Vehicle information pushing method and system

Abstract

The invention relates to the technical field of vehicles, and provides a vehicle information pushing method and system. The method comprises the following steps: S1, determining a vehicle strain based on a collected data stream; s2, calculating the total parking time according to the vehicle speed and the engine state; s3, inputting the vehicle strain, the total parking duration and a position type obtained based on the vehicle positioning coordinates into a pre-trained intention recognition model, and outputting a current parking intention scene; and S4, based on the intention scene, obtaining an optimal touch push opportunity by adopting a preset strategy, and performing push display at a display end. According to the invention, the technical problems of improper pushing opportunity and inaccurate content in the prior art are solved, so that the effectiveness and user satisfaction of information service are really improved.

Description

Technical Field

[0001] This application relates to the field of vehicle technology, and in particular to a vehicle information push method and system. Background Technology

[0002] In the commercial vehicle operation sector, commercial vehicles, as production tools, exhibit significant differences in user operating scenarios, usage behaviors, and parking scenarios compared to passenger vehicles. Parking behavior in commercial vehicles is often closely related to specific business operations (mandatory rest, vehicle maintenance, waiting for loading and unloading, etc.). In recent years, with the popularization of vehicle-to-everything (V2X) big data technology, location-based information push has become an important means to improve the operational efficiency and value-added services of commercial vehicles.

[0003] Existing information push technologies, especially when targeting commercial vehicle scenarios, have the following problems:

[0004] Poor timing accuracy of push notifications: Most current push notification strategies are triggered when a vehicle enters a specific geofence (such as a service area or logistics park). This ignores the vehicle's status and the driver's real-time needs. If a message is pushed while the vehicle is in motion, not only will the driver be unable to pay attention, posing a safety hazard, but the message may also be instantly ignored, resulting in an extremely low effective reach rate.

[0005] Weak relevance of push notification content: Current technology fails to effectively differentiate the operational characteristics and driver needs of different types of commercial vehicles. The working scenarios and rest schedules of tractor-trailer and dump truck drivers differ significantly. The current "one-size-fits-all" push notification model cannot provide content strongly relevant to the driver's current situation, resulting in low click-through and conversion rates.

[0006] Lack of in-depth understanding of driver intent: Existing solutions generally lack a deep understanding of the "parking" behavior. Parking scenarios include short breaks, meals, long overnight rests, waiting in line, loading and unloading operations, or vehicle repairs, and the information needs corresponding to different intents are completely different.

[0007] Therefore, there is an urgent need in this field for a technical solution that can intelligently identify the parking intentions of commercial vehicle drivers and push highly relevant information at the optimal time to solve the technical problems of inappropriate push timing and inaccurate content in existing technologies, thereby truly improving the effectiveness of information services and user satisfaction. Summary of the Invention

[0008] The purpose of this invention is to provide a vehicle information push method and system, solving the technical problems of inappropriate push timing and inaccurate content in the prior art. The specific solution is as follows:

[0009] A method for pushing vehicle information, the method comprising the following steps:

[0010] S1: Determine the vehicle type based on the collected data stream;

[0011] S2: Calculate the total parking time based on vehicle speed and engine status;

[0012] S3: Input the vehicle type, total parking time, and location type obtained based on the vehicle's location coordinates into the pre-trained intent recognition model, and output the intent scenario of the current parking;

[0013] S4: Based on the intent scenario, a preset strategy is used to obtain the best time to push notifications, and the notifications are displayed on the display end.

[0014] Optionally, step S1 specifically includes:

[0015] The vehicle network cloud platform receives data streams uploaded by vehicle terminals in real time via its data interface; the data streams include at least: static attribute data and dynamic behavior data; the static attribute data includes at least the vehicle VIN code and vehicle model; the dynamic behavior data includes at least: vehicle latitude and longitude information, instantaneous speed, engine status, engine speed, and corresponding timestamps;

[0016] The data stream is preprocessed; the preprocessing includes at least: data cleaning and noise reduction, removal of null and outlier values, latitude and longitude data correction, smoothing and filtering of abnormal speed jumps, and mutual speed correction between instrument speed and GPS speed.

[0017] Identify vehicle type and category from the preprocessed data stream.

[0018] Optionally, step S2 specifically includes:

[0019] Based on instantaneous speed and engine status, obtain the start and end times of the parking event;

[0020] Calculate the total parking duration based on the start and end times of the parking event.

