Internet of vehicles data compression method, device and equipment based on satellite communication and medium

Abstract

The invention relates to an Internet of Vehicles data compression method and device based on satellite communication, equipment and a medium, and the method comprises the steps: obtaining the operation scene data of a target vehicle, dividing the operation scene data into multiple groups of original data according to the triggering scene type, and enabling each group of original data to correspond to one or more Internet of Vehicles functions; obtaining a preset multi-modal large model, and constructing a satellite communication Internet of Vehicles application recommendation model according to each group of original data and the multi-modal large model; according to the Internet of Vehicles data obtained by the satellite communication Internet of Vehicles application recommendation model, constructing an Internet of Vehicles data compression model and performing iterative training until the Internet of Vehicles data compression model reaches a preset convergence condition, and obtaining a target compression model; and according to the target compression model, compressing the acquired Internet of Vehicles real-time data of the target vehicle to obtain compressed Internet of Vehicles real-time data. By adopting the method, the satellite communication data volume can be reduced, so that the satellite communication flow cost is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of automobiles, and in particular, to a method, device, equipment and medium for compressing vehicle networking data based on satellite communication. Background Art

[0002] With the development of automotive technology, the application demand for commercial vehicle satellite communication vehicle networking has increased. However, the satellite communication traffic cost is relatively high. It is necessary to reduce costs while meeting communication requirements and improve the user operation experience.

[0003] In the related art, an on-vehicle low-earth orbit communication system includes a low-earth orbit satellite communication constellation, a background command and management station, and several two-vehicle on-vehicle satellite communication working groups; each low-earth orbit satellite communication vehicle is equipped with a low-earth orbit satellite communication terminal system, which covers a host module, a wireless communication module, a power supply module and an antenna module. Among them, the wireless communication module includes a 4G / 5G communication module and a low-earth orbit satellite communication module, and the two can perform two-way data communication, which can solve the problem of data service processing of the data sharing group at the terminal source of the low-earth orbit communication system by the on-vehicle satellite communication terminal system. However, although this system combines with the terrestrial network communication system for integration, due to the relatively large vehicle networking data, the satellite communication traffic cost is usually high. Summary of the Invention

[0004] Based on this, in view of the technical problem of the relatively high communication traffic cost of the on-vehicle low-earth orbit communication system in the above-mentioned related art, it is necessary to provide a method, device, computer equipment, computer-readable storage medium and computer program product for compressing vehicle networking data based on satellite communication.

[0005] In a first aspect, the present application provides a method for compressing vehicle networking data based on satellite communication, the method comprising:

[0006] Obtain the operation scenario data of the target vehicle, and divide the operation scenario data into multiple groups of original data according to the triggered scenario type, and each group of original data corresponds to one or more vehicle networking functions;

[0007] Obtain a preset multi-modal large model, and construct a satellite communication vehicle networking application recommendation model according to each group of original data and the multi-modal large model;

[0008] Construct a vehicle networking data compression model based on the vehicle networking data obtained by the satellite communication vehicle networking application recommendation model and perform iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtain a target compression model;

[0009] Compress the real-time vehicle networking data of the target vehicle obtained according to the target compression model to obtain the compressed real-time vehicle networking data.

[0010] In one embodiment, the method further includes:

[0011] Obtain a preset error function, and calculate the error function value according to the error function, the vehicle networking data, and the vehicle networking data after each iteration;

[0012] When the error function value is less than a preset difference threshold, determine that the vehicle networking data compression model reaches the preset convergence condition, and obtain the target compression model.

[0013] In one embodiment, constructing and iteratively training a vehicle networking data compression model based on the vehicle networking data obtained by the satellite communication vehicle networking application recommendation model includes:

[0014] Classify the vehicle networking data according to the vehicle networking functions to obtain data training sets corresponding to different vehicle networking functions;

[0015] Construct data compression models corresponding to various vehicle networking functions one by one according to the data training sets, and iteratively train each data compression model. Each data compression model is based on a convolutional autoencoder architecture.

[0016] In one embodiment, after obtaining the compressed real-time vehicle networking data, the method further includes:

[0017] Group the compressed real-time vehicle networking data according to the vehicle networking functions;

[0018] Store and send the compressed real-time vehicle networking data by group.

[0019] In one embodiment, the operation scenario data includes vehicle status information, driver status information, and environmental status information. The vehicle status information includes vehicle speed, fuel quantity, and tire pressure. The driver status information includes fatigue level and driving habits. The environmental status information includes weather conditions, road congestion level, and surrounding traffic signs.

