Information sending method and device, and electronic device

Abstract

The application provides an information sending method and device and electronic equipment, wherein the method comprises the following steps: determining a plurality of driving paths of a vehicle and driving risks of each driving path in different road sections; determining signal strengths in road sections with driving risks exceeding a first preset value in the driving paths, and taking road sections with signal strengths lower than a second preset value as target road sections; calculating a reaching probability and a reaching remaining time of a user to the target road sections; after the reaching probability and the reaching remaining time both meet a first alert value, performing encryption processing on rescue information of the vehicle to obtain an encrypted data packet and a decryption key; after the reaching probability and the reaching remaining time both meet a second alert value, sending the encrypted data packet to a rescue vehicle; and after the reaching probability and the reaching remaining time both meet a third alert value, sending the decryption key to the rescue vehicle. Through the method, the rescue vehicle can more easily obtain complete data, and the security of private data is taken into account.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of computers, and in particular to an information sending method and device and electronic equipment. BACKGROUND

[0002] Vehicle rescue technology is a comprehensive technology system that integrates rapid diagnosis, emergency repair, professional traction and towing, special environment disposal and other means, with the help of professional equipment such as power-on equipment, tow trucks, cranes and breaking tools, combined with GPS / Beidou positioning, intelligent scheduling, remote collaboration and other technologies, through standardized processes to achieve on-site rapid repair, safe towing or transportation to the repair site, the core goal is to efficiently resolve road emergencies, ensure personnel safety and road smoothness, covering the full scene from common fault emergency handling to complex environment special rescue, and is upgrading towards intelligence and efficiency. SUMMARY

[0003] Therefore, the purpose of the present application is to provide an information sending method and device and electronic equipment to make it easier for rescue vehicles to obtain complete data while ensuring the security of private data.

[0004] In a first aspect, an information sending method is provided, comprising:

[0005] After detecting that the vehicle is in a driving state, determine a plurality of driving paths according to the user's driving habit data and real-time acquired traffic information data;

[0006] Input the road information of each driving path into a pre-trained driving risk identification model to determine the driving risk of each driving path in different road sections;

[0007] Determine the signal strength of the road section in each driving path where the driving risk exceeds a first preset value, and take the road section where the signal strength is lower than a second preset value as a target road section;

[0008] According to the real-time acquired vehicle driving information, calculate the arrival probability and arrival remaining time of the user to each target road section;

[0009] After the arrival probability and arrival remaining time meet a first alert value, encrypt the rescue information of the vehicle to obtain an encrypted data packet and a decryption key;

[0010] After the arrival probability and arrival remaining time meet a second alert value, establish a close-range communication connection with the rescue vehicle through close-range networking technology, and send the encrypted data packet to the rescue vehicle within a preset distance range;

[0011] When the arrival probability and the arrival remaining time both meet the third alert value, the decryption key is sent to the rescue vehicle through the short-distance communication connection, so that the rescue vehicle decrypts the encrypted data packet through the decryption key and completes the rescue according to the decryption result.

[0012] With reference to the first aspect, in a first possible implementation manner of the first aspect, the method further includes:

[0013] calculating the moving speed of the vehicle according to the real-time acquired position signal;

[0014] if the continuity of the acquired moving signal is normal, determining whether the vehicle is in the driving state according to whether the moving speed is greater than a third preset value;

[0015] if the continuity of the acquired moving signal is abnormal, triggering the camera of the driving recorder to acquire the in-vehicle image and the out-of-vehicle image;

[0016] if the facial feature is recognized from the in-vehicle image and the image moving condition is recognized from the out-of-vehicle image, it is determined that the current is in the driving state.

[0017] With reference to the first aspect, in a second possible implementation manner of the first aspect, the method further includes:

[0018] acquiring historical accident data, the historical accident data at least including accident occurrence positions and accident severities;

[0019] segmenting the natural road according to the historical accident data to determine a plurality of road segments, the accident occurrence frequencies of two adjacent road segments being different;

[0020] acquiring training samples, the training samples including the following data of each road segment: road segment basic data, historical accident data, real-time traffic information data, environment dynamic data, and driving habit data;

[0021] inputting the training samples into an untrained target mathematical model to complete the training of the target mathematical model, the target mathematical model being a logistics regression model;

[0022] inputting the road information into the trained target mathematical model to generate the driving risk in each different road segment.

[0023] With reference to the first aspect, the embodiments of the present application provide a third possible implementation manner of the first aspect, and the method comprises the following steps.

[0024] obtaining signal strength information and vehicle position information uploaded by vehicles on different road segments;

[0025] generating a first signal distribution map according to the signal strength information and the vehicle position information;

[0026] correcting the first signal distribution map based on signal strength information uploaded by a road side signal detector and a signal strength distribution rule determined based on geographic information, to determine a second signal distribution map;

[0027] determining signal strength in each road segment with driving risk exceeding a first preset value according to the second signal distribution map, and taking a road segment with signal strength lower than a second preset value as a target road segment.

[0028] With reference to the first aspect, the embodiments of the present application provide a fourth possible implementation manner of the first aspect, and the rescue information comprises the following contents: position information, accident information, emergency demand information, on-site environment information, vehicle information and personnel information.

[0029] With reference to the fourth possible implementation manner of the first aspect, the embodiments of the present application provide a fifth possible implementation manner of the first aspect, and the method further comprises the following steps.

[0030] generating first rescue information of the current accident according to historical accident conditions that have occurred on the target road segment, wherein the first rescue information comprises the position information, the accident information, the emergency demand information and the on-site environment information;

[0031] adjusting the first rescue information according to driving habit data of a driver of the vehicle;

[0032] generating second rescue information according to vehicle information and information of a driver and a passenger in the vehicle, wherein the second rescue information comprises the vehicle information and the personnel information.

[0033] With reference to the fourth possible implementation manner of the first aspect, the embodiments of the present application provide a sixth possible implementation manner of the first aspect, and the rescue information further comprises image information, wherein the image information comprises video information of an in-vehicle driving recorder, video information of an out-vehicle driving recorder and image information of a user; and the method further comprises the following steps.

[0034] extract foreground information from the image information in the rescue information to determine a background image and a foreground image;

[0035] extract key frames from the video information in the rescue information to determine a key frame image and a non-key frame image;

[0036] encrypt the background image and the non-key frame image carried in the rescue information to obtain an encrypted data packet, and encrypt the foreground image and the key frame image in a decryption key.

