TWI823195B - Intelligent recommendation method and system - Google Patents

Abstract

The present application provides an intelligent recommendation method and a system. The method includes: determining occupant attributes of passengers in a vehicle; collecting voice information of passengers in the vehicle; sending the voice information and the occupant attributes to a cloud server; using the cloud server to obtain user intention based on the voice information; when determining that the user intention is suitable for adding the occupant attributes in the vehicle, obtaining updated user intention based on the occupant attributes in the vehicle; and using the cloud server to obtain recommendation information based on the updated user intention, and sending the recommendation information to a on-board device. The present application can make relevant recommendations according to the occupant attributes in the vehicle.

Description

Smart recommendation methods and systems

The present application relates to the field of vehicle management technology, and in particular to a smart recommendation method and system.

More and more smart cars have provided smart voice assistants like SIRI and AMAZON ALEXA. Drivers can communicate with the on-board voice assistant through natural language and control related functions of the on-board devices. However, during the implementation of the present application, the inventor found that the existing vehicle-mounted voice assistants did not consider the conditions of other vehicle occupants when controlling the relevant functions of the vehicle-mounted device based on the driver's voice.

In view of the above, it is necessary to provide an intelligent recommendation method and system that can make relevant recommendations based on the attributes of the occupants in the vehicle.

The first aspect of this application provides a smart recommendation method, applied to a vehicle-mounted device. The vehicle-mounted device includes a microphone and a camera module. The method includes: using the camera module to capture images of occupants in the vehicle, and confirming based on the images of the occupants. the attributes of the occupants in the vehicle; using the microphone to collect the voice information of the occupants in the vehicle; using the vehicle-mounted device to send the voice information and the attributes of the occupants in the vehicle to the cloud server; using the cloud server based on the Obtain user intent from voice information; when the cloud server determines that the user intent is suitable for adding occupant attributes in the vehicle, obtain updated user intent based on the occupant attributes in the vehicle; and utilize the cloud server Recommended information is obtained based on the updated user intention, and the recommended information is sent to the vehicle-mounted device.

Optionally, the vehicle's occupant attributes include: the total number of occupants in the vehicle, Driver's identity, driver's age, driver's gender.

Optionally, the attributes of the occupants in the vehicle also include: the identity information of each occupant, the age and gender of each occupant, and the composition relationship of all occupants in the vehicle.

Optionally, the method uses the vehicle-mounted device to confirm the attributes of the occupants in the vehicle based on the occupant images.

Optionally, the method uses the cloud server to confirm the attributes of the occupants in the vehicle based on the occupant images.

Optionally, the method further includes: using the camera module to capture images of the members in the vehicle when the door of the vehicle is closed.

Optionally, the method further includes: using the camera module to capture images of members in the vehicle when the vehicle starts the engine.

Optionally, the method further includes: when it is determined that the user intention is not suitable for adding attributes of the occupants in the vehicle, displaying the recommended information on a display screen of the vehicle-mounted device; or using a speaker of the vehicle-mounted device to broadcast The recommended information.

Optionally, the method further includes: obtaining each occupant's preference information based on each occupant's identity information; and obtaining recommendation information based on the user intention and each occupant's preference information.

A second aspect of this application provides a smart recommendation system for implementing the smart recommendation method. The system includes a vehicle-mounted device and a cloud server that are communicatively connected to each other.

Compared with the existing technology, the smart recommendation method and system can make relevant recommendations based on the attributes of the occupants in the vehicle, that is, the recommendations are more accurate.

100:Vehicle

3: Vehicle-mounted device

32: Processor

31:Memory

30, 40: Recommendation system

300: Confirm module

301: Get the module

302:Execute module

401:Receive module

402:Response module

33:Detection equipment

34:Microphone

35:Communication equipment

36:Display screen

37: Speaker

4:Cloud server

41:Memory

42: Processor

43:Communication equipment

S1~S11, S20~S30, S41~S49, S51~S60: steps

Figure 1 is an application environment diagram of the smart recommendation method according to the preferred embodiment of the present application.

Figure 2 is a functional module diagram of the recommendation system according to the first preferred embodiment of the present application.

Figure 3 is a functional module diagram of the recommendation system according to the second preferred embodiment of the present application.

