WO2020199963A1

WO2020199963A1 - Method for food ordering robot to identify food ordering intention of user, and robot

Info

Publication number: WO2020199963A1
Application number: PCT/CN2020/080725
Authority: WO
Inventors: 彭瑞刚
Original assignee: 时时同云科技(成都)有限责任公司
Priority date: 2019-03-29
Filing date: 2020-03-23
Publication date: 2020-10-08
Also published as: CN109979453A

Abstract

A smart intention identification human-machine interaction method for a food ordering robot, comprising the following steps: summarizing food ordering intentions on the basis of service features (step 1); calibrating which food ordering intentions common food ordering terms belong to (step 2); on the basis of the number of summarized food ordering intentions and the number of common food ordering terms, selecting and training a food ordering intention identification model (step 3); and a food ordering robot identifying, by means of the intention identification model, a food ordering intention of a user expressed by means of speech (step 4).

Description

Method and robot for ordering robot to recognize user's ordering intention

Cross reference of related applications

This application claims the priority of a Chinese patent application filed with the Chinese Patent Office on March 29, 2019, the application number is 201910250987.5, and the name is "An Intelligent Intent Recognition Human-Computer Interaction Method for Food Ordering Robots". The reference is incorporated in this application.

Technical field

The present disclosure relates to the technical field of robot intention recognition, and in particular, to a method and a robot for an ordering robot to recognize a user's ordering intention.

Background technique

With the development of social economy and the improvement of industrialization level, people are using more and more robots. Intelligent robot technology has become a hot spot in social research, and robot attendants have entered ordinary restaurants. In order to improve customer dining efficiency and save shop costs, more and more shops begin to use ordering robots, but the intent recognition scheme of traditional ordering robots is to configure a certain intent, and generally need to configure the verb and noun corresponding to the intent , It is extremely inconvenient to configure. One more word in the middle can’t be recognized without one less word. It does not have the model training function, which makes it difficult to configure the voice conversion. The machine often fails to recognize the ordering intention. Therefore, the key to solving the above problems is to develop a method for the ordering robot to accurately recognize the user's ordering intention.

The application publication number is CN 109146717 A and the publication date of the application publication date is 2019.01.04 discloses a method for self-service ordering and delivery of food in an unmanned restaurant; the method includes: the user obtains it through his own personal mobile terminal The identification information provided by the dining table; the user opens the login interface of the ordering system through his own personal mobile terminal and logs into the ordering system; the ordering system authenticates the received identification information, and the ordering system provides the ordering interface to the user The user orders food and forms an order in the ordering system; the ordering system automatically shares the information with the restaurant robot, which receives the order and identification information, and picks up the food in the back kitchen according to the order of the order; The robot locates the table position according to the position information of the table in the identification information, and delivers the meal ordered by the user to the designated meal. ; This disclosure realizes the formation of intelligent automatic management between the user order and the meal, reduces the number of restaurant waiters, and helps reduce restaurant operating costs.

However, this unmanned restaurant self-service ordering and delivery method does not have a model training function, which makes it difficult to configure voice conversion, and the machine often fails to recognize the ordering intention.

Summary of the invention

(1) Technical problems solved

In view of the above-mentioned shortcomings in the prior art, the present disclosure provides a method and a robot for an ordering robot to recognize a user’s ordering intention, which can effectively overcome the existing technology that does not have a model training function, resulting in voice conversion. Configuration is difficult, and the machine often fails to recognize the defects of ordering intentions.

(2) Technical solution

In order to achieve the above objective, the present disclosure provides a method for a food ordering robot to recognize a user's ordering intention, which includes the following steps:

Summarize ordering intentions based on business characteristics;

Mark the ordering intention to which it belongs to the daily ordering language;

Select and train an ordering intention recognition model based on the summed up number of ordering intentions and the number of daily ordering terms; and

The ordering robot recognizes the user's ordering intention expressed by voice through the intention recognition model.

The present disclosure also provides an ordering robot using the above method for recognizing a user’s ordering intention. The ordering robot is provided with a controller, a wireless communication module, a voice recognition module, a speaker and a display screen; wherein, wireless communication The module, voice recognition module, speaker and display screen are all electrically connected to the controller.

