CN119058502A - Vehicle control method, device, equipment and storage medium - Google Patents

Abstract

The application relates to a vehicle control method, a device, equipment and a storage medium. The vehicle control method comprises the steps of obtaining vehicle data of a vehicle, carrying out vehicle control operation prediction according to the vehicle data, obtaining object data of a current use object of the vehicle when a prediction result indicates that seat ventilation control operation is to be carried out, wherein the object data comprises an object portrait for reflecting seat ventilation preference, generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, and carrying out seat ventilation control on the vehicle according to the seat ventilation control instruction. By adopting the method, the automatic and intelligent control process for ventilation of the vehicle seat can be realized.

Description

Vehicle control method, device, equipment and storage medium

Technical Field

The present application relates to the field of automotive technologies, and in particular, to a vehicle control method, apparatus, device, and storage medium.

Background

In the riding process, if a passenger wants to ventilate the seat, the seat ventilation function needs to be started. Typically, the driver presses the seat ventilation switch to activate seat ventilation. The seat ventilation function typically has different gear settings, with the appropriate gear being manually selected according to the personal comfort requirements.

However, this manual adjustment of the seat ventilation reduces the ride experience of the relevant occupants in some situations, such as situations where the driver is not able to manually adjust the switch while driving or situations where the driver is unfamiliar with the riding vehicle and is not aware of how the seat ventilation adjustment is to be performed.

Disclosure of Invention

The embodiment of the application provides a vehicle control method, a device, computer equipment and a storage medium, which can realize automatic and intelligent seat ventilation control on a vehicle and improve riding experience of a user.

According to a first aspect of an embodiment of the application, a vehicle control method is provided, and the method comprises the steps of obtaining vehicle data of a vehicle, predicting vehicle control operation according to the vehicle data, obtaining object data of a current use object of the vehicle when a prediction result indicates that seat ventilation control operation is to be executed, wherein the object data comprises an object portrait for reflecting seat ventilation preference, generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, and carrying out seat ventilation control on the vehicle according to the seat ventilation control instruction.

With reference to the first aspect, in a first possible implementation manner of the first aspect, the vehicle data includes at least one of outside temperature data, inside temperature data, weather data of a region where the vehicle is located, voice data of a use object, and status data of the use object, the object data further includes at least one of a first riding record and a second riding record, the first riding record includes an operation record of adjusting an inside air temperature using the object during the present riding and/or a voice record of adjusting an inside air temperature using the object during the present riding, the operation record of adjusting the inside air temperature does not include an operation record of manually adjusting a seat ventilation, and the second riding record includes an operation record of manually adjusting the seat ventilation each time using the object during the history riding.

With reference to the first implementation manner of the first aspect, in a first implementation manner of the first aspect, the predicting the vehicle control operation according to the vehicle data includes inputting the vehicle data into a first neural network of a pre-trained vehicle control model, performing vehicle control intention recognition on the vehicle data by using the first neural network to obtain an intention recognition result, wherein the intention recognition result indicates a control operation to be performed on the vehicle, and determining the intention recognition result as a prediction result.

With reference to the first possible implementation manner of the first aspect, in a third possible implementation manner of the first aspect, the generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result, and the object data includes constructing a vehicle control instruction prompt text according to the vehicle data, the prediction result, and the object data, inputting the vehicle control instruction prompt text into a second neural network of a vehicle control model, wherein the second neural network is not the same neural network as the first neural network, and performing vehicle control instruction prediction on the vehicle control instruction prompt text by using the second neural network to obtain the seat ventilation control instruction for the vehicle.

With reference to the second possible implementation manner or the third possible implementation manner of the first aspect, in a fourth possible implementation manner of the first aspect, after the seat ventilation control is performed on the vehicle according to the seat ventilation control instruction, the method further includes acquiring a third passenger record of the use object, where the third passenger record includes at least one of an operation record of manually adjusting the seat ventilation by the use object during the current passenger and/or a voice record of reflecting the seat ventilation condition by the use object during the current passenger, calculating an instruction prediction error of the seat ventilation control instruction according to the third passenger record, and updating the vehicle control model by using a back propagation algorithm according to the instruction prediction error.

With reference to the first aspect, in a fifth possible implementation manner of the first aspect, after the seat ventilation control is performed on the vehicle according to the seat ventilation control instruction, the method further includes obtaining a seat ventilation adjustment policy corresponding to the usage object, where the seat ventilation adjustment policy is determined according to a seat ventilation adjustment habit of the usage object, and performing dynamic seat ventilation adjustment on the vehicle according to a relationship that a seat ventilation parameter indicated by the seat ventilation adjustment policy changes with a seat ventilation time.

With reference to the first aspect, in a sixth implementation manner of the first aspect, after the vehicle data of the vehicle is obtained, the method further includes, before the prediction result is obtained, determining whether the vehicle data meets a trigger condition of the vehicle control operation prediction, and when the vehicle data is detected to meet the trigger condition, triggering the vehicle control operation prediction according to the vehicle data to obtain the prediction result.

According to a second aspect of the embodiment of the application, a vehicle control device is provided, and the device comprises an acquisition module, a processing module and a processing module, wherein the acquisition module is used for acquiring vehicle data of a vehicle, the processing module is used for predicting vehicle control operation according to the vehicle data, the acquisition module is further used for acquiring object data of a current use object of the vehicle when a prediction result indicates that seat ventilation control operation is to be executed, the object data comprises an object image used for reflecting seat ventilation preference, and the processing module is further used for generating a seat ventilation control instruction aiming at the vehicle according to the vehicle data, the prediction result and the object data and carrying out seat ventilation control on the vehicle according to the seat ventilation control instruction.

In a third aspect of the embodiments of the present application, there is provided an apparatus including a processor and a memory storing a program or instructions that when executed by the processor implement the foregoing vehicle control method.

In a fourth aspect of the embodiments of the present application, there is provided a computer-readable storage medium including a program or instructions that, when executed on a device, cause the device to perform the aforementioned vehicle control method.

