Vehicle control method and apparatus based on behavior prediction

Abstract

This invention relates to a vehicle control method and apparatus based on behavior prediction. The method includes: acquiring current driving operation data of a target vehicle driven by a user; inputting the current driving operation data into a pre-trained driving operation model to obtain behavior prediction information, wherein the behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period; generating control commands based on the target device corresponding to the behavior prediction information and the state of the target device, wherein the target device is a device in the target vehicle; and sending the control commands to the target device to cause the target device to operate according to the target device state indicated by the control commands. This application overcomes the technical problem that existing methods often require additional sensors to be installed in the vehicle for prediction and do not incorporate user driving behavior, thus failing to accurately predict user needs for vehicle use.

Description

Technical Field

[0001] This invention relates to the field of vehicle intelligent control technology, and specifically to a vehicle control method and device based on behavior prediction. Background Technology

[0002] Currently, intelligent mobility services have become a crucial factor in people's choice of transportation. For mobility service providers, understanding users' travel needs and habits is essential for improving service quality and reducing resource waste. Therefore, methods and systems for predicting user behavior based on vehicle-based embedded data have received widespread attention.

[0003] Currently, many studies have used machine learning algorithms and models to model and predict users' travel behavior. For example, by using information such as GPS signals and sensor data, and through methods such as classification, clustering and regression analysis, they can predict users' travel needs and behaviors.

[0004] The inventors discovered that existing methods often require additional sensors to be installed in the vehicle to make predictions, and they do not take into account the user's driving behavior, thus failing to accurately predict the user's needs for vehicle use.

[0005] It is evident that the relevant technologies suffer from the technical problems described above. Summary of the Invention

[0006] One objective of this invention is to provide a vehicle control method and apparatus based on behavior prediction, in order to solve the technical problem that existing technologies often require the installation of additional sensors in the vehicle to make predictions, and do not combine user driving behavior with the vehicle, thus failing to accurately predict user needs for vehicle use; a second objective is to provide a vehicle control apparatus based on behavior prediction; a third objective is to provide an electronic device; and a fourth objective is to provide a storage medium.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0008] A vehicle control method based on behavior prediction includes:

[0009] Obtain the current driving operation data of the target vehicle being driven by the user;

[0010] The current driving operation data is input into a pre-trained driving operation model to obtain behavior prediction information, wherein the behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period.

[0011] Control commands are generated based on the target device and the state of the target device corresponding to the behavior prediction information, wherein the target device is a device in the target vehicle;

[0012] The control command is sent to the target device so that the target device operates according to the target device state indicated by the control command.

[0013] The method in this embodiment determines individual driving behavior prediction information through the user's current driving operation data, and then generates control commands for controlling the target vehicle. It can predict the prediction information and the user's desired control of the target vehicle without installing additional sensors on the vehicle. By combining the user's driving behavior in the process of determining the control commands, this method can more accurately predict and meet the user's vehicle usage needs. This overcomes the technical problem that existing methods often require the installation of additional sensors on the vehicle for prediction and do not combine the user's driving behavior, thus failing to accurately predict the user's vehicle usage needs.

[0014] Optionally, as in the aforementioned behavior prediction-based vehicle control method, before acquiring the current driving operation data of the target vehicle driven by the user, the method further includes:

[0015] Acquire multiple historical driving operation data and multiple historical controller data, wherein the historical controller data is used to indicate the operating status of each controller in the vehicle;

[0016] The multiple historical driving operation data and multiple historical controller data are preprocessed to obtain preprocessed data;

[0017] The preprocessed data is used to train the model to be trained, thereby obtaining the driving operation model.

[0018] The method in this embodiment uses preprocessed data obtained by preprocessing multiple historical driving operation data and multiple historical controller data to train the model to be trained, thereby obtaining the driving operation model. This enables the driving operation model to make predictions based on driving operation data and controller data. Furthermore, by preprocessing historical data before training, the problem of low model accuracy caused by training with erroneous data can be effectively avoided.

[0019] Optionally, as in the aforementioned vehicle control method based on behavior prediction, acquiring multiple historical driving operation data and multiple historical controller data includes:

[0020] The message queue of the vehicle terminal service platform obtains the message corresponding to each candidate vehicle by acquiring the historical driving operation data and historical controller data uploaded by each candidate vehicle using a preset transmission method. The preset transmission method includes SDK and / or data pass-through.

[0021] The messages corresponding to each candidate vehicle in the message queue are transferred to the target database to obtain the multiple historical driving operation data and multiple historical controller data.

[0022] The method in this embodiment provides an implementation for acquiring a large amount of historical driving operation data and historical controller data, and can effectively ensure the breadth of data sources, thereby facilitating the improvement of model training accuracy.

