Map service demand adaptation method and device, electronic equipment and storage medium

Abstract

The application relates to a map service demand adaptation method and device, electronic equipment and a storage medium. The method comprises the following steps: obtaining running data of a plurality of target vehicles and map service request logs of the plurality of target vehicles, and obtaining multi-dimensional characteristics of each target vehicle according to the running data of each target vehicle and the map service request logs of each target vehicle; determining a map data request amount of each target vehicle in a target time period based on the multi-dimensional characteristics of each target vehicle, and obtaining a total map data request amount of a map service server in the target time period based on the map data request amount of each target vehicle; and adjusting a pushing strategy and / or resource allocation parameter of the map service server for pushing map data to each target vehicle according to the total map data request amount. Thus, the problem that a map service server lacks multi-dimensional customer data support and is difficult to predict demand fluctuations, resulting in high peak resource shortage and low valley resource waste in the related art is solved.

Description

Technical Field

[0001] This application relates to the field of vehicle networking technology, and in particular to a method, apparatus, electronic device and storage medium for adapting to map data service requirements. Background Technology

[0002] High-precision maps serve as core foundational data for intelligent driving and in-vehicle navigation, and the demand for them continues to grow with the popularization of intelligent driving technologies. How to rationally plan map data service resources based on vehicle-side data while ensuring service quality has become a crucial technical challenge in the field of connected vehicles.

[0003] In related technologies, macro-demand prediction is performed through map rasterization and geographic feature extraction, or user habit models are trained using vehicle driving data to optimize navigation route recommendations.

[0004] However, the map data service terminals in the above methods lack multi-dimensional customer data support and are difficult to predict demand fluctuations, resulting in resource shortages during peak periods and resource waste during off-peak periods, which urgently needs to be addressed. Summary of the Invention

[0005] This application provides a method, apparatus, electronic device, and storage medium for adapting map data service demands, in order to solve the problems in related technologies where the map data service terminal lacks multi-dimensional customer data support and is unable to predict demand fluctuations, resulting in peak resource shortages and off-peak resource waste. It achieves dynamic adaptation between the resource supply of the map data service terminal and the actual demand of the vehicle terminal, thereby improving resource utilization efficiency and service stability.

[0006] To achieve the above objectives, the first aspect of this application proposes a method for adapting map data service requirements, comprising the following steps: Obtain the operational data and map service request logs of multiple target vehicles, and obtain the multi-dimensional features of each target vehicle based on the operational data and map service request logs of each target vehicle. Based on the multi-dimensional characteristics of each target vehicle, the map data request volume of each target vehicle within the target time period is determined, and based on the map data request volume of each target vehicle, the total map data request volume of the map data server within the target time period is obtained. Based on the total map data request volume, adjust the map data push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle.

[0007] According to one embodiment of this application, obtaining the multi-dimensional features of each target vehicle includes: Extract a first feature from the operational data of each target vehicle, and extract a second feature from the map service request log of each target vehicle; By combining the first feature and the second feature, multi-dimensional features of each target vehicle are generated.

[0008] According to one embodiment of this application, the first feature includes at least one of driving behavior features and map request scenario features, and the second feature is a request frequency feature.

[0009] According to one embodiment of this application, obtaining the total map data request volume of the map data server within the target time period includes: The vehicle weight coefficient for each target vehicle is determined based on its vehicle model location information and / or owner profile information. The total map data request amount is obtained based on the vehicle weight coefficient of each target vehicle.

[0010] According to one embodiment of this application, obtaining the total map data request volume includes: The map data request amount of each target vehicle is multiplied by the corresponding vehicle weight coefficient to obtain the weighted request amount of each target vehicle. The weighted request amounts of each target vehicle are summed to obtain the total map data request amount.

[0011] According to the map data service demand adaptation method proposed in this application, by acquiring the operating data and map service request logs of multiple target vehicles, multi-dimensional characteristics of each target vehicle can be obtained based on the operating data and map service request logs of each target vehicle. Based on the multi-dimensional characteristics of each target vehicle, the map data request volume of each target vehicle within a target time period is determined, and the total map data request volume of the map data server within the target time period is obtained based on the map data request volume of each target vehicle. Based on the total map data request volume, the push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle are adjusted. This solves the problems in related technologies where the map data server lacks multi-dimensional customer data support, making it difficult to predict demand fluctuations, leading to peak resource shortages and off-peak resource waste. It achieves dynamic adaptation between the resource supply of the map data server and the actual demand of the vehicle, improving resource utilization efficiency and service stability.

[0012] To achieve the above objectives, a second aspect of this application provides an adaptation device for map data service requirements, comprising: The acquisition module is used to obtain the operating data and map service request logs of multiple target vehicles, and to obtain the multi-dimensional features of each target vehicle based on the operating data and map service request logs of each target vehicle. The processing module is used to determine the map data request volume of each target vehicle within a target time period based on the multi-dimensional characteristics of each target vehicle, and to obtain the total map data request volume of the map data server within the target time period based on the map data request volume of each target vehicle. The adjustment module is used to adjust the map data push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle based on the total map data request volume.

