Collision early warning method, system and equipment

Abstract

The invention relates to the technical field of intelligent transportation, and discloses a collision early warning method, system and device. The method comprises the steps of obtaining a high-precision map and performing fragmentation processing on the high-precision map to obtain a preset map database; determining a first range according to the real-time position of the target vehicle, and obtaining a plurality of first map fragments in the first range based on a preset map database; in a preset geographic feature library, obtaining geographic feature tags corresponding to the first map fragments; determining a collision early warning range based on the geographic feature tags corresponding to all the first map fragments; acquiring a plurality of corresponding target traffic element data in the collision early warning range, and determining a collision state of the target vehicle based on all the target traffic element data; and based on the collision state, collision warning information is sent to the target vehicle. According to the scheme of the invention, the problems of low efficiency and poor real-time performance in collision early warning are solved, the access and calculation efficiency of traffic data is optimized, real-time collision early warning is realized, and the occurrence rate of traffic accidents is reduced.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a collision warning method, system and equipment. Background Art

[0002] At present, when vehicles are driving, the main way to control vehicle activities is through the driver observing the surrounding environment and obtaining traffic information from the vehicle's internal data collection equipment. However, the driver and the vehicle data collection equipment are easily affected by factors such as the traffic environment and weather, resulting in the inability to obtain information about traffic participants in a timely manner, which poses a certain potential danger to the driving safety of traffic participants and the vehicle itself. However, many factors of traffic accident risks cannot be effectively overcome by regulating driver behavior. Therefore, it is proposed to develop an intelligent transportation system to assist drivers in perceiving the status information of surrounding traffic and vehicles. At present, the collision warning technology in the existing intelligent transportation system has problems such as large data volume, slow loading and low matching when dealing with collision warnings between vehicles and surrounding traffic participants, which has a certain impact on the reliability of the warning.

[0003] In summary, there is room for further improvement in the warning reliability of existing collision warning technologies. Summary of the invention

[0004] In view of this, the present invention proposes a collision warning method, system and device, which solve the problems of low efficiency and poor real-time performance in collision warning, optimize the access and calculation efficiency of traffic-related data, realize real-time collision warning, and reduce the incidence of traffic accidents.

[0005] Based on the above purpose, an embodiment of the present invention provides a collision warning method, which specifically includes the following steps: Obtain a high-precision map and slice it to obtain a preset map database; Determine a first range according to the real-time position of the target vehicle, and acquire a plurality of first map slices located within the first range based on the preset map database; In a preset geographic feature library, obtain the geographic feature labels corresponding to each of the first map slices; Determining a collision warning range based on the geographic feature tags corresponding to all the first map slices; Acquire a plurality of target traffic element data corresponding to the collision warning range, and determine the collision state of the target vehicle based on all the target traffic element data; Based on the collision state, collision warning information is sent to the target vehicle.

[0006] In some implementations, the step of determining the collision warning range based on the geographic feature tags corresponding to all the first map slices includes: Based on the geographic feature labels corresponding to the first maps, determining whether they are first type labels; If it is a first type tag, the first map slice is used as the target map slice; If it is not a first type of label, recursively search for other map slices that are adjacent to the first map slice and have the same corresponding geographic feature label, and use the first map slice and the other map slices as target map slices respectively; Based on all the target map slices, the collision warning range is determined.

[0007] In some implementations, the step of obtaining data of a number of target traffic elements corresponding to the collision warning range includes: Based on each traffic sensing device, real-time collection of first traffic element data; According to the latitude and longitude coordinates and the impact range corresponding to each of the first traffic element data, determining whether there is the target map segment matching the first traffic element data in the collision warning range; In response to the existence of the target map slice matching the first traffic element data, the traffic element data is used as the target traffic element data.

[0008] In some implementations, the step of determining the collision state of the target vehicle based on all the target traffic element data includes: Determining a traffic flow state based on all of the target traffic element data; Filtering out target traffic participants and their location information from all the target traffic element data, and determining the distance and direction between the target traffic participants and the target vehicle based on the location information; A collision state of the target vehicle is determined based on the distance, the orientation, and the traffic flow state.

[0009] In some implementations, the step of obtaining a high-precision map and slicing it to obtain a preset map database includes: The high-precision map is obtained, and starting from the latitude and longitude origin in the high-precision map, grid division is performed according to a preset interval to obtain a plurality of second map slices; Obtaining the tile description information corresponding to each of the second map tiles, and constructing the preset map database based on all the tile description information; The tile description information includes a map tile identifier, a map sheet number, a map tile graphic, and an associated perception group number.