[0023] Optionally, as in the aforementioned vehicle control method based on behavior prediction, the preprocessing of the multiple historical driving operation data and multiple historical controller data to obtain preprocessed data includes:

[0024] Missing data is removed from the multiple historical driving operation data and multiple historical controller data to obtain the data after removal. The missing data refers to the data in the multiple historical driving operation data and multiple historical controller data that lack time information and / or location information.

[0025] Among all candidate vehicle datasets, a target vehicle dataset is determined that has a data volume greater than a preset lower limit and a time distribution that meets preset requirements. Each candidate vehicle dataset includes all the removed candidate vehicles corresponding to each candidate vehicle data.

[0026] For each target vehicle dataset, outlier removal and fitting are performed on all removed data in each target vehicle dataset to obtain the fitted vehicle data corresponding to each target vehicle dataset.

[0027] The preprocessed data is obtained based on all the fitted vehicle data.

[0028] The method in this embodiment removes missing data and outliers before fitting, which makes the distribution of the preprocessed data more regular and thus more conducive to obtaining an accurate model in later training.

[0029] Optionally, as in the aforementioned vehicle control method based on behavior prediction, the message queue of the on-board terminal service platform obtains a message corresponding to each candidate vehicle by acquiring historical driving operation data and historical controller data uploaded by each candidate vehicle using a preset transmission method, including:

[0030] The message queue of the vehicle terminal service platform obtains the voice tracking data, UI click tracking data, music tracking data and video tracking data of each candidate vehicle.

[0031] The message queue of the vehicle terminal service platform merges the voice tracking data, UI click tracking data, music tracking data, and video tracking data of each candidate vehicle to obtain historical driving operation data and historical controller data corresponding to each candidate vehicle.

[0032] The method in this embodiment acquires voice tracking data, UI click tracking data, music tracking data, and video tracking data, thereby enabling multi-faceted training of the model based on different tracking data to achieve the goal of predicting user needs in multiple ways.

[0033] Optionally, as in the aforementioned vehicle control method based on behavior prediction, the step of inputting the current driving operation data into a pre-trained driving operation model to obtain behavior prediction information includes:

[0034] Each sub-operation information in the current driving operation data is input into the driving operation model, wherein the sub-operation information is the user's operation information on the target vehicle;

[0035] The driving operation model identifies the behavior prediction information corresponding to each of the sub-operation information.

[0036] The method in this embodiment can predict each sub-operation information in the current driving operation information, and then predict the behavior prediction information corresponding to each sub-operation information. This can more comprehensively predict the user's needs and further improve the user experience.

[0037] Optionally, as in the aforementioned behavior prediction-based vehicle control method, the method further includes:

[0038] Obtain the current controller data of the target vehicle being driven by the user;

[0039] The current controller data is input into the driving operation model to obtain controller state prediction information, wherein the controller state prediction information is information that predicts whether the controller will malfunction within a second preset time period.

[0040] If the controller status prediction information indicates that the target controller will fail within the second preset time period, an alarm message is generated.

[0041] The alarm information is sent to the target vehicle so that the target vehicle displays the alarm information.

[0042] The method in this embodiment obtains controller state prediction information by acquiring current controller data, and then determines whether the target controller will fail within a second preset time period based on the controller state prediction information, thereby effectively improving the safety of vehicle driving.

[0043] According to another aspect of this application, a behavior prediction-based vehicle control device is also provided, comprising:

[0044] The acquisition module is used to acquire the current driving operation data of the target vehicle driven by the user;

[0045] The prediction module is used to input the current driving operation data into a pre-trained driving operation model to obtain behavior prediction information, wherein the behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period.

[0046] The generation module is used to generate control commands based on the target device corresponding to the behavior prediction information and the state of the target device, wherein the target device is a device in the target vehicle;

[0047] The sending module is used to send the control command to the target device so that the target device operates according to the target device state indicated by the control command.

[0048] According to another aspect of this application, an electronic device is also provided, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus.

[0049] The memory is used to store computer programs;

[0050] The processor is configured to perform the steps of the method as described in any of the preceding methods by running the computer program stored in the memory.

[0051] According to another aspect of this application, a computer-readable storage medium is also provided, wherein a computer program is stored therein, wherein the computer program is configured to perform the steps of the method described in any of the preceding claims when executed.