[0020] In a second aspect, the present application further provides a vehicle networking data compression device based on satellite communication. The device includes:

[0021] An operation scenario acquisition module, configured to acquire operation scenario data of a target vehicle, and divide the operation scenario data into multiple groups of original data according to the trigger scenario type. Each group of original data corresponds to one or more vehicle networking functions;

[0022] A recommendation model construction module, configured to acquire a preset multimodal large model, and construct a satellite communication vehicle networking application recommendation model according to each group of original data and the multimodal large model;

[0023] A compression model generation module, configured to construct a vehicle networking data compression model based on the vehicle networking data obtained by the satellite communication vehicle networking application recommendation model and perform iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtain a target compression model;

[0024] A vehicle networking data compression module, configured to compress the real-time vehicle networking data of a target vehicle obtained according to the target compression model to obtain compressed real-time vehicle networking data.

[0025] Thirdly, the present application further provides a satellite communication-based vehicle networking data compression system, including the satellite communication-based vehicle networking data compression device provided in the second aspect. The system further includes:

[0026] A network status monitoring module, configured to periodically monitor the ground network signal strength, and enable satellite communication connection when the ground network signal strength is monitored to not meet the preset standard for a continuous preset number of times;

[0027] A vehicle networking application recommendation module, configured to, after the satellite communication connection is started, collect the current vehicle status, driver status, and environmental status in real time, and generate one or more vehicle networking function recommendation messages according to the current vehicle status, driver status, and environmental status.

[0028] In one embodiment, the system further includes:

[0029] A compressed data transmission module, configured to compress vehicle networking data and transmit it via satellite, and restore the compressed vehicle networking data at the vehicle networking platform end to obtain restored vehicle networking data.

[0030] Fourthly, the present application further provides a computer device, which includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0031] Obtain the operation scenario data of a target vehicle, and divide the operation scenario data into multiple groups of original data according to the triggered scenario type. Each group of original data corresponds to one or more vehicle networking functions;

[0032] Obtain a preset multi-modal large model, and construct a satellite communication vehicle networking application recommendation model according to each group of original data and the multi-modal large model;

[0033] Construct a vehicle networking data compression model based on the vehicle networking data obtained by the satellite communication vehicle networking application recommendation model and perform iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtain a target compression model;

[0034] Compress the obtained real-time vehicle networking data of the target vehicle according to the target compression model to obtain the compressed real-time vehicle networking data.

[0035] In a fifth aspect, the present application also provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented:

[0036] Obtain the operation scenario data of the target vehicle, and divide the operation scenario data into multiple groups of original data according to the triggered scenario type. Each group of original data corresponds to one or more vehicle networking functions;

[0037] Obtain a preset multimodal large model, and construct a satellite communication vehicle networking application recommendation model according to each group of original data and the multimodal large model;

[0038] Construct a vehicle networking data compression model based on the vehicle networking data obtained from the satellite communication vehicle networking application recommendation model and perform iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtain the target compression model;

[0039] Compress the obtained real-time vehicle networking data of the target vehicle according to the target compression model to obtain the compressed real-time vehicle networking data.

[0040] In a sixth aspect, the present application also provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the following steps are implemented:

[0041] Obtain the operation scenario data of the target vehicle, and divide the operation scenario data into multiple groups of original data according to the triggered scenario type. Each group of original data corresponds to one or more vehicle networking functions;

[0042] Obtain a preset multimodal large model, and construct a satellite communication vehicle networking application recommendation model according to each group of original data and the multimodal large model;

[0043] Construct a vehicle networking data compression model based on the vehicle networking data obtained from the satellite communication vehicle networking application recommendation model and perform iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtain the target compression model;

[0044] Compress the obtained real-time vehicle networking data of the target vehicle according to the target compression model to obtain the compressed real-time vehicle networking data.

[0045] The above vehicle networking data compression method based on satellite communication first obtains the operation scenario data of the target vehicle, and divides the operation scenario data into multiple groups of original data according to the triggered scenario types, where each group of original data corresponds to one or more vehicle networking functions; then obtains a preset multi-modal large model, and constructs a satellite communication vehicle networking application recommendation model according to each group of original data and the multi-modal large model; then constructs a vehicle networking data compression model based on the vehicle networking data obtained by the satellite communication vehicle networking application recommendation model and performs iterative training until the vehicle networking data compression model reaches a preset convergence condition and obtains a target compression model; finally, compresses the obtained real-time vehicle networking data of the target vehicle according to the target compression model to obtain compressed real-time vehicle networking data. By adopting the above method, the present application can adaptively trigger vehicle networking functions according to external conditions, has a certain degree of intelligence, and can also effectively reduce the amount of satellite communication data, thereby reducing satellite communication traffic costs and reducing costs. Description of the Drawings

[0046] Figure 1 It is an application environment diagram of the vehicle networking data compression method based on satellite communication in an embodiment;