[0037] With reference to the first aspect, a seventh possible implementation of the first aspect is provided in the embodiments of the application, and the sending of the encrypted data packet to the rescue vehicle within the preset distance range through the near-distance networking technology to establish the near-distance communication connection with the rescue vehicle comprises the following steps.

[0038] The vehicle initiates a networking broadcast at a predetermined time interval.

[0039] After obtaining the rescue response request sent by the other vehicle, the position information and the verification information in the rescue response request are read.

[0040] The position information is used to determine whether each other vehicle is in a same-direction moving state.

[0041] According to the same-direction moving state of the other vehicle and the verification information, the rescue vehicle is selected from the other vehicles, and the near-distance networking is completed with the rescue vehicle.

[0042] The encrypted data packet is sent to the multiple rescue vehicles through the Internet or the completed near-distance networking.

[0043] In the second aspect, the embodiments of the application further provide an information sending device, comprising:

[0044] A first determination module is configured to determine a plurality of driving paths according to the driving habit data of the user and the real-time acquired traffic information data after detecting that the vehicle is in a driving state.

[0045] An input module is configured to input road information of each driving path into a driving risk identification model that is pre-trained to determine driving risks of each driving path in different road sections.

[0046] A second determination module is configured to determine signal strengths in road sections in which the driving risks exceed a first preset value in each driving path, and to determine road sections in which the signal strengths are lower than a second preset value as target road sections.

[0047] A first calculation module is configured to calculate arrival probabilities and arrival remaining times of the user to each target road section according to real-time acquired vehicle driving information.

[0048] The encryption module is configured to encrypt rescue information of the vehicle to obtain an encrypted data packet and a decryption key when the arrival probability and the arrival remaining time both meet a first alert value.

[0049] The establishment module is configured to establish a short-distance communication connection with the rescue vehicle by using a short-distance networking technology when the arrival probability and the arrival remaining time both meet a second alert value, and send the encrypted data packet to the rescue vehicle within a preset distance range.

[0050] The sending module is configured to send the decryption key to the rescue vehicle through the short-distance communication connection when the arrival probability and the arrival remaining time both meet a third alert value, so that the rescue vehicle decrypts the encrypted data packet by using the decryption key and completes the rescue according to a decryption result.

[0051] In a third aspect, an electronic device is provided, including a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps in any possible implementation manner of the first aspect.

[0052] The information sending method, device and electronic device provided in the embodiments of the present application can ensure that the user privacy is not easily disclosed, and at the same time, before the vehicle moves to an area with poor signal, most of the data (encrypted data packet) is sent to the rescue vehicle, and then when the vehicle moves to the area with poor signal, the decryption key with less data is sent to the rescue vehicle, so that the rescue vehicle can more easily obtain complete data, and the security of the private data is taken into account.

[0053] In order to make the above objectives, characteristics and advantages of the present application more apparent, clear and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are referred to. BRIEF DESCRIPTION OF DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some of the embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0055] Figure 1 A flow chart of an information sending method provided by an embodiment of the present application is shown;

[0056] Figure 2 A schematic diagram of a map provided by an embodiment of the present application is shown;

[0057] Figure 3 A schematic diagram of a road segment division provided by an embodiment of the present application is shown;

[0058] Figure 4 A structural schematic diagram of an information sending device provided by an embodiment of the present application is shown;

[0059] Figure 5 A structural schematic diagram of an electronic device provided by an embodiment of the present application is shown. DETAILED DESCRIPTION

[0060] In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the following will combine the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. The components of the embodiments of the present application described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the claimed present application, but only represents selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of the present application.

[0061] The vehicle rescue technology in the related art is usually after the occurrence of vehicle accidents, that is, after the occurrence of vehicle accidents, the driver of the accident vehicle seeks rescue and sends the location thereof to the rescue vehicle, and the rescue vehicle arrives at the accident location according to the location to perform rescue.

[0062] But this rescue mode is post-rescue, and in some areas with poor signal, using this rescue mode will cause the rescue signal to be difficult to send, and further cause the rescue to be delayed and failed.

[0063] In view of the above, this application provides an information sending method, apparatus and electronic device, which are described below through embodiments.

[0064] To facilitate understanding of this embodiment, a method for sending information disclosed in this application will first be described in detail. This method is applied to in-vehicle systems, such as... Figure 1 As shown, the process includes the following steps S101-S107:

[0065] S101: After detecting that the vehicle is in motion, determine multiple driving routes based on the user's driving habit data and real-time traffic information data;

[0066] S102: Input the road information of each driving route into the pre-trained driving risk identification model to determine the driving risk of each driving route in different road sections;

[0067] S103: Determine the signal strength of road segments in each driving path where the driving risk exceeds a first preset value, and designate road segments with signal strength lower than a second preset value as target road segments;

[0068] S104: Based on the real-time vehicle driving information, calculate the probability of the user reaching each target road segment and the remaining time to reach the target road segment;

[0069] S105: After the arrival probability and remaining arrival time both meet the first warning value, the vehicle rescue information is encrypted to obtain an encrypted data packet and a decryption key;

[0070] S106: After the arrival probability and remaining arrival time both meet the second warning value, establish a short-range communication connection with the rescue vehicle through short-range networking technology, and send encrypted data packets to the rescue vehicle within the preset distance range;

[0071] S107: After the arrival probability and remaining arrival time both meet the third warning value, the decryption key is sent to the rescue vehicle through a short-range communication connection so that the rescue vehicle can decrypt the encrypted data packet using the decryption key and complete the rescue based on the decryption result.

[0072] In step S101, there are many ways to detect when a vehicle has entered a driving state, mainly divided into two types: one is a judgment based on the detection signals of vehicle mechanical data, and the other is a judgment based on position signals. Vehicle mechanical data includes vehicle speed (km / h), engine speed (rpm), gear status (D / R), ABS working status, and acceleration. Position signals are relatively simple; they determine the vehicle's speed based on the real-time acquired vehicle's current position, and thus determine whether the vehicle has moved. Typically, position signals refer to position signals fed back by satellites, or they can be obtained through technologies such as base station positioning, and then the vehicle speed is calculated to further determine whether the vehicle has moved.