[0021] Optionally, obtaining the start and end times of the parking event based on instantaneous speed and engine status specifically includes:

[0022] When the instantaneous speed of the vehicle is continuously lower than the first speed threshold and the engine is in a shut-off state for a first set time, the start timestamp T-start and start position P-start of the parking event are recorded.

[0023] Continuously monitor the vehicle status. When the vehicle's instantaneous speed increases and exceeds the second speed threshold, or when the engine status changes from off to on, record the end timestamp T-end of the parking event.

[0024] The total parking duration of the parking event is obtained by calculating the difference between the end timestamp T-end and the start timestamp T-start.

[0025] Optionally, step S3 involves inputting the vehicle type, total parking time, and location type obtained based on the vehicle's positioning coordinates into a pre-trained intent recognition model to output the current parking intent scenario, specifically including:

[0026] Match the starting location P-start of the parking event with the preset electronic fence database and output the location type of the current vehicle parking location;

[0027] Based on a preset configurable threshold, the total parking time is divided into corresponding time category intervals; wherein, the division value of the time category interval can be dynamically configured according to the operating characteristics of the vehicle series;

[0028] The vehicle type, total parking time, and location type are input into a pre-trained intent recognition model, which outputs the optimal intent scenario for the current parking. The intent recognition model is trained based on labeled data of historical parking events, taking into account vehicle type, parking time category range, and parking location type.

[0029] Optionally, the classification values ​​for the category intervals are dynamically configured based on the operational characteristics of the vehicle series, specifically including:

[0030] Based on vehicle models and their corresponding usage scenarios, the parking duration of each vehicle model is statistically analyzed using big data.

[0031] Configure the corresponding time range threshold based on the parking duration of the vehicle series;

[0032] Based on vehicle driving data, update the time interval threshold library corresponding to various vehicle types.

[0033] Optionally, step S4 specifically includes:

[0034] The pushed information in the information database is tagged; the tags include at least: intent recognition scenario, target vehicle type, geographical range, information priority, and effective time limit;

[0035] When the intent recognition module outputs the intent scenario of the current vehicle, it filters all information entries that match the intent scenario tag in the information database based on the vehicle type, location type and intent scenario of the current vehicle.

[0036] The selected information items are comprehensively scored and sorted based on scene matching degree and information priority to obtain the first information item in the message list;

[0037] Based on the first information item, a delayed push strategy is adopted to obtain the best time to push and display the information on the display device.

[0038] A vehicle information push system, the system comprising:

[0039] The data acquisition module is configured to determine the vehicle type based on the acquired data stream;

[0040] The calculation module is configured to calculate the total parking time based on vehicle speed and engine status;

[0041] The output module is configured to input the vehicle type, total parking time, and location type obtained based on the vehicle's positioning coordinates into a pre-trained intent recognition model, and output the intent scenario of the current parking.

[0042] The processing module is configured to use a preset strategy to obtain the best time to push notifications based on intent scenarios, and then display the notifications on the display device.

[0043] An electronic device includes: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus; characterized in that the memory stores a computer program, which, when executed by the processor, causes the processor to perform the steps of the method.

[0044] A computer-readable storage medium storing a computer program executable by an electronic device, which, when run on the electronic device, causes the electronic device to perform the steps of the method described herein.

[0045] A simulation platform, comprising:

[0046] An electronic device for implementing the steps of the method;

[0047] A processor that runs a program, and when the program runs, it executes the steps of the method from data output by the electronic device.

[0048] A storage medium for storing a program that, when run, executes the steps of the method on data output from an electronic device.

[0049] The above solution achieves the following beneficial technical effects:

[0050] This application provides a vehicle information push method and system; it deeply integrates vehicle type, parking duration, and location information to construct a dedicated intent recognition model for commercial vehicles, thereby achieving accurate identification of parking intent scenarios for vehicles of different types and parking states. This realizes the transformation of information push from a broad-based approach to highly relevant and precise push, effectively solving the technical problem of mismatch between push content and the driver's actual needs. Furthermore, this embodiment considers the appropriateness of information push timing, abandoning the fixed-time push or blind push during driving methods in the prior art. Push is selected during a time when the driver is parked and resting, has the conditions to receive information, and is in the corresponding demand scenario. This ensures that the vehicle is stationary when push is made, avoiding the safety hazards of pushing information while driving, and highly matches the driver's receiving state, effectively solving the technical problem of low information reach rate caused by improper push timing. Attached Figure Description

[0051] Figure 1 This is a flowchart illustrating a vehicle information push method. Detailed Implementation

[0052] To make the purpose, technical solution, and advantages of this application clearer, the following will be described in conjunction with the appendix. Figure 1 This application will be described in further detail. It is obvious that the described embodiments are merely some, not all, of the embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments described herein without inventive effort are within the scope of protection of this application.