In sum, the control operation to be executed on the vehicle is predicted through the vehicle data, and when the seat ventilation operation is to be executed on the vehicle, the vehicle data, the prediction result and the object data are combined to generate the seat ventilation control instruction, so that the automation and the intellectualization of the vehicle seat ventilation control process are realized, the use process of the seat ventilation function is more convenient, and the riding experience of passengers is effectively improved.

Drawings

FIG. 1 is a schematic illustration of a vehicle control process in one embodiment;

FIG. 2 is a flow chart of a method of controlling a vehicle in one embodiment;

FIG. 3 is a flow chart of a method of controlling a vehicle in another embodiment;

FIG. 4 is a schematic structural view of a vehicle control apparatus in one embodiment;

fig. 5 is a schematic diagram of the structure of the apparatus in one embodiment.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application will be described below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the application.

It should be noted that, the illustrations provided in the present embodiment merely illustrate the basic concept of the present invention by way of illustration, and only the components related to the present invention are shown in the drawings and are not drawn according to the number, shape and size of the components in actual implementation, and the form, number and proportion of the components in actual implementation may be arbitrarily changed, and the layout of the components may be more complex. The structures, proportions, sizes, etc. shown in the drawings herein are shown in detail for purposes of illustration only, and are not intended to limit the scope of the invention, which is defined in the claims, any structural modification, proportional change or size adjustment should still fall within the scope of the disclosure without affecting the efficacy and achievement of the present invention. Also, the terms such as "upper," "lower," "left," "right," "middle," and "a" and the like recited in the present specification are merely for descriptive purposes and are not intended to limit the scope of the invention, but are intended to provide relative positional changes or modifications without materially altering the technical context in which the invention may be practiced.

The application provides a vehicle control method which can realize automatic and intelligent control of ventilation of a vehicle seat and improve riding experience of passengers. The vehicle control method can be applied to a server, a vehicle and other equipment. The server may be a single server or a cluster of servers. The vehicle may be an unmanned vehicle or a manned vehicle. In some embodiments, the vehicle control method may also be applied to a vehicle-mounted electronic device or system for performing the vehicle control method.

The vehicle control method comprises the steps of obtaining vehicle data of a vehicle, and carrying out vehicle control operation prediction according to the vehicle data to obtain a prediction result. The prediction result indicates a control operation to be performed on the vehicle. When the prediction result indicates that the seat ventilation control operation is to be performed on the vehicle, object data of a current use object of the vehicle is acquired. And generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, and performing seat ventilation control on the vehicle. The vehicle control scheme remarkably improves the intelligentization and individuation level of the seat ventilation control, can better meet the actual demands of passengers, and improves the use experience of the passengers on the vehicle.

In some embodiments, the vehicle control method may involve one or more of the techniques of FIG. 1. The techniques shown in fig. 1 are described below:

And acquiring corresponding types of data through sensors respectively corresponding to the different types of data to obtain vehicle data.

Data recognition, namely recognizing vehicle data by using a large model to obtain a recognition result as a prediction result of vehicle control operation. The large model may be a large language model. For example, a vehicle control model as referred to herein may be provided. Specifically, the large model may be used to perform vehicle control intention recognition based on vehicle data, and the intention recognition result may be obtained as a prediction result. The intention recognition is carried out by using the large model, so that the requirements of passengers can be more accurately understood, and the method is not limited to simple preset condition control.

And (3) data supplementation, namely acquiring object data for generating a seat ventilation control instruction.

And prompting enhancement, namely constructing a vehicle control instruction prompting text according to the vehicle data, the prediction result and the object data. The application vehicle control instruction prompt text can be used for guiding the large model to generate instructions which are more in line with requirements. In some embodiments, in the stage of constructing the vehicle control instruction hint text, a function call technique may be utilized to construct the vehicle control instruction hint text from vehicle data, prediction results, and object data. Where functionality call techniques can call specific functions or modules to handle complex logic and computations. The vehicle control instruction prompt text language expression generated by the technology is more standard.

And (3) task planning, namely analyzing the prompt text of the vehicle control instruction to obtain the vehicle control instruction. Specifically, the analysis task can be executed through the large model, and a control instruction for the vehicle is obtained. Specifically, the large model can be utilized to analyze the prompt text of the vehicle control instruction, so as to obtain the vehicle control instruction.

The tool uses transmitting vehicle control commands to the cabin controller (Cockpit Domain Controller, CDC) via a specific communication protocol. The cabin domain controller may be an intelligent cabin domain controller. In some embodiments, the target vehicle control interface may also be invoked to transmit seat ventilation control instructions to the cabin controller. The target vehicle control interface is a control interface of the cabin area controller for receiving instructions such as a seat ventilation instruction. According to different actual requirements, the target vehicle control interface may also implement other functions, which are not described herein.

And issuing an instruction, namely controlling the seat ventilation of the vehicle according to the seat ventilation control instruction through the cabin area controller. Specifically, the intelligent cabin domain controller may issue a seat ventilation control instruction to the seat adjustment execution device, so that the seat adjustment execution device performs seat ventilation control by using the seat ventilation instruction. It should be noted that, according to the embodiment of the application, the seat ventilation control instruction can be quickly and accurately generated and adjusted in time by advanced technical architecture and algorithm. By adopting the embodiment of the application, the effect of seat ventilation control is obviously improved, and more comfortable riding feeling is brought to passengers.

The vehicle control method described above is briefly described below in connection with an actual application scenario.

In the practical application scene, the temperature in the vehicle is assumed to be 32 ℃, and a driver says that the vehicle is a bit of stuffy and needs to strengthen ventilation in the vehicle. The control intention of the vehicle is accurately analyzed from the vehicle data (in-vehicle temperature data and voice data of the driver) using the large model, thereby determining that the control operation to be performed on the vehicle is a seat ventilation operation. Then, the large model can combine the data such as the image of the driver to generate accurate vehicle control instructions, wherein the vehicle control instructions are seat ventilation control instructions. And then, the vehicle control instruction is issued to the cabin controller, so that the ventilation intensity of the seat is enhanced.