[0013] According to one embodiment of this application, the module for obtaining the multi-dimensional features of each target vehicle is specifically used for: Extract a first feature from the operational data of each target vehicle, and extract a second feature from the map service request log of each target vehicle; By combining the first feature and the second feature, multi-dimensional features of each target vehicle are generated.

[0014] According to one embodiment of this application, the first feature includes at least one of driving behavior features and map request scenario features, and the second feature is a request frequency feature.

[0015] According to one embodiment of this application, the processing module for obtaining the total map data request volume of the map data server within the target time period includes: The determination unit is used to determine the vehicle weight coefficient of each target vehicle based on the vehicle model positioning information and / or vehicle owner profile information of each target vehicle. The obtaining unit is used to obtain the total map data request amount based on the vehicle weight coefficient of each target vehicle.

[0016] According to one embodiment of this application, the unit for obtaining the total map data request volume is specifically used for: The map data request amount of each target vehicle is multiplied by the corresponding vehicle weight coefficient to obtain the weighted request amount of each target vehicle. The weighted request amounts of each target vehicle are summed to obtain the total map data request amount.

[0017] According to the map data service demand adaptation device proposed in this application embodiment, by acquiring the operating data and map service request logs of multiple target vehicles, multi-dimensional characteristics of each target vehicle can be obtained based on the operating data and map service request logs of each target vehicle. Based on the multi-dimensional characteristics of each target vehicle, the map data request volume of each target vehicle within a target time period is determined, and the total map data request volume of the map data server within the target time period is obtained based on the map data request volume of each target vehicle. Based on the total map data request volume, the push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle are adjusted. This solves the problems in related technologies where the map data server lacks multi-dimensional customer data support, making it difficult to predict demand fluctuations, leading to peak resource shortages and off-peak resource waste. It achieves dynamic adaptation between the resource supply of the map data server and the actual demand of the vehicle, improving resource utilization efficiency and service stability.

[0018] To achieve the above objectives, a third aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the program to implement the method for adapting to map service requirements as described in the above embodiments.

[0019] To achieve the above objectives, a fourth aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the method for adapting map and data service requirements as described in the above embodiments.

[0020] To achieve the above objectives, a fifth aspect of this application provides a computer program product comprising a computer program that, when executed by a processor, is used to implement the adaptation method for map service requirements as described in the above embodiments.

[0021] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0022] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a method for adapting map data service requirements according to an embodiment of this application; Figure 2 This is a block diagram of an adaptation system for map data service requirements according to one embodiment of the present application; Figure 3A flowchart illustrating another method for adapting map data service requirements according to an embodiment of this application; Figure 4 This is a block diagram of an adaptation device for map data service requirements provided according to embodiments of this application; Figure 5 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0023] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0024] The following description, with reference to the accompanying drawings, describes a method, apparatus, electronic device, and storage medium for adapting map data service requirements according to embodiments of this application.

[0025] Figure 1 This is a flowchart of a method for adapting map data service requirements according to an embodiment of this application.

[0026] Before introducing the method for adapting map data service requirements proposed in the embodiments of this application, we will first introduce the map data service requirement adaptation system involved in this method. For example... Figure 2As shown, the adaptation system for this map data service requirement includes an in-vehicle communication module, a cloud platform, and a map data service system (i.e., the map data server). The in-vehicle communication module is installed in each target vehicle and is used to collect key signal data of vehicle components via the CAN (Controller Area Network) bus and collect usage data of in-vehicle applications through the vehicle's infotainment system, reporting this data to the cloud platform. The cloud platform receives the operational data and map service request logs reported by each target vehicle, extracts multi-dimensional features based on the operational data and map service request logs of each target vehicle, predicts the map data request volume of each target vehicle within a target time period based on these multi-dimensional features, sums the map data request volumes of all target vehicles to obtain the total map data request volume, and sends the total map data request volume to the map data service system. The map data service system receives the total map data request volume sent by the cloud platform and adjusts the map data push strategy and / or resource allocation parameters to each target vehicle based on the total map data request volume. The data flow between the components includes: vehicle signal data collected by the in-vehicle communication module is uploaded to the cloud platform; the cloud platform sends a service request to the map data service system to obtain map data; the map data service system returns map data related to road information, road events, emergencies, and driving behavior to the cloud platform; and finally, the cloud platform or the in-vehicle communication module distributes the map data to the vehicle. This achieves a complete closed loop from vehicle-side data collection and cloud-based demand prediction to dynamic resource adaptation by the map data service.

[0027] Next, the adaptation method for map data service requirements proposed in the embodiments of this application will be described in detail.

[0028] For example, such as Figure 1 As shown, the adaptation method for this map data service requirement includes the following steps: In step S101, the operation data of multiple target vehicles and the map service request logs of multiple target vehicles are obtained, and the multi-dimensional features of each target vehicle are obtained based on the operation data of each target vehicle and the map service request logs of each target vehicle.

