each service.

The target service parameter of the target service belongs to at least one service parameter associated with the target service, and the target parameter value of the target service parameter belongs to at least one corresponding parameter value under the target service parameter associated with the target service.

The process of determining the target service based on the large language model may be referred to in the foregoing description, and will not be described herein.

The target parameter value of the target service parameter to be referenced for determining the target service based on the large language model may be combined with a selectable parameter value of the service parameter of each target service, and the service parameter of the target service and the corresponding parameter value contained in the voice signal are analyzed, or the matching relationship between the semantic analysis result of the voice signal and the parameter value of the service parameter is combined.

For example, assuming that the content of the voice signal input by the user is "wash clothes of a baby, the large language model can not only identify that the service desired by the user is a clothes washing service, but also determine that the target service is a clothes washing service according to the clothes washing mode of the clothes washing service including soft washing, and the parameter value of the clothes washing mode of the baby is soft washing.

It will be appreciated that for any target service, if the speech signal input by the user does not relate to the parameter value of the service parameter associated with the target service, the target service parameter and the parameter value thereof that the large language model determines to be referenced for executing the target service may be null.

Of course, for a target service, if the voice signal input by the user includes service parameter values of multiple service parameters related to the target service, the target service parameters required to be referred to by the target service determined by the large language model may be multiple, and each target service parameter corresponds to a corresponding target service parameter value.

S205, for each item mark service, if the target service is provided with the necessary service parameter and the necessary service parameter of the target service is not included in the target service parameter which is required to be referred by the target service and determined by the large language model, generating a parameter reminding statement by using the large language model based on the necessary service parameter of the target service.

The parameter reminding statement is used for prompting the user to provide parameter values of the necessary service parameters of the target service. For example, assuming that the absence of a critical service parameter in the clothes washing service is temperature, the parameter alert statement generated by the large language model may be "please provide the temperature of the next washed clothes".

As described above, some services are set with an optional service parameter, in which case, if the parameter value of the optional service parameter of the service is absent, execution of the service is affected.

Based on this, after obtaining the target parameter values of at least one target service and the target service parameters associated with the target service output by the large language model, the server needs to detect whether each target service is set with a mandatory service parameter, but the large language model does not identify the mandatory service parameter of the target service and the target parameter value thereof from the voice signal. If there is a target service for which the alternative service parameter is set and the parameter value of the alternative service parameter of the target service is not recognized from the voice signal based on the large language model, this step S205 needs to be performed.

It can be understood that the large language model can generate an interactive sentence based on an input phrase or sentence, etc., and therefore, based on the alternative service parameters of the target service, an interactive sentence, i.e., a parameter alert sentence, prompting the user to input the alternative service of the target service can be generated by means of the large language model.

For example, the server may input the necessary service parameters of the target service as prompt words into the large language model, and the large language model may output the parameter prompt statement.

In particular, in the case where service processing modules of different services are configured in the server, for each service, if the service has an optional service parameter, the optional service parameter of the service may be configured in the service processing module of the service, and of course, parameter values of which the optional service parameter is optional may also be configured.

Correspondingly, for each target service, a service processing module of the target service can be called, and target parameter values of all target service parameters of the target service determined by the large language model are input into the service processing module. Judging whether the target parameter value of each target service parameter of the target service contains the target parameter value of the necessary service parameter or not based on the configured necessary service parameter of the target service by the service processing module of the target service, and if so, inputting the necessary service parameter of the target service into a large language model as a prompt by the service processing module, and obtaining a parameter prompt statement by using the large language model.

As can be seen in connection with fig. 3, a large model processing center is included in the server, which may be considered as a large language model or a module for processing speech signals based on a large language model. Based on this, the voice signal input by the user or the text converted from the voice signal can be input to the large model processing center, so as to determine the parameter values of the target service and the related service parameters which the user desires to execute.

Meanwhile, a service processing center is also configured in the server, and the service processing center comprises service processing modules corresponding to different services. Based on this, the target service and the target parameter value of the target service parameter thereof output by the large language model may be input into the service processing module of the target service, so that the service processing module determines whether there is a lack of the necessary service parameter, and generates a service execution instruction indicating execution of the target service.

S206, returning the parameter input prompt matched with the parameter prompt statement to the intelligent terminal so as to output the parameter input prompt through the target terminal.

Wherein the parameter input reminder is used for prompting the user to provide the parameter value of the necessary service parameter of the target service.