Figure 4 is a first flow chart of the smart recommendation method according to the preferred embodiment of the present application.

Figure 5 is a second flow chart of the smart recommendation method according to the preferred embodiment of the present application.

Figure 6 is a third flow chart of the smart recommendation method according to the preferred embodiment of the present application.

Figure 7 is a fourth flow chart of the smart recommendation method according to the preferred embodiment of the present application.

In order to more clearly understand the above objects, features and advantages of the present application, the present application will be described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, as long as there is no conflict, the embodiments of the present application and the features in the embodiments can be combined with each other.

Many specific details are set forth in the following description to facilitate a full understanding of the present application. The described embodiments are only some, rather than all, of the embodiments of the present application. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the scope of protection of this application.

Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the technical field to which this application belongs. The terminology used herein in the description of the application is for the purpose of describing specific embodiments only and is not intended to limit the application.

Refer to Figure 1, which is an application environment diagram of the smart recommendation method provided by the preferred embodiment of the present application.

In this embodiment, the smart recommendation method is applied in the application environment where the smart recommendation system 200 is composed of the vehicle-mounted device 3 and the cloud server 4 that are communicated with each other. The smart recommendation method uses the vehicle-mounted device 3 to collect voice information and the attributes of the occupants in the vehicle, and uses the cloud server 4 to identify the user's intention based on the voice information and the attributes of the occupants in the vehicle, and perform relevant recommendations based on the user's intention. Specific details are introduced below.

In this embodiment, the vehicle-mounted device 3 is installed on the vehicle 100 . The vehicle-mounted device 3 can also be called a vehicle-mounted computer, and includes a memory 31 electrically connected to each other, at least one processor 32 , a detection device 33 , a microphone 34 , a communication device 35 , a display 36 , and a speaker 37 .

In this embodiment, the cloud server 4 includes a memory 41, at least one processor 42, and a communication device 43. The cloud server 4 and the vehicle-mounted device 3 establish communication connections using the communication device 43 and the communication device 35 respectively. The communication device 43 and the communication device 35 may be wireless communication devices.

Persons skilled in the art should understand that the structures of the vehicle-mounted device 3 and the cloud server 4 shown in Figure 1 do not constitute a limitation of the embodiments of the present application. The vehicle-mounted device 3 and the cloud server 4 may also include additional components than those shown in Figure 1 . More or less other hardware or software, or different components. For example, the vehicle-mounted device 3 may also include equipment such as a speed sensor. The cloud server 4 may also include a display screen, etc.

It should be noted that the above-mentioned vehicle-mounted device 3 and cloud server 4 are only examples. If other existing or future vehicle-mounted devices and cloud servers can be adapted to this application, they should also be included in the protection scope of this application. , and is incorporated herein by reference.

In some embodiments, the memory 31 and the memory 41 can be used to store program codes and various data of computer programs respectively. For example, the memory 31 can be used to store the recommendation system 30 installed in the vehicle-mounted device 3 , and realize high-speed and automatic access to programs or data during the operation of the vehicle-mounted device 3 . The memory 41 can be used to store the recommendation system 40 installed in the cloud server 4, and realize high-speed and automatic access to programs or data during the operation of the cloud server 4. The memory 31 and the memory 41 may include read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), erasable programmable read-only memory. Memory (Erasable Programmable Read-Only Memory, EPROM), One-time Programmable Read-Only Memory (OTPROM), Electronically Erasable Programmable Read-Only Memory (Electrically-Erasable Programmable Read -Only Memory (EEPROM), Compact Disc Read-Only Memory (CD-ROM) or other optical disk memory, magnetic disk memory, tape memory, or any other non-transitory device that can be used to carry or store data A non-transferable computer-readable storage medium.