[0029] It is understood that, in this embodiment, operational data refers to the state parameters and event records generated by various components and systems during vehicle operation, including signal data such as vehicle speed, acceleration, steering, and braking, as well as trigger event information such as the activation and deactivation of intelligent driving functions, used to characterize the vehicle's driving status, driving operations, and function triggering. Map service request logs refer to records of various map data requests initiated by the vehicle to the map data service provider during operation, including information such as request time, requested map data type, and requested data volume, used to characterize the vehicle's map data demand behavior. Multi-dimensional features refer to various types of information extracted from operational data and map service request logs used to characterize the vehicle's map data service needs.

[0030] Specifically, the cloud platform can collect key signal data of vehicle components via the CAN bus through the vehicle's onboard terminal, and collect the opening and closing information of vehicle applications and the continuous running time information of each application through the vehicle's in-vehicle infotainment system, thereby obtaining the operating data of multiple target vehicles; the cloud platform can obtain the map data request records and corresponding request data volume initiated by each target vehicle to the map data service at each historical time point from the in-vehicle navigation application and / or TSP (Telematics Service Provider) cloud system, as a map service request log.

[0031] Based on the operational data and map service request logs of each target vehicle, the cloud platform can obtain multi-dimensional characteristics of each target vehicle. Therefore, the cloud platform can comprehensively characterize the map service demand features of each target vehicle from both operational data and map service request logs, providing a data foundation for predicting individual vehicle needs and allocating overall resources in subsequent steps.

[0032] As one possible implementation method, in some embodiments, the multi-dimensional features of each target vehicle are obtained by: extracting a first feature from the operating data of each target vehicle and extracting a second feature from the map service request log of each target vehicle; and concatenating the first feature and the second feature to generate the multi-dimensional features of each target vehicle.

[0033] It is understood that, in the embodiments of this application, the first feature refers to the feature information extracted from the operating data that is used to characterize the vehicle's driving status and driving behavior pattern; the second feature refers to the feature information extracted from the map service request log that is used to characterize the vehicle's map data request behavior pattern.

[0034] Specifically, the cloud platform can extract a first feature from the operational data of each target vehicle and a second feature from the map service request logs of each target vehicle. By concatenating the first and second features, that is, connecting the feature vectors of the first and second features along the feature dimension, they are merged into a unified feature vector. This unified feature vector contains both behavioral and scene dimension information from the operational data and request frequency dimension information from the map service request logs, so as to comprehensively represent the vehicle's map data service needs.

[0035] Optionally, in some embodiments, the first feature includes at least one of driving behavior features and map request scenario features, and the second feature is a request frequency feature.

[0036] Driving behavior characteristics refer to features reflecting a user's driving style and operating habits, such as speed change patterns and acceleration / deceleration behavior. These characteristics can be obtained through statistical analysis of signals such as vehicle speed, acceleration, steering, and braking in the operational data. Map request scenario characteristics reflect the changing patterns of map data requests by the vehicle under different driving scenarios. For example, the different map data request patterns triggered when the vehicle switches between different road types or when the intelligent driving function is activated / deactivated. Map request scenario characteristics can be obtained by identifying driving area switching events and intelligent driving function triggering events in the operational data and statistically analyzing the changes in map data requests before and after each event. For example, identifying a scenario where a vehicle enters a city road from an elevated road from the operational data, and statistically analyzing the changes in the amount and type of map data requests within a specified time window before and after this scenario switch, can serve as a feature reflecting the changing patterns of map data demand under different driving scenarios. Request frequency characteristics reflect the number of times and the amount of data requested by the vehicle for various types of map data in different time periods. Request frequency characteristics can be obtained by statistically analyzing map service request logs by time and map data type. For example, the number of times and the amount of data requested by each target vehicle for various types of map data in different time periods can be counted as characteristics that can reflect the patterns of vehicle map data request behavior.

[0037] Therefore, the multi-dimensional features obtained by fusing multi-source data and stitching together various types of features can more comprehensively and accurately represent the map data service needs of each target vehicle, providing a more accurate data foundation for predicting the number of vehicle requests on the map data service side, thereby helping to improve the accuracy of demand prediction and the rationality of resource allocation.

[0038] In step S102, based on the multi-dimensional characteristics of each target vehicle, the map data request volume of each target vehicle within the target time period is determined, and based on the map data request volume of each target vehicle, the total map data request volume of the map data server within the target time period is obtained.

[0039] It is understood that, in this embodiment, the target time period refers to a specified future time interval for which map data service demand needs to be predicted, such as a future hour, day, or peak period, used to determine the time range for prediction. Map data request volume refers to the total number of map data requests that vehicles are expected to initiate to the map data service provider within the target time period, used to quantify the size of the vehicle's map data service demand. Total map data request volume refers to the sum of all map data requests received by the map data service provider from all target vehicles within the target time period, used to characterize the overall service demand scale of the map data service provider.

[0040] In other words, the cloud platform can use the multi-dimensional characteristics of each target vehicle as input to estimate the map data request volume of each target vehicle within a target time period. Then, the cloud platform can sum up the map data request volumes of each target vehicle to obtain the total map data request volume of the map data server within the target time period. Thus, the cloud platform can achieve the aggregation of the overall demand scale of the map data server from the micro-level demand prediction of individual vehicles, providing a quantitative basis for subsequent adjustments to the resource allocation and push strategies of the map data server.