For example, considering that a user interacts with a server in a voice manner, the application can generate a parameter input reminder in a voice manner based on the parameter reminding statement, for example, the parameter reminding statement in a text manner is converted into a voice manner, and the parameter input reminder in the voice manner is obtained.

Of course, the parameter input reminder may be a text reminder or other reminders, which is not limited.

It can be understood that after the user obtains the parameter input prompt through the intelligent terminal, if the user inputs the parameter value of the necessary service parameter of the target service based on the parameter input prompt, the server still determines the parameter value of the target service parameter of the target service that the user desires to execute by using the large language model, and finally feeds back the parameter value to the service processing module of the target service, and the specific process is as described above, and will not be repeated here.

It will be appreciated that the process is similar to that of the intelligent home robot, and will not be described in detail.

S207, if the target parameter values of the target service parameters associated with each target service are not lack of the target parameter values of the respective necessary service parameters, the intelligent home robot is instructed to execute at least one item target service according to the target parameter values of the target service parameters corresponding to each target service.

If the voice signal is obtained after the voice signal of the user is obtained initially and the reminding is fed back to the intelligent terminal of the user, and finally, the target parameter value of the necessary service parameter of the target service is identified by means of the large language model, wherein the target parameter value of the necessary service parameter of the target service is included in the parameter values of the target service parameter of the target service, then the server determines the target parameter value of the necessary service parameter of the target service is absent.

In practical application, if at least one target service has an execution sequence, the server may determine a target service that needs to be processed currently according to the execution sequence, and determine whether the target service lacks a target parameter value of the necessary service parameter. If the target parameter value of the essential service parameter is not found out in the project label service, the parameter input prompt is returned to the intelligent terminal of the user through the introduction, after the essential service parameter is found out in the project label service, the intelligent robot can be instructed to execute the target service, and then the next target service to be processed is determined according to the execution sequence.

As shown in fig. 2, in actual application, after determining the target service that needs to be processed currently, the server may also call the service processing module of the target service to determine whether the target parameter value of the necessary service parameter of the target service is absent, and execute related processing. In the case where the service processing module confirms the target parameter value of the optional service parameter not missing for the target service, the service processing module may generate a corresponding service execution instruction and add to the message queue as in fig. 2.

Correspondingly, after confirming that the service processing module of the target service has generated the service execution instruction of the target service, the server determines the next target service to be processed, invokes the service processing module of the target service to process the target service according to the description, and finally controls the intelligent home robot to execute one or more target services.

It will be appreciated that some service intelligent home robots may accomplish this by themselves for different services, but many services, intelligent home robots also need to accomplish this by means of intelligent home devices inside the home. For example, the switching lamp needs to be completed by means of a smart lamp, and washing laundry needs to be completed by means of a smart washing machine.

Based on the above, for each target service, before the service processing module of the target service generates a service execution instruction corresponding to the target service, the service processing module can also determine the target equipment type of the intelligent home equipment which needs to be controlled by the intelligent home robot to execute the target service; there is also a need to determine a target smart home device that is associated by the user and that belongs to the target device type. As shown in fig. 2, the server may determine the smart home devices controllable within each user's home.

On the basis, the service execution instruction generated by the service processing module of the target service can be used for instructing the intelligent home robot to control the target intelligent home device to execute the target service.

To facilitate an understanding of the benefits of the present disclosure, the following description is provided in connection with one example:

taking the control of turning off the light as an example, according to the existing control mode of the intelligent home robot, if the user inputs voice: the intelligent home robot can only recognize a single voice command, cannot recognize a plurality of commands, and cannot recognize services indicated by voice signals of a user if the voice signals of the user are not matched with the configured single voice command.

Moreover, the intelligent home robot cannot understand which lamps are the other lamps at present, and the intelligent home robot cannot respond to the instruction input by the user through the voice signal.

In the present application, a service list of the intelligent home robot may be configured in advance in the server according to a usage specification or the like of each intelligent lamp that the intelligent home robot can control, the service list includes a service for switching on and off the lamps, and a parameter value (specific lamp name value) corresponding to a service parameter (lamp name) associated with the service for switching on and off the lamps is assumed to be configured with lamp names that can be turned on and off, including: parlor lights, primary bedroom lights, secondary bedroom lights and restaurant lights.