[0047] Figure 2 It is a flowchart of the vehicle networking data compression method based on satellite communication in an embodiment;

[0048] Figure 3 It is a structural diagram of the vehicle networking data compression model in an embodiment;

[0049] Figure 4 It is a flowchart of obtaining the target compression model in an embodiment;

[0050] Figure 5 It is a schematic diagram of the vehicle networking data compression model training process in an embodiment;

[0051] Figure 6 It is a flowchart of iteratively training each data compression model in an embodiment;

[0052] Figure 7 It is a flowchart of grouping the real-time vehicle networking data in an embodiment;

[0053] Figure 8 It is a structural block diagram of the vehicle networking data compression device based on satellite communication in an embodiment;

[0054] Figure 9 It is a structural block diagram of the vehicle networking data compression system based on satellite communication in an embodiment;

[0055] Figure 10 It is an internal structural diagram of a computer device in an embodiment;

[0056] Figure 11 It is the internal structure diagram of a computer device in another embodiment. Detailed implementation manners

[0057] In order to make the objectives, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0058] The vehicle networking data compression method based on satellite communication provided by the embodiments of the present application can be applied to an application environment as Figure 1 shown. Among them, the terminal 102 communicates with the server 104 through a network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or can be placed in the cloud or on other network servers. Among them, the terminal 102 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers and Internet of Things devices, and the Internet of Things devices can be intelligent vehicle-mounted devices, etc. The terminal 102 is used to obtain the operation scenario data of the vehicle, including but not limited to vehicle status information, driver status information and environmental status information. The server 104 can be implemented by an independent server or a server cluster composed of multiple servers.

[0059] In one embodiment, as Figure 2 shown, taking the application of this method to the Figure 1 terminal in it as an example for illustration, it can be understood that this method can also be applied to the server, and can also be applied to a system including a terminal and a server, and is realized through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0060] Step 202, obtain the operation scenario data of the target vehicle, and divide the operation scenario data into multiple groups of original data according to the triggered scenario type, and each group of original data corresponds to one or more vehicle networking functions.

[0061] Among them, the target vehicle is the vehicle selected to participate in the vehicle networking data compression system based on satellite communication. The operation scenario data is a series of data related to the vehicle itself, the driving environment and the driver status generated by the target vehicle during the actual operation process. The triggered scenario type is a classification method for the vehicle operation scenario, and is divided according to different characteristics and conditions of the vehicle operation. For example, high-speed driving scenario, urban congestion scenario, parking scenario, etc.

[0062] The original data is a data set obtained by grouping the running scenario data according to the trigger scenario type. Each group of original data represents the data generated by the vehicle running under a specific trigger scenario, and this data will be used to construct a satellite communication vehicle networking application recommendation model subsequently. For example, the original data in the high-speed driving scenario may include data such as the vehicle speed, engine speed, and vehicle distance continuously recorded in this scenario. That is to say, different trigger scenarios, as the original data, take various specific situations that can trigger the vehicle networking function during the vehicle running process as the data source. For example, when the vehicle is driving at a high speed and the fuel level is lower than a certain threshold, the fueling reminder function is triggered. This specific "high-speed driving and low fuel level" scenario is a trigger scenario, and the relevant data (such as vehicle speed, fuel level, etc.) is collected as the original data.

[0063] The vehicle networking function refers to various services and applications provided for vehicles and users based on vehicle networking technology. Common vehicle networking functions include safety functions, convenience functions, and infotainment functions. Safety functions include collision warning, lane departure warning, fatigue driving reminder, etc., aiming to improve driving safety; convenience functions include real-time traffic condition query, intelligent navigation, remote control of the vehicle (unlocking, locking, starting the engine, etc.), automatic parking, etc., aiming to enhance driving convenience; infotainment functions include online music playing, video playing, news and information push, etc., aiming to provide entertainment and information services for the people in the vehicle.

[0064] Specifically, after dividing the running scenario data into multiple groups of original data according to the trigger scenario type, the collected original data needs to be classified according to different trigger scenarios, and then the corresponding vehicle networking function is assigned to each group of data as an identifier. For example, all the scenario data triggered by vehicle faults for the fault diagnosis function are grouped together, and "fault diagnosis" is used as the label for this group of data. Briefly speaking, grouping can be carried out according to the similarity of trigger scenarios, and dimensions such as vehicle state, environmental conditions, and time can be considered. For example, the scenario data related to the vehicle networking function triggered in bad weather is grouped into one group, and the function scenario data triggered during night driving is grouped into another group. For each group of data, determine one or more vehicle networking functions corresponding to the scenario of this group, and use these function names as the label for this group of data. For example, if the scenario corresponding to a group of data is that the vehicle deviates from the lane and triggers two vehicle networking functions, lane departure warning and automatic direction correction, then "lane departure warning, automatic direction correction" is used as the label for this group of data.