[0041] As one possible approach, in some embodiments, obtaining the total map data request volume of the map data server within a target time period includes: determining the vehicle weight coefficient of each target vehicle based on the vehicle model positioning information and / or vehicle owner profile information of each target vehicle; and obtaining the total map data request volume based on the vehicle weight coefficient of each target vehicle.

[0042] Among them, vehicle model positioning information refers to the feature information used to characterize the vehicle's own attribute category, such as vehicle model, price range, and configuration level. Vehicles with different model positioning usually have different map data service needs. Owner profile information refers to the feature information used to characterize the owner's individual attribute category, such as the owner's age and geographical region. Different owner profiles usually have different car usage behaviors and map usage frequencies. The vehicle weight coefficient is a quantitative parameter used to reflect the contribution of different target vehicles to the overall service demand of the map data service. The higher the vehicle model positioning (such as higher-priced or more feature-rich models) or the higher the map data demand tendency corresponding to the owner profile (such as the owner frequently using navigation or intelligent driving functions in daily travel), the higher the weight coefficient of the vehicle. The weight coefficient serves as an adjustment factor for the map data request volume of each target vehicle, used to make the final cumulative result more accurately reflect the overall demand scale of the map data service.

[0043] Specifically, the cloud platform can retrieve vehicle model, configuration level, and price range data for each target vehicle from the vehicle registration information pre-stored in the TSP cloud system, as vehicle model positioning information. Simultaneously, the cloud platform can obtain personal information registered by car owners when purchasing a vehicle or activating in-vehicle infotainment services from the vehicle manufacturer's user management system, including data such as age and geographic region, as owner profile information. Furthermore, the cloud platform can supplement and correct the owner profile information by analyzing behavioral data such as historical mileage, common route types, and navigation usage frequency in vehicle operation data, making the profile information more closely reflect actual driving conditions. It should be noted that the acquisition of the above-mentioned owner information is conducted with the user's authorization and consent, and complies with relevant data security and privacy protection regulations.

[0044] Then, the cloud platform can comprehensively evaluate the contribution of each target vehicle to the map data service demand based on vehicle location information and / or owner profile information, and assign a corresponding vehicle weight coefficient to each target vehicle, so that different vehicles account for different proportions in the total accumulated requests. That is, when using both vehicle location information and owner profile information simultaneously, the cloud platform can quantify and score the vehicle location information and owner profile information separately. In terms of vehicle location information, the vehicle location score is determined based on factors such as vehicle model, price range, and configuration level. The higher the price range or the more comprehensive the configuration level of the vehicle, the more complete its intelligent driving functions, and the greater its demand for high-precision map data. For example, vehicles equipped with advanced driver assistance systems have a higher map data demand than vehicles without such systems. In terms of owner profile information, the owner profile score is determined based on tags such as owner age and geographic region. The more frequently an owner uses navigation and intelligent driving functions in their daily travel, the greater their map data demand. For example, owners who frequently commute across cities or live in areas with complex road conditions have a higher map data demand than owners who commute short distances along fixed routes. The cloud platform can weight and sum the vehicle location score and the owner profile score to obtain a comprehensive score for each target vehicle. The higher the comprehensive score, the greater the contribution of the vehicle to the overall service demand of the map data service. The cloud platform can assign a corresponding vehicle weight coefficient to each target vehicle according to the comprehensive score. The higher the comprehensive score, the greater the weight coefficient, so that vehicles with greater contribution occupy a higher proportion in the total request volume.

[0045] For example, a target vehicle is a high-end model equipped with an advanced driver assistance system, with a model positioning score of 0.9. The owner is a young or middle-aged business professional who commutes daily across cities, with an owner profile score of 0.8. The cloud platform weights and sums these two scores to obtain a comprehensive score of 0.85 for the vehicle, which is the highest among all target vehicles. Therefore, the cloud platform assigns the largest weight coefficient of 1.2 to this target vehicle. Another target vehicle is a basic model without intelligent driving functions, with a model positioning score of 0.3. The owner is a retiree who only travels short distances each week, with an owner profile score of 0.2. The comprehensive score is 0.25. The cloud platform can assign a smaller weight coefficient of 0.6 to this vehicle.

[0046] When relying solely on vehicle model location information, the cloud platform can directly determine the comprehensive score for each target vehicle based on its model location score. A higher model location score indicates a greater contribution to the demand for mapping services. The cloud platform assigns a corresponding vehicle weight coefficient to each target vehicle based on its model location score. For example, when assessing the contribution of demand based solely on vehicle model, configuration level, and price tier, high-end models equipped with advanced driver assistance systems (ADAS) are assigned a higher weight coefficient, while basic models without intelligent driving functions are assigned a lower weight coefficient. Similarly, when relying solely on owner profile information, the cloud platform can directly determine the comprehensive score for each target vehicle based on its owner profile score. A higher owner profile score indicates a greater contribution to the demand for mapping services. The cloud platform assigns a corresponding vehicle weight coefficient to each target vehicle based on its owner profile score. For example, when assessing the contribution of demand based solely on tags such as owner age and geographic region, owners who frequently commute across cities are assigned a higher weight coefficient, while owners who commute short distances on fixed routes are assigned a lower weight coefficient.