[0052] The beneficial effects of this invention are:

[0053] This method determines individual driving behavior prediction information by using the user's current driving operation data, and then generates control commands for controlling the target vehicle. It can predict the information and the user's desired control of the target vehicle without installing additional sensors on the vehicle. By combining the user's driving behavior in the process of determining the control commands, this method can more accurately predict and meet the user's vehicle usage needs. This overcomes the technical problem that existing methods often require the installation of additional sensors on the vehicle for prediction and do not combine the user's driving behavior, thus failing to accurately predict the user's vehicle usage needs. Attached Figure Description

[0054] Figure 1 This is a flowchart illustrating a behavior prediction-based vehicle control method according to an embodiment of this application.

[0055] Figure 2 This is a schematic diagram of data acquisition according to one embodiment of this application;

[0056] Figure 3 This is a schematic diagram of a data preprocessing process according to an embodiment of this application;

[0057] Figure 4 This application provides a service architecture diagram for a behavior prediction-based vehicle control method according to one embodiment of the present application.

[0058] Figure 5 This is a data selection diagram of one embodiment of this application;

[0059] Figure 6 This is a fitting result diagram of one embodiment of this application;

[0060] Figure 7 This is a diagram showing the data preprocessing results of one embodiment of this application;

[0061] Figure 8 This is a block diagram of an integrated docking and berthing device according to one embodiment of this application;

[0062] Figure 9 This is a schematic diagram of an electronic device in one embodiment of this application. Detailed Implementation

[0063] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.

[0064] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0065] According to one aspect of the embodiments of this application, a vehicle control method based on behavior prediction is provided. Optionally, in this embodiment, the above-described vehicle control method based on behavior prediction can be applied to a hardware environment consisting of a terminal and a server. The server is connected to the terminal via a network and can be used to provide services (such as advertising push services, application services, etc.) to the terminal or clients installed on the terminal. A database can be set up on the server or independently of the server to provide data storage services to the server.

[0066] The aforementioned network may include, but is not limited to, at least one of the following: wired network, wireless network. The aforementioned wired network may include, but is not limited to, at least one of the following: wide area network, metropolitan area network, local area network. The aforementioned wireless network may include, but is not limited to, at least one of the following: Wi-Fi (Wireless Fidelity), Bluetooth. The terminal is not limited to PC, mobile phone, tablet computer, etc.

[0067] The behavior prediction-based vehicle control method of this application embodiment can be executed by a server, a terminal, or both. Alternatively, the behavior prediction-based vehicle control method of this application embodiment can be executed by a client installed on the terminal.

[0068] Taking the behavior prediction-based vehicle control method in this embodiment as an example, which is jointly executed by a server and / or a vehicle (i.e., a terminal), Figure 1 A vehicle control method based on behavior prediction provided in this application includes the following steps:

[0069] Step S101: Obtain the current driving operation data of the target vehicle driven by the user;

[0070] The behavior prediction-based vehicle control method in this embodiment can be applied to scenarios where the user's intended vehicle control is predicted during vehicle use. Examples include predicting whether the user will play music, predicting activities the user will undertake at their destination, predicting whether the user will use the air conditioning, and identifying other vehicle usage scenarios. The above-described behavior prediction-based vehicle control method is also applicable to other types of behavior prediction-based vehicle control scenarios, provided there is no contradiction.

[0071] The target vehicle can collect information from user control actions or information recorded by the vehicle's own sensors. The target vehicle then uploads the current driving operation data to the server implementing the method of this embodiment, thereby allowing the server to determine the current driving operation data.

[0072] Current driving operation data may include, but is not limited to: user braking, steering wheel turning, accelerator pressing, current time, current interior temperature, current exterior temperature, etc.

[0073] Step S102: Input the current driving operation data into the pre-trained driving operation model to obtain behavior prediction information. The behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period.

[0074] After obtaining the current driving operation data, the current driving operation data can be input into a pre-trained driving operation model, so that the driving operation model can identify the behavior prediction information corresponding to the current driving operation data.

[0075] Behavioral prediction information can be used to predict the control behaviors that a user may need to perform on a target vehicle within a first preset time period in the future.

[0076] For example, by inputting current driving operation data into the driving operation model, the driving operation model can obtain behavioral prediction information to predict whether the user needs to use the air conditioner within 1 minute (i.e., an optional first preset time period, or 2 minutes, 15 seconds, etc.).

[0077] Step S103: Generate control commands based on the target device corresponding to the behavior prediction information and the status of the target device, wherein the target device is the device in the target vehicle.

[0078] After obtaining the behavior prediction information, the target device to be controlled by the behavior prediction information and the target device state to be adjusted can be determined. Then, control commands can be generated based on the target device and the target device state.

[0079] Optionally, the server can generate control commands based on behavior prediction information and send the control commands to the target vehicle.