[0065] Step 204, obtain a preset multi-modal large model, and construct a satellite communication vehicle networking application recommendation model according to each group of original data and the multi-modal large model.

[0066] Among them, the preset multimodal large model is a pre-trained large artificial intelligence model that can process various different types of data (such as text, images, audio, sensor data, etc.). In the field of vehicle networking, the preset multimodal large model can comprehensively analyze data from multiple sources to achieve more accurate application recommendations. The satellite communication vehicle networking application recommendation model is a model obtained by fine-tuning the preset multimodal large model using the divided groups of original data. It combines the characteristics of satellite communication and is specifically used for application recommendations in vehicle networking scenarios. The main role of the satellite communication vehicle networking application recommendation model is to accurately recommend one or more vehicle networking functions applicable in the input vehicle operation scenario data. For example, when the input scenario data shows that the vehicle is driving at night and the road lighting conditions are poor, the model will recommend turning on functions such as the vehicle's night vision assistance system and adaptive high beams.

[0067] Step 206, based on the vehicle networking data obtained from the satellite communication vehicle networking application recommendation model, construct a vehicle networking data compression model and perform iterative training until the vehicle networking data compression model reaches the preset convergence condition and obtain the target compression model.

[0068] Among them, the target compression model is a vehicle networking data compression model that has been trained and reaches the preset convergence condition, has good compression performance, and can effectively compress and process vehicle networking real-time data. The structure of the vehicle networking data compression model in this embodiment is as Figure 3 shown.

[0069] Specifically, input the operation scenario data of the target vehicle into the satellite communication vehicle networking application recommendation model, and the model will output a list of vehicle networking functions applicable in this scenario.

[0070] For each recommended vehicle networking function, determine the vehicle networking data required for its normal operation. This data can be obtained from various vehicle sensors, in-vehicle systems, and relevant external data sources. For example, if the recommended function is real-time traffic condition query, the required data includes the vehicle's current location, traffic flow information of surrounding roads, etc.; for the fatigue driving reminder function, physiological state data of the driver (such as eye movement, heart rate, etc.) and driving duration data need to be obtained. In this way, according to the recommended vehicle networking functions, obtain the corresponding vehicle networking data for subsequent construction of the vehicle networking data compression model.

[0071] Step 208, according to the target compression model, compress the vehicle networking real-time data of the target vehicle obtained to obtain the compressed vehicle networking real-time data.

[0072] Specifically, after inputting the real-time vehicle networking data into the corresponding target compression model, the compressed real-time vehicle networking data is obtained. In this way, the amount of satellite communication data can be effectively reduced, thereby reducing the satellite communication traffic cost and lowering the cost.

[0073] In the above vehicle networking data compression method based on satellite communication, the vehicle networking function can be adaptively triggered according to external conditions, with a certain degree of intelligence. It can also effectively reduce the amount of satellite communication data, thereby reducing the satellite communication traffic cost and lowering the cost.

[0074] In one embodiment, the operation scenario data includes vehicle status information, driver status information, and environmental status information. The vehicle status information includes vehicle speed, fuel quantity, and tire pressure. The driver status information includes fatigue level and driving habits. The environmental status information includes weather conditions, road congestion level, and surrounding traffic signs.

[0075] Among them, various sensors are installed on the target vehicle, such as vehicle speed sensors, acceleration sensors, cameras, GPS locators, etc., and there is also an in-vehicle system, such as an on-board diagnostic system, which can continuously collect the real-time vehicle networking data of the target vehicle. The vehicle status information mainly includes vehicle speed, engine speed, tire pressure, fuel quantity or battery power, brake status, etc. The driver status information specifically includes the driving duration of the driver, whether fatigued (which can be judged by monitoring eye movement, head posture, etc. through a camera), driving habits (such as the frequency of hard acceleration and hard braking). The environmental status information specifically includes weather conditions (sunny, rainy, foggy, etc.), road congestion conditions, the positions and movements of surrounding vehicles and pedestrians, road types (highway, urban road, rural road, etc.), and surrounding traffic signs (speed limit, direction indication, etc.).

[0076] In one embodiment, as Figure 4 shown, the method further includes:

[0077] Step 402, obtain a preset error function, and calculate the error function value according to the error function, the vehicle networking data, and the vehicle networking data after each iteration.

[0078] Among them, the error function, also known as the loss function or cost function, is a function that measures the degree of difference between the predicted value and the true value of the model. During the training process of the vehicle networking data compression model, the error function is used to evaluate the effect of the model in compressing data after each iteration, that is, the difference between the compressed data and the original vehicle networking data. Common error functions include mean square error (MSE), mean absolute error (MAE), etc.