[0047] Based on the vehicle weight coefficient of each target vehicle, the cloud platform can obtain the total map data request volume of the map data server within the target time period.

[0048] Therefore, by introducing weighting coefficients related to vehicle attributes and individual owner characteristics, the cloud platform can make the calculation of total map data requests more closely match the actual demand distribution, avoiding the cumulative results from deviating from the real situation due to large differences in demand among different vehicles, and providing a more accurate data foundation for subsequent resource allocation.

[0049] Optionally, in some embodiments, obtaining the total map data request volume includes: multiplying the map data request volume of each target vehicle by the corresponding vehicle weight coefficient to obtain the weighted request volume of each target vehicle; and summing the weighted request volumes of each target vehicle to obtain the total map data request volume.

[0050] Specifically, the cloud platform can multiply the map data request volume of each target vehicle by the corresponding vehicle weight coefficient to obtain the weighted request volume of each target vehicle. This weighted request volume can represent the equivalent contribution value of each target vehicle in the total request volume. Subsequently, the cloud platform can sum up the weighted request volumes of all target vehicles to obtain the total map data request volume of the map data server within the target time period.

[0051] Therefore, the cloud platform can differentiate the weights of different vehicles by introducing weight coefficients before accumulating them, making the calculation of the total map data request volume more in line with the actual demand distribution, and providing a more accurate data foundation for subsequent resource allocation.

[0052] In step S103, the map data server adjusts the push strategy and / or resource allocation parameters for pushing map data to each target vehicle based on the total map data request volume.

[0053] It is understood that, in this embodiment, the push strategy refers to the method and rules by which the map data server sends map data to the vehicle, including push frequency, push timing, and push data volume, which are used to control the transmission behavior of map data from the server to the vehicle. Resource allocation parameters refer to the relevant configurations of hardware and network resources allocated by the map data server to support the map data push service, including data transmission bandwidth and computing resources, which are used to ensure the processing capacity and transmission efficiency of the map data push service.

[0054] Specifically, when the total map data request volume is large, the map data server adaptively adjusts the push strategy and / or resource allocation parameters to cope with the high demand. This includes, but is not limited to, adjustments to the push strategy such as the frequency and amount of map data pushed to vehicles, as well as adjustments to resource allocation parameters such as data transmission bandwidth and computing resources used to process map data requests and transmission. When the total map data request volume is small, the map data server adaptively reduces the strength of the push strategy or reduces resource allocation parameters to avoid resource idleness and waste.

[0055] For example, the map data server has a first threshold (e.g., 1000 times / second) and a second threshold (e.g., 200 times / second), where the first threshold is greater than the second threshold. The map data server can compare the total number of map data requests with the first and second thresholds and execute corresponding adjustment strategies based on the comparison results. That is, assuming the current push frequency is 10 times per second to each vehicle, the current bandwidth is 100Mbps, and the current computing resources are a 4-core CPU (Central Processing Unit). When the total number of map data requests exceeds the first threshold, it indicates that the total demand for map data from each target vehicle within the current prediction period is high. The map data server will adjust according to the first service configuration scheme: In terms of push strategy, the map data server will increase the push frequency of map data to each target vehicle from the current frequency to the first target frequency (e.g., 20 times per second), and determine the push priority of each vehicle based on the multi-dimensional characteristics of each target vehicle, pushing map data in descending order of priority to ensure that vehicles with high demand can obtain map data first; In terms of resource allocation, the map data server will increase the data transmission bandwidth from the current bandwidth to the first target bandwidth (e.g., 200Mbps), and increase the computing resources used to process map data requests from the current amount to the first target amount (e.g., 8-core CPU) to ensure processing capacity and transmission efficiency under high concurrency requests.

[0056] When the total map data request volume is lower than the second threshold, it indicates that the total demand for map data from each target vehicle within the current prediction period is low. The map data server will adjust according to the second service configuration scheme: In terms of push strategy, the map data server will reduce the push frequency of map data to each target vehicle from the current frequency to the second target frequency (e.g., 5 times per second), and identify low-demand vehicles based on the multi-dimensional characteristics of each target vehicle, prioritizing the reduction of the push frequency for low-demand vehicles; in terms of resource allocation, the map data server will reduce the data transmission bandwidth from the current bandwidth to the second target bandwidth (e.g., 50Mbps), and reduce the computing resources used to process map data requests from the current resource amount to the second target resource amount (e.g., 2-core CPU) to avoid resource idleness and waste. It should be noted that the above values ​​are only illustrative examples and do not constitute a limitation on the embodiments of this application. In actual applications, they can be flexibly set according to the processing capacity of the map data server and actual business needs.

[0057] When the total map data request volume is between the second threshold and the first threshold, it indicates that the total demand for map data by each target vehicle in the current prediction period is at a normal level, and the map data server maintains the current push frequency, data transmission bandwidth and computing resource configuration unchanged.

[0058] Therefore, the map data server can dynamically adjust the push strategy and / or resource allocation parameters according to the changes in the total map data request volume, so that the server's push behavior and resource investment can match the actual needs, and avoid resource waste while meeting the map data needs of the vehicle side.