[0053] The terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms “a,” “said,” and “the” used in the embodiments of this application and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0054] It should be understood that the term "and / or" used in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this article generally indicates that the preceding and following related objects have an "or" relationship.

[0055] It should be understood that although the terms first, second, third, etc., may be used in the embodiments of this application, these descriptions should not be limited to these terms. These terms are only used to distinguish the descriptions. For example, first may also be referred to as second without departing from the scope of the embodiments of this application, and similarly, second may also be referred to as first.

[0056] Depending on the context, the words “if” or “suppose” as used here can be interpreted as “when” or “in response to determination” or “in response to detection.” Similarly, depending on the context, the phrases “if determination” or “if detection (of the stated condition or event)” can be interpreted as “when determination” or “in response to determination” or “when detection (of the stated condition or event)” or “in response to detection (of the stated condition or event).”

[0057] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or device. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or device that includes said element.

[0058] It should be noted that any symbols and / or numbers present in the specification that are not marked in the accompanying drawings are not reference numerals.

[0059] The optional embodiments of this application are described in detail below with reference to the accompanying drawings.

[0060] Figure 1 The method for pushing vehicle information shown includes the following steps:

[0061] S1: Determine the vehicle type based on the collected data stream;

[0062] S2: Calculate the total parking time based on vehicle speed and engine status;

[0063] S3: Input the vehicle type, total parking time, and location type obtained based on the vehicle's location coordinates into the pre-trained intent recognition model, and output the intent scenario of the current parking;

[0064] S4: Based on the intent scenario, a preset strategy is used to obtain the best time to push notifications, and the notifications are displayed on the display end.

[0065] Specifically, this embodiment deeply integrates vehicle type, parking duration, and location information to construct a dedicated intent recognition model for commercial vehicles. This breaks through the limitations of traditional broad-based information push, enabling precise identification of parking intent scenarios for vehicles of different types and parking states. It transforms information push from a broad-based approach to highly relevant and precise push, effectively solving the technical problem of mismatch between push content and drivers' actual needs, and significantly improving click-through rates and driver satisfaction. Furthermore, this embodiment considers the appropriateness of information push timing, abandoning the fixed-time push or blind push during driving methods found in existing technologies. Pushes are selected during periods when drivers are resting, have the conditions to receive information, and are in the corresponding demand scenario. This ensures that the vehicle is stationary when pushes, avoiding safety hazards associated with pushing information while driving, and closely matches the driver's receiving state, effectively solving the technical problem of low information reach caused by inappropriate push timing.

[0066] In one specific embodiment, step S1 specifically includes:

[0067] The vehicle network cloud platform receives data streams uploaded by vehicle terminals in real time via its data interface; the data streams include at least: static attribute data and dynamic behavior data; the static attribute data includes at least the vehicle VIN code and vehicle model; the dynamic behavior data includes at least: vehicle latitude and longitude information, instantaneous speed, engine status, engine speed, and corresponding timestamps;

[0068] The data stream is preprocessed; the preprocessing includes at least: data cleaning and noise reduction, removal of null and outlier values, latitude and longitude data correction, smoothing and filtering of abnormal speed jumps, and mutual speed correction between instrument speed and GPS speed.

[0069] Identify vehicle type and category from the preprocessed data stream.

[0070] For example, based on vehicle model, vehicles can be identified as different types such as tractor trucks, dump trucks, cargo trucks, and special-purpose vehicles. The typical parking scenarios and intentions corresponding to different vehicle types differ significantly.

[0071] In one specific embodiment, step S2 specifically includes:

[0072] Based on instantaneous speed and engine status, obtain the start and end times of the parking event;

[0073] Calculate the total parking duration based on the start and end times of the parking event.

[0074] In one specific embodiment, obtaining the start and end times of the parking event based on instantaneous speed and engine status specifically includes:

[0075] When the instantaneous speed of the vehicle is continuously lower than the first speed threshold and the engine is in a shut-off state for a first set time, the start timestamp T-start and start position P-start of the parking event are recorded.