[0080] Alternatively, the target vehicle can directly obtain the behavior prediction information generated by the server and generate control instructions based on the behavior prediction information to control the target device to operate according to the target device's state.

[0081] For example, if the target device corresponding to the behavior prediction information is the audio equipment in the target vehicle and the target device is in the state of playing music, then the control command is to instruct the audio equipment to play music.

[0082] Step S104: Send control commands to the target device so that the target device operates according to the target device state indicated by the control commands.

[0083] After generating the control command, the server can send the control command to the target vehicle.

[0084] Therefore, after receiving the control command, the target vehicle can control the target device according to the control command so that the target device operates according to the target device state.

[0085] Furthermore, in step S101, the target user's identity information, such as gender and age, can be identified. This allows for the input of the target user's identity information and current driving operation data into the driving operation model (which has at least been pre-trained using the user's identity information and driving operation data) to obtain behavior prediction information based on the user's identity and current driving operation data. This enables more precise consideration of the needs of different users.

[0086] The method in this embodiment determines individual driving behavior prediction information through the user's current driving operation data, and then generates control commands for controlling the target vehicle. It can predict the prediction information and the user's desired control of the target vehicle without installing additional sensors on the vehicle. By combining the user's driving behavior in the process of determining the control commands, this method can more accurately predict and meet the user's vehicle usage needs. This overcomes the technical problem that existing methods often require the installation of additional sensors on the vehicle for prediction and do not combine the user's driving behavior, thus failing to accurately predict the user's vehicle usage needs.

[0087] As an optional embodiment, the vehicle control method based on behavior prediction described above, before obtaining the current driving operation data of the target vehicle driven by the user in step S101, further includes the following steps:

[0088] Acquire multiple historical driving operation data and multiple historical controller data, wherein the historical controller data is used to indicate the operating status of each controller in the vehicle.

[0089] In other words, the types of data used to train the driving operation model include driving operation data and controller data. Historical driving operation data refers to the historical data of different users operating different vehicles, such as historical brake data and historical steering wheel data. Historical controller data refers to the historical data of each controller of different vehicles, such as historical engine data and historical motor data.

[0090] Multiple historical driving operation data and multiple historical controller data are preprocessed to obtain preprocessed data.

[0091] As an optional embodiment, multiple historical driving operation data and multiple historical controller data are preprocessed to obtain preprocessed data, including the following steps:

[0092] Missing data is removed from multiple historical driving operation data and multiple historical controller data to obtain the data after removal. The missing data refers to data in multiple historical driving operation data and multiple historical controller data that lack time information and / or location information.

[0093] Under normal circumstances, each historical driving operation data and historical controller data needs to include time information and / or location information. If a certain historical driving operation data or a certain historical controller data does not include time information and / or location information, then the corresponding historical driving operation data or a certain historical controller data is identified as missing data and is removed.

[0094] Among all candidate vehicle datasets, a target vehicle dataset is identified that has a data volume greater than a preset lower limit and a time distribution that meets preset requirements after data removal. Each candidate vehicle dataset includes all removed candidate vehicles corresponding to each candidate vehicle dataset.

[0095] After obtaining the data after elimination, the candidate vehicle dataset corresponding to each candidate vehicle can be determined.

[0096] And it is necessary to select the target vehicle dataset that conforms to the historical travel pattern from all candidate vehicle datasets.

[0097] Matplotlib (a Python 2D plotting library that generates publication-quality graphics in various hardcopy formats and cross-platform interactive environments) can be used to display the data, with time as the x-axis and latitude as the y-axis. Typically, candidate vehicle datasets with abundant data and relatively regular data distribution are selected as the target vehicle dataset.

[0098] For each target vehicle dataset, outlier removal and fitting are performed on all removed data in each target vehicle dataset to obtain the fitted vehicle data corresponding to each target vehicle dataset.

[0099] After obtaining the target vehicle dataset, outliers can be removed to obtain cleaned data. Then, the cleaned data can be fitted using the Gaussian function to obtain the fitted vehicle data corresponding to each target vehicle dataset.

[0100] Based on all the fitted vehicle data, the preprocessed data is obtained.

[0101] Once the fitted vehicle data corresponding to each target vehicle dataset is obtained, the preprocessed data including all fitted vehicle data can be determined.

[0102] The method in this embodiment removes missing data and outliers before fitting, which makes the distribution of the preprocessed data more regular and thus more conducive to obtaining an accurate model in later training.

[0103] The driving operation model is obtained by training the model to be trained using the preprocessed data.

[0104] After obtaining the preprocessed data, the preprocessed data can be divided into a training set and a validation set. The training set is used to train the model to be trained, and the trained model is obtained. When the trained model is validated by the validation set and meets the preset accuracy, the trained model is determined as the driving operation model.