[0079] According to the above content, the Internet of Vehicles data is a variety of data generated by the vehicle during operation, including vehicle status data (such as vehicle speed, engine speed, tire pressure, etc.), driver status data (such as driving time, fatigue, etc.), and environmental status data (such as weather conditions, road congestion, etc.). These data are collected through various sensors on the vehicle (such as speed sensors, acceleration sensors, cameras, GPS locators, etc.) and vehicle systems (such as on-board diagnostic systems). In the process of training the Internet of Vehicles data compression model, the model will be continuously iterated and updated. In each iteration, the model will compress the original Internet of Vehicles data according to the current parameters, and the compressed data obtained is the Internet of Vehicles data after each iteration. As the number of iterations increases, the parameters of the model will continue to be optimized, and the compressed data will become closer and closer to the ideal compression effect.

[0080] Step 404, when the error function value is less than a preset difference threshold, it is determined that the Internet of Vehicles data compression model reaches a preset convergence condition, and a target compression model is obtained.

[0081] The preset difference threshold is a pre-set value used to determine whether the Internet of Vehicles data compression model has reached the preset convergence condition. When the error function value is less than this threshold, it means that the difference between the compressed data and the original data is small enough, and the performance of the model has reached a relatively ideal state, and the model can be considered to have converged.

[0082] The IoV data compression model is a model used to compress IoV data. Its purpose is to reduce the storage space and transmission bandwidth of data while retaining the important information of the original data as much as possible. Common IoV data compression models include compression models based on autoencoders, which compress the original data into a low-dimensional representation through an encoder, and then reconstruct the low-dimensional representation into an approximate version of the original data through a decoder. The target compression model refers to an IoV data compression model that has reached the preset convergence conditions after training. This model can be used to effectively compress real-time IoV data to improve the efficiency of data transmission and storage.

[0083] Specifically, first obtain a preset error function: Before training the Internet of Vehicles data compression model, it is necessary to select a suitable error function according to the specific application scenario and requirements. You can choose a common error function, such as the mean square error (MSE). In this embodiment, the mean square error function is selected as an example, and the error function is the mean square error L:

[0084] ;

[0085] in, It is the input of the IoV data compression model; It is the output of the Internet of Vehicles data compression model.

[0086] Then, use the vehicle networking data obtained according to the satellite communication vehicle networking application recommendation model to train the vehicle networking data compression model. The training framework is as Figure 5 shown. The input for each batch is multiple data samples. Encode and compress the data in each sample, and then perform reconstruction. Process the error between the output and the input of each iteration training of the network to obtain the error value between the two, and optimize the parameters in the network by means of gradient update. The network parameters are characterized by the weights and biases of each convolution kernel in the convolutional layer. Through N iterations of training, continuously update the network parameters until the loss function converges.

[0087] In this embodiment, by obtaining a preset error function, calculating the error function value, judging whether the preset convergence condition is reached, and finally obtaining the target compression model, the training process of the vehicle networking data compression model is completed, which helps to ensure the compression effect of the vehicle networking data.

[0088] In one embodiment, as Figure 6 shown, construct a vehicle networking data compression model based on the vehicle networking data obtained from the satellite communication vehicle networking application recommendation model and perform iterative training, including:

[0089] Step 602, classify the vehicle networking data according to the vehicle networking functions to obtain data training sets corresponding to different vehicle networking functions.

[0090] Among them, the data training set is obtained by classifying the vehicle networking data according to different vehicle networking functions. Each data training set contains data related to specific vehicle networking functions, and these data will be used to train the corresponding compression model.

[0091] Step 604, construct data compression models corresponding to various vehicle networking functions one by one according to the data training sets, and iteratively train each data compression model. Each data compression model is based on the convolutional autoencoder architecture.

[0092] Among them, the convolutional autoencoder architecture is a deep learning architecture, which consists of an encoder and a decoder. The encoder maps the input data to a low-dimensional feature space to achieve data compression; the decoder reconstructs the low-dimensional features into an approximate version of the original data. The convolutional autoencoder uses convolutional layers to extract the features of the data, which is particularly suitable for processing data with spatial structures, such as images, sequence data, etc., and can also effectively extract the features of the data and perform compression in vehicle networking data compression.

[0093] Specifically, when constructing the vehicle networking data compression model, each vehicle networking function corresponds to a data compression model, and each model is based on the convolutional autoencoder architecture. The model architecture is as Figure 3As shown. The training set of each model is made from all the vehicle networking data required for this vehicle networking function. The model is mainly composed of an autoencoder. The encoder is mainly composed of convolutional layers and activation layers. The input X of size M×T×N is passed through the encoder to obtain the compressed representation F(X) of X. Subsequently, F(X) is used as the input of the decoder to reconstruct the original data X'.