In some embodiments, the at least one processor 32 and the at least one processor 42 may each be composed of integrated circuits. For example, it can be composed of a single packaged integrated circuit, or it can be composed of multiple integrated circuits packaged with the same function or different functions, including one or A combination of multiple central processing units (CPUs), microprocessors, digital signal processing chips, graphics processors, and various control chips. The at least one processor 32 is the control core (Control Unit) of the vehicle-mounted device 3. It uses various interfaces and lines to connect various components of the entire vehicle-mounted device 3, and executes programs or models stored in the memory 31. Groups or instructions, and call the data stored in the memory 31 to perform various functions and process data of the vehicle-mounted device 3, for example, perform the function of smart recommendation (see Figure 4, Figure 5, and Introduction to Figure 6). The at least one processor 42 is the control core (Control Unit) of the cloud server 4, using various interfaces and lines to connect various components of the entire cloud server 4, by executing programs stored in the memory 41 Or modules or instructions, and call the data stored in the memory 41 to perform various functions of the cloud server 4 and process data, for example, perform the function of smart recommendation (see Figure 4 and Figure 4 below for details. 5, and introduction to Figure 6).

In this embodiment, the detection device 33 includes, but is not limited to, one or more cameras (also referred to as “camera modules”), pressure sensors, ultrasonic sensors, etc. installed on the vehicle 100 . device, gravity sensor.

In this embodiment, the one or more cameras may be used to photograph each occupant in the vehicle 100 . The one or more cameras may also be used to capture the scene in front of the vehicle 100 .

In this embodiment, the one or more cameras may include a 2DCMOS (Complementary Metal Oxide Semiconductor) camera with infrared function or a ToF (Time of Flight) camera with a laser. The pressure sensor, ultrasonic sensor, or gravity sensor may be used to sense whether there is an occupant on each seat of the vehicle 100 .

The microphone 34 may be used to collect voice information, for example, the voice information of the occupants in the vehicle 100 .

In this embodiment, the display screen 36 may be a touch display screen, used to display various data of the vehicle-mounted device 3 , such as the user interface of the recommendation system 30 . The speaker 37 can be used to output voice signals.

In this embodiment, the recommendation system 30 can be divided into one or more modules. The one or more modules are stored in the memory 31 and processed by one or more processors (eg, 32) to implement the functions provided by this application. Referring to FIG. 2 , in this embodiment, the recommendation system 30 can be divided into a determination module 300 , an acquisition module 301 , and an execution module 302 . The module referred to in this application is a computer program segment that can complete a specific function. The detailed functions of each module will be described in detail below with reference to Figures 4, 5, and 6.

In this embodiment, the recommendation system 40 can be divided into one or more modules. The one or more modules are stored in the memory 41 and processed by one or more processors (eg, 42) to implement the functions provided by this application. Referring to FIG. 3 , in this embodiment, the recommendation system 40 can be divided into a receiving module 401 and a response module 402 . The module referred to in this application is a computer program segment that can complete a specific function. The detailed functions of each module will be described in detail below with reference to Figures 4, 5, and 6.

Figure 4 is a first flow chart of the smart recommendation method provided by the preferred embodiment of the present application.

In this embodiment, the smart recommendation method can be applied in an application environment composed of the vehicle-mounted device 3 and the cloud server 4 . For the vehicle-mounted device 3 and the cloud server 4 that require smart recommendation, the functions provided by this application for smart recommendation can be directly integrated on the vehicle-mounted device 3 and the cloud server 4, or a software development kit (Software Development Kit) can be used. Kit, SDK) runs on the vehicle-mounted device 3 and the cloud server 4.

As shown in Figure 4, the smart recommendation method specifically includes the following steps. According to different needs, the order of the steps in the flow chart can be changed, and some steps can be omitted.

Step S1: The determination module 300 of the vehicle-mounted device 3 determines whether the vehicle 100 meets preset conditions. When the vehicle 100 meets the preset conditions, step S2 is executed.

In one embodiment, the vehicle 100 meeting the preset condition may mean that the door of the vehicle 100 is closed and/or the vehicle 100 starts the engine.

For example, when the door of the vehicle 100 is closed, the door lock of the vehicle-mounted device 3 A detector (not shown in the figure) will send a door lock signal to the vehicle-mounted device 3 . Therefore, the determination module 300 can determine that the vehicle 100 meets the preset condition when detecting the door lock signal. It should be noted that if the vehicle 100 has passengers getting on and off at different times, and then the vehicle 100 meets the preset conditions, the process steps of this method can be executed.

Step S2: The execution module 302 of the vehicle-mounted device 3 uses the camera module of the detection device 33 to capture images of the occupants in the vehicle 100, and confirms the attributes of the occupants in the vehicle 100 based on the occupant images.