In some application scenarios, after the vehicle is subjected to seat ventilation control, for example, the driver voice feedback is "too weak in ventilation", the large model can adjust network parameters of the large model through a back propagation algorithm based on the voice data of the feedback, so that optimization of the large model is realized. Here, the feedback information of the driver's use of the seat ventilation may be recorded within 1 hour, and the seat ventilation intensity is adjusted based thereon, and then optimization of the large model is performed based on the habit of using the subject being continuously learned based on the feedback information. The feedback information may be, for example, an operation of manually adjusting the ventilation level of the seat or a voice reflecting the ventilation condition of the seat.

In a complex environment, the temperature in the vehicle fluctuates frequently, and feedback from passengers is various. The large model depends on a powerful neural network architecture and fine tuning optimization, so that riding demands of passengers are accurately understood, proper vehicle control instructions are generated, meanwhile, stable operation of the large model can be guaranteed by means of agent technology, and comfortable ventilation experience is provided for the passengers.

The foregoing vehicle control method is specifically described next with reference to the embodiments of fig. 2 and 3.

Referring to fig. 2, a flow chart of a vehicle control method according to an embodiment of the present application is shown. The method may be performed by a device. The method will be specifically described below taking the apparatus as an example of a vehicle. Specifically, the method comprises the following steps:

s201, acquiring vehicle data of a vehicle.

The vehicle may acquire vehicle data. In particular, the vehicle may continuously collect vehicle data. Or the vehicle may collect vehicle data at regular intervals. Or the vehicle may read the vehicle data that has been acquired during the target period at regular time. For example, the vehicle may read vehicle data that has been collected in approximately 1 minute from the area where the vehicle data is stored every 1 minute.

Wherein the vehicle data includes at least one of outside temperature data and inside temperature data. The off-vehicle temperature data may be, for example, an off-vehicle temperature value. The temperature value outside the vehicle can be detected by means of installed weather software and the like, or can be acquired by a temperature sensor arranged outside the vehicle. The in-vehicle temperature data may be, for example, an in-vehicle temperature value. The temperature value in the vehicle can be acquired by a temperature sensor arranged in the vehicle.

In some embodiments, the vehicle data may further include at least one of weather data of an area in which the vehicle is located, voice data of the usage object, and status data of the usage object. It should be noted that using richer vehicle data for the prediction of vehicle control operations and the generation of seat ventilation control commands facilitates the generation of more accurate prediction results and seat ventilation control commands.

Wherein the weather data includes at least one of seasonal data and weather data. Season data, which may be data describing the current season in spring, summer, autumn, etc. Weather data, which may be data describing the current weather, such as sunny days, rainy days, cloudy days, etc. In some embodiments, the weather data further includes abnormal weather warning information, such as high temperature yellow warning information, high Wen Chengse warning information, high temperature red warning information, heavy rain blue warning information, heavy rain yellow warning information, heavy rain orange warning information, and heavy rain red warning information.

Wherein the object of use refers to a passenger. The object of use may be, for example, a primary driver, a secondary driver or a passenger in the rear seat. It should be noted that, depending on the application scenario, the usage object may be any one or more of all the passengers, for example, all the passengers, or one or more specified by all the passengers, for example, a primary driver and a secondary driver of all the passengers.

The voice data can be collected by a voice collector, such as a microphone, arranged in the vehicle during the riding process. Because the types of the prediction results obtained based on different in-vehicle temperature data and/or out-of-vehicle temperature data may be different, the introduction of voice data plays an important role in the accurate prediction of the vehicle control operation, and the vehicle data containing the voice data has more definite direction to the vehicle control operation, so that the prediction results are more close to the real operation requirements.

Wherein the status data includes at least one of data reflecting whether the usage object is in a commute state and data reflecting whether the usage object is in a specified face state. The specified face state may be, for example, a sweating state or an overheat state. In some embodiments, the data of whether the vehicle is in a commute state may be obtained after analyzing whether the current location of the vehicle is a workplace or whether the destination to which the vehicle is traveling is a workplace. The data specifying the face state may be determined by analyzing the facial features of the portrait image after the portrait image of the user is captured by the camera.

In some embodiments, the vehicle may acquire vehicle data of the vehicle in a case where a use object exists inside the vehicle to avoid an operation of performing seat ventilation control without taking a vehicle, thereby reducing energy consumption of the vehicle. Whether the vehicle has a use object or not can be judged by whether a face image of the use object is detected or not, or can be judged by whether prompt information indicating successful reception of a passenger is received or not and whether prompt information indicating the end of the present journey is received or not.

S202, vehicle control operation prediction is performed according to vehicle data.

The vehicle may generate a prediction result regarding the vehicle control operation based on the vehicle data. Wherein the prediction result indicates a control operation to be performed on the vehicle, the prediction result may be, for example, seat ventilation.

In some embodiments, the vehicle may create a first correspondence between the vehicle data and the vehicle control operation before performing step S202, so that the vehicle control operation to which the vehicle data corresponds may be determined as a prediction result according to the first correspondence in step S202. Wherein the first correspondence relationship refers to a correspondence relationship between vehicle data and vehicle control operations.

In some embodiments, the vehicle may determine whether the vehicle data satisfies the trigger condition predicted by the vehicle control operation, and after detecting that the vehicle data satisfies the trigger condition, step S202 is performed. In view of the fact that in the case where the vehicle frequently collects vehicle data, if the vehicle control operation prediction is performed once every time the vehicle data is acquired, a great computational load is imposed on the vehicle, and therefore, here, the vehicle control operation prediction is performed after the vehicle data is detected to satisfy the trigger condition, it is possible to reduce the waste of computational resources caused by the excessively frequent vehicle control operation prediction.

Specifically, when the vehicle data includes outside-vehicle temperature data, the triggering condition may be detecting that the outside-vehicle temperature data is in the first temperature range. For example, the first temperature range may be a range of greater than or equal to 20 degrees. When the temperature data outside the vehicle is greater than or equal to 20 degrees, it is indicated that the use object may have a need for seat ventilation or the like. When the vehicle data includes in-vehicle temperature data, the trigger condition may be detecting that the in-vehicle temperature data is in a second temperature range. For example, the second temperature range may be a range of greater than or equal to 20 degrees. When the temperature data in the vehicle is greater than or equal to 20 degrees, the use object is indicated that the seat ventilation requirement may exist. It should be noted that the first temperature range may be the same as the second temperature range, or may be different from the second temperature range, and may be set in combination with actual situations. When the vehicle data includes voice data of the usage object, the trigger condition may be that the voice data is detected to reflect that the current in-vehicle temperature is high. The voice data reflecting the higher current temperature in the vehicle may be, for example, "good heat inside the vehicle", "how much too hot the air conditioner is for use". The voice reflects that the current temperature in the vehicle is high, indicating that the use object may have a seat ventilation requirement.