[0094] Exemplarily, first, using the satellite communication vehicle networking application recommendation model, combining the characteristics of satellite communication and the application requirements of vehicle networking, the vehicle networking data suitable for use in the satellite communication environment is filtered out. This data may come from various sensors of the vehicle, in-vehicle systems, and communication interactions with the outside world. For example, in areas with good satellite communication signals, more detailed vehicle status data and real-time traffic information may be obtained; while in areas with weak signals, only key vehicle position and safety-related data may be obtained.

[0095] Secondly, according to the data requirements of different vehicle networking functions, the obtained vehicle networking data is classified. For example, for the vehicle navigation function, the relevant data includes map information, vehicle real-time position, destination information, etc.; for the remote diagnosis function, the data may involve engine fault codes, abnormal data of vehicle sensors, etc.; and the classified data is respectively sorted into data training sets corresponding to different vehicle networking functions. Each data training set should contain enough samples to ensure the effectiveness of subsequent model training.

[0096] Then, a convolutional autoencoder architecture is adopted to construct a data compression model, and for each data training set of vehicle networking functions, a convolutional autoencoder model is constructed respectively. Then, the parameters of each convolutional autoencoder model are initialized, such as the weights and biases of the convolutional kernels, and a suitable loss function is selected to measure the reconstruction error of the model. Commonly used loss functions include mean square error (MSE), that is, the average of the squares of the differences between the predicted value and the true value.

[0097] Then, each model is iteratively trained using the corresponding data training set. In each iteration, the data is input into the model, compressed by the encoder, and then reconstructed by the decoder, and the loss value between the reconstructed data and the original data is calculated. Then, according to the loss value, the parameters of the model are updated using an optimization algorithm (such as stochastic gradient descent) to reduce the loss value. When the loss value reaches the preset threshold or the number of training times reaches the maximum number of iterations, the training is stopped. At this time, each data compression model has achieved a good compression effect and can effectively compress the data of the corresponding vehicle networking function.

[0098] In this embodiment, through the above method, a vehicle networking data compression model corresponding to various vehicle networking functions can be constructed, and it has good compression performance, so as to improve the transmission efficiency and reduce the cost of vehicle networking data in the satellite communication environment.

[0099] In one embodiment, as Figure 7 shown, after obtaining the compressed real-time vehicle networking data, the method further includes:

[0100] Step 702, group the compressed real-time vehicle networking data according to vehicle networking functions.

[0101] Among them, according to the foregoing content, the real-time vehicle networking data has tags, and the compressed real-time vehicle networking data can be grouped according to the tags. Grouping can facilitate the quick retrieval of data.

[0102] Specifically, in the vehicle networking system, data usually has specific identification information, that is, tags, which are used to indicate the functional category to which it belongs. For example, function tags such as "navigation data" and "diagnostic data" will be included in the header or metadata of the data. By parsing this identification information, the data can be accurately assigned to the corresponding functional groups.

[0103] Step 704, store and send the compressed real-time vehicle networking data by group.

[0104] Specifically, according to the characteristics and usage requirements of different vehicle networking function data, select a suitable storage method. For some data that needs to be quickly accessed and processed, such as the real-time environment data of the autonomous driving assistance function, it can be stored locally in the vehicle's local hard disk or solid-state drive. This can reduce data transmission latency and improve the system's response speed. For some data that needs to be stored and shared for a long time, such as the historical fault data of the remote diagnosis function and the historical driving records of the navigation function, they can be stored in the cloud server. Cloud storage has the advantages of large capacity, high reliability, and easy sharing.

[0105] According to the real-time requirements and importance of different vehicle networking function data, formulate a reasonable sending strategy. For some data with high real-time requirements, such as the real-time environment data of the autonomous driving assistance function and the real-time traffic information of the navigation function, a real-time sending method needs to be adopted to ensure that the data can be transmitted to the relevant processing systems in a timely manner. The rapid transmission of data can be achieved through a high-speed wireless communication network, such as a 5G network. For some data with low real-time requirements, such as the regular health reports of the remote diagnosis function and the update data of the infotainment function, a timed sending method can be adopted to send the data to the specified receiving end at a preset time interval. This can reduce the frequency of data transmission and reduce the occupancy of network bandwidth.

[0106] In this embodiment, the compressed real-time vehicle networking data is grouped according to vehicle networking functions, and the compressed real-time vehicle networking data is stored and sent in groups, which can ensure the sufficiency and orderliness when data is extracted.