In this embodiment, the execution module 302 uses the camera module to capture images of the occupants in the vehicle 100, and obtains attributes of the occupants in the vehicle 100 based on the captured images.

In one embodiment, the attributes of the occupants in the vehicle 100 include, but are not limited to, the total number of occupants in the vehicle 100 , the identity of the driver, the age of the driver, and the gender of the driver. In one embodiment, the attributes of the occupants in the vehicle 100 also include the identity information of each occupant, the age and gender of each occupant, and the composition relationship of all the occupants in the vehicle 100 .

In one embodiment, the composition relationship of all the occupants in the vehicle 100 may refer to a couple relationship, a family relationship, a parent-child relationship, or other relationships such as a stranger relationship.

In this embodiment, when the vehicle 100 meets the preset conditions, the execution module 302 can control the camera module of the detection device 33 to photograph the occupants in the vehicle 100 to obtain the required information. The captured image (ie, the image of the occupants in the vehicle 100). The execution module 302 of the vehicle-mounted device 3 can use a gender and age recognition algorithm based on facial images to determine the age and gender of each occupant based on the captured images. In one embodiment, the vehicle-mounted device 3 pre-stores in the memory 41 the identity information of each occupant who has ridden in the vehicle 100 , as well as the relationship between the occupants (i.e., each occupant who has ridden in the vehicle 100 The relationship between the occupants and other occupants who have ridden in the vehicle 100 is, for example, a couple, a parent-child relationship, or other relationships). In one embodiment, the identity information of each occupant includes, but is not limited to, each occupant's name, facial photo, and role of each occupant as an occupant in the vehicle 100 (that is, each occupant rides in When the vehicle 100 is described, each occupant is a driver or a passenger), and the driving mode of the vehicle 100 when each occupant is an occupant in the vehicle 100 .

In one embodiment, the driving mode of the vehicle 100 may include, but is not limited to, an economic mode (normal mode), a sport mode (sport mode), and a highway mode (highway mode). It should be noted that different driving modes have different power requirements for the vehicle 100 . Among them, the economy mode has the lowest power requirement for the vehicle 100 , the sports mode has the second highest power requirement for the vehicle 100 , and the highway mode has the highest power requirement for the vehicle 100 .

In one embodiment, when the vehicle-mounted device 3 cannot retrieve the identity information of any occupant from the memory 41 , the vehicle-mounted device 3 determines that any occupant is a stranger. In one embodiment, the vehicle-mounted device 3 can use a face recognition algorithm to determine whether each occupant is based on the facial photos of the occupants pre-stored in the memory 41 and the captured images of each occupant. for strangers.

Specifically, when it is recognized that the face photo of any occupant does not match the face photo pre-stored in the memory 41 , the vehicle-mounted device 3 can determine that the any occupant is a stranger. On the contrary, when the face photo of any occupant matches the face photo pre-stored in the memory 41 , the vehicle-mounted device 3 can determine that any occupant is an occupant who has ridden in the vehicle 100 . Therefore, the vehicle-mounted device 3 can also obtain other identity information of any occupant from the memory 41 based on the face photo of the occupant, such as name, role as an occupant in the vehicle 100, As an occupant in the vehicle 100 , the driving mode of the vehicle, the relationship between any occupant and other occupants, etc.

In other embodiments, the execution module 302 of the vehicle-mounted device 3 may also obtain the images of the occupants in the vehicle 100 and then send the obtained images of the occupants to the cloud server 4, and the cloud server 4 to confirm the attributes of the occupants in the vehicle 100 based on the occupant images.

Step S3: The acquisition module 301 of the vehicle-mounted device 3 uses the microphone 34 to collect the voice information of the occupants in the vehicle 100.

For example, when an occupant of the vehicle 100 such as the driver says "Please recommend a nearby restaurant," the microphone 34 can collect the corresponding voice information.

Step S4: The execution module 302 of the vehicle-mounted device 3 sends the voice information to the cloud server 4 through the communication device 35.

Step S5: The receiving module 401 of the cloud server 4 receives the voice information through the communication device 43. The response module 402 of the cloud server 4 analyzes the voice information to obtain the user's intention and generates recommendation information based on the user's intention. The response module 402 sends the user intention and the recommendation information to the vehicle-mounted device 3 through the communication device 43 .