And S203, when the prediction result indicates that the seat ventilation control operation is to be performed on the vehicle, acquiring the object data of the current use object of the vehicle.

The vehicle may acquire locally stored object data of a current use object of the vehicle or query the object data of the current use object of the vehicle from the specified server when the prediction result indicates that the seat ventilation control operation is to be performed on the vehicle.

Wherein the object data includes an object representation. The object representation reflects the seat ventilation preference of the use object. In some embodiments, the object representation may include seat ventilation parameter preference data. The seat ventilation parameter preference data is, for example, a preferred seat ventilation level or a score for each seat ventilation level provided by the vehicle, a higher score indicating a higher preference for that seat ventilation level. It should be noted that, according to the vehicles, the seat ventilation parameters may be other seat ventilation parameters besides the seat ventilation level, which is not limited by the embodiments of the present application.

In some embodiments, the object data further includes at least one of a first ride record and a second ride record. The first riding record comprises an operation record for adjusting the temperature in the car by using the object in the riding process. The operation record of the use object for adjusting the air temperature in the vehicle includes each operation of the use object for adjusting the air temperature in the vehicle, and the operation of adjusting the air temperature in the vehicle may be, for example, an operation of opening a window for ventilation or an operation of opening an air conditioner. Wherein the operation record for adjusting the temperature in the vehicle does not include an operation record for manually adjusting the ventilation of the seat. The first ride record may be used to analyze seat ventilation requirements.

In some embodiments, the first ride record may further include a voice record of the use of the subject to adjust the air temperature within the vehicle. The voice record for adjusting the air temperature in the vehicle may include a voice for indicating that the vehicle is adjusting the air temperature in the vehicle, and the voice for indicating that the vehicle is ventilating by opening windows of the vehicle, a voice for indicating that the vehicle is turning on an air conditioner, a voice for indicating that the vehicle is increasing the ventilation intensity of the seat, and a voice for indicating that the ventilation intensity of the seat of the vehicle is too weak, for example. The second ride record includes an operational record of each manual adjustment of seat ventilation by the subject during the historical ride. The historical riding process may be, for example, one or more riding processes before the current riding process, or may be a specified number of riding processes (such as the last 5 riding processes) of which the usage object is closest to the current riding process, or may be all riding processes of the usage object before the current riding process. In some embodiments, the historical ride process may include the current ride process in addition to the ride process preceding the current ride process.

In some embodiments, the current use object of the vehicle is obtained by obtaining a first object identifier of the use object according to an unlocking result when the riding mode of the use object is unlocking riding, and inquiring an object portrait of the use object according to the first object identifier. The unlocking riding mode comprises, but is not limited to, riding through a code or riding through a virtual key. The first object identifier may be information that uniquely identifies the usage object, such as a name, a mobile phone number, or an identification card number of the usage object. Or the current use object of the vehicle can be obtained by opening a camera arranged in the vehicle, carrying out face recognition on a face image of the use object shot by the camera to obtain a recognition result, carrying a second object identifier of the use object, and inquiring the object portrait of the use object according to the second object identifier. The second object identifier may be information that uniquely identifies the usage object, such as a name, a phone number, or an identification card number of the usage object. It should be noted that the first object identifier and the second object identifier may be the same or different, and may be set in combination with actual situations.

S204, generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, and performing seat ventilation control on the vehicle according to the seat ventilation control instruction.

The vehicle can analyze the vehicle data, the prediction result and the object data to obtain the seat ventilation instruction aiming at the vehicle, the object data is introduced in the process to generate the seat ventilation instruction, so that the seat ventilation instruction is more personalized and more targeted, the vehicle data, the prediction result and the object data are combined, the seat ventilation instruction is more comprehensive and accurate, and the riding experience of passengers can be greatly improved.

The seat ventilation control command is a command for controlling the seat ventilation of the vehicle. The seat ventilation instructions may carry seat ventilation parameters to be adjusted to. The seat ventilation parameter may be a seat ventilation level. The ventilation level of the seat can be 1 level, 2 level, 3 level and the like according to different application scenes, and can also be low level, medium level, high level and the like. Here, the higher the seat ventilation level, the more remarkable the ventilation effect. It should be noted that, according to the vehicle model, other parameters besides the seat ventilation level may be used as the seat ventilation parameter, which is not listed here. In some embodiments, the seat ventilation control instructions may also carry a seat identifier to be adjusted, which may be information that uniquely identifies the seat inside the vehicle, such as the number of the seat or the orientation information of the seat inside the vehicle. The seat mark can realize targeted control of seat ventilation.

In some embodiments, the vehicle may create a second correspondence between the three of the vehicle data, the prediction result, and the object data and the seat ventilation command before executing step S204, so that the seat ventilation control command corresponding to the three of the vehicle data, the prediction result, and the object data may be determined as the seat ventilation command for the vehicle according to the second correspondence in step S204. The second correspondence refers to correspondence between vehicle data, a prediction result, object data and a seat ventilation instruction.

In some embodiments, the vehicle performs seat ventilation control of the vehicle according to the seat ventilation command after generating the seat ventilation control command for the vehicle. Here, the server generates a seat ventilation control command, and then issues the seat ventilation command to the vehicle, and the vehicle performs seat ventilation control based on the seat ventilation command.

In some embodiments, when the number of the usage objects is one, the embodiment of the application can realize the seat ventilation control on the seat where the usage object is located based on the object image of the one usage object, and can also realize the seat ventilation control on the seats where other usage objects are located in the vehicle based on the object image of the one usage object, thereby achieving the purpose of uniformly performing the seat ventilation control. When the number of the objects to be used is plural, the object images of the objects to be used are acquired, and it is understood that the object images of the plurality of objects to be used are acquired, respectively, and the purpose of controlling the seat ventilation of the seat in which the respective objects to be used are located based on the respective object images is achieved.