[0105] The method in this embodiment uses preprocessed data obtained by preprocessing multiple historical driving operation data and multiple historical controller data to train the model to be trained, thereby obtaining a driving operation model. This allows the driving operation model to make predictions based on driving operation data and controller data. Furthermore, by preprocessing historical data before training, the problem of low model accuracy caused by training with erroneous data can be effectively avoided.

[0106] As an optional embodiment, as described in the aforementioned vehicle control method based on behavior prediction, the step of acquiring multiple historical driving operation data and multiple historical controller data includes the following steps:

[0107] The message queue of the vehicle terminal service platform obtains the message corresponding to each candidate vehicle by acquiring the historical driving operation data and historical controller data uploaded by each candidate vehicle using a preset transmission method. The preset transmission method includes SDK and / or data pass-through.

[0108] like Figure 2 As shown, optionally, the vehicle, as the terminal that generates data, uploads the historical driving operation data generated by thousands of candidate vehicles, as well as the historical controller data of the candidate vehicles themselves, to the message queue service such as MQTT or Kafka of the vehicle terminal service platform (TSP) through technologies such as SDK or data pass-through.

[0109] The messages corresponding to each candidate vehicle in the message queue are transferred to the target database to obtain multiple historical driving operation data and multiple historical controller data.

[0110] Then, the messages in the message queue are consumed, and the messages corresponding to each candidate vehicle in the message queue are transferred to the target database to obtain multiple historical driving operation data and multiple historical controller data.

[0111] By doing so, through long-term accumulation, a large amount of historical driving operation data and historical controller data can be obtained, thus completing the preparation for the first stage of data collection.

[0112] The method in this embodiment provides an implementation for acquiring a large amount of historical driving operation data and historical controller data, and can effectively ensure the breadth of data sources, thereby facilitating the improvement of model training accuracy.

[0113] As an optional embodiment, as described above in the vehicle control method based on behavior prediction, the message queue of the on-board terminal service platform obtains a message corresponding to each candidate vehicle by acquiring historical driving operation data and historical controller data uploaded by each candidate vehicle using a preset transmission method, including:

[0114] The message queue of the vehicle terminal service platform obtains voice tracking data, UI click tracking data, music tracking data, and video tracking data for each candidate vehicle.

[0115] The message queue of the vehicle terminal service platform merges the voice tracking data, UI click tracking data, music tracking data, and video tracking data of each candidate vehicle to obtain the historical driving operation data and historical controller data corresponding to each candidate vehicle.

[0116] In this embodiment, after acquiring the voice tracking data, UI click tracking data, music tracking data, and video tracking data for each candidate vehicle, multiple data sources can be integrated using Python libraries such as NumPy (Numerical Python is an open-source numerical computing extension of Python), Pandas (a tool based on NumPy created to solve data analysis tasks), or big data processing frameworks such as Spark. This step can integrate the historical driving operation data and historical controller data corresponding to each candidate vehicle.

[0117] The method in this embodiment acquires voice tracking data, UI click tracking data, music tracking data, and video tracking data, thereby enabling multi-faceted training of the model based on different tracking data to achieve the goal of predicting user needs in multiple ways.

[0118] As an optional embodiment, as described above in the vehicle control method based on behavior prediction, step S102 inputs the current driving operation data into a pre-trained driving operation model to obtain behavior prediction information, including:

[0119] Input each sub-operation information in the current driving operation data into the driving operation model, where the sub-operation information is the user's operation information on the target vehicle;

[0120] The driving operation model identifies the behavioral prediction information corresponding to each sub-operation information.

[0121] In other words, the driving operation model can identify each sub-operation information and then obtain the prediction information corresponding to each sub-operation.

[0122] For example, based directly on time (i.e., the type of sub-operation information is time), the user's current action and location can be predicted. For instance, if it is currently 9 PM, the server can calculate based on the time and this driving operation model that the user is highly likely to start the car and turn on the air conditioning and music in five minutes, generating corresponding behavioral prediction information and sending this control command to the HBase database.

[0123] The vehicle control module in the target vehicle can quickly and randomly read and write the behavior prediction information through the background service, and then generate corresponding control commands based on the behavior prediction information. For example, it can generate control commands that can directly turn on the air conditioner and music in advance, thereby providing more intelligent services to users. In addition, it can further process the predicted location that the user is about to arrive at, such as generating corresponding recommendation information based on the predicted location, such as entertainment and dining services, and quickly responding to the user through the interactive interface.

[0124] The method in this embodiment can predict each sub-operation information in the current driving operation information, and then predict the behavior prediction information corresponding to each sub-operation information. This can more comprehensively predict the user's needs and further improve the user experience.