[0059] To help those skilled in the art further understand the adaptation method for map data service requirements proposed in the embodiments of this application, further explanation is provided below in conjunction with specific implementation methods.

[0060] Combination Figure 3 As shown, during vehicle operation, the in-vehicle communication module can collect vehicle operating data in real time and report it to the cloud platform. Specifically, in the smart cockpit, data collection is mainly achieved through TBOX (Telematics Box, in-vehicle remote information processing terminal) embedded points and vehicle system embedded points. TBOX embedded points refer to the collection of key signal data of various components by TBOX components via the CAN bus, divided into low-frequency (collected once every ten seconds) and high-frequency (collected once per second) frequencies; vehicle system embedded points refer to the collection of information such as the opening, closing, and continuous operation time of applications in the vehicle system, divided into local and online dimensions. The collected embedded point signals are reported to the TSP cloud system. After receiving the data, the TSP cloud system parses the signals according to the protocol and then uploads the parsed data to the data platform. The data platform mainly uses data from the Information Computing Center (ICC) and the Autonomous Driving Computing Center (ADCC). For example, when a user uses the intelligent driving function, the ADCC requests map data from the ICC. The ICC then sends a request to the map data server through the vehicle network and returns the result to the ADCC, thus enabling on-demand acquisition of map data.

[0061] Based on the received operational data, the cloud platform determines the driver's current driving scenario, such as elevated highway driving, urban road driving, and automatic parking. Combining this with the driver's historical driving habits, it uses scenario models to analyze and predict the driver's map data needs in different scenarios. Specifically, high-precision map data can be divided into four layers based on update frequency: static basic data layer, quasi-static data layer, dynamic event data layer, and real-time interactive data layer. The static basic data layer is updated weekly and includes basic information such as road geometry, lane topology, and fixed traffic signs. The quasi-static data layer is updated hourly and includes event data such as construction areas, temporary control points, and long-term road occupancy. The dynamic event data layer is updated every 5 minutes and includes data such as traffic accidents, sudden congestion, and abnormal weather warnings. The real-time interactive data layer is updated every second and includes data such as traffic light phases, vehicle emergency braking warnings, and cooperative lane change requests. Drivers will generate differentiated map data requests when switching driving areas or triggering intelligent driving functions. For example, when a driver transitions from an elevated highway to a city road, the vehicle speed decreases, and the demand for data at each level changes accordingly. The request volume for the dynamic event data layer increases, while the request volume for the quasi-static data layer decreases. Similarly, when a driver enters a parking lot and uses the automatic parking function, the request volume for real-time parking data in the real-time interactive data layer increases. Furthermore, drivers' daily driving habits also influence the demand for map data services. Drivers accustomed to driving on elevated highways request relatively less data related to city roads; drivers accustomed to traveling during rush hour request more data from the dynamic event data layer and the real-time interactive data layer; and drivers accustomed to driving on highways at night request more data from the quasi-static data layer, such as data related to temporary construction control in highway areas.

[0062] The cloud platform can also acquire vehicle model location information and owner profile information for each target vehicle, and combine this with data request statistics to use a big data model to assess the data request volume, obtaining a data request volume assessment result related to a time period. After the model is established, it is continuously debugged and iterated based on big data prediction results and actual request results to improve the model's prediction accuracy. Specifically, the cloud platform can also acquire new sample data according to a preset period (such as quarterly) to incrementally train or retrain the trained prediction model, enabling the model to adapt to changes in data distribution and maintain prediction accuracy. After each iteration, the coefficient of determination and root mean square error are recalculated using the latest test set to verify whether the accuracy of the updated model has improved compared to the previous version.

[0063] It should be noted that the big data models used in this application are either random forest regression models or gradient boosting decision tree models. Both models are ensemble learning algorithms based on decision trees, suitable for processing multidimensional feature inputs containing both continuous and categorical variables, and outputting continuous predicted values, which matches local demand prediction scenarios (inputs are multidimensional heterogeneous features, outputs are continuous request quantity values). In practical applications, the model selection can be based on the data scale, training cost, and prediction accuracy requirements. When the number of training samples is large (e.g., exceeding 100,000) and high training speed is required, the random forest regression model can be preferred; when high prediction accuracy is required and the quality of the training data labels is good, the gradient boosting decision tree model can be preferred.

[0064] The model's input data consists of multi-dimensional features for each target vehicle. These features are numerical vectors, including user driving behavior features, map request scenario features, and request frequency features. Before being input into the model, the cloud platform standardizes the multi-dimensional features, mapping each feature value to a distribution range with a mean of 0 and a standard deviation of 1, to eliminate scale differences between features of different dimensions. For categorical features (such as vehicle type positioning level, geographic region labels, etc.), one-hot encoding can be used to convert them into numerical vectors. The model's output data is the predicted map data request volume (in seconds or hours) for a single target vehicle within the target time period, and the output is a continuous numerical value.