[0073] Since the solution provided in this application is designed for areas with poor signal, the method of determining vehicle speed using only satellite or base station signals may fail in such scenarios. Therefore, it is preferable to use vehicle mechanical data to detect whether the vehicle has entered a driving state. Furthermore, a fusion detection technique combining image recognition can be used. Specifically, it can be a combination of location signal and image recognition. That is, the method provided in this application also includes: calculating the vehicle's speed based on the real-time acquired location signal; if the continuity of the acquired movement signal is normal, determining whether the vehicle is in a driving state based on whether the movement speed is greater than a third preset value; if the continuity of the acquired movement signal is abnormal, triggering the dashcam's camera to acquire images inside and outside the vehicle; if facial features are identified from the image inside the vehicle and image movement is detected from the image outside the vehicle, then it is determined that the vehicle is currently in a driving state.

[0074] Specifically, facial feature recognition can be achieved directly using pre-trained open-source models, regardless of whether the facial features appear in the driver's position or other locations. Identifying moving targets from images outside the vehicle involves extracting feature points from two adjacent images (typically images acquired at intervals of 0.5 or more seconds), calculating the distance of each feature point in each image, and finally determining whether image movement has occurred based on the distance between the feature points. Image movement occurs when the movement vectors (distance and direction) of each feature point are approximately the same. This method is primarily used to handle irregular movements of certain associated feature points. For example, when a tree in a foreground image sways in the wind, multiple feature points on the tree are in a state of irregular movement, which can easily lead to misidentification. Therefore, verification through the consistency of feature point movement vectors is necessary.

[0075] The process of determining multiple driving routes based on user driving habit data and traffic information data generally involves two scenarios: user-specified destination and user-unspecified destination. User-specified destination means the user has already entered the destination during the driving process; unspecified destination means the user has not entered the destination during the driving process, in which case the user's historical information is used to pre-determine possible driving destinations.

[0076] In the absence of a specified driving destination, the driving route in this application can be determined as follows:

[0077] S1011: Based on the target user's (vehicle driver's) historical driving information and the target user's user type's historical driving information, determine multiple driving destinations for the target user;

[0078] S1012: Determine the main driving route to the driving destination based on the location information of the multiple driving destinations;

[0079] S1013: Based on the main driving route, the topological relationship of roads in the navigation map, the target user's current location information, the location information of each driving destination, and traffic information data, calculate the detour ratio of the target user to each driving destination.

[0080] S1014: Use the driving destination with a detour ratio lower than the preset value as the final driving destination;

[0081] S1015: Calculate the driving path for the target user to reach each final driving destination.

[0082] In step S1011, the determined driving destination is the possible driving destination of the target user. The reason for determining multiple destinations is to improve the fault tolerance rate. This solution does not necessarily need to determine the most accurate destination, but rather to find the possible optimal solution as comprehensively as possible. Therefore, the path of the target user to each driving destination can be determined first. These paths can cover the possible paths to different driving destinations.

[0083] Furthermore, in step S1011, based on the target user's historical driving information (the target user's own historical driving habits) and the historical driving information of the user type to which the target user belongs (other users' own historical driving habits), the most likely driving destinations for the user are determined. Here, the user type to which the target user belongs is obtained after pre-classifying all users. That is, before this solution is implemented, user classification and preference information collection must be completed. Specifically, user identity information can be collected in advance, which may include age, gender, occupation, education, spending power, work location, historical consumption information, etc. After obtaining this identity information, users can be classified using clustering. There are two specific classification strategies: one is to use clustering, such as using the K-Means algorithm, where "K cluster numbers" can be pre-set, and the distance between users and "cluster centers" (such as Euclidean distance) is calculated, iteratively optimizing to make users of the same type closest and users of different types farthest. In specific implementation, to ensure computational efficiency in subsequent steps, the number of types should not be too large, generally 4-8 types are more appropriate, that is, the core number of clusters should be controlled within 4-8. Alternatively, the DBSCAN algorithm can be used for clustering without limiting the number of clusters. While the results of this method are less controllable, it's a good starting point to determine the appropriate number of clusters before using the K-Means algorithm. In other words, the DBSCAN algorithm can be used for the first clustering, and the number of clusters determined based on the initial results. Then, based on the determined number of clusters, the K-Means algorithm can be used for the final clustering to determine the type of each user. Determining the number of clusters based on the initial clustering results primarily involves removing categories with too few individuals or individuals that are too far from the cluster core.

[0084] Another classification method is through supervised learning, such as Logistic Regression, Decision Tree, and Random Forest. These methods require users to first determine the core of the classification, that is, to define the classification logic. This is a targeted classification approach, generally used to purposefully guide users to a specific location. This method is suitable for initial implementation in specific regions (such as cities or areas with clearly defined business establishment categories, like shopping malls or restaurants offering only a certain type of product). However, it is not suitable if the types of business establishments in the implementation area are complex. In this solution, "establishments" refers to commercial establishments such as gas stations, parking lots, restaurants, and service areas.

[0085] Regardless of the classification method, once the user type is determined, the type of any given user can be identified. Then, the preferences of users of this type can be determined based on their behavior. Determining user preferences primarily utilizes the frequency and number of times a user visits a specific location (historical visit information), as well as the logical correlation between different locations or different pieces of information (logical correlation can be determined through pre-design or AI analysis). For example, the correlation between a user's age and visiting a bubble tea shop is negative (the older the user, the lower the probability of visiting a bubble tea shop); similarly, there is a negative correlation between visiting a farmers' market and visiting a large supermarket. Factors such as time and weather can also be considered. For instance, the frequency and number of times a certain type of user visits different locations at different times and in different regions can be statistically analyzed to form a sample. Training based on this sample can then reveal the preference information of a certain type of user, thereby determining multiple possible driving destinations for the current user.

[0086] Then, in step S1012, these multiple driving destinations are used to determine the driving route, which allows the target user to conveniently drive to the multiple driving destinations. Specifically, there are two methods for calculating the driving route: The first method calculates the shortest driving route from the target user's current location to each driving destination (the shortest driving route is the shortest distance or shortest time path from the current location to the corresponding driving destination); then, for each road, a reference score is calculated based on whether the shortest driving route includes that road (the more shortest driving routes pass through / include that road, the higher the reference score; that is, the number of times the shortest driving route passes through / includes that road is positively correlated with the reference score). Finally, the driving routes are formed based on the top-ranked roads according to their reference scores. A driving route generally consists of multiple consecutive roads connected end-to-end. Therefore, in practice, the reference score for each road can be calculated using the method described above, resulting in multiple reference driving routes that roughly point to multiple driving destinations. These reference driving routes cannot involve traveling back and forth on the same road; for example, road A cannot simultaneously travel east-west and west-east. Then, for each reference driving route, the total reference score of the reference route is calculated based on the reference score of each road in the reference driving route, and finally the reference driving route with the largest total reference score is selected as the main driving route.