[0076] Continuously monitor the vehicle status. When the vehicle's instantaneous speed increases and exceeds the second speed threshold, or when the engine status changes from off to on, record the end timestamp T-end of the parking event.

[0077] The total parking duration of the parking event is obtained by calculating the difference between the end timestamp T-end and the start timestamp T-start.

[0078] Specifically, when the vehicle's instantaneous speed remains below the first speed threshold and the engine is off for a set time, the delay confirmation within the set time effectively eliminates interference from non-real parking scenarios such as temporary braking and idling, thus accurately determining the start point of the parking event and avoiding false triggering of parking judgments due to instantaneous state fluctuations. Continuous monitoring of the vehicle's status is also performed. When the vehicle's instantaneous speed increases and exceeds the second speed threshold, or the engine changes from off to on, the end timestamp T-end of the parking event is recorded, achieving comprehensive coverage of the parking end time point and ensuring that no parking status monitoring is missed.

[0079] In one specific embodiment, step S3 involves inputting the vehicle type, total parking time, and location type obtained based on the vehicle's positioning coordinates into a pre-trained intent recognition model, and outputting the current parking intent scenario, specifically including:

[0080] Match the starting location P-start of the parking event with the preset electronic fence database and output the location type of the current vehicle parking location;

[0081] For example, the electronic fence database includes at least: highway service areas, logistics parks, ports and wharves, mines, construction sites, gas stations, repair shops, and fruit and vegetable wholesale markets;

[0082] Based on a preset configurable threshold, the total parking time is divided into corresponding time category intervals; wherein, the division value of the time category interval can be dynamically configured according to the operating characteristics of the vehicle series;

[0083] The vehicle type, parking duration category range, and location type are input into a pre-trained intent recognition model, which outputs the optimal intent scenario for the current parking. The intent recognition model is trained based on labeled data of historical parking events, taking into account the vehicle type, parking duration category range, and parking location type.

[0084] Specifically, this embodiment achieves precise location type matching by matching the parking start location with an electronic fence database. The electronic fence data includes scenarios such as highway service areas and logistics parks, providing accurate geographical scene support for the intent recognition model. Then, the threshold for classifying parking duration categories is dynamically configured for different vehicle models, breaking through the limitation of a single duration and making the duration classification more in line with the operating characteristics of different vehicle models. By constructing a dedicated intent recognition model that integrates vehicle model, duration category range, and location type, and training the model based on historical labeled data, deep coupling of multiple types of data is achieved. This model can output the optimal intent scenario that matches the actual parking behavior of the vehicle, thereby ensuring accurate information push in the future.

[0085] In one specific embodiment, the classification values ​​of the category intervals are dynamically configured based on the operational characteristics of the vehicle series, specifically including:

[0086] Based on vehicle models and their corresponding usage scenarios, the parking duration of each vehicle model is statistically analyzed using big data.

[0087] Configure the corresponding time range threshold based on the parking duration of the vehicle series;

[0088] Based on vehicle driving data, update the time interval threshold library corresponding to various vehicle types.

[0089] For example, this embodiment divides the calculated total parking time (Duration) into different logical intervals based on preset configurable thresholds. These interval values ​​can be dynamically configured according to the operating characteristics of different vehicle models. For different vehicle models and corresponding operating scenarios, the distribution results of vehicle parking time are statistically analyzed based on big data, and interval thresholds corresponding to short, medium, and long parking times are established based on this. The time interval threshold library for each vehicle model is continuously and regularly updated based on accumulated vehicle operation data.

[0090] The construction of a multi-dimensional intent recognition model based on "vehicle type-duration-location" first requires domain experts to mine and analyze a large number of historical vehicle network data, combining their business experience, and then labeling the intent scenarios. For some scenarios that are difficult to label manually, the K-Means unsupervised clustering algorithm is used to cluster the features of parking events (vehicle type, duration, location, etc.), while incorporating some business knowledge to summarize the relevant intent scenarios. For the labeled dataset, a random forest model is used to generate a mapping logic model. The current vehicle type, duration range, and location type are input into the mapping logic model, which then outputs the most likely intent scenario.