[0065] During the model training phase, the cloud platform acquires sample operation data and sample map service request logs from multiple sample vehicles, extracts corresponding multi-dimensional features of the samples, and constructs a training sample set using the actual map data request volume of each sample vehicle as a label. The training sample set is divided by time dimension, using historical time periods (e.g., the past 6 months) as the training and validation sets, and future time periods (e.g., the next month) as the test set to verify the model's generalization ability. During training, the mean squared error and mean absolute error between predicted and actual values ​​are used as loss functions, and model parameters are optimized by minimizing the loss function. For the random forest regression model, hyperparameters such as the number of decision trees, maximum depth, and minimum number of leaf node samples are tuned using a grid search method; for the gradient boosting decision tree model, hyperparameters such as the learning rate, number of trees, tree depth, and number of leaf nodes are tuned using a grid search method. K-fold cross-validation (e.g., K=5 or K=10) is used to evaluate the model's performance stability on different data subsets to prevent overfitting. After training is completed, the coefficient of determination and root mean square error are used as evaluation indicators of model accuracy. When the coefficient of determination reaches 0.85 or above and the root mean square error is less than the preset threshold, the model is considered to have passed training and can be deployed online.

[0066] The cloud platform sends the predicted map data service demand results to the map data service system. Based on the total map data requests received, the map data service system determines the total amount of resources required within the target time period and dynamically adjusts the map data push strategy and / or resource allocation parameters for each target vehicle accordingly. Specifically, OEMs propose resource subscription plans to the map data service provider that meet actual needs by refining the specific request volume for map data services in each time period, thus avoiding navigation response delays due to insufficient resources during peak hours and impacting user experience. The map data service provider can also dynamically configure corresponding service resources based on the specific request volume provided by the OEM, avoiding resource waste while providing stable and reliable map data service quality. Based on the service agreement adjustment results and combined with the historical map data request volume of different car owners in different scenarios, the OEM assesses the map data service demand patterns corresponding to each car owner, thereby determining the map data service level configuration for different car owners. Specifically, car owners who do not frequently use intelligent driving functions or have low demand for real-time interactive data have lower map data service demand; while car owners who frequently use intelligent driving functions or have high demand for real-time interactive data have higher map data service demand.

[0067] In practice, in-vehicle navigation services typically include the cost of map data services for the first three years after vehicle delivery in the overall vehicle purchase price. OEMs can assess a car owner's map data request volume and service demand patterns across different time periods based on their actual usage data from the first three years, and then provide differentiated map data service configuration plans starting in the fourth year. Simultaneously, OEMs can present car owners with the assessment results of their map data service needs and corresponding service configuration recommendations, enabling them to understand the correlation between their driving behavior and map data service requirements, and guiding them to develop reasonable driving habits.

[0068] Thus, starting from the driver's driving, through the cloud platform's scenario-based statistical prediction of map data demand, and the map data service system's dynamic adaptation based on resource demand, a complete closed-loop processing flow is formed, achieving precise matching between map data service resources and actual vehicle needs.

[0069] According to the map data service demand adaptation method proposed in this application, by acquiring the operating data and map service request logs of multiple target vehicles, multi-dimensional characteristics of each target vehicle can be obtained based on the operating data and map service request logs of each target vehicle. Based on the multi-dimensional characteristics of each target vehicle, the map data request volume of each target vehicle within a target time period is determined, and the total map data request volume of the map data server within the target time period is obtained based on the map data request volume of each target vehicle. Based on the total map data request volume, the push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle are adjusted. This solves the problems in related technologies where the map data server lacks multi-dimensional customer data support, making it difficult to predict demand fluctuations, leading to peak resource shortages and off-peak resource waste. It achieves dynamic adaptation between the resource supply of the map data server and the actual demand of the vehicle, improving resource utilization efficiency and service stability.

[0070] Next, with reference to the accompanying drawings, an adaptation device for map data service requirements proposed according to embodiments of this application is described.

[0071] Figure 4 This is a block diagram of an adaptation device for map data service requirements according to an embodiment of this application.

[0072] like Figure 4 As shown, the adaptation device 10 for the map data service requirement includes: an acquisition module 100, a processing module 200, and an adjustment module 300.

[0073] The module 100 is used to acquire the operating data of multiple target vehicles and the map service request logs of multiple target vehicles, and to obtain the multi-dimensional features of each target vehicle based on the operating data of each target vehicle and the map service request logs of each target vehicle. The processing module 200 is used to determine the map data request volume of each target vehicle within a target time period based on the multi-dimensional characteristics of each target vehicle, and to obtain the total map data request volume of the map data server within the target time period based on the map data request volume of each target vehicle. The adjustment module 300 is used to adjust the push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle based on the total map data request volume.

[0074] Optionally, in some embodiments, multi-dimensional features of each target vehicle are obtained, and module 100 is specifically used for: Extract the first feature from the operational data of each target vehicle, and extract the second feature from the map service request log of each target vehicle; By combining the first and second features, multi-dimensional features for each target vehicle are generated.

[0075] Optionally, in some embodiments, the first feature includes at least one of driving behavior features and map request scenario features, and the second feature is a request frequency feature.

[0076] Optionally, in some embodiments, the total map data request volume of the map data server within the target time period is obtained. The processing module 200 includes: The determination unit is used to determine the vehicle weight coefficient of each target vehicle based on the vehicle model positioning information and / or vehicle owner profile information of each target vehicle. The acquisition unit is used to obtain the total map data request amount based on the vehicle weight coefficient of each target vehicle.