[0087] The second method does not divide the map according to roads, but according to abstract areas. Specifically, the user's current location is fixed, multiple driving destinations are relatively fixed, and when most driving destinations are roughly in the same direction, or most driving destinations are in a few fixed directions, a target direction can be formed on the map, which points to the center of multiple driving destinations. In other words, during implementation, multiple driving endpoints can first be clustered according to their locations, and it can be determined whether a target cluster exists in the clustering results. The proportion of driving endpoints contained in the target cluster exceeds a predetermined value (generally, when there is only one target cluster, the driving endpoints contained in that target cluster should account for more than 70% of the total driving endpoints; when there are two target clusters, the driving endpoints contained in each target cluster should account for more than 35% of the total driving endpoints; generally, there will not be three or more target clusters, or if it is impossible to find a target cluster that accounts for more than 70% of the total driving endpoints, and it is impossible to find two target clusters that account for more than 35% of the total driving endpoints, the subsequent process can be skipped). Then, for each target cluster, the main driving route is determined based on the location information of the target cluster and the current location information of the target user.

[0088] The location information of the target clustering result can be the geometric midpoint of all driving endpoints included in the target clustering result, or a location where the distance to each driving endpoint of the target clustering result is approximately the same. The primary driving path determined in this way is the path from the user's current location to the location of the target clustering result (a navigation route composed of multiple consecutive roads). Alternatively, it can refer to an area from the user's current location to the location of the target clustering result, a region of a certain width on the map, with the user's current location as the starting point and the target clustering result location as the endpoint. For example... Figure 2 As shown, a schematic diagram of a map is presented. Figure 2 In the diagram, triangles represent the target user's current location, circles represent the location of the target clustering results, squares represent locations on the map, and slanted rectangular areas represent the main driving paths that indicate the region type.

[0089] Besides location clustering, this step can also be achieved through directional clustering. In implementation, the relative angle between each driving endpoint and the target user's current location is first calculated. Then, multiple driving endpoints are clustered according to the degree of the angle. It is then determined whether a target cluster exists, where the proportion of driving endpoints included in the target cluster exceeds a predetermined value (generally, if there is only one target cluster, its driving endpoints should account for more than 80% of the total driving endpoints; if there are two target clusters, each should contain more than 45% of the total driving endpoints). Afterward, for each target cluster, the main driving route is determined based on its location information and the target user's current location.

[0090] The location information of the target clustering result can be the geometric midpoint of all driving endpoints included in the target clustering result, or a location where the distance to each driving endpoint of the target clustering result is basically the same.

[0091] The main difference between location clustering and angle clustering is that angle clustering may ignore location attributes, leading to a decrease in the final reference value. However, this method can still be used, primarily because locations with high driving risks are generally far from the driving destination. Therefore, even if there are two driving destinations, one 5km and the other 20km away, the impact on subsequent steps is minimal. In other words, when the user's current location is close to the driving destination, the angle calculation method is not applicable.

[0092] Next, in step S1013, detour scenarios need to be calculated based on the main driving route and road topology, the target user's current location information, the location information of each driving destination, and traffic information data. Specifically, firstly, based on the target user's current location information, the location information of the driving destination, and the road topology, the route for the target user to move to the driving destination is calculated. Then, the overlap and route reversal situations between this route and the main driving route are compared. Finally, the detour ratio is calculated based on the overlap and route reversal situations. The overlap here can be calculated by determining the image similarity between the route and the main driving route and calculating the detour ratio based on the similarity (the higher the similarity, the lower the detour ratio), or by calculating the distance between the two and calculating the detour ratio based on the distance (the greater the distance, the greater the detour ratio). Route reversal situations refer to whether the same road is used for reversal during the movement from the driving destination back to the main driving route. The more frequent these situations, the more it indicates that the corresponding driving destination should not be used as the final driving destination. Traffic information data mainly reflects road congestion and can also be used to determine the probability of road selection and the probability of detours.

[0093] When the main driving path is a region-type path, the central axis of that region (the line segment with the user's current location as the starting point and the target cluster result's location as the ending point) can be directly used as a reference for calculating similarity. Alternatively, the region itself can be used as the calculation reference. If the region is used, as long as the user's route to the driving destination returns to that region, it is no longer considered a detour. Conversely, if the route exceeds the region, the length and distance of the excess route (the distance beyond the region's boundary) are calculated, and then the detour rate is calculated based on the length and distance of the excess route.

[0094] In steps S1014 and S1015, after determining the detour ratio, the driving destinations with the lowest detour ratios can be taken as the final driving destinations, and the driving path of the target user to the final driving destinations can be taken as the final output of this step.

[0095] In fact, steps S1013-S1015 are still filtering, selecting the most suitable driving destination for the target user from multiple driving destinations as the final driving destination, specifically based on the travel route that the user generally will not choose to take a detour.

[0096] After determining the driving route, step S102 allows for the assessment of driving risks for each route across different road segments. Specifically, training samples need to be generated first. These training samples include the following data: basic road segment data, historical accident data, real-time traffic information data, environmental dynamics data, and driving habit data. Basic road segment data includes road geometry parameters (slope, radius of curvature, lane width), road type (bridges, tunnels, sharp bends, etc.), and traffic facilities (guardrails, traffic signs, traffic lights). Historical accident data includes the time, location, frequency, severity, and type of accidents (rear-end collisions, rollovers, etc.). Real-time traffic information data includes real-time traffic flow, speed distribution, vehicle spacing, and vehicle type ratio. Environmental dynamics data includes real-time weather (rain, snow, fog), visibility, and road surface dryness / iciness. Driving habit data includes average vehicle speed, frequency of rapid acceleration / braking, and frequency of lane changes. These data mainly reflect the correlation between historical accidents and other types of data. Among them, driving habit data can reflect the probability of accidents corresponding to different driving styles. Before using this data, it is necessary to obtain the driving behavior habits of the drivers to be analyzed (the drivers of the vehicles in this solution). The acquisition of this data needs to be limited to a minimum time; otherwise, due to the small sample size, the determination of the user's driving style may be inaccurate.