In the embodiment shown in fig. 2, the vehicle may analyze the vehicle data to determine that the seat ventilation control operation is to be performed on the vehicle, and then perform the seat ventilation control on the vehicle in combination with the data such as the object data of the current use object of the vehicle, so that an automatic process of the seat ventilation control of the vehicle is realized, and the introduction of the object data enables the seat ventilation of the vehicle to be more flexible and targeted, so that the riding experience of the vehicle by the passengers is greatly improved.

Referring to fig. 3, a flow chart of another vehicle control method according to an embodiment of the application is shown. The method may be performed by a device. The method will be specifically described below taking the apparatus as an example of a vehicle. Specifically, the method comprises the following steps:

S301, acquiring vehicle data of a vehicle.

Step S301 may refer to step S201 in the embodiment of fig. 2, and the embodiment of the present application is not described herein.

S302, inputting the vehicle data into a first neural network of a pre-trained vehicle control model, and carrying out vehicle control intention recognition on the vehicle data by utilizing the first neural network to obtain an intention recognition result.

S303, determining the intention recognition result as a prediction result.

In step S301 to step S303, the vehicle inputs vehicle data into the vehicle control model to process, and obtains an intention recognition result as a prediction result. Wherein the intention recognition result indicates a control operation to be performed on the vehicle. Specifically, the vehicle control model includes a first neural network, and the vehicle may input vehicle data into the vehicle control model, and then perform intent recognition on the vehicle data using the first neural network to obtain an intent recognition result, thereby taking the intent recognition result as a prediction result. The method can infer an accurate prediction result according to the vehicle data.

In some embodiments, the first neural network includes an input layer, a hidden layer, and an output layer. The input layer is for receiving vehicle data. And the hidden layer is used for extracting the characteristic vector of the vehicle data. And the output layer is used for processing according to the feature vectors to obtain an intention recognition result. It should be noted that, according to different practical application scenarios, the first neural network may be one or a plurality of first neural networks.

In some embodiments, since the vehicle data includes data of a type such as audio type (e.g., voice data of the object) in addition to text type (e.g., in-vehicle temperature data), the vehicle data may be multimodal through the first neural network to obtain the intention recognition result. For example, a pattern recognition method corresponding to each type of data in the vehicle data can be determined, then a pattern recognition method corresponding to each type of data is adopted to extract a feature vector of the type of data, and then fusion processing is carried out on the feature vectors of each type of data to obtain a fusion feature vector, so that classification processing is carried out on the fusion feature vector to obtain an intention recognition result.

S304, when the prediction result indicates that the seat ventilation control operation is to be performed on the vehicle, acquiring the object data of the current use object of the vehicle.

Step S304 may refer to step S203 in the embodiment of fig. 2, which is not described in detail in the embodiment of the present application.

And S305, generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, and performing seat ventilation control on the vehicle according to the seat ventilation control instruction.

In step S305, the vehicle may input vehicle data into a vehicle control model for processing to obtain a seat ventilation control instruction for the vehicle. In some embodiments, the vehicle control model includes a second neural network, and step S305 may specifically be to construct a vehicle control instruction prompt text according to the vehicle data, the prediction result, and the object data, and then input the vehicle control instruction prompt text into the vehicle control model, so as to predict the vehicle control instruction for the vehicle control instruction prompt text by using the second neural network in the vehicle control model, and obtain the seat ventilation control instruction for the vehicle. The process vehicle control instruction prompts the introduction of text, so that the input data of the second neural network is more professional standard, and further, the seat ventilation control instruction is more accurate.

Wherein the second neural network is not the same neural network as the first neural network. The vehicle control instruction prompt text is text for instructing the vehicle control model to generate the seat control instruction. The vehicle control instruction prompt text may include vehicle data, a prediction result, object data, or include data which is converted from the vehicle data, the prediction result, and the object data and is convenient for the second neural network to understand and process. In some embodiments, the vehicle control instruction hint text may also include character information and command information associated with seat ventilation control. For example, the character information may be a car control expert or a seat ventilation control expert. The command information may be a request to generate a seat ventilation control instruction. The vehicle control instruction prompt text is that supposing that you are an expert responsible for vehicle control, you can use vehicle data to carry out intelligent control on seat ventilation, and the seat ventilation control instruction is generated. It should be noted that the above is merely an example of the vehicle control prompt text, and the vehicle control prompt text may be in other forms.

In some embodiments, the vehicle control model includes a third neural network, and step S305 is specifically to input vehicle data, a prediction result, and object data into the vehicle control model, and then predict vehicle control instructions for the vehicle data, the prediction result, and the object data by using the third neural network, so as to obtain a seat ventilation control instruction for the vehicle. Wherein the first neural network, the second neural network, and the third neural network are not the same neural network. The manner in which the seat ventilation control instructions are generated based on the third neural network is logically simpler in that the seat ventilation control instructions are generated, and the speed at which the seat ventilation control instructions are generated is increased, as compared to the former manner in which the seat ventilation control instructions are generated based on the second neural network.

In some embodiments, the vehicle can also optimize the vehicle control model by acquiring a third passenger record of the used object, and then calculating the instruction prediction error of the seat ventilation control instruction according to the third passenger record, so that the vehicle control model is updated by adopting a back propagation algorithm according to the instruction prediction error, and the vehicle control model is optimized by adopting the instruction prediction error and the back propagation algorithm in the process, so that the prediction accuracy of the vehicle control model is greatly improved. The third passenger record comprises at least one of an operation record for manually adjusting the ventilation of the seat by using the object in the current passenger process and/or a voice record for reflecting the ventilation condition of the seat by using the object in the current passenger process. The manual adjustment of the operational record of the seat ventilation may for example comprise the manually adjusted seat ventilation parameters. For example, after the user manually adjusts the seat ventilation level from level 3 to level 4, the manual adjustment of the seat ventilation operation record includes level 4 of the seat ventilation level. In some embodiments, the manual adjustment of the operational record of seat ventilation may also include manual adjustment of the in-vehicle temperature data after seat ventilation. In some embodiments, the manual adjustment of the operational record of seat ventilation may also include manual adjustment of the in-vehicle temperature data and seat ventilation parameters prior to seat ventilation. Wherein the data in the third passenger record is recorded after the seat ventilation control of the vehicle by the seat ventilation control instruction, for example, the data in the third passenger record is recorded within one hour after the seat ventilation control of the vehicle by the seat ventilation control instruction. The voice recording reflecting the ventilation condition of the seat may be, for example, a voice recording reflecting a lower or higher ventilation intensity of the seat.