On the basis, after the server obtains the voice input by the user, the voice signal is analyzed by means of a large language model based on the parameter value of the service parameter of the configured switching lamp service, so that the user hopes to turn on the living room lamp first, and then turn off the main lying lamp, the secondary lying lamp and the restaurant lamp. Accordingly, the server may instruct the intelligent home robot to perform a light on service and the light that needs to be turned on is a living room light, and then perform a light off service, and the light that needs to be turned off is a primary bedroom light, a secondary bedroom light, and a restaurant light. Correspondingly, the intelligent home robot can sequentially send an opening instruction to the living room lamp and send a closing instruction to the main bedroom lamp, the secondary bedroom lamp and the dining room lamp, so that the intelligent home robot can not only recognize voice signals outside the non-voice instructions input by the user, but also enable the user to input the voice signals for controlling the intelligent home robot to have higher flexibility, and recognize a plurality of voice instructions of the user.

It will be appreciated that in embodiments of the present application, the intelligent home robot may also perform other functions not mentioned above, as well as those that are currently available with conventional home robots.

As shown in fig. 2, the intelligent home robot includes a video module, a sensor module, a storage module, and the like, in addition to an audio module that can be used to realize voice interaction with a user.

The audio module can collect the voice input by the user and play the audio fed back by the server. For example, when the user speaks "help me turn off the light," the audio module will upload the piece of audio to the large model processing center in the server for processing. If the server feeds back a reply sentence, e.g. "good", the audio module will output the speech of the feedback sentence.

The video module is used for acquiring video images of family members; starting or closing video monitoring based on voice instruction of family members; and applications for synchronizing real-time video to terminal devices such as mobile phones or computers of family members, etc.

The sensor module is used for sensing information such as the position where the intelligent home robot is located, obstacles and the like, for example, the intelligent home robot can scan by means of the sensor model to form a map inside a home, and an indoor inspection path and the like are generated.

And the storage module is used for storing some local information of the intelligent home robot, such as user information of individual users in the home, and sensitive information such as maps in the home.

In the application, the server has a large model processing center and a service processing module, and can also have modules such as an encryption user data storage center, and the like, without limitation.

For example, an encrypted user data storage center may be used to store user identification agreed to by a user, user security information, and the like in encrypted form.

The embodiment of the application also provides a control device of the intelligent home robot.

As shown in fig. 4, which is a schematic diagram illustrating a composition structure of a control device of an intelligent home robot according to an embodiment of the present application, the device is applied to a server, and the device includes:

a signal obtaining unit 401 for obtaining a voice signal input by a user;

a robot determining unit 402, configured to determine an intelligent home robot associated with the user;

a list determining unit 403 configured to determine a service list configured for the intelligent home robot, the service list including service description information of a plurality of services that the intelligent home robot can provide;

A service determining unit 404, configured to determine, based on the voice signal and service description information of the plurality of services, at least one target service that the user desires to execute, using a large language model, the target service belonging to the plurality of services;

and a service indication unit 405, configured to instruct the intelligent home robot to execute the at least one project label service.

In one possible implementation, the service list determined by the list determining unit further includes: at least one service parameter associated with the service and at least one parameter value corresponding to each service parameter;

the service determination unit includes:

a service determining subunit, configured to determine, based on the voice signal, service description information of the multiple services and at least one parameter value corresponding to each service parameter associated with the service, and determine, by using a large language model, at least one target service that the user desires to execute and a target parameter value of a target service parameter to which the target service needs to be referred;

the service instruction unit includes:

and the first service indicating subunit is used for indicating the intelligent home robot to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service.

In a further possible implementation manner, at least part of the services in the service list determined by the list determining unit are provided with a mandatory service parameter, and the mandatory service parameter of the services belongs to at least one service parameter associated with the services;

the apparatus further comprises:

a first sentence generating unit, configured to generate, for each target server, a parameter alert sentence by using the large language model based on the candidate service parameter of the target service if the candidate service parameter of the target service is set in the target service and the candidate service parameter of the target service to be referred to by the large language model is not included in the target service parameter of the target service determined by the large language model before the first service instruction subunit instructs the intelligent home robot to execute the at least one target service, where the parameter alert sentence is used to prompt a user to provide a parameter value of the candidate service parameter of the target service;

the first reminding feedback unit is used for returning a parameter input reminder matched with the parameter reminding statement to target terminal equipment so as to output the parameter input reminder through the target terminal, wherein the parameter input reminder is used for prompting a user to provide a parameter value of a necessary service parameter of the target service, and the target terminal equipment is an intelligent terminal or an intelligent home robot for sending the voice signal.