[0097] There are two ways to divide road segments: one is by natural road segments, and the other is by accident segments. Dividing by natural road segments utilizes road connectivity; that is, when there is a fork in the road or a traffic light, a road is divided into two roads based on the location of the traffic light or fork. Dividing by accident segments requires first determining the location of traffic accidents. Based on the distribution of traffic accident locations, all road segments are divided into high-frequency accident segments, low-frequency accident segments, and accident-free segments. This division is not based on road connectivity, but rather on accident density. For example, even within a minimum road segment defined by road connectivity, the frequency of accidents varies at different locations, so it can be further divided according to density. Figure 3 As shown, the classification strategy is illustrated. The roads in this diagram are connected roads without traffic lights or intersections. Black dots represent roads where accidents have occurred. Therefore, we can deduce that Class A roads have no accidents, Class B roads have a low accident frequency, and Class C roads have a high accident probability. In practice, the accident level dimension can be further increased, and the scheme can be implemented using a weighted calculation method. That is, step S102 can be implemented as follows:

[0098] Obtain historical accident data; historical accident data should at least include the location and severity of the accident.

[0099] Natural roads are segmented based on historical accident data to identify multiple road segments, where the accident frequency differs between adjacent road segments.

[0100] Obtain training samples, which include the following data for each road segment: basic road segment data, historical accident data, real-time traffic information data, environmental dynamic data, and driving habit data;

[0101] The training samples are input into the untrained target mathematical model to complete the training of the target mathematical model; the target mathematical model is a logistic regression model.

[0102] Road information is input into the trained target mathematical model to generate driving risks for each different road segment.

[0103] After determining the driving risk of each road segment, in step S103, the signal strength of each high-risk road segment (the road segment whose driving risk exceeds the first preset value) can be analyzed.

[0104] Specifically, the method for analyzing the signal strength of each road segment mainly relies on the signal strength uploaded by each vehicle under the vehicle-to-everything (V2X) network. That is, step S103 can be implemented as follows:

[0105] Obtain signal strength information and vehicle location information uploaded by vehicles on different road sections;

[0106] A first signal distribution map is generated based on signal strength information and vehicle location information;

[0107] The first signal distribution map is corrected based on the signal strength information uploaded by the roadside signal detectors and the signal strength distribution pattern determined based on geographical information, so as to determine the second signal distribution map;

[0108] Based on the second signal distribution map, determine the signal strength in each road segment where the driving risk exceeds the first preset value, and designate road segments with signal strength below the second preset value as target road segments.

[0109] There are two types of roadside signal detectors: one is a fixed roadside device, and the other is a dynamic detection vehicle.

[0110] If the deployment of fixed roadside monitoring points (static reference points) is to be carried out, professional signal monitoring equipment (such as Rohde & Schwarz fixed monitoring terminals) can be deployed at key locations such as urban main roads, highway sections, and tunnel entrances and exits. This equipment is calibrated and the signal acquisition accuracy error is ≤±2dBm, serving as a "static reference station".

[0111] Calibration logic: Periodically (e.g., weekly), the baseline signal strength of a fixed monitoring point on a certain road segment is compared with the average signal strength uploaded by vehicles on that road segment, and the deviation coefficient is calculated (e.g., if the average vehicle data is 5dBm lower than the baseline value, then 5dBm is added to the data of all vehicles on that road segment for calibration).

[0112] Advantages: It covers key scenarios, and the calibrated data can offset errors caused by differences in hardware between different vehicles.

[0113] If dynamic testing vehicles are used, a small number of "calibration vehicles" (e.g., 10-20 vehicles per city) can be deployed in conjunction with operators or third-party testing organizations. These vehicles are equipped with calibrated professional signal acquisition equipment and ordinary on-board units (OBUs). The calibration vehicles travel along a preset route, simultaneously collecting "professional equipment data" and "on-board OBU data".

[0114] Calibration logic: Establish a mapping model (such as a linear regression model) between "vehicle OBU data and professional equipment data" to generate specific calibration coefficients for different brands and models of vehicle OBU modules. For example, if the RSRP value collected by a certain brand of OBU is generally 8dBm lower than that of professional equipment, then all subsequent data uploaded from vehicles of that brand will be calibrated by adding 8dBm.

[0115] Advantages: Covers a wider range of road sections, dynamically calibrates hardware deviations of different vehicle models, and improves the accuracy of data across all scenarios.

[0116] Further verification can be performed using data from the operator. The specific logic is as follows: if a road segment is within the coverage area of ​​a base station (≤500 meters away from the base station) and the base station load is ≤70% (no congestion), the signal strength collected by the vehicle network should be ≥-90dBm (4G scenario). If it is lower than -110dBm, it is necessary to check whether it is due to data abnormality or base station failure. If the base station is in a congested state (load ≥90%), the signal strength may be normal, but the transmission rate is low. It is necessary to separately mark it as "base station congestion" to avoid misjudging it as a weak signal.

[0117] The signal strength distribution pattern determined based on geographic information depends on the fact that the signal strength distribution pattern is essentially influenced by "spatial correlation" and "propagation attenuation characteristics":

[0118] Spatial correlation, such as the signal strength of adjacent road sections (e.g., within 10-50 meters), should be continuous and there should be no unexplained sudden changes (except in special scenarios such as tunnel entrances and exits).

[0119] Propagation attenuation characteristics include signal strength decreasing with increasing distance from the base station, and the attenuation rate being strongly correlated with the scene (e.g., slow attenuation in open roads and fast attenuation in dense urban areas).

[0120] Based on the above, the obtained first signal distribution map can be corrected (such as reducing or increasing certain areas; the main purpose of the correction is to smooth out the differences caused by the detection equipment and the differences caused by the inconsistent standards in different areas, as well as to adjust some obviously incorrect values).

[0121] Then the target road segment can be selected based on the values ​​of the second signal distribution map.

[0122] In step S104, the probability of the user reaching each target road segment and the remaining time to reach it are calculated in real time based on the vehicle's driving information. This step can be implemented in a simpler way: the closer the vehicle is to the target road segment, the higher the probability of arrival. The distance can be calculated as either a straight-line distance or an actual distance (first calculate the path, then calculate the distance traveled along that path). The remaining time to reach also requires calculating the path first, then the distance, and finally the remaining time to reach. Furthermore, since the driving path has already been calculated in the preceding step S101, this calculated path can be used directly to complete the calculation in this step, without needing to repeat the calculation.