[0077] Optionally, in some embodiments, the unit that obtains the total map data request amount is specifically used for: The map data request volume of each target vehicle is multiplied by the corresponding vehicle weight coefficient to obtain the weighted request volume of each target vehicle. The weighted request amounts for each target vehicle are summed to obtain the total map data request amount.

[0078] It should be noted that the explanation of the above-described method for adapting map and data service requirements also applies to the device for adapting map and data service requirements in this embodiment, and will not be repeated here.

[0079] According to the map data service demand adaptation device proposed in this application embodiment, by acquiring the operating data and map service request logs of multiple target vehicles, multi-dimensional characteristics of each target vehicle can be obtained based on the operating data and map service request logs of each target vehicle. Based on the multi-dimensional characteristics of each target vehicle, the map data request volume of each target vehicle within a target time period is determined, and the total map data request volume of the map data server within the target time period is obtained based on the map data request volume of each target vehicle. Based on the total map data request volume, the push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle are adjusted. This solves the problems in related technologies where the map data server lacks multi-dimensional customer data support, making it difficult to predict demand fluctuations, leading to peak resource shortages and off-peak resource waste. It achieves dynamic adaptation between the resource supply of the map data server and the actual demand of the vehicle, improving resource utilization efficiency and service stability.

[0080] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device may include: The memory 501, the processor 502, and the computer program stored on the memory 501 and capable of running on the processor 502.

[0081] When the processor 502 executes the program, it implements the adaptation method for the map and data service requirements provided in the above embodiments.

[0082] Furthermore, electronic devices also include: Communication interface 503 is used for communication between memory 501 and processor 502.

[0083] The memory 501 is used to store computer programs that can run on the processor 502.

[0084] The memory 501 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0085] If the memory 501, processor 502, and communication interface 503 are implemented independently, then the communication interface 503, memory 501, and processor 502 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0086] Optionally, in a specific implementation, if the memory 501, processor 502, and communication interface 503 are integrated on a single chip, then the memory 501, processor 502, and communication interface 503 can communicate with each other through an internal interface.

[0087] The processor 502 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of this application.

[0088] This application also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the above-described method for adapting to map and data service requirements.

[0089] This application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described adaptation method for map and data service requirements.

[0090] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0091] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0092] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for adapting map data service requirements, characterized in that, Includes the following steps: Obtain the operational data and map service request logs of multiple target vehicles, and obtain the multi-dimensional features of each target vehicle based on the operational data and map service request logs of each target vehicle. Based on the multi-dimensional characteristics of each target vehicle, the map data request volume of each target vehicle within the target time period is determined, and based on the map data request volume of each target vehicle, the total map data request volume of the map data server within the target time period is obtained. Based on the total map data request volume, adjust the map data push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle.

2. The method according to claim 1, characterized in that, The obtained multi-dimensional features of each target vehicle include: Extract a first feature from the operational data of each target vehicle, and extract a second feature from the map service request log of each target vehicle; By combining the first feature and the second feature, multi-dimensional features of each target vehicle are generated.

3. The method according to claim 2, characterized in that, The first feature includes at least one of driving behavior features and map request scenario features, and the second feature is the request frequency feature.

4. The method according to claim 1, characterized in that, The total map data request volume obtained from the map data server within the target time period includes: The vehicle weight coefficient for each target vehicle is determined based on its vehicle model location information and / or owner profile information. The total map data request amount is obtained based on the vehicle weight coefficient of each target vehicle.

5. The method according to claim 4, characterized in that, The process of obtaining the total map data request volume includes: The map data request amount of each target vehicle is multiplied by the corresponding vehicle weight coefficient to obtain the weighted request amount of each target vehicle. The weighted request amounts of each target vehicle are summed to obtain the total map data request amount.

6. An adaptation device for map data service requirements, characterized in that, include: The acquisition module is used to obtain the operating data and map service request logs of multiple target vehicles, and to obtain the multi-dimensional features of each target vehicle based on the operating data and map service request logs of each target vehicle. The processing module is used to determine the map data request volume of each target vehicle within a target time period based on the multi-dimensional characteristics of each target vehicle, and to obtain the total map data request volume of the map data server within the target time period based on the map data request volume of each target vehicle. The adjustment module is used to adjust the map data push strategy and / or resource allocation parameters of the map data server to push map data to each target vehicle based on the total map data request volume.

7. The apparatus according to claim 6, characterized in that, The module for obtaining the multi-dimensional features of each target vehicle is specifically used for: Extract a first feature from the operational data of each target vehicle, and extract a second feature from the map service request log of each target vehicle; By combining the first feature and the second feature, multi-dimensional features of each target vehicle are generated.

8. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and capable of running on the processor, the processor executing the program to implement the mapping service requirement adaptation method as described in any one of claims 1-5.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the adaptation method for map and data service requirements as described in any one of claims 1-5.

10. A computer program product, characterized in that, Includes a computer program, which, when executed by a processor, is used to implement the adaptation method for map and data service requirements as described in any one of claims 1-5.

Images