[0125] As an optional embodiment, the vehicle control method based on behavior prediction described above further includes the following steps:

[0126] Obtain the current controller data of the target vehicle being driven by the user;

[0127] Input the current controller data into the driving operation model to obtain controller state prediction information, which is the information that predicts whether the controller will fail within a second preset time period.

[0128] An alarm message is generated when the controller status prediction information indicates that the target controller will fail within a second preset time period.

[0129] The alarm information is sent to the target vehicle so that the target vehicle can display the alarm information.

[0130] In this embodiment, the target vehicle itself may obtain the operating parameters of its various controllers to obtain the current controller data.

[0131] As previously known, the driving operation model is trained using historical controller data. Therefore, current controller data can be input into this driving operation model to obtain controller state prediction information, which can then predict whether the controller will malfunction within a second preset time period. For example, information such as current engine noise can be used to determine whether the engine will malfunction within the next minute (i.e., one of the second preset time periods; alternatively, it could be 2 minutes, 15 seconds, etc.).

[0132] When the controller status prediction information indicates that the target controller will fail within a second preset time period, alarm information such as image alarm information or audio alarm information can be generated.

[0133] The alarm information is sent to the corresponding module of the target vehicle. For example, the image alarm information is sent to the display interface and displayed, and the audio alarm information is sent to the audio module and played, so as to make the target vehicle display the alarm information.

[0134] The method in this embodiment obtains controller state prediction information by acquiring current controller data, and then determines whether the target controller will fail within a second preset time period based on the controller state prediction information, thereby effectively improving the safety of vehicle driving.

[0135] The following describes an application example that applies any of the foregoing embodiments:

[0136] 1. Data collection, such as Figure 2 As shown, vehicles, acting as data-generating terminals, upload historical driving operation data and historical controller data from thousands of vehicles to the MQTT or Kafka message queue service of the Vehicle Terminal Service Platform (TSP) via SDKs or data pass-through technologies. The data is then transferred from the TSP platform to the target database. This process allows for the acquisition of a large amount of vehicle location and speed information, user operation data, and other related information. Through long-term accumulation, a vast amount of historical user operation data records can be compiled, thus completing the first phase of data collection preparation.

[0137] 2. Data preprocessing and model training: Collecting large amounts of data does not directly contribute to model training; sometimes, erroneous data can even lead to failure. Therefore, data needs to be cleaned before use. For example... Figure 3 As shown, multiple data sources are integrated using Python libraries such as Pandas and NumPy, or big data processing frameworks such as Spark. This step integrates data such as user operation data and vehicle condition data, while removing missing data related to time and location. Next, a target vehicle dataset that conforms to historical travel patterns needs to be selected. Matplotlib is used for visualization, with time as the x-axis and latitude as the y-axis. Typically, a target vehicle dataset with abundant data and a relatively regular data distribution is chosen, such as... Figure 5 As shown in the top right sub-image; according to Figure 7 The process involves removing outliers, adding missing data, and cleaning the data. Figure 6 As shown, the cleaned data is fitted using the Gaussian function to obtain preprocessed data. Finally, the model is trained based on the preprocessed data to obtain the vehicle's driving operation model.

[0138] 3. Application of the model, such as Figure 4As shown, this user-specific driving operation model is deployed to the Flink streaming engine. Based on time, the model can predict the user's current actions and location. For example, if the driving operation model determines it's 9 PM based on the target vehicle's uploaded driving operation data to the TSP, and calculates that the user is likely to start the car and turn on the air conditioning and music in five minutes, this result is sent to the HBase database. The target vehicle quickly retrieves the behavior prediction information through background services. This behavior prediction information is then combined with the services provided by various backend modules, such as the vehicle control module which controls the air conditioning and audio. Control commands corresponding to the behavior prediction information can be sent to the vehicle control module, directly turning on the air conditioning and music in advance, providing more intelligent services to the user. For instance, predicting the user's upcoming location can quickly provide recommendations such as entertainment and dining services.

[0139] The rich data collected by in-vehicle devices can provide users with more personalized, scenario-based, and intelligent services, while also enabling enterprises to achieve more refined operations and management. Therefore, data-driven business models will become a crucial trend in the future development of the Internet of Vehicles.

[0140] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0141] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods according to the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of the various embodiments of this application.