[0123] Step S105 is the preparation for the rescue operation, which is when the arrival probability reaches the minimum required value and the remaining time for arrival is less than the fourth preset value (usually 5 minutes; this value can also be adjusted according to the different roads, such as setting different values ​​for urban roads, rural roads, highways, and rural roads; furthermore, this fourth preset value can be adjusted according to the density of the target road segment in the area where the vehicle is located). The encrypted object in this step is the rescue information, which includes the following: location information, accident information, emergency demand information, on-site environment information, vehicle information, and personnel information.

[0124] The information includes: location information (latitude and longitude, altitude, and surrounding landmarks); personnel information (number of people, names, contact information, drug allergies, etc., which is helpful for emergency rescue); accident information (accident type, rollover, fall, loss of contact, trapped, etc.) and equipment information (vehicle status, remaining mobile phone battery, etc.); emergency needs information (missing supplies, special personnel (children, pregnant women, etc.); on-site environmental information (natural environment: weather (heavy rain, heavy snow, strong winds, visibility), terrain complexity, and whether there are surrounding dangers); and vehicle information (vehicle type, license plate number, color, and remaining fuel / battery power). Of the above information, vehicle and personnel information can be prepared in advance, meaning the driver needs to input it beforehand. The other information cannot be entered in advance, but can be prepared based on statistical data. This is mainly because the target road segment the vehicle is about to reach is known, and there are historical accident records for that target road segment. Therefore, the location information, accident information, emergency needs information, and on-site environmental information can be generated using historical accidents on that target road segment.

[0125] That is, the method provided in this application also includes the following: generating first rescue information for the current accident based on historical accident information that has occurred on the target road segment. The first rescue information includes: location information, accident information, emergency demand information, and on-site environmental information.

[0126] The first rescue information is adjusted based on the driver's driving habits data.

[0127] Secondary rescue information is generated based on the vehicle information and the information of the driver and passengers in the vehicle. The secondary rescue information includes vehicle information and personnel information.

[0128] The use of driving habit data to adjust the first rescue information mainly refers to the fact that driving habit data can reflect the driver's driving style. Based on this driving style, it is easier to reflect what kind of accident may occur, and thus this type of accident can be used as the main information to be prepared.

[0129] Once the rescue information is confirmed, it can be broken down, which may include the following: generating a dynamic AES key based on the vehicle processor chip hardware code, the current timestamp, and the identification code of the target road segment (this AES key is automatically updated each time the vehicle approaches the target road segment); encrypting the rescue information using the AES key to obtain an encrypted data packet and a decryption key (since AES encryption technology is symmetric encryption, the decryption key is also this key).

[0130] If asymmetric encryption algorithms such as RSA or ECC are used, an encryption key needs to be generated first, followed by a corresponding decryption key, and finally the encryption key is used to encrypt the rescue information.

[0131] Furthermore, the rescue information can also include image information, including video information from the in-vehicle dashcam and the external dashcam, as well as the user's image information (information stored in image form, such as ID card information). When encrypting the vehicle's rescue information to obtain the encrypted data packet and decryption key, the specific processing can be carried out as follows:

[0132] Foreground extraction is performed on the image information in the rescue information to determine the background and foreground images;

[0133] Keyframes are extracted from video information in the rescue data to identify keyframe and non-keyframe images;

[0134] Background and non-keyframe images are encrypted within the rescue information to obtain encrypted data packets, while foreground and keyframe images are carried in the decryption key.

[0135] In this way, the encrypted data packet no longer directly carries the foreground image and keyframe image, which contain a lot of important information. This makes it difficult for image information to be leaked. At the same time, the foreground image and keyframe image are carried in the decryption key, which avoids the problem of data loss caused by sending the foreground image and keyframe image separately.

[0136] In step S106, after detecting that the vehicle is closer to the target road segment (closer than in step S105) and the remaining time to reach it is shorter, a short-range network can be established. Typically, a P2P connection can be chosen as the short-range network method. Then, encrypted data packets are sent to nearby rescue vehicles, which are those that have established a short-range network. However, the encrypted data packets can be sent without using the established P2P connection, but rather through a general transmission method (Internet transmission). This is mainly because the encrypted data packets are relatively large, so a more robust transmission method is preferable. Alternatively, the established short-range communication connection can be used, as the most dangerous situation has not yet been reached, and therefore, traditional methods can still be employed.

[0137] It should be noted that the encrypted data packets sent in this step are not sent to just one rescue vehicle, but to multiple rescue vehicles separately. This is mainly because it is uncertain which rescue vehicles will ultimately receive the decryption key (due to poor signal at the location of the rescue vehicles, they may not be able to receive it).

[0138] Step S106 can be achieved in the following way:

[0139] Vehicles initiate network broadcasts at predetermined intervals;

[0140] After receiving a rescue response request from another vehicle, read the location information and verification information from the rescue response request;

[0141] Determine whether each other vehicle is moving in the same direction based on its location information;

[0142] Based on whether other vehicles are moving in the same direction and verification information, select a rescue vehicle from the other vehicles and form a close-range network with the rescue vehicle;

[0143] The encrypted data packets are sent to multiple rescue vehicles via the Internet or through a completed short-range network.

[0144] The location information filtering mainly determines which rescue vehicles are closest to the vehicle being rescued, and those vehicles should be selected as rescue vehicles. Information verification primarily verifies the legitimacy of the rescue response request, preventing devices that steal information from being mistakenly identified as rescue vehicles. This can be done by using the acquired location information to verify whether other vehicles are moving, or further verify whether they are moving in the same direction, and selecting vehicles moving in the same direction as the rescue vehicle. This is mainly because there may be several minutes between steps S106 and S107, and vehicles moving in the opposite direction may not be able to provide timely assistance.

[0145] In step S107, when the vehicle is detected to be closer to the target road segment (closer than in step S106) and the remaining time to reach it is shorter (reaching the maximum warning value), the Internet signal is already difficult to transmit. At this point, the established short-range communication connection can be used to send the decryption key to the rescue vehicle, so that the rescue vehicle can use the decryption key to decrypt the encrypted data packet obtained in step S106, thereby obtaining the required rescue information and completing the rescue.