In yet another possible implementation, the apparatus further includes:

the module calling unit is used for calling a service processing module corresponding to the target service for each target service after the service determining unit determines at least one target service expected to be executed by the user by utilizing a large language model, and generating a service execution instruction corresponding to the target service by the service processing module, wherein the service execution instruction is used for indicating the intelligent home robot to instruct to execute the target service;

the service instruction unit includes:

and the second service instruction subunit is used for sending the service execution instruction corresponding to the at least one project label service to the intelligent home robot so that the intelligent home robot executes the at least one project label service.

Further, the device further comprises:

the device type determining unit is used for determining the target device type of the intelligent home device which is required to be controlled by the intelligent home robot to execute the target service through the service processing module before the module calling unit generates the service execution instruction corresponding to the target service through the service processing module;

the device determining unit is used for determining target intelligent home devices which are associated by the user and belong to the target device type;

The service execution instruction generated by the second service instruction subunit is used for instructing the intelligent home robot to control the target intelligent home device to execute the target service.

In yet another possible implementation manner, the service determining unit includes:

a service order determining unit configured to determine, based on the voice signal and service description information of the plurality of services, at least one landmark service that the user desires to execute and an execution order of the at least one landmark service using a large language model;

a service indication unit comprising:

and the third service instruction subunit is used for sequentially instructing the intelligent home robot to execute the at least one project label service according to the execution sequence of the at least one project label service.

In yet another possible implementation, the apparatus further includes:

an abnormal sentence determining unit configured to output, if a target service that the user desires to execute cannot be determined using the large language model, an abnormal alert sentence for prompting that the service indicated by the speech signal cannot be recognized through the large language model;

the abnormal reminding unit is used for returning an abnormal reminding matched with the abnormal reminding statement to the target terminal equipment so as to output the abnormal reminding through the target terminal equipment, wherein the abnormal reminding is used for reminding the user that the service indicated by the voice signal cannot be identified, and the target terminal equipment is an intelligent terminal or an intelligent home robot for sending the voice signal.

It is to be understood that, in the present application, the terms "first," "second," "third," "fourth," and the like in the description and in the claims and the above figures, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in other sequences than those illustrated herein.

It should be noted that, in the present specification, each embodiment is described in a progressive manner, and each embodiment is mainly described as different from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other. Meanwhile, the features described in the embodiments in the present specification may be replaced with or combined with each other to enable those skilled in the art to make or use the present application. For the apparatus class embodiments, the description is relatively simple as it is substantially similar to the method embodiments, and reference is made to the description of the method embodiments for relevant points.

Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises an element.

The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the application. Thus, the present application is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application, which are intended to be comprehended within the scope of the present application.

Claims (10)

1. A control method of an intelligent home robot, applied to a server, the method comprising:

obtaining a voice signal input by a user;

determining the intelligent home robot associated with the user;

determining a service list configured for the intelligent home robot, wherein the service list comprises service description information of a plurality of services which can be provided by the intelligent home robot;

Determining at least one target service expected to be executed by the user by utilizing a large language model based on the voice signal and service description information of the plurality of services, wherein the target service belongs to the plurality of services;

and instructing the intelligent home robot to execute the at least one project label service.

2. The control method of an intelligent home robot according to claim 1, wherein the service list further includes: at least one service parameter associated with the service and at least one parameter value corresponding to each service parameter;

the determining, based on the speech signal and the service description information of the plurality of services, at least one target service that the user desires to execute by using a large language model includes:

based on the voice signals, determining at least one project label service expected to be executed by the user and target parameter values of target service parameters required to be referenced for executing the target service by utilizing a large language model according to service description information of the plurality of services and at least one parameter value corresponding to each service parameter associated with the service;

the instructing the intelligent home robot to perform the at least one landmark service includes:

And the intelligent home robot is instructed to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service.

3. The control method of an intelligent home robot according to claim 2, wherein at least part of services in the service list are provided with mandatory service parameters, and the mandatory service parameters of the services belong to at least one service parameter associated with the services;

before the intelligent home robot is instructed to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service, the intelligent home robot further comprises:

for each target service, if the target service is provided with a necessary service parameter and the necessary service parameter of the target service is not included in the target service parameters required to be referred to by the target service determined by the large language model, generating a parameter reminding statement by using the large language model based on the necessary service parameter of the target service, wherein the parameter reminding statement is used for prompting a user to provide a parameter value of the necessary service parameter of the target service;

and returning a parameter input prompt matched with the parameter prompt statement to target terminal equipment so as to output the parameter input prompt through the target terminal, wherein the parameter input prompt is used for prompting a user to provide a parameter value of a necessary service parameter of the target service, and the target terminal equipment is an intelligent terminal or the intelligent home robot for sending the voice signal.