[0091] In one specific embodiment, step S4 specifically includes:

[0092] The pushed information in the information database is tagged; the tags include at least: intent recognition scenario, target vehicle type, geographical range, information priority, and effective time limit;

[0093] When the intent recognition module outputs the intent scenario of the current vehicle, it filters all information entries that match the intent scenario tag in the information database based on the vehicle type, parking location and intent scenario of the current vehicle.

[0094] The selected information items are comprehensively scored and sorted based on scene matching degree and information priority to obtain the first information item in the message list;

[0095] Based on the first information item, a delayed push strategy is adopted to obtain the best time to push and display the information on the display device.

[0096] Specifically, this embodiment constructs a comprehensive information tagging system by binding tags such as intent recognition scenario, target vehicle type, and geographical range to the information pushed in the information database. This provides a structured data foundation for subsequent accurate filtering. In the filtering stage, tags are matched based on vehicle type, parking location, and intent scenario. The top-ranked information is determined by a comprehensive score ranking based on scenario matching degree and information priority. This abandons the traditional broad-based push mode and achieves precise adaptation of information to vehicle scenarios and model requirements, improving the relevance and click-through rate of the information. Based on the above, by adopting a delayed push strategy, the optimal delivery time can be selected according to scenario characteristics. This avoids interference with driver operations in the early stages of parking and ensures effective delivery of push information when the driver can receive it, avoiding safety hazards of push notifications while driving and ensuring the effectiveness of information push and driving safety.

[0097] This can be understood as follows: by tagging various types of information in the database, corresponding dimensional tags are defined, such as intent recognition scenarios, target vehicle types, applicable geographical ranges, information priority, and effective timeframes. When the intent recognition module outputs an intent scenario, it quickly filters all information entries with matching tags from the content database based on the current vehicle type, precise parking location, and intent scenario. The filtered information is then comprehensively scored and ranked based on scenario matching degree and information priority to ensure that the most relevant information is placed first in the generated message list.

[0098] For example, take long-haul trucks as an example.

[0099] When the freight truck in this application travels to a service area on the Beijing-Shanghai Expressway, it is determined that the intended scenario is to rest at the expressway service area (the determination criteria are: the vehicle type is a long-distance freight truck, the estimated parking time is 25 minutes, which falls within the medium duration range, and the location type is an expressway service area).

[0100] In comparison, the traditional push notification model cannot accurately match tags. Existing technology may push information to the truck driver that is irrelevant to the long-haul freight scenario, such as promotions in urban supermarkets and discounts on home charging stations. At the same time, it may push the information when the vehicle has just stopped and the driver is getting out of the vehicle to check the condition, which will lead to inaccurate push content. Secondly, the inappropriate timing of the push can also interfere with the driver's operation.

[0101] The tags for obtaining the first information (such as discounts at service area restaurants) in this application are: the intent to identify the scenario is resting at a highway service area, the target vehicle type is a long-distance freight truck, the geographical range is a service area on the Beijing-Shanghai Expressway, the information priority is medium, and the effective time is 08:00-20:00 on the same day.

[0102] The second piece of information (such as the guide to loading and unloading efficiency in logistics parks) is labeled as follows: the intent is to identify the scenario as loading and unloading in a logistics park, the target vehicle is a long-haul truck, the geographical scope is a certain logistics park, the information priority is high, and the effective period is the next 2 days.

[0103] The second piece of information (such as a discount on car maintenance) is labeled as follows: the intended scenario is daily parking in a residential area, the target vehicle type is a family sedan, the geographical scope is a certain community, the information priority is low, and the effective period is within one week.

[0104] After filtering: When the intended scenario of resting at a highway service area is output, information matching the tags is filtered from the information database based on the vehicle type (long-haul truck), location type (highway service area), and real-time attributes of the intended scenario (resting at a highway service area). At this point, only the first piece of information simultaneously meets the tag conditions of the intended scenario being resting at a highway service area, the target vehicle type being a long-haul truck, the geographical range covering the current service area, and the effective time period including the current time period; the second piece of information's intended scenario is loading and unloading goods at a logistics park, and its geographical range does not match the current service area, so it is excluded; the third piece of information's vehicle type and intended scenario do not match, so it is also excluded, directly solving the technical problem of inaccurate content.