[0146] Specifically, in addition to short-range networking (such as LoRa networking and Bluetooth networking), the decryption key can also be transmitted via BeiDou short messages. The reason for adopting this transmission method is that BeiDou short messages are more practical, easier to transmit small amounts of data, and easier to carry out rescue operations.

[0147] In practice, the preferred approach is to use close-range networking and simultaneous transmission of BeiDou short messages.

[0148] Based on the same technical concept, this application embodiment also provides an information transmitting device, as shown in Figure 4, the device comprising:

[0149] The first determining module 401 is used to determine multiple driving routes based on the user's driving habit data and real-time traffic information data after detecting that the vehicle is in a driving state.

[0150] The input module 402 is used to input the road information of each driving path into the pre-trained driving risk identification model to determine the driving risk of each driving path in different road segments.

[0151] The second determining module 403 is used to determine the signal strength of road segments in each driving path where the driving risk exceeds the first preset value, and to take road segments with signal strength lower than the second preset value as target road segments.

[0152] The first calculation module 404 is used to calculate the probability of a user reaching each target road segment and the remaining time of arrival based on the real-time vehicle driving information.

[0153] The encryption module 405 is used to encrypt the vehicle's rescue information after the arrival probability and the remaining arrival time both meet the first warning value, so as to obtain an encrypted data packet and a decryption key.

[0154] The module 406 is used to establish a short-range communication connection with the rescue vehicle through short-range networking technology after the arrival probability and the remaining arrival time both meet the second warning value, and send the encrypted data packet to the rescue vehicle within a preset distance range.

[0155] The sending module 407 is used to send the decryption key to the rescue vehicle via a short-range communication connection after the arrival probability and the remaining arrival time both meet the third warning value, so that the rescue vehicle can decrypt the encrypted data packet using the decryption key and complete the rescue based on the decryption result.

[0156] Optional, also includes:

[0157] The second calculation module is used to calculate the vehicle's speed based on the real-time acquired position signal;

[0158] The judgment module is used to determine whether the vehicle is in motion based on whether the moving speed is greater than a third preset value if the continuity of the moving signal acquisition is normal.

[0159] The acquisition module is used to trigger the dashcam's camera to acquire images inside and outside the vehicle if the continuity of mobile signal acquisition is abnormal.

[0160] The third determining module is used to determine that the vehicle is currently in motion if facial features are identified from the in-vehicle image and image movement is identified from the external image.

[0161] Optionally, when the input module 402 inputs the road information of each driving path into the pre-trained driving risk identification model to determine the driving risk of each driving path in different road segments, it is specifically used for:

[0162] Obtain historical accident data; historical accident data should at least include the location and severity of the accident.

[0163] Natural roads are segmented based on historical accident data to identify multiple road segments, where the accident frequency differs between adjacent road segments.

[0164] Obtain training samples, which include the following data for each road segment: basic road segment data, historical accident data, real-time traffic information data, environmental dynamic data, and driving habit data;

[0165] The training samples are input into the untrained target mathematical model to complete the training of the target mathematical model; the target mathematical model is a logistic regression model.

[0166] Road information is input into the trained target mathematical model to generate driving risks for each different road segment.

[0167] Optionally, when the second determining module 403 determines the signal strength in road segments where the driving risk exceeds a first preset value in each driving path, and designates road segments with signal strength below a second preset value as target road segments, it specifically performs the following:

[0168] Obtain signal strength information and vehicle location information uploaded by vehicles on different road sections;

[0169] A first signal distribution map is generated based on signal strength information and vehicle location information;

[0170] The first signal distribution map is corrected based on the signal strength information uploaded by the roadside signal detectors and the signal strength distribution pattern determined based on geographical information, so as to determine the second signal distribution map;

[0171] Based on the second signal distribution map, determine the signal strength in each road segment where the driving risk exceeds the first preset value, and designate road segments with signal strength below the second preset value as target road segments.

[0172] Optionally, the rescue information includes the following: location information, accident information, emergency demand information, on-site environment information, vehicle information, and personnel information.

[0173] Optional, also includes:

[0174] The first generation module is used to generate the first rescue information for this accident based on the historical accident information that has occurred on the target road segment. The first rescue information includes: location information, accident information, emergency demand information, and on-site environment information.

[0175] The adjustment module is used to adjust the first rescue information based on the driver's driving habits data.

[0176] The second generation module is used to generate second rescue information based on the vehicle information and the information of the driver and passengers in the vehicle. The second rescue information includes vehicle information and personnel information.

[0177] Optionally, the rescue information also includes image information, including video information from the in-vehicle dashcam and the external dashcam, as well as the user's image information; the encryption module, when used to encrypt the vehicle's rescue information to obtain encrypted data packets and decryption keys, is specifically used for:

[0178] Foreground extraction is performed on the image information in the rescue information to determine the background and foreground images;

[0179] Keyframes are extracted from video information in the rescue data to identify keyframe and non-keyframe images;

[0180] Background and non-keyframe images are encrypted within the rescue information to obtain encrypted data packets, while foreground and keyframe images are carried in the decryption key.

[0181] Optionally, when the establishment module 406 is used to establish a short-range communication connection with the rescue vehicle through short-range networking technology and send the encrypted data packet to the rescue vehicle within a preset distance range, it is specifically used for:

[0182] Vehicles initiate network broadcasts at predetermined intervals;

[0183] After receiving a rescue response request from another vehicle, read the location information and verification information from the rescue response request;

[0184] Determine whether each other vehicle is moving in the same direction based on its location information;

[0185] Based on whether other vehicles are moving in the same direction and verification information, select a rescue vehicle from the other vehicles and form a close-range network with the rescue vehicle;

[0186] The encrypted data packets are sent to multiple rescue vehicles via the Internet or through a completed short-range network.

[0187] Figure 5 A schematic diagram of an electronic device provided in this application embodiment includes: a processor 501, a memory 502, and a bus 503. The memory 502 stores machine-readable instructions executable by the processor 501. When the electronic device runs the above-described information processing method, the processor 501 and the memory 502 communicate through the bus 503. The processor 501 executes the machine-readable instructions to perform the steps of the method described in Embodiment 1.

[0188] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described device and electronic equipment can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0189] In the several embodiments provided in this application, it should be understood that the disclosed methods, apparatus, and electronic devices can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or modules may be electrical, mechanical, or other forms.