In some embodiments, the vehicle may calculate the instruction prediction error of the seat ventilation control instruction according to the third passenger record by acquiring a standard seat ventilation instruction corresponding to the third passenger record by the vehicle, and then determining a deviation between the standard seat ventilation instruction and the generated seat ventilation control instruction as the instruction prediction error of the generated seat ventilation control instruction. For example, the vehicle may acquire a third passenger record after the vehicle adjusts the seat ventilation level of the primary driver to level 3 using the generated seat ventilation control instruction using the subject as the primary driver. The third car record is assumed to include a voice record that the primary driver reflects "seat ventilation strength is too low" and/or an operation record that the primary driver manually adjusts the seat ventilation level from level 3to level 4. The vehicle may acquire a standard seat ventilation control command corresponding to the seat ventilation level 4 according to the third operation record, and then calculate a deviation between the standard seat ventilation control command and the generated seat ventilation command as a command prediction error of the generated seat ventilation control command.

In some embodiments, the third ride record may be used to update the object representation in addition to optimizing the vehicle control model. In some embodiments, the vehicle updating the target image with the third passenger record may specifically be determining a target seat ventilation parameter that is preferred by the user during the current passenger process according to the third passenger record, and updating the target image with the target seat ventilation parameter. In some embodiments, the vehicle updates the target portrait with the target seat ventilation parameters by accumulating the frequency of use of the target seat ventilation parameters or increasing the score of the target seat ventilation parameters, and if the target seat ventilation parameters are the most frequently used of all the seat ventilation parameters or the target seat ventilation parameters are the most scored of all the seat ventilation parameters, updating the original seat ventilation parameters in the seat ventilation preference data to the target seat ventilation parameters. For example, if it is determined from the third passenger record that the seat ventilation level preferred by the user in the current passenger is 4, the frequency of use of the seat ventilation level 4 may be accumulated or the score for the seat ventilation level 4 may be increased. When the seat ventilation level 4 is the seat ventilation level with the highest use frequency in all seat ventilation levels or the seat ventilation level 4 is the seat ventilation level with the highest score in all seat ventilation levels, the original seat ventilation level in the seat ventilation parameter preference data is updated to be the seat ventilation level 4. In some embodiments, the vehicle updates the object representation with the third passenger record, and the updated object representation may be used in a subsequent seat ventilation control instruction generation process.

In some embodiments, after the seat ventilation control command is used to control the seat ventilation of the vehicle, the seat ventilation parameters of the vehicle can be dynamically adjusted according to the seat ventilation adjustment habit of the user. Specifically, the vehicle may determine a seat ventilation adjustment strategy for the subject of use, and dynamically adjust the seat ventilation of the vehicle according to a relationship of seat ventilation parameters as indicated by the seat ventilation adjustment strategy as a function of seat ventilation time. Wherein the seat ventilation adjustment strategy is preset. Or the seat ventilation adjustment strategy is determined according to the seat ventilation adjustment habit of the user. In some embodiments, the seat ventilation adjustment habits of the subject may be generated based on the operational records of the subject manually adjusting the seat ventilation within the target time frame, and/or may be generated based on the operational records of the vehicle automatically adjusting the seat ventilation within the target time frame. The target time range may or may not be the same time range as the aforementioned specified time range.

For example, the vehicle may adjust the seat ventilation level of the seat to 4 levels according to the seat ventilation instruction using the object as the driver, and then the vehicle may acquire the seat ventilation adjustment strategy corresponding to the driver. And the vehicle dynamically adjusts the seat ventilation according to the relation that the seat ventilation parameters indicated by the seat ventilation adjustment strategy change along with the seat ventilation time. It is assumed that the seat ventilation adjustment strategy indicates that seat ventilation level 4 is maintained for half an hour, after which seat ventilation level 3 is maintained at all times. Then the vehicle can switch the seat ventilation level to 3 after maintaining the seat ventilation level for 4 steps for half an hour, and then maintain the seat ventilation level for 3 steps all the time.

In the embodiment shown in fig. 3, the vehicle may introduce a large model to implement prediction of the vehicle control operation and generation of the seat ventilation control command, and since a slight difference in data may also cause the vehicle to perform different vehicle control operations, introducing a large model may implement refined prediction of the vehicle control operation and the seat ventilation control command instead of simple condition judgment according to the configuration, thereby making the seat ventilation control of the vehicle more accurate.

It should be understood that, although the steps in the flowcharts of fig. 1-3 are shown in order as indicated by the arrows, these steps are not necessarily performed in order as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in fig. 1-3 may include multiple sub-steps or multiple stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily occur in sequence, but may be performed alternately or alternately with at least a portion of the other steps or sub-steps or stages of other steps.

Referring to fig. 4, a schematic structural diagram of a vehicle control device according to an embodiment of the present application is provided. The apparatus may be provided in a device. The device comprises an acquisition module 401 and a processing module 402. Specific:

The vehicle seat ventilation control system comprises an acquisition module 401 for acquiring vehicle data of a vehicle, a processing module 402 for predicting vehicle control operation according to the vehicle data, the acquisition module 401 further for acquiring object data of a current use object of the vehicle when a prediction result indicates that seat ventilation control operation is to be performed, wherein the object data comprises an object portrait for reflecting seat ventilation preference, and the processing module 402 further is used for generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data and performing seat ventilation control on the vehicle according to the seat ventilation control instruction.