In this embodiment, the response module 402 may first use speech recognition technology to convert the speech information into text; and then use an intention recognition algorithm, such as a rule method based on a dictionary and template, to analyze the text to obtain the user's intention.

For example, assuming that the voice information is "Please recommend a nearby restaurant," the response module 402 uses an intent recognition algorithm to obtain the user's intention: "search" or "nearby restaurant." The response module 402 obtains restaurant information within a preset distance, such as 500 meters, from a designated APP such as Google Maps software based on the user's intention, that is, "search" or "nearby restaurant", and uses the restaurant information as the recommended information. .

Step S6: The acquisition module 301 of the vehicle-mounted device 3 receives the user intention and recommendation information through the communication device 35.

Step S7: The execution module 302 of the vehicle-mounted device 3 determines whether the user intention is suitable for adding attributes of the occupants in the vehicle 100. When it is determined that the user intention is not suitable for adding attributes of the occupants in the vehicle 100 , step S8 is performed. When it is determined that the user intention is suitable for adding attributes of the occupants in the vehicle 100, step S9 is performed.

In one embodiment, the execution module 302 may respond to user input to determine whether the user intent is appropriate for adding occupant attributes within the vehicle 100 .

In one embodiment, the execution module 302 can generate a dialog window and display the dialog window on the display screen 36 . The execution module 302 may determine whether the user intention is suitable for adding occupant attributes in the vehicle 100 according to the user's selection on the dialogue window.

For example, because the server 4 currently does not consider the attributes of the occupants in the vehicle 100 when recommending restaurants, the recommended restaurants include many restaurants suitable for couples, and the current relationship between the occupants in the vehicle 100 is a parent-child relationship, which the user finds inconvenient. Choose from family-friendly restaurants. Therefore, the user may choose to determine on the dialog box dialog window that the user's intention is suitable for adding occupant attributes in the vehicle 100 . Of course, the execution module 302 can also determine whether the user's intention is suitable for adding attributes of the occupants in the vehicle 100 by analyzing the user's voice input.

In other embodiments, the execution module 302 of the vehicle-mounted device 3 can store various user intentions suitable for adding attributes of the occupants in the vehicle in advance; If there is any one of the various user intentions for the occupant attributes, it is determined that the user intention obtained by the current analysis is suitable for adding the occupant attributes in the vehicle; if the user intention obtained by the current analysis does not belong to the various pre-stored user intentions that are suitable for adding the occupant attributes in the vehicle. If there is any one of the user intentions, it is determined that the user intention obtained by the current analysis is not suitable for adding attributes of the occupants in the vehicle.

In this embodiment, the various user intentions suitable for adding attributes of the occupants in the vehicle may refer to user intentions including keywords related to people such as "restaurant", "hotel", or "attractions".

Step S8: When it is determined that the user intention is not suitable for adding attributes of the occupants in the vehicle 100, the execution module 302 of the vehicle-mounted device 3 displays the recommended information. The execution module 302 of the vehicle-mounted device 3 can also broadcast the recommended information.

For example, the execution module 302 can display the recommended information on the display screen 36 . The execution module 302 can call the speaker 37 to broadcast the recommended information.

Step S9: When it is determined that the user intention is suitable for adding the attributes of the occupants in the vehicle 100, the execution module 302 of the vehicle-mounted device 3 updates the user intention by adding the attributes of the occupants in the vehicle 100 to the user intention. The user intention is determined and the updated user intention is sent to the cloud server 4 through the communication device 35 .

In other embodiments, the execution module 302 may also only add the composition relationships of all occupants of the vehicle 100 to the user intention to obtain the updated user intention. For example, would The composition relationships of all the occupants of the vehicle 100 , such as the parent-child relationship, are added to the user intention to obtain the updated user intention: "search", "nearby restaurant", and "parent-child". In one embodiment, if the execution module 302 only adds the composition relationship of all occupants of the vehicle 100 to the user intention, the execution module 302 can also add the attributes of the occupants in the vehicle 100 Other attributes such as the identity information of the occupant are sent to the cloud server 4 .

Step S10: The response module 402 of the cloud server 4 generates updated recommendation information based on the updated user intention. The response module 402 of the cloud server 4 also sends the updated recommendation information to the vehicle-mounted device 3 through the communication device 43 .