The present disclosure also provides an electronic device, which includes:

At least one processor; and,

A memory communicatively connected with the at least one processor; wherein,

The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute any of the foregoing first aspect or any implementation of the first aspect A method for the ordering robot to recognize the user's ordering intention.

The present disclosure also provides a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium stores at least one executable instruction, and the executable instruction is used to make a processor execute the aforementioned first aspect or the first A method for the ordering robot in any implementation manner of the aspect to recognize the user's ordering intention.

The present disclosure also provides a computer program product. The computer program product includes a calculation program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions that, when executed by a processor, cause the The processor executes the method for the ordering robot in the foregoing first aspect or any implementation of the first aspect to recognize the user's ordering intention.

(3) Beneficial effects

Compared with the prior art, the present disclosure provides a method and robot for an ordering robot to recognize the user's ordering intention, and the beneficial effect produced is: creatively applying the method of natural language text classification to the ordering robot The field of intention recognition in the ordering stage greatly improves the robot's intelligence. It does not need to be divided into verbs and nouns manually. It has a model training function. After the model is trained, the corresponding intention can be accurately identified for different sentences. Greatly increase the flexibility and simplicity of configuration and optimize the user experience.

Description of the drawings

In order to explain the embodiments of the present disclosure or the technical solutions in the prior art more clearly, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only These are some embodiments of the present disclosure. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

Figure 1 is a schematic flow diagram of the disclosed method;

2 is a schematic diagram of the system structure of the ordering robot of the present disclosure;

FIG. 3 is a schematic structural diagram of an electronic device 30 provided by an embodiment of the disclosure.

detailed description

In order to make the objectives, technical solutions, and advantages of the embodiments of the present disclosure clearer, the technical solutions in the embodiments of the present disclosure will be described clearly and completely in conjunction with the accompanying drawings in the embodiments of the present disclosure. Obviously, the described embodiments It is a part of the embodiments of the present disclosure, but not all the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present disclosure.

A method for an ordering robot to recognize a user's ordering intention, as shown in Figures 1 to 2, includes the following steps:

Step 1: Intent preparation. The developer summarizes the intent of ordering according to business characteristics;

Step 2: Data preparation, summarizing and calibrating the intent of daily ordering terms;

Step three, model training, select and train an ordering intention recognition model according to the number of intentions and the scale of training data;

Step 4, online application, the ordering robot obtains the voice information of the user's order, converts the text through voice recognition technology, enters the text into the intention recognition model, obtains the probability corresponding to each intention, and applies the appropriate probability threshold set according to the business characteristics Determine the intention of the conversation;

Step five, online optimization, for the situation where the user’s ordering intention cannot be identified through comparison or the identified ordering intention is wrong, training data is added, and steps 3 and 4 are repeated to make the intention recognition more and more accurate. Among them, the user's ordering intention cannot be recognized, that is, the online intention recognition score is low, and the preset probability threshold is not reached.

Specifically, the ordering robot in step 4 is equipped with a controller, the controller model is STM32F103, and the controller is powered by a small battery; the ordering robot in step 4 is equipped with a wireless communication module, and the wireless communication module model is URS-GPRS- 730, the wireless communication module is electrically connected to the controller; the ordering robot in step 4 is equipped with a voice recognition module, the model of the speech recognition module is LD3320, and the speech recognition module is electrically connected to the controller; the ordering robot in step 4 is equipped with The speaker for the voice feedback information is electrically connected to the controller; the ordering robot in step 4 is provided with a display screen for confirming the ordering information for customers, and the display screen is electrically connected to the controller.