[0105] By selecting the first piece of information, there is no need to sort multiple pieces of information; it is directly determined as the first piece of information. If multiple matching pieces of information are selected at the same time (such as adding a fourth piece of information (discount on tire inspection at a service area auto repair shop), tagged as intent recognition scenario being rest at a highway service area, target vehicle type being a long-haul freight truck, geographical range being a service area on the Beijing-Shanghai Expressway, information priority being high, and effective time being 09:00-18:00 on the same day), then the scene matching degree (the matching degree between the first piece of information and the rest scenario is 80%, and the matching degree of the fourth piece of information is 60%) and the information priority (the fourth piece of information is high, and the first piece of information is medium) will be combined to make a comprehensive score. Finally, the fourth piece of information with the higher comprehensive score will be pushed first to further ensure the accuracy of the content.

[0106] This embodiment adopts a delayed push strategy based on the intention scenario of resting at highway service areas. The confirmed information is pushed 10 minutes after the vehicle has come to a complete stop, when the driver is likely to enter a resting state. This avoids interfering with the driver's initial operation of getting out of the car to check the vehicle condition, while also ensuring that the driver is in a state where he can receive information, thus solving the technical problem of inappropriate push timing.

[0107] An intelligent delayed push strategy is adopted to select the best delivery time. First, a dynamic delay parameter is preset for each intent scenario. For non-emergency scenarios such as "rest" and "waiting", the push is delayed by a preset time after the parking event begins to avoid interfering with the driver's activities after parking. For "urgent or high-value" information related to safety and efficiency, the push can be sent immediately.

[0108] Push Execution: Send the matched information items from S4 to the driver through the most appropriate channel (such as push notifications from the driver's mobile app linked to the vehicle, messages on the vehicle's screen, etc.).

[0109] Driver behavioral feedback data is recorded, including information click-through rate, browsing duration, and subsequent behaviors. This behavioral feedback data is collected and fed back to the big data analysis platform for regular evaluation and optimization of the accuracy of the intent recognition model and the relevance of the information matching strategy.

[0110] In summary, the commercial vehicle precision information push method based on parking behavior analysis provided in this application deeply integrates vehicle type, quantified parking duration and location information to construct a dynamically adjustable intelligent intent scene recognition model. It intelligently pushes the driver's parking intent and the best time to reach the information, realizing the shift from "broad-spectrum push" to "scenario-based precision reach". While ensuring high relevance of information, it improves the effective click-through rate and user satisfaction.

[0111] Furthermore, the data stream in this embodiment is not limited to signals such as engine status, vehicle speed, and latitude and longitude location. It can also be expanded in terms of data dimensions: when determining intent, it can not only consider engine shutdown, but also incorporate vehicle load status, door opening and closing signals, and combine environmental, real-time weather, temperature and other information to analyze the vehicle's scenario and intent.

[0112] 2. The intent-based scenario matching information mentioned in this invention can also be integrated with user profiles to build a long-term profile for each driver, and combined with their historical preferences to achieve personalized recommendations during information matching.

[0113] 3. The information push mentioned in this invention is not limited to text and image information. It can also be combined with the vehicle system to provide richer interactive forms such as voice broadcast, navigation to the push destination, and gas station recommendations.

[0114] On the other hand, this application provides a vehicle information push system, the system comprising:

[0115] The data acquisition module is configured to determine the vehicle type based on the acquired data stream;

[0116] The calculation module is configured to calculate the total parking time based on vehicle speed and engine status;

[0117] The output module is configured to input the vehicle type, total parking time, and location type obtained based on the vehicle's positioning coordinates into a pre-trained intent recognition model, and output the intent scenario of the current parking.

[0118] The processing module is configured to use a preset strategy to obtain the best time to push notifications based on intent scenarios, and then display the notifications on the display device.

[0119] On the other hand, this application provides an electronic device, including: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0120] The memory stores a computer program that, when executed by a processor, causes the processor to perform the steps of the method.

[0121] On the other hand, this application provides a computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implement the steps of the method.

[0122] A simulation platform, comprising:

[0123] An electronic device for implementing the steps of the method;

[0124] A processor that runs a program, which, when running, executes the steps of the method claimed in the electronic device from data output by the program.

[0125] A storage medium for storing a program that, when run, executes the steps of the method on data output from an electronic device.

[0126] The technical features of the above embodiments can be combined in any way. For the sake of brevity, 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, they should be considered to be within the scope of this specification.