[0107] The vehicle networking data compression method based on satellite communication of the present application is based on the commercial vehicle satellite communication vehicle networking application recommendation and data compression technology, which can adaptively identify the connection activation of satellite communication, automatically recommend and enable the vehicle networking functions that should be used for users according to the actual situation, and reduce the number of satellite communications by means of data compression, thereby greatly reducing the traffic cost of satellite communication and reducing the user operation volume, and further saving costs.

[0108] It should be understood that although the steps in the flowcharts involved in the above-mentioned embodiments are shown in sequence according to the indications of the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless there is a clear indication in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments may include multiple steps or multiple stages. These steps or stages are not necessarily executed at the same moment, but can be executed at different moments. The execution order of these steps or stages is not necessarily sequential, but can be executed alternately or alternately with at least a part of other steps or steps or stages in other steps.

[0109] Based on the same inventive concept, the embodiment of the present application also provides a vehicle networking data compression device based on satellite communication for implementing the vehicle networking data compression method based on satellite communication involved above. The solution provided by this device to solve the problem is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the vehicle networking data compression device based on satellite communication provided below can refer to the limitations on the vehicle networking data compression method based on satellite communication in the above text, and will not be repeated here.

[0110] In one embodiment, as Figure 8 shown, a vehicle networking data compression device based on satellite communication is provided, including: an operation scenario acquisition module 802, a recommendation model construction module 804, a compression model generation module 806, and a vehicle networking data compression module 808, where:

[0111] The operation scenario acquisition module 802 is configured to acquire the operation scenario data of the target vehicle, and divide the operation scenario data into multiple groups of original data according to the trigger scenario type, and each group of original data corresponds to one or more vehicle networking functions.

[0112] The recommended model construction module 804 is configured to obtain a preset multimodal large model, and construct a satellite communication vehicle networking application recommendation model based on each group of original data and the multimodal large model.

[0113] The compression model generation module 806 is configured to construct a vehicle networking data compression model based on the vehicle networking data obtained by the satellite communication vehicle networking application recommendation model, and perform iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtain a target compression model.

[0114] The vehicle networking data compression module 808 is configured to compress the real-time vehicle networking data of the target vehicle obtained according to the target compression model to obtain compressed real-time vehicle networking data.

[0115] In one embodiment, the compression model generation module 806 is further configured to: obtain a preset error function, and calculate an error function value according to the error function, the vehicle networking data, and the vehicle networking data after each iteration; when the error function value is less than a preset difference threshold, determine that the vehicle networking data compression model reaches a preset convergence condition, and obtain a target compression model.

[0116] In one embodiment, the compression model generation module 806 is further configured to: classify the vehicle networking data according to vehicle networking functions to obtain data training sets corresponding to different vehicle networking functions; construct data compression models corresponding to various vehicle networking functions one by one according to the data training sets, and iteratively train each data compression model, and each data compression model is based on a convolutional autoencoder architecture.

[0117] In one embodiment, the apparatus is further configured to: group the compressed real-time vehicle networking data according to vehicle networking functions; store and send the compressed real-time vehicle networking data by group.

[0118] In one embodiment, the operation scenario acquisition module 802 is further configured to: define that the operation scenario data includes vehicle status information, driver status information, and environmental status information, the vehicle status information includes vehicle speed, fuel quantity, and tire pressure, the driver status information includes fatigue level and driving habits, and the environmental status information includes weather conditions, road congestion level, and surrounding traffic signs.

[0119] Each module in the above vehicle networking data compression apparatus based on satellite communication can be implemented in whole or in part by software, hardware, and their combination. The above modules can be embedded in or independent of the processor in the computer device in the form of hardware, or stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0120] In one embodiment, as Figure 9As shown, a vehicle networking data compression system based on satellite communication is provided, including the above-mentioned vehicle networking data compression device based on satellite communication. The system further includes:

[0121] A network status monitoring module 902, configured to periodically monitor the ground network signal strength, and enable satellite communication connection when the ground network signal strength is detected to not meet the preset standard for a continuous preset number of times;

[0122] A vehicle networking application recommendation module 904, configured to, after the satellite communication connection is started, collect the current vehicle status, driver status, and environmental status in real time, and generate one or more vehicle networking function recommendation messages according to the current vehicle status, driver status, and environmental status.

[0123] In one embodiment, the system further includes:

[0124] A compressed data transmission module 906, configured to compress vehicle networking data and transmit it via satellite, and restore the compressed vehicle networking data at the vehicle networking platform end to obtain the restored vehicle networking data.

[0125] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as Figure 10 shown. The computer device includes a processor, a memory, and a network interface connected via a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store vehicle networking data of target vehicles, operation scenario data, compressed vehicle networking real-time data, etc. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements a vehicle networking data compression method based on satellite communication.