When using, the developer summarizes the intent of ordering according to the business characteristics and completes the intent preparation; then summarizes and calibrates the intent of the daily ordering language to complete the data preparation; secondly, select the appropriate machine learning according to the number of intents and the scale of training data Model, train the intention classification model, and complete the model training; again, the ordering robot obtains the voice information of the user’s order, converts the voice signal into text through the voice recognition module, and enters the text into the intention recognition model to obtain the probability corresponding to each intent. Apply Set appropriate thresholds according to the business characteristics to determine the intent in the dialogue, and complete the online application; for the low score in online intent recognition, or to recognize the wrong intent, add training data, repeat steps 3 and 4, and make the intent recognition more and more The more accurate, the online optimization is completed; and feedback information can be played to the user through the speaker, and the ordering information can be confirmed to the user through the display screen to ensure that the ordering information is correct. The controller controls the wireless communication module to send the ordering information to the background for processing To complete the ordering work;

The intent recognition scheme of the existing general platform is to configure a certain intent. Generally, it is necessary to configure the verb and noun corresponding to the intent, such as "lai a bowl of noodles" in the intent of ordering. The verb "lai" and nouns need to be configured during configuration. "A bowl of noodles" is extremely inconvenient to configure. One more word in the middle cannot be recognized. This disclosure only needs to tell the model that "coming a bowl of noodles" is an ordering intention during the data preparation stage, and does not need to be manually divided into The verb "lai" and the noun "one bowl of noodles", after training the model, it can accurately identify the corresponding intentions for the different sentences like "lai one noodle", which greatly increases the flexibility and simplicity of configuration, and optimizes Improve the user experience.

This disclosure has disclosed the models of the controller, the wireless communication module, and the voice recognition module. The internal structure and pin functions of the electrical components involved in this disclosure are the common knowledge of those skilled in the art, and those skilled in the art have the ability to establish The circuit connection relationship between the above-mentioned electrical components, in addition, the use of a wireless communication module to establish a wireless connection is common knowledge of those skilled in the art.

The present disclosure provides a method and a robot for an ordering robot to recognize a user's ordering intention, which produces beneficial effects: creatively applying the method of natural language field text classification to the field of intention recognition at the ordering stage of the ordering robot, It greatly improves the robot's intelligence. It does not need to be manually split into verbs and nouns. It has a model training function. After the model is trained, the corresponding intent can be accurately identified for different sentences, which greatly increases the flexibility and configuration Convenience, optimized user experience.

3 is a schematic structural diagram of an electronic device 30 provided by an embodiment of the present disclosure. The electronic device 30 includes at least one processor 301 (for example, a CPU), at least one input and output interface 304, a memory 302, and at least one communication bus 303 for implementing The connection and communication between these components. The at least one processor 301 is configured to execute the computer instructions stored in the memory 302, so that the at least one processor 301 can execute any of the foregoing embodiments of the method for ordering robots to recognize the user's ordering intention. The memory 302 is a non-transitory memory (non-transitory memory), which may include volatile memory, such as high-speed random access memory (RAM: Random Access Memory), and may also include non-volatile memory (non-volatile memory) , Such as at least one disk storage. The communication connection with at least one other device or unit is realized through at least one input and output interface 304 (which may be a wired or wireless communication interface).

In some embodiments, the memory 302 stores a program 3021, and the processor 301 executes the program 3021, which is used to execute any of the aforementioned food ordering robots in the method embodiment for recognizing the user's ordering intention.

The electronic device can exist in many forms, including but not limited to:

(1) Mobile communication equipment: This type of equipment is characterized by mobile communication functions, and its main goal is to provide voice and data communications. Such terminals include: smart phones (such as iPhone), multimedia phones, functional phones, and low-end phones.

(2) Ultra-mobile personal computer equipment: This type of equipment belongs to the category of personal computers, has calculation and processing functions, and generally also has mobile Internet features. Such terminals include: PDA, MID and UMPC devices, such as iPad.

(3) Portable entertainment equipment: This type of equipment can display and play multimedia content. Such devices include: audio, video players (such as iPod), handheld game consoles, e-books, as well as smart toys and portable car navigation devices.

(4) Specific server: a device that provides computing services. The composition of a server includes a processor, hard disk, memory, system bus, etc. The server is similar to a general computer architecture, but due to the need to provide highly reliable services, it is High requirements in terms of performance, reliability, security, scalability, and manageability.

(5) Other electronic equipment with data interaction function.

It should be noted that in this article, relational terms such as first and second are only used to refer to

One entity or operation is distinguished from another entity or operation without necessarily requiring or implying these

There is any such actual relationship or sequence between entities or operations. Moreover, the terms "include", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device that includes a series of elements includes not only those elements, but also includes Other elements of, or also include elements inherent to this process, method, article or equipment. If there are no more restrictions, the element defined by the sentence "including a..." does not exclude the existence of other same elements in the process, method, article, or equipment including the element.