4. The method of claim 1, further comprising, after the determining, using a large language model, at least one landmark service that the user desires to perform:

for each target service, a service processing module corresponding to the target service is called, and a service execution instruction corresponding to the target service is generated through the service processing module, wherein the service execution instruction is used for indicating the intelligent home robot to execute the target service;

the instructing the intelligent home robot to perform the at least one landmark service includes:

and sending a service execution instruction corresponding to the at least one project label service to the intelligent home robot so that the intelligent home robot executes the at least one project label service.

5. The control method of an intelligent home robot according to claim 4, further comprising, before generating a service execution instruction corresponding to the target service by the service processing module:

determining a target equipment type of intelligent home equipment which is required to be controlled by the intelligent home robot to execute the target service through the service processing module;

Determining a target intelligent home device associated by the user and belonging to the target device type;

the service execution instruction is used for instructing the intelligent home robot to control the target intelligent home device to execute the target service.

6. The method of claim 1, wherein the determining at least one landmark service desired to be performed by the user using a large language model comprises:

determining at least one project label service expected to be executed by the user and the execution sequence of the at least one project label service by utilizing a large language model;

the instructing the intelligent home robot to perform the at least one landmark service includes:

and sequentially indicating the intelligent home robot to execute the at least one project label service according to the execution sequence of the at least one project label service.

7. The control method of an intelligent home robot according to claim 1, further comprising:

if the target service expected to be executed by the user cannot be determined by utilizing the large language model, outputting an abnormal reminding statement through the large language model, wherein the abnormal reminding statement is used for prompting that the service indicated by the voice signal cannot be identified;

And returning an abnormal prompt matched with the abnormal prompt statement to target terminal equipment so as to output the abnormal prompt through the target terminal equipment, wherein the abnormal prompt is used for prompting that the user cannot recognize the service indicated by the voice signal, and the target terminal equipment is an intelligent terminal or an intelligent home robot for sending the voice signal.

8. A control device of an intelligent home robot, applied to a server, the device comprising:

a signal obtaining unit for obtaining a voice signal input by a user;

a robot determining unit, configured to determine an intelligent home robot associated with the user;

a list determining unit configured to determine a service list configured for the intelligent home robot, the service list including service description information of a plurality of services that the intelligent home robot can provide;

a service determining unit, configured to determine, based on the voice signal and service description information of the plurality of services, at least one target service that the user desires to execute, using a large language model, the target service belonging to the plurality of services;

and the service indicating unit is used for indicating the intelligent home robot to execute the at least one project label service.

9. The control device of the intelligent home robot according to claim 8, wherein the service list determined by the list determining unit further includes: at least one service parameter associated with the service and at least one parameter value corresponding to each service parameter;

the service determination unit includes:

a service determining subunit, configured to determine, based on the voice signal, service description information of the multiple services and at least one parameter value corresponding to each service parameter associated with the service, and determine, by using a large language model, at least one target service that the user desires to execute and a target parameter value of a target service parameter to which the target service needs to be referred;

the service instruction unit includes:

and the first service indicating subunit is used for indicating the intelligent home robot to execute the at least one target service according to the target parameter value of the target service parameter corresponding to each target service.

10. The control device of an intelligent home robot according to claim 9, wherein at least part of the services in the service list determined by the list determining unit are provided with a mandatory service parameter, and the mandatory service parameter of the services belongs to at least one service parameter associated with the services;

The apparatus further comprises:

a first sentence generating unit, configured to generate, for each target server, a parameter alert sentence by using the large language model based on the candidate service parameter of the target service if the candidate service parameter of the target service is set in the target service and the candidate service parameter of the target service to be referred to by the large language model is not included in the target service parameter of the target service determined by the large language model before the first service instruction subunit instructs the intelligent home robot to execute the at least one target service, where the parameter alert sentence is used to prompt a user to provide a parameter value of the candidate service parameter of the target service;

the first reminding feedback unit is used for returning a parameter input reminder matched with the parameter reminding statement to target terminal equipment so as to output the parameter input reminder through the target terminal, wherein the parameter input reminder is used for prompting a user to provide a parameter value of a necessary service parameter of the target service, and the target terminal equipment is an intelligent terminal or an intelligent home robot for sending the voice signal.