[0142] According to another aspect of the embodiments of this application, a behavior prediction-based vehicle control device for implementing the above-described behavior prediction-based vehicle control method is also provided. Figure 8 This is a structural block diagram of a behavior prediction-based vehicle control device, as an optional embodiment of this application. Figure 8 As shown, the device may include:

[0143] Module 1 is used to acquire the current driving operation data of the target vehicle driven by the user;

[0144] Prediction module 2 is used to input the current driving operation data into the pre-trained driving operation model to obtain behavior prediction information, wherein the behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period.

[0145] The generation module 3 is used to generate control commands based on the target device corresponding to the behavior prediction information and the status of the target device, wherein the target device is the device in the target vehicle;

[0146] The sending module 4 is used to send control commands to the target device so that the target device operates according to the target device state indicated by the control commands.

[0147] It should be noted that the acquisition module 1 in this embodiment can be used to perform the above step S101, the prediction module 2 in this embodiment can be used to perform the above step S102, the generation module 3 in this embodiment can be used to perform the above step S103, and the sending module 4 in this embodiment can be used to perform the above step S10.

[0148] In addition to the modules described above, the apparatus in this embodiment may also include modules that execute any method as described in any of the foregoing embodiments of the behavior prediction-based vehicle control method.

[0149] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the content disclosed in the above embodiments. It should be noted that the above modules, as part of the device, can operate in ways such as... Figure 1 The method shown can be implemented in either software or hardware within a hardware environment, where the hardware environment includes a network environment.

[0150] According to another aspect of the embodiments of this application, an electronic device for implementing the above-described behavior prediction-based vehicle control method is also provided, the electronic device being a server, a terminal, or a combination thereof.

[0151] According to another embodiment of this application, an electronic device is also provided, comprising: Figure 9As shown, the electronic device may include: a processor 1501, a communication interface 1502, a memory 1503, and a communication bus 1504, wherein the processor 1501, the communication interface 1502, and the memory 1503 communicate with each other through the communication bus 1504.

[0152] Memory 1503 is used to store computer programs;

[0153] When processor 1501 executes the program stored in memory 1503, it performs the following steps:

[0154] Step S101: Obtain the current driving operation data of the target vehicle driven by the user.

[0155] Step S102: Input the current driving operation data into the pre-trained driving operation model to obtain behavior prediction information. The behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period.

[0156] Step S103: Generate control commands based on the target device corresponding to the behavior prediction information and the status of the target device, wherein the target device is the device in the target vehicle.

[0157] Step S104: Send control commands to the target device so that the target device operates according to the target device state indicated by the control commands.

[0158] Optionally, in this embodiment, the communication bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. This communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, only one thick line is used to represent it in the figure, but this does not mean that there is only one bus or one type of bus. The communication interface is used for communication between the aforementioned electronic device and other devices.

[0159] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0160] The processors mentioned above can be general-purpose processors, including but not limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; they can also be DSP (Digital Signal Processor), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0161] This application also provides a computer-readable storage medium, which includes a stored program, wherein the program executes the method steps of the above method embodiments when it runs.

[0162] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, ROMs, RAMs, portable hard drives, magnetic disks, or optical disks.

[0163] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0164] If the integrated units in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in the aforementioned computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause one or more computer devices (which may be personal computers, servers, or network devices, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application.

[0165] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0166] In the several embodiments provided in this application, it should be understood that the disclosed client can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, indirect coupling or communication connection between units or modules, and may be electrical or other forms.

[0167] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the solution provided in this embodiment, depending on actual needs.

[0168] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0169] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A vehicle control method based on behavior prediction, characterized by, include: Acquire the current driving operation data of the target vehicle driven by the user, wherein the current driving operation data includes at least the user's control data of the target vehicle; The current driving operation data is input into a pre-trained driving operation model to obtain behavior prediction information, wherein the behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period. Control commands are generated based on the target device and its status corresponding to the behavior prediction information, wherein the target device is a device in the target vehicle; The control command is sent to the target device so that the target device operates according to the target device state indicated by the control command; Prior to acquiring the current driving operation data of the target vehicle driven by the user, the method further includes: Acquire multiple historical driving operation data and multiple historical controller data, wherein the historical controller data is used to indicate the operating status of each controller in the vehicle; The multiple historical driving operation data and multiple historical controller data are preprocessed to obtain preprocessed data; The preprocessed data is used to train the model to be trained, and the driving operation model is obtained. The preprocessing of the multiple historical driving operation data and multiple historical controller data to obtain preprocessed data includes: Missing data is removed from the multiple historical driving operation data and multiple historical controller data to obtain the data after removal. The missing data refers to the data in the multiple historical driving operation data and multiple historical controller data that lack time information and / or location information. Among all candidate vehicle datasets, a target vehicle dataset is determined that has a data volume greater than a preset lower limit and a time distribution that meets preset requirements. Each candidate vehicle dataset includes all the removed candidate vehicles corresponding to each candidate vehicle data. For each target vehicle dataset, outlier removal and fitting are performed on all removed data in each target vehicle dataset to obtain fitted vehicle data corresponding to each target vehicle dataset. The fitting process is implemented based on the Gaussian function. The preprocessed data is obtained based on all the fitted vehicle data.