[0126] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as Figure 11As shown in the figure. The computer device includes a processor, a memory, a communication interface, a display screen, and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be implemented through WIFI, a mobile cellular network, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for compressing vehicle networking data based on satellite communication. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covered on the display screen, or a button, a trackball, or a touchpad provided on the housing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0127] Those skilled in the art can understand that Figure 10 and Figure 11 the structure shown in the figure is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0128] In one embodiment, a computer device is provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0129] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0130] In one embodiment, a computer program product is provided, including a computer program. When the computer program is executed by the processor, the steps in the above method embodiments are implemented.

[0131] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data that have been authorized by the user or fully authorized by all parties.

[0132] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile and volatile memories. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, data processing logics based on quantum computing, etc., without limitation.

[0133] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope described in this specification.

[0134] The above-described embodiments merely represent several implementation manners of the present application. Their descriptions are relatively specific and detailed, but they should not be construed as limiting the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the appended claims.

Claims

1. A vehicle networking data compression method based on satellite communication, characterized in that, The method includes: Obtaining the operation scenario data of the target vehicle, and dividing the operation scenario data into multiple groups of original data according to the triggered scenario type, where each group of the original data corresponds to one or more vehicle networking functions; Obtaining a preset multimodal large model, and constructing a satellite communication vehicle networking application recommendation model according to each group of the original data and the multimodal large model; Constructing a vehicle networking data compression model based on the vehicle networking data obtained from the satellite communication vehicle networking application recommendation model and performing iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtaining a target compression model; Compressing the real-time vehicle networking data of the target vehicle obtained according to the target compression model to obtain compressed real-time vehicle networking data.

2. The method according to claim 1, characterized in that The method further includes: Obtaining a preset error function, and calculating an error function value according to the error function, the vehicle networking data, and the vehicle networking data after each iteration; When the error function value is less than a preset difference threshold, determining that the vehicle networking data compression model reaches the preset convergence condition, and obtaining the target compression model.

3. The method according to claim 1, wherein The constructing a vehicle networking data compression model based on the vehicle networking data obtained from the satellite communication vehicle networking application recommendation model and performing iterative training includes: Classifying the vehicle networking data according to vehicle networking functions to obtain data training sets corresponding to different vehicle networking functions; Constructing data compression models corresponding to various vehicle networking functions one by one according to the data training sets, and iteratively training each of the data compression models, where each of the data compression models is based on a convolutional autoencoder architecture.

4. The method according to claim 1, wherein After obtaining the compressed real-time vehicle networking data, the method further includes: Grouping each of the compressed real-time vehicle networking data according to vehicle networking functions; Storing and sending the compressed real-time vehicle networking data in groups.

5. The method according to claim 1, wherein The operation scenario data includes vehicle state information, driver state information, and environmental state information. The vehicle state information includes vehicle speed, fuel quantity, and tire pressure. The driver state information includes fatigue degree and driving habits. The environmental state information includes weather conditions, road congestion degree, and surrounding traffic signs.

6. A vehicle networking data compression device based on satellite communication, characterized in that, The device includes: An operation scenario acquisition module, configured to obtain the operation scenario data of the target vehicle, and divide the operation scenario data into multiple groups of original data according to the triggered scenario type, where each group of the original data corresponds to one or more vehicle networking functions; A recommendation model construction module, configured to obtain a preset multimodal large model, and construct a satellite communication vehicle networking application recommendation model according to each group of the original data and the multimodal large model; A compression model generation module, configured to construct a vehicle networking data compression model based on the vehicle networking data obtained from the satellite communication vehicle networking application recommendation model and perform iterative training until the vehicle networking data compression model reaches a preset convergence condition, and obtain a target compression model; A vehicle networking data compression module, configured to compress the real-time vehicle networking data of the target vehicle obtained according to the target compression model to obtain compressed real-time vehicle networking data.

7. A vehicle networking data compression system based on satellite communication, characterized in that The system includes the vehicle networking data compression device based on satellite communication described in claim 6, and the system further includes: A network status monitoring module, configured to periodically monitor the ground network signal strength, and enable satellite communication connection when the ground network signal strength that is continuously monitored for a preset number of times does not meet the preset standard; A vehicle networking application recommendation module, configured to, after the satellite communication connection is started, collect the current vehicle status, driver status, and environmental status in real time, and generate one or more vehicle networking function recommendation messages according to the current vehicle status, the driver status, and the environmental status.

8. The system according to claim 7, characterized in that, The system further includes: A compressed data transmission module, configured to compress vehicle networking data and transmit it via satellite, and restore the compressed vehicle networking data at the vehicle networking platform side to obtain the restored vehicle networking data.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method described in any one of claims 1 to 5 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, the steps of the method described in any one of claims 1 to 5 are implemented.

Images