In an alternative embodiment, the vehicle data comprises at least one of outside temperature data, inside temperature data, weather data of a region where the vehicle is located, voice data of a use object and state data of the use object, the object data further comprises at least one of a first riding record and a second riding record, the first riding record comprises an operation record for adjusting the inside air temperature of the vehicle by using the object in the riding process and/or a voice record for adjusting the inside air temperature of the vehicle by using the object in the riding process, the operation record for adjusting the inside air temperature of the vehicle does not comprise an operation record for manually adjusting the ventilation of a seat, and the second riding record comprises an operation record for manually adjusting the ventilation of the seat by using the object each time in the history riding process.

In an alternative embodiment, the processing module 402 predicts a vehicle control operation based on vehicle data, specifically, inputs the vehicle data into a first neural network of a pre-trained vehicle control model, performs vehicle control intention recognition on the vehicle data by using the first neural network to obtain an intention recognition result, indicates a control operation to be performed on the vehicle, and determines the intention recognition result as a prediction result.

In an alternative embodiment, the processing module 402 generates a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, specifically constructs a vehicle control instruction prompt text according to the vehicle data, the prediction result and the object data, inputs the vehicle control instruction prompt text into a second neural network of the vehicle control model, wherein the second neural network is not the same neural network as the first neural network, and predicts the vehicle control instruction prompt text by using the second neural network to obtain the seat ventilation control instruction for the vehicle.

In an alternative embodiment, the obtaining module 401 is further configured to obtain a third passenger record of the usage object after the vehicle is subjected to the seat ventilation control according to the seat ventilation control command, where the third passenger record includes at least one of an operation record of manually adjusting the seat ventilation by the usage object during the current passenger and/or a voice record of reflecting the seat ventilation condition by the usage object during the current passenger, calculate a command prediction error of the seat ventilation control command according to the third passenger record, and update the vehicle control model by adopting a back propagation algorithm according to the command prediction error.

In an alternative embodiment, the processing module 402 obtains a seat ventilation adjustment strategy corresponding to the use object after performing seat ventilation control on the vehicle according to the seat ventilation control instruction, wherein the seat ventilation adjustment strategy is determined according to the seat ventilation adjustment habit of the use object, and performs dynamic seat ventilation adjustment on the vehicle according to the relationship that the seat ventilation parameters indicated by the seat ventilation adjustment strategy change along with the seat ventilation time.

In an alternative embodiment, the processing module 402 is further configured to determine, after acquiring the vehicle data of the vehicle, whether the vehicle data meets a trigger condition of the vehicle control operation prediction before performing the vehicle control operation prediction according to the vehicle data to obtain a prediction result, and when detecting that the vehicle data meets the trigger condition, trigger the step of performing the vehicle control operation prediction according to the vehicle data to obtain the prediction result.

In the embodiment shown in fig. 4, the vehicle control device may analyze the vehicle data to determine that the seat ventilation control operation is to be performed on the vehicle, and then combine the data such as the object data of the current use object of the vehicle to perform the seat ventilation control on the vehicle, so as to implement an automatic and intelligent process of the seat ventilation control of the vehicle, and the introduction of the object data makes the seat ventilation of the vehicle more flexible and targeted, so that the riding experience of the passengers on the vehicle is greatly improved.

Referring to fig. 5, a schematic structural diagram of an apparatus according to an embodiment of the present application is provided. The device may be a vehicle or a server. The device may include a processor 501 and a memory 502. The processor 501 and the memory 502 described above include, but are not limited to, being connected by way of a bus (not labeled in fig. 5) or the like. The processor 501 is configured to execute a plurality of instructions, and the memory 502 stores a program or instructions that when executed by the processor 501 implement the vehicle control method of the above-described embodiment.

The processor 501 may be an electronic tuning unit (Electronic Control Unit, ECU), a central processing unit (central processing unit, CPU), a general purpose processor, a co-processor, a digital signal processor (digital signalprocessor, DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (fieldprogrammable GATE ARRAY, FPGA) or other programmable logic device, transistor logic device, hardware components, or any combination thereof. The processor 501 may also be a combination that implements computing functionality, such as a combination comprising one or more microprocessors, a combination of 5SP and microprocessors, and the like. In this embodiment, the processor 501 may use a single-chip microcomputer, and various control functions may be implemented by programming the single-chip microcomputer. In the embodiment of the present application, the processor 501 is configured to perform the functions of the acquisition module 401 and also to perform the functions of the processing module 402.

The memory 501 is used to provide a storage space in which a program or instructions, and/or data may be stored. The memory 501 may be one or more of random access memory (random access memory, RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), or portable read-only memory (compact disc read-only memory, CD-ROM), etc.

Specifically, the processor 502 invokes a program or instruction stored in the memory 1203 to perform operations of acquiring vehicle data of a vehicle and predicting a vehicle control operation according to the vehicle data, acquiring object data of a current use object of the vehicle when a prediction result indicates that a seat ventilation control operation is to be performed, wherein the object data includes an object representation for reflecting a seat ventilation preference, generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result, and the object data, and performing seat ventilation control on the vehicle according to the seat ventilation control instruction.

In one possible implementation, the vehicle data comprises at least one of outside temperature data, inside temperature data, weather data of a region where the vehicle is located, voice data of a use object and state data of the use object, the object data further comprises at least one of a first riding record and a second riding record, the first riding record comprises an operation record for adjusting the inside air temperature of the vehicle by using the object in the riding process and/or a voice record for adjusting the inside air temperature of the vehicle by using the object in the riding process, the operation record for adjusting the inside air temperature of the vehicle does not comprise an operation record for manually adjusting the ventilation of a seat, and the second riding record comprises an operation record for manually adjusting the ventilation of the seat by using the object each time in the history riding process.

In one possible implementation, the processor 502 is specifically configured to input vehicle data into a first neural network of a pre-trained vehicle control model, perform vehicle control intent recognition on the vehicle data using the first neural network to obtain an intent recognition result, the intent recognition result indicates a control operation to be performed on the vehicle, and determine the intent recognition result as a prediction result.

In one possible implementation, the processor 502 is specifically configured to construct a vehicle control instruction prompt text according to vehicle data, a prediction result, and object data, input the vehicle control instruction prompt text into a second neural network of a vehicle control model, where the second neural network is not the same neural network as the first neural network, and predict the vehicle control instruction prompt text by using the second neural network to obtain a seat ventilation control instruction for the vehicle.