The various embodiments in this specification are described in a related manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments.

In particular, as for the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and for related parts, please refer to the partial description of the method embodiment.

The logic and/or steps represented in the flowchart or described in other ways herein, for example, can be considered as a sequenced list of executable instructions for implementing logic functions, and can be embodied in any computer-readable medium, For use by instruction execution systems, devices, or equipment (such as computer-based systems, systems including processors, or other systems that can fetch instructions from instruction execution systems, devices, or equipment and execute instructions), or combine these instruction execution systems, devices Or equipment. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by an instruction execution system, device, or device or in combination with these instruction execution systems, devices, or devices. More specific examples (non-exhaustive list) of computer readable media include the following: electrical connections (electronic devices) with one or more wiring, portable computer disk cases (magnetic devices), random access memory (RAM), Read only memory (ROM), erasable and editable read only memory (EPROM or flash memory), fiber optic devices, and portable compact disk read only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable media on which the program can be printed, because it can be used, for example, by optically scanning the paper or other media, and then editing, interpreting, or other suitable media if necessary. The program is processed in a manner to obtain the program electronically and then stored in the computer memory.

It should be understood that each part of the present disclosure can be implemented by hardware, software, firmware or a combination thereof.

In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if it is implemented by hardware, as in another embodiment, it can be implemented by any one or a combination of the following technologies known in the art: a logic gate circuit for implementing logic functions on data signals Discrete logic circuits, application-specific integrated circuits with suitable combinational logic gates, programmable gate array (PGA), field programmable gate array (FPGA), etc.

The above are only specific implementations of the present disclosure, but the protection scope of the present disclosure is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present disclosure. All should be covered within the protection scope of this disclosure. Therefore, the protection scope of the present disclosure should be subject to the protection scope of the claims.

Claims

A method for an ordering robot to recognize a user's ordering intention, which is characterized in that it includes the following steps:

Summarize ordering intentions based on business characteristics;

Mark the ordering intention to which it belongs to the daily ordering language;

Select and train an ordering intention recognition model according to the summed-up number of ordering intentions and the number of daily ordering terms; and

The ordering robot recognizes the user's ordering intention expressed by voice through the intention recognition model.
The method according to claim 1, wherein the recognizing the user's ordering intention expressed by voice through the intention recognition model further comprises:

Converting the voice into text information;

Inputting the text information into the intention recognition model to obtain the probability value of the text information corresponding to each of the ordering intentions; and

The probability value is compared with a preset probability threshold for each of the ordering intentions to identify the user's ordering intention.
The method according to claim 2, wherein when the user’s ordering intention cannot be recognized by comparison or the recognized ordering intention is wrong, the ordering intention recognition is trained by adding the daily ordering language model.
The method according to any one of claims 1-3, wherein the method further comprises:

The identified ordering intention is fed back to the user through the speaker.
The method according to any one of claims 1-3, wherein the method further comprises:

The recognized ordering intention is displayed to the user through the display screen for confirmation by the user.
The method according to any one of claims 1-3, wherein the method further comprises:

The ordering robot sends the identified ordering intention to the background for processing.
An ordering robot using the method for recognizing a user’s ordering intention according to any one of claims 1 to 6, wherein the ordering robot is provided with a controller, a wireless communication module, and a voice Identification module, speaker and display screen;

Wherein, the wireless communication module, voice recognition module, speaker and display screen are all electrically connected to the controller.
An electronic device, characterized in that, the electronic device includes:

At least one processor; and,

A memory communicatively connected with the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute any of the preceding claims 1-6. A method for the ordering robot to recognize the user's ordering intention.
A non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium stores at least one executable instruction, and the executable instruction is used to make a processor execute any of the preceding claims 1-6 Ordering robot recognizes the user’s ordering intention.
A computer program product comprising a calculation program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, when the program instructions are executed by a processor, cause the processing The device executes the method for recognizing the user’s ordering intention by the ordering robot described in any one of claims 1-6.