For example, the response module 402 sends multiple restaurants suitable for parents and children to the vehicle-mounted device 3 in the form of a recommendation list.

In one embodiment, the cloud server 4 pre-stores the identity information of each occupant who has ridden in the vehicle 100 and the preference information of each occupant.

In other embodiments, the response module 402 may obtain the preference information of each occupant based on the identity information of each occupant.

In other embodiments, the response module 402 updates the recommendation information based on the updated user intention and each occupant's preference information.

Step S11: The acquisition module 301 of the vehicle-mounted device 3 receives the updated recommendation information through the communication device 35. The execution module 302 of the vehicle-mounted device 3 can also display the updated recommendation information on the display screen 36 . The execution module 302 of the vehicle-mounted device 3 can also use the speaker 37 to broadcast the updated recommendation information.

Figure 5 is a second flow chart of the smart recommendation method provided by the preferred embodiment of the present application.

In this embodiment, the smart recommendation method can be applied in an application environment composed of the vehicle-mounted device 3 and the cloud server 4 . For the vehicle-mounted device 3 and the cloud server 4 that require smart recommendation, the functions provided by this application for smart recommendation can be directly integrated on the vehicle-mounted device 3 and the cloud server 4, or a software development kit (Software Development Kit) can be used. Kit, SDK) runs on the vehicle-mounted device 3 and the cloud server 4.

As shown in Figure 5, the smart recommendation method specifically includes the following steps. According to different needs, the order of the steps in the flow chart can be changed, and some steps can be omitted.

Step S20 is the same as step S1 and will not be described again here.

Step S21 is the same as step S2 and will not be described again here.

Step S22 is the same as step S3 and will not be described again here.

Step S23 is the same as step S4 and will not be described again here.

Step S24 is the same as step S5 and will not be described again here.

Step S25 is the same as step S6 and will not be described again here.

Step S26: The execution module 302 of the vehicle-mounted device 3 determines whether the user intention is suitable for adding attributes of the occupants in the vehicle. When it is determined that the user intention is not suitable for adding attributes of the occupants in the vehicle, step S27 is performed. When it is determined that the user intention is suitable for adding attributes of the occupants in the vehicle, step S28 is performed.

The specific implementation of determining whether the user intention is suitable for adding attributes of the occupants in the vehicle is the same as the previous description in step S7, and therefore will not be described again.

Step S27 is the same as step S8 and will not be described again here.

Step S28: When it is determined that the user's intention is suitable for adding the attributes of the occupants in the vehicle 100, the execution module 302 of the vehicle-mounted device 3 sends the attributes of the occupants in the vehicle 100 to the occupants through the communication device 35. Described cloud server 4.

Step S29: The response module 402 of the cloud server 4 updates the user intention based on the attributes of the occupants in the vehicle 100, and generates updated recommendation information based on the updated user intention. The response module 402 of the cloud server 4 also sends the updated recommendation information to the vehicle-mounted device 3 through the communication device 43 .

In one embodiment, the response module 402 obtains the updated user intent by adding attributes of the occupants in the vehicle 100 to the user intent.

In other embodiments, the response module 402 may also only add the composition relationships of all occupants of the vehicle 100 to the user intention to obtain the updated user intention. For example, would The composition relationships of all the occupants of the vehicle 100 , such as the parent-child relationship, are added to the user intention to obtain the updated user intention: "search", "nearby restaurant", and "parent-child".

Step S30: The acquisition module 301 of the vehicle-mounted device 3 receives the updated recommendation information through the communication device 35. The execution module 302 of the vehicle-mounted device 3 can also display the updated recommendation information on the display screen 36 . The execution module 302 of the vehicle-mounted device 3 can also use the speaker 37 to broadcast the updated recommendation information.

It should be noted that the difference between the smart recommendation methods illustrated in Figure 4 and Figure 5 is that the smart recommendation method illustrated in Figure 5 is determined by the server 4 based on the user's intention to add attributes of the occupants in the vehicle. The attributes of the occupants in the vehicle 100 are updated with the user intention and recommendations are made based on the updated user intention. Since the computing power of the server 4 is stronger than that of the vehicle-mounted device 3, it can respond to the user's needs more quickly.