In a possible implementation manner, the processor 502 is further configured to obtain a third passenger record of the use object, where the third passenger record includes at least one of an operation record of manually adjusting the seat ventilation by the use object during the present passenger and/or a voice record of reflecting the seat ventilation condition by the use object during the present passenger, calculate an instruction prediction error of the seat ventilation control instruction according to the third passenger record, and update the vehicle control model by using a back propagation algorithm according to the instruction prediction error.

In a possible implementation manner, the processor 502 is specifically configured to obtain a seat ventilation adjustment policy corresponding to the usage object according to the seat ventilation control instruction, where the seat ventilation adjustment policy is determined according to a seat ventilation adjustment habit of the usage object, and dynamically adjust the seat ventilation of the vehicle according to a relationship that a seat ventilation parameter indicated by the seat ventilation adjustment policy changes with a seat ventilation time.

In a possible implementation manner, the processor 502 is further configured to determine whether the vehicle data meets a triggering condition of the vehicle control operation prediction, and trigger the step of performing the vehicle control operation prediction according to the vehicle data to obtain a prediction result when the vehicle data is detected to meet the triggering condition.

In addition, the embodiment of the application also provides a computer readable storage medium, which is used for storing a program or instructions, and the program or instructions enable the device to execute the vehicle control method when the program or instructions are run on the device. For technical details not disclosed in the embodiments of the computer readable storage medium, please refer to the description of the method embodiments of the present application. In addition, the description of the beneficial effects of the same method in the embodiments of the present application may also refer to the description of the embodiments of the method of the present application, and no further description is given here.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples illustrate only a few embodiments of the application, which are described in detail and are not to be construed as limiting the scope of the application. It should be noted that it will be apparent to those skilled in the art that several variations and modifications can be made without departing from the spirit of the application, which are all within the scope of the application. Accordingly, the scope of protection of the present application is to be determined by the appended claims.

Claims (10)

1. A vehicle control method, characterized by comprising:

Acquiring vehicle data of a vehicle, and carrying out vehicle control operation prediction according to the vehicle data;

When the prediction result indicates that the seat ventilation control operation is to be executed, acquiring object data of a current use object of the vehicle, wherein the object data comprises an object portrait for reflecting the seat ventilation preference;

And generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, and performing seat ventilation control on the vehicle according to the seat ventilation control instruction.

2. The method of claim 1, wherein the vehicle data includes at least one of outside temperature data, inside temperature data, weather data of a region in which the vehicle is located, voice data of the usage object, and status data of the usage object;

The object data further comprises at least one of a first riding record and a second riding record, wherein the first riding record comprises an operation record for adjusting the temperature in the vehicle by the object in the riding process and/or a voice record for adjusting the temperature in the vehicle by the object in the riding process, the operation record for adjusting the temperature in the vehicle does not comprise an operation record for manually adjusting the ventilation of the seat, and the second riding record comprises an operation record for manually adjusting the ventilation of the seat by the object in the riding process.

3. The method of claim 2, wherein said predicting vehicle control operations based on said vehicle data comprises:

inputting the vehicle data into a first neural network of a pre-trained vehicle control model;

Performing vehicle control intention recognition on the vehicle data by using the first neural network to obtain an intention recognition result, wherein the intention recognition result indicates a control operation to be performed on the vehicle;

and determining the intention recognition result as a prediction result.

4. The method of any one of claim 2, wherein generating a seat ventilation control instruction for the vehicle from the vehicle data, the prediction result, the object data, comprises:

Constructing a vehicle control instruction prompt text according to the vehicle data, the prediction result and the object data;

Inputting the vehicle control instruction prompt text into a second neural network of the vehicle control model, wherein the second neural network is not the same neural network as the first neural network;

And predicting the vehicle control instruction according to the vehicle control instruction prompt text by using the second neural network to obtain a seat ventilation control instruction aiming at the vehicle.

5. The method according to claim 3 or 4, wherein after the vehicle is subjected to the seat ventilation control according to the seat ventilation control instruction, the method further comprises:

The third passenger record comprises at least one of an operation record of manually adjusting the seat ventilation by the use object in the current passenger process and/or a voice record of reflecting the seat ventilation condition by the use object in the current passenger process;

Calculating an instruction prediction error of the seat ventilation control instruction according to the third passenger record;

and updating the vehicle control model by adopting a back propagation algorithm according to the instruction prediction error.

6. The method of claim 1, wherein after seat ventilation control of the vehicle in accordance with the seat ventilation control instruction, the method further comprises:

Acquiring a seat ventilation adjustment strategy corresponding to the use object, wherein the seat ventilation adjustment strategy is determined according to the seat ventilation adjustment habit of the use object;

and dynamically adjusting the seat ventilation of the vehicle according to the relation that the seat ventilation parameters indicated by the seat ventilation adjustment strategy change along with the seat ventilation time.

7. The method of claim 1, wherein after the acquiring the vehicle data of the vehicle, the predicting the vehicle control operation according to the vehicle data, before obtaining the prediction result, further comprises:

judging whether the vehicle data meets triggering conditions predicted by vehicle control operation or not;

and triggering the step of predicting the vehicle control operation according to the vehicle data to obtain a prediction result when the vehicle data is detected to meet the triggering condition.

8. A vehicle control apparatus, characterized in that the apparatus comprises:

the acquisition module is used for acquiring vehicle data of the vehicle;

The processing module is used for carrying out vehicle control operation prediction according to the vehicle data;

the acquisition module is further used for acquiring object data of a current use object of the vehicle when the prediction result indicates that seat ventilation control operation is to be executed, wherein the object data comprises an object portrait for reflecting seat ventilation preference;

the processing module is further used for generating a seat ventilation control instruction for the vehicle according to the vehicle data, the prediction result and the object data, and performing seat ventilation control on the vehicle according to the seat ventilation control instruction.

9. An apparatus comprising a processor and a memory storing a program or instructions which when executed by the processor implement the method of any one of claims 1 to 7.

10. A computer readable storage medium, characterized in that the computer readable storage medium comprises the program or the instructions, which when run on a device, cause the device to perform the method of any of claims 1 to 7.