[0127] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A vehicle information push method characterized by, The method comprises the following steps: S1: determining the vehicle type based on the collected data stream; S2: calculating the total parking time according to the vehicle speed and engine state; S3: inputting the vehicle type, total parking time, and location type obtained based on the vehicle positioning coordinates into a pre-trained intention recognition model to output the intention scenario of the current parking; S4: based on the intention scenario, using a preset strategy to obtain the best touch push opportunity and pushing and displaying on the display end.

2. The method of claim 1, wherein, The step S1 specifically comprises: real-time receiving the data stream uploaded by the vehicle terminal through the data interface of the Internet of Vehicles cloud platform; wherein the data stream at least comprises static attribute data and dynamic behavior data; the static attribute data at least comprises vehicle VIN code and vehicle model; the dynamic behavior data at least comprises vehicle latitude and longitude information, instantaneous speed, engine state, engine speed, and corresponding time stamp; preprocessing the data stream; the preprocessing at least comprises data cleaning and denoising, removing null values and abnormal values, latitude and longitude data rectification processing, abnormal jump value smoothing filtering processing of speed, and mutual correction processing of vehicle speed through the instrument vehicle speed and GPS vehicle speed; identifying the vehicle type of the preprocessed data stream.

3. The method of claim 2, wherein, The step S2 specifically comprises: obtaining the start time and end time of the parking event according to the instantaneous speed and engine state; calculating the total parking time according to the start time and end time of the parking event.

4. The method of claim 3, wherein, The step of obtaining the start time and end time of the parking event according to the instantaneous speed and engine state specifically comprises: when the instantaneous speed of the vehicle continuously falls below the first speed threshold and the engine state is the off state and maintains the first set time, the start time stamp T-start and the start position P-start of the parking event are recorded; continuously monitoring the vehicle state, and when the instantaneous speed of the vehicle rises and exceeds the second speed threshold or the engine state changes from the off state to the start state, the end time stamp T-end of the parking event is recorded; obtaining the total parking time of the parking event according to the difference between the end time stamp T-end and the start time stamp T-start.

5. The method of claim 4, wherein, The step S3 of inputting the vehicle type, total parking time, and location type obtained based on the vehicle positioning coordinates into a pre-trained intention recognition model to output the intention scenario of the current parking specifically comprises: matching the start position P-start of the parking event with a preset electronic fence database to output the location type of the current vehicle parking site; dividing the total parking time into corresponding time interval categories according to a preset configurable threshold; wherein the division value of the time interval category can be dynamically configured according to the running characteristics of the vehicle type; inputting the vehicle type, total parking time, and location type into a pre-trained intention recognition model to output the optimal intention scenario of the current parking; wherein the intention recognition model is trained based on the vehicle type, parking time category interval, and parking location type through the labeled data of historical parking events.

6. The method of claim 5, wherein, The division value of the category interval is dynamically configured according to the running characteristics of the vehicle type, specifically comprising: According to the vehicle type and its corresponding use scene, the parking time of the vehicle type is counted through big data; According to the parking time of the vehicle type, the time interval threshold of the corresponding vehicle type is configured; According to the vehicle driving data, the time interval threshold library corresponding to each type of vehicle type is updated.

7. The method of claim 6, wherein, The step S4 specifically includes: Labeling the push information in the information library; The label at least includes: intention recognition scene, target vehicle type, geographical range, information priority, validity time limit; When the intention recognition module outputs the intention scene of the current vehicle, all information entries matching the intention scene label in the information library are filtered according to the vehicle type, location type and intention scene of the current vehicle; Based on the scene matching degree and information priority, the filtered information entries are comprehensively scored and sorted to obtain the first information entry in the message list; Based on the first information entry, a delay push strategy is adopted to obtain the best touch push opportunity and push display on the display end.

8. A vehicle information push system characterized by comprising: The system includes: The acquisition module is configured to determine the vehicle type based on the collected data stream; The calculation module is configured to calculate the total parking time according to the vehicle speed and engine state; The output module is configured to input the vehicle type, total parking time and location type obtained based on the vehicle positioning coordinates into the pre-trained intention recognition model, and output the intention scene of the current parking; The processing module is configured to obtain the best touch push opportunity based on the intention scene and a preset strategy, and push display on the display end.

9. An electronic device comprising: The processor, the communication interface, the memory and the communication bus, wherein the processor, the communication interface, the memory and the communication bus complete mutual communication through the communication bus; characterized in that the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer program is stored and can be executed by the electronic device, and when the computer program runs on the electronic device, the electronic device executes the steps of the method in any one of claims 1 to 7.

Images