[0190] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0191] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0192] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a processor-executable, non-volatile, computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0193] Finally, it should be noted that the above-described embodiments are merely specific implementations of this application, used to illustrate the technical solutions of this application, and not to limit them. The scope of protection of this application is not limited thereto. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments, or make equivalent substitutions for some of the technical features, within the scope of the technology disclosed in this application. Such modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of protection of the claims.

Claims

1. A method for sending information, characterized in that, include: After detecting that the vehicle is in motion, multiple driving routes are determined based on the user's driving habits data and real-time traffic information data. The road information for each driving route is input into a pre-trained driving risk identification model to determine the driving risk of each driving route in different road segments; Determine the signal strength of road segments in each driving route where the driving risk exceeds a first preset value, and designate road segments with signal strength below a second preset value as target road segments; Based on the real-time vehicle driving information, calculate the probability of a user reaching each target road segment and the remaining time to reach the target road segment; After the arrival probability and remaining arrival time both meet the first warning value, the vehicle rescue information is encrypted to obtain an encrypted data packet and a decryption key. Once the arrival probability and remaining arrival time both meet the second warning value, a short-range communication connection is established with the rescue vehicle through short-range networking technology, and the encrypted data packet is sent to the rescue vehicle within a preset distance range; Once the arrival probability and remaining arrival time both meet the third warning value, the decryption key is sent to the rescue vehicle via a short-range communication connection, so that the rescue vehicle can decrypt the encrypted data packet using the decryption key and complete the rescue based on the decryption result.

2. The method according to claim 1, characterized in that, Also includes: The vehicle's speed is calculated based on the real-time location signals acquired. If the continuity of the mobile signal acquisition is normal, then the vehicle is judged to be in motion based on whether the mobile speed is greater than the third preset value. If the continuity of mobile signal acquisition is abnormal, the dashcam's camera will be triggered to acquire images of the inside and outside of the vehicle. If facial features are identified from the in-vehicle image and image movement is detected from the external image, then it is determined that the vehicle is currently in motion.

3. The method according to claim 1, characterized in that, The step of inputting road information for each driving path into a pre-trained driving risk identification model to determine the driving risk of each driving path in different road segments includes: Obtain historical accident data; historical accident data should at least include the location and severity of the accident. Natural roads are segmented based on historical accident data to identify multiple road segments, where the accident frequency differs between adjacent road segments. Obtain training samples, which include the following data for each road segment: basic road segment data, historical accident data, real-time traffic information data, environmental dynamic data, and driving habit data; The training samples are input into the untrained target mathematical model to complete the training of the target mathematical model; the target mathematical model is a logistic regression model. Road information is input into the trained target mathematical model to generate driving risks for each different road segment.

4. The method according to claim 1, characterized in that, The step of determining the signal strength of road segments in each driving path where the driving risk exceeds a first preset value, and designating road segments with signal strength below a second preset value as target road segments, includes: Obtain signal strength information and vehicle location information uploaded by vehicles on different road sections; A first signal distribution map is generated based on signal strength information and vehicle location information; The first signal distribution map is corrected based on the signal strength information uploaded by the roadside signal detectors and the signal strength distribution pattern determined based on geographical information, so as to determine the second signal distribution map; Based on the second signal distribution map, determine the signal strength in each road segment where the driving risk exceeds the first preset value, and designate road segments with signal strength below the second preset value as target road segments.

5. The method according to claim 1, characterized in that, The rescue information includes the following: location information, accident information, emergency needs information, on-site environment information, vehicle information, and personnel information.

6. The method according to claim 5, characterized in that, Also includes: The first rescue information for this accident is generated based on the historical accident information that has occurred on the target road section. The first rescue information includes: location information, accident information, emergency demand information and on-site environment information. The first rescue information is adjusted based on the driver's driving habits data. Secondary rescue information is generated based on the vehicle information and the information of the driver and passengers in the vehicle. The secondary rescue information includes vehicle information and personnel information.

7. The method according to claim 5, characterized in that, The rescue information also includes image information, which includes video information from the in-vehicle dashcam and the external dashcam, as well as the user's image information; The process of encrypting the vehicle's rescue information to obtain an encrypted data packet and a decryption key includes: Foreground extraction is performed on the image information in the rescue information to determine the background and foreground images; Keyframes are extracted from video information in the rescue data to identify keyframe and non-keyframe images; Background and non-keyframe images are encrypted within the rescue information to obtain encrypted data packets, while foreground and keyframe images are carried in the decryption key.

8. The method according to claim 1, characterized in that, The step of establishing a short-range communication connection with the rescue vehicle through short-range networking technology and sending the encrypted data packet to the rescue vehicle within a preset distance range includes: Vehicles initiate network broadcasts at predetermined intervals; After receiving a rescue response request from another vehicle, read the location information and verification information from the rescue response request; Determine whether each other vehicle is moving in the same direction based on its location information; Based on whether other vehicles are moving in the same direction and verification information, select a rescue vehicle from the other vehicles and form a close-range network with the rescue vehicle; The encrypted data packets are sent to multiple rescue vehicles via the Internet or through a completed short-range network.

9. An information transmitting device, characterized in that, include: The first determining module is used to determine multiple driving routes based on the user's driving habit data and real-time traffic information data after detecting that the vehicle is in motion. The input module is used to input the road information of each driving path into the pre-trained driving risk identification model to determine the driving risk of each driving path in different road segments. The second determining module is used to determine the signal strength of road segments in each driving path where the driving risk exceeds the first preset value, and to take road segments with signal strength lower than the second preset value as target road segments. The first calculation module is used to calculate the probability of a user reaching each target road segment and the remaining time of arrival based on the real-time vehicle driving information. An encryption module is used to encrypt the vehicle's rescue information after the arrival probability and remaining arrival time both meet the first warning value, to obtain an encrypted data packet and a decryption key; The module is used to establish a short-range communication connection with the rescue vehicle through short-range networking technology after the arrival probability and the remaining arrival time both meet the second warning value, and send the encrypted data packet to the rescue vehicle within a preset distance range. The sending module is used to send a decryption key to the rescue vehicle via a short-range communication connection after the arrival probability and the remaining arrival time both meet the third warning value, so that the rescue vehicle can decrypt the encrypted data packet using the decryption key and complete the rescue based on the decryption result.

10. An electronic device, characterized in that, include: The device includes a processor, a memory, and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is in operation, the processor communicates with the memory via the bus, and the machine-readable instructions, when executed by the processor, perform the steps of the method as described in any one of claims 1 to 8.

Images