Figure 6 is a third flow chart of the smart recommendation method provided by the preferred embodiment of the present application.

In this embodiment, the smart recommendation method can be applied in an application environment composed of the vehicle-mounted device 3 and the cloud server 4 . For the vehicle-mounted device 3 and the cloud server 4 that require smart recommendation, the functions provided by this application for smart recommendation can be directly integrated on the vehicle-mounted device 3 and the cloud server 4, or a software development kit (Software Development Kit) can be used. Kit, SDK) runs on the vehicle-mounted device 3 and the cloud server 4.

As shown in Figure 6, the smart recommendation method specifically includes the following steps. According to different needs, the order of the steps in the flow chart can be changed, and some steps can be omitted.

Step S41 is the same as step S22 and will not be described again here.

Step S42 is the same as step S23 and will not be described again here.

Step S43 is the same as step S24 and will not be described again here.

Step S44 is the same as step S25 and will not be described again here.

Step S45: The execution module 302 of the vehicle-mounted device 3 determines whether the user intention is suitable for adding attributes of the occupants in the vehicle. When it is determined that the user intent is not appropriate to add within the vehicle When the passenger attribute is selected, step S46 is executed. When it is determined that the user intention is suitable for adding attributes of the occupants in the vehicle, step S47 is performed.

The implementation method of determining whether the user intention is suitable for adding attributes of the occupants in the vehicle is the same as the description of the aforementioned steps S7 and S26, and therefore will not be described again.

Step S46 is the same as step S27 and will not be described again here.

Step S47: The execution module 302 of the vehicle-mounted device 3 uses the camera module of the detection device 33 to capture images of the occupants in the vehicle 100, and confirms the attributes of the occupants in the vehicle 100 based on the occupant images, and through The communication device 35 sends the attributes of the occupants in the vehicle 100 to the cloud server 4 .

It should be noted that in this step S47, the camera module of the detection device 33 is used to capture images of the occupants in the vehicle 100, and based on the occupant images, it is confirmed that the attributes of the occupants of the vehicle 100 are the same as those in step S2. The recorded method of using the camera module of the detection device 33 to capture images of the occupants in the vehicle 100 and confirming the attributes of the occupants in the vehicle 100 based on the occupant images will not be described again here. Step S48 is the same as step S29 and will not be described again here.

Step S49 is the same as step S30 and will not be described again here.

It should be noted that the difference between the smart recommendation methods illustrated in Figure 4 and Figure 6 is that the smart recommendation method illustrated in Figure 6 only detects the contents of the vehicle 100 when it is determined that the user's intention is suitable for adding attributes of the occupants in the vehicle. Crew attributes. Since the vehicle-mounted device 3 does not need to detect the attributes of the occupants in the vehicle 100 in advance, it does not need to detect the attributes of the occupants when there is no need to add the attributes of the occupants. This saves data processing resources and can respond to user needs more quickly.

Figure 7 is a fourth flow chart of the smart recommendation method provided by the preferred embodiment of the present application.

In this embodiment, the smart recommendation method can be applied in an application environment composed of the vehicle-mounted device 3 and the cloud server 4 . For the vehicle-mounted device 3 and the cloud server 4 that require smart recommendation, the functions provided by this application for smart recommendation can be directly integrated on the vehicle-mounted device 3 and the cloud server 4, or a software development kit (Software Development Kit) can be used. Kit, SDK) runs on the vehicle-mounted device 3 and the cloud server 4.

As shown in Figure 7, the smart recommendation method specifically includes the following steps. According to different needs, the order of the steps in the flow chart can be changed, and some steps can be omitted.

Step S51 is the same as step S1 and will not be described again here.

Step S52 is the same as step S2 and will not be described again here.

Step S53 is the same as step S3 and will not be described again here.

Step S54: The vehicle-mounted device 3 sends the voice information and the attributes of the occupants in the vehicle 100 to the cloud server 4.

Step S55: The receiving module 401 of the cloud server 4 receives the voice information through the communication device 43. The response module 402 of the cloud server 4 analyzes the voice information to obtain the user's intention.