2. The vehicle control method based on behavior prediction according to claim 1, characterized by, The acquisition of multiple historical driving operation data and multiple historical controller data includes: The message queue of the vehicle terminal service platform obtains the message corresponding to each candidate vehicle by acquiring the historical driving operation data and historical controller data uploaded by each candidate vehicle using a preset transmission method. The preset transmission method includes SDK and / or data pass-through. The messages corresponding to each candidate vehicle in the message queue are transferred to the target database to obtain the multiple historical driving operation data and multiple historical controller data.

3. The vehicle control method based on behavior prediction according to claim 2, characterized in that, The message queue of the vehicle terminal service platform obtains the message corresponding to each candidate vehicle by acquiring the historical driving operation data and historical controller data uploaded by each candidate vehicle using a preset transmission method, including: The message queue of the vehicle terminal service platform obtains the voice tracking data, UI click tracking data, music tracking data and video tracking data of each candidate vehicle. The message queue of the vehicle terminal service platform merges the voice tracking data, UI click tracking data, music tracking data, and video tracking data of each candidate vehicle to obtain historical driving operation data and historical controller data corresponding to each candidate vehicle.

4. The vehicle control method based on behavior prediction according to claim 1, characterized by, The step of inputting the current driving operation data into a pre-trained driving operation model to obtain behavior prediction information includes: Each sub-operation information in the current driving operation data is input into the driving operation model, wherein the sub-operation information is the user's operation information on the target vehicle; The driving operation model identifies the behavior prediction information corresponding to each of the sub-operation information.

5. The vehicle control method based on behavior prediction according to claim 1, characterized by, The method further includes: Obtain the current controller data of the target vehicle being driven by the user; The current controller data is input into the driving operation model to obtain controller state prediction information, wherein the controller state prediction information is information that predicts whether the controller will malfunction within a second preset time period. If the controller status prediction information indicates that the target controller will fail within the second preset time period, an alarm message is generated. The alarm information is sent to the target vehicle so that the target vehicle displays the alarm information.

6. A vehicle control device based on behavior prediction, characterized by, include: The acquisition module is used to acquire the current driving operation data of the target vehicle driven by the user, wherein the current driving operation data includes at least the user's control data of the target vehicle; The prediction module is used to input the current driving operation data into a pre-trained driving operation model to obtain behavior prediction information, wherein the behavior prediction information is used to predict the control behavior that the user will perform on the target vehicle within a first preset time period. The generation module is used to generate control commands based on the target device corresponding to the behavior prediction information and the state of the target device, wherein the target device is a device in the target vehicle; A sending module is configured to send the control command to the target device so that the target device operates according to the target device state indicated by the control command; Specifically, the behavior prediction-based vehicle control device is used for: Acquire multiple historical driving operation data and multiple historical controller data, wherein the historical controller data is used to indicate the operating status of each controller in the vehicle; The multiple historical driving operation data and multiple historical controller data are preprocessed to obtain preprocessed data; The preprocessed data is used to train the model to be trained, and the driving operation model is obtained. Specifically, the behavior prediction-based vehicle control device is also used for: Missing data is removed from the multiple historical driving operation data and multiple historical controller data to obtain the data after removal. The missing data refers to the data in the multiple historical driving operation data and multiple historical controller data that lack time information and / or location information. Among all candidate vehicle datasets, a target vehicle dataset is determined that has a data volume greater than a preset lower limit and a time distribution that meets preset requirements. Each candidate vehicle dataset includes all the removed candidate vehicles corresponding to each candidate vehicle data. For each target vehicle dataset, outlier removal and fitting are performed on all removed data in each target vehicle dataset to obtain fitted vehicle data corresponding to each target vehicle dataset. The fitting process is implemented based on the Gaussian function. The preprocessed data is obtained based on all the fitted vehicle data.

7. An electronic device comprising a processor, a communication interface, a memory and a communication bus, wherein, The processor, the communication interface, and the memory communicate with each other via the communication bus, characterized in that... The memory is used to store computer programs; The processor is configured to perform the method steps of any one of claims 1 to 5 by running the computer program stored in the memory.

8. A computer readable storage medium, characterized in that, The storage medium stores a computer program, wherein the computer program is configured to execute the steps of the method described in any one of claims 1 to 5 when it is run.

Images