Step S56: The response module 402 of the cloud server 4 determines whether the user intention is suitable for adding attributes of the occupants in the vehicle 100. When it is determined that the user intention is not suitable for adding the attributes of the occupants in the vehicle 100, step S57 is executed; when it is determined that the user intention is suitable for adding the attributes of the occupants in the vehicle 100, step S58 is executed.

In one embodiment, the cloud server 4 can store various user intentions suitable for adding attributes of occupants in the vehicle in advance; if the user intention obtained by the current analysis belongs to the various pre-stored intentions suitable for adding attributes of occupants in the vehicle If there is any one of the user intentions, it is determined that the user intention obtained by the current analysis is suitable for adding the attributes of the occupants in the vehicle; if the user intention obtained by the current analysis does not belong to the various pre-stored user intentions that are suitable for adding the attributes of the occupants in the vehicle. In any case, it is determined that the user intention obtained by the current analysis is not suitable for adding attributes of the occupants in the vehicle.

In this embodiment, the various user intentions suitable for adding attributes of the occupants in the vehicle may refer to user intentions including keywords related to people such as "restaurant", "hotel", or "attractions".

Step S57: When it is determined that the user intention is not suitable for adding attributes of the occupants in the vehicle 100, the response module 402 of the cloud server 4 generates recommendation information based on the user intention. After step S57 is executed, step S59 is executed.

Step S58: When it is determined that the user intention is suitable for adding the attributes of the occupants in the vehicle 100, the response module 402 of the cloud server 4 obtains the result by adding the attributes of the occupants in the vehicle 100 to the user intention. Updated user intent, and generate recommended information based on the updated user intent.

Step S59: The response module 402 of the cloud server 4 sends the recommendation information to the vehicle-mounted device 3 through the communication device 43.

Step S60: The acquisition module 301 of the vehicle-mounted device 3 receives the recommended information through the communication device 35. The execution module 302 of the vehicle-mounted device 3 can also display the recommended information on the display screen 36 . The execution module 302 of the vehicle-mounted device 3 can also use the speaker 37 to broadcast the recommended information.

This invention meets the requirements for an invention patent, and a patent application must be filed in accordance with the law. However, the above are only the preferred embodiments of the present invention. Any equivalent modifications or changes made by those familiar with the art of this invention based on the creative spirit of this invention should be included in the scope of the following patent applications.

S1~S11: steps

Claims (6)

A smart recommendation method applied to a vehicle-mounted device. The vehicle-mounted device includes a microphone and a camera module. The method includes: using the camera module to capture images of occupants in the vehicle, and confirming the vehicle based on the images of the occupants. The attributes of the occupants in the vehicle; using the microphone to collect the voice information of the occupants in the vehicle; using the vehicle-mounted device to send the voice information and the attributes of the occupants in the vehicle to the cloud server; using the cloud server Obtain user intent based on the voice information; when the cloud server determines that the user intent is suitable for adding occupant attributes in the vehicle, obtain updated user intent based on the occupant attributes in the vehicle; and utilize the The cloud server obtains recommendation information based on the updated user intention, and sends the recommendation information to the vehicle-mounted device. The smart recommendation method according to claim 1, wherein the occupant attributes of the vehicle include: the total number of occupants in the vehicle, the identity of the driver, the age of the driver, and the gender of the driver. The smart recommendation method as described in claim 2, wherein the attributes of the occupants in the vehicle also include: the identity information of each occupant, the age and gender of each occupant, and the composition relationship of all occupants in the vehicle. The smart recommendation method according to claim 1, wherein the method further includes: determining whether the vehicle satisfies preset conditions, wherein the preset conditions include closing the door of the vehicle and/or starting the vehicle. The engine; wherein the using the camera module to capture images of the occupants in the vehicle, and confirming the attributes of the occupants in the vehicle based on the images of the occupants also includes: after confirming that the vehicle meets the preset conditions, using The camera module captures images of the occupants in the vehicle and confirms attributes of the occupants in the vehicle based on the images of the occupants. The smart recommendation method as described in request item 1, wherein the method also includes: Display the recommended information on the display screen of the vehicle-mounted device; or use a speaker of the vehicle-mounted device to broadcast the recommended information. A smart recommendation system for implementing the smart recommendation method described in any one of claims 1-5, the system includes the vehicle-mounted device and the cloud server that are communicatively connected to each other.

Images