[0010] In some implementations, the step of constructing the preset geographic feature library further includes: Traversing all geographic attribute information in the high-precision map; Based on each of the geographic attribute information, determining whether there is a third map slice matching it in the preset map database; In response to the existence of a third map slice matching the third map slice, using the geographic attribute information as a geographic feature tag of the third map slice; The preset geographic feature library is constructed based on the third map slice and the corresponding geographic feature tag.

[0011] In some implementations, the step of acquiring a plurality of first map slices within the first range based on the preset map database includes: Based on geometric intersection calculation, all map slices intersecting with the first range are searched in the preset map database and are used as the first map slices.

[0012] In some implementations, the step of screening out target traffic participants and their location information from all the target traffic element data includes: Determining whether the type of traffic participant in each of the target traffic element data is a vulnerable traffic participant; In response to the type of the traffic participant being a vulnerable traffic participant, the vulnerable traffic participant is taken as the target traffic participant, and the location information of the vulnerable traffic participant is acquired.

[0013] In some embodiments, the step of determining the collision state of the target vehicle based on the distance and the orientation and the traffic flow state comprises: In response to the distance and the orientation meeting a collision threshold condition and the traffic flow state being a preset traffic state, determining that the collision state is a first state in which a collision warning message should be sent to the target vehicle; In response to the distance and the orientation not meeting a collision threshold condition or the traffic flow state not being a preset traffic state, determining the collision state as a second state in which collision warning information is not sent to the target vehicle.

[0014] Another aspect of the embodiment of the present invention further provides a collision warning system, including: A map slicing processing unit configured to obtain a high-precision map and perform slicing processing on it to obtain a preset map database; An initial slice acquisition unit, configured to determine a first range according to a real-time position of the target vehicle, and acquire a plurality of first map slices located within the first range based on a preset map database; A geographic feature determination unit, configured to obtain, from a preset geographic feature library, geographic feature labels corresponding to each of the first map slices; a warning range determining unit, configured to determine a collision warning range based on the geographic feature tags corresponding to all the first map slices; A collision state determination unit configured to obtain a plurality of target traffic element data corresponding to the collision warning range, and determine the collision state of the target vehicle based on all the target traffic element data; The collision warning unit is configured to send collision warning information to the target vehicle based on the collision state.

[0015] According to another aspect of an embodiment of the present invention, a computer device is provided, comprising: at least one processor; and a memory, wherein the memory stores a computer program executable on the processor, and the computer program implements the steps of the above method when executed by the processor.

[0016] The present invention has at least the following beneficial technical effects: based on the real-time position of the target vehicle, the first map slice that is around the target vehicle and within a fixed range is quickly loaded, and combined with the geographic feature label corresponding to the first map slice, the data loading range is dynamically adjusted to obtain the collision warning range, and the map slice and its target traffic element data are loaded on demand according to the collision warning scene, avoiding the loading of full data and unnecessary consumption of computing resources, reducing the loading time of traffic data around the vehicle, and realizing efficient spatial indexing and dynamic update mechanism of map slices. The collision warning method of the present invention improves the real-time and efficiency of collision warning calculation of target traffic participants by combining map slicing technology, geographic feature labeling and traffic element data binding, meets the needs of intelligent networked vehicles for efficient and accurate data processing, realizes real-time collision warning, provides reliable basic support for intelligent driving, and reduces the incidence of traffic accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A block diagram of an embodiment of a collision warning method provided by the present invention; Figure 2 A schematic diagram of an embodiment of the execution process of the collision warning method provided by the present invention; Figure 3A schematic diagram of an embodiment of a collision warning system provided by the present invention; Figure 4 A schematic structural diagram of an embodiment of a computer device provided by the present invention. DETAILED DESCRIPTION

[0019] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the embodiments of the present invention are further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0020] It should be noted that all expressions using "first" and "second" in the embodiments of the present invention are for distinguishing two non-identical entities with the same name or non-identical parameters. It can be seen that "first" and "second" are only for the convenience of expression and should not be understood as limitations on the embodiments of the present invention. The subsequent embodiments will not explain this one by one.

[0021] With the development of intelligent connected vehicles, the Intelligent Connected Vehicle Roadmap Figure 2 .0》 defines three levels of connected data services - connected auxiliary information interaction, connected collaborative perception, and connected collaborative decision-making and control, aiming to improve the safety and capability expansion of intelligent driving. Among them, collision warning with vulnerable traffic participants is one of the important application scenarios of connected collaborative perception services, and vulnerable traffic participants include pedestrians, bicycles, etc. In order to achieve high real-time collision warning services, it is necessary to rely on the powerful computing power of the cloud platform and intelligent and efficient data processing methods to quickly obtain data such as the status of traffic participants around the vehicle, traffic flow, and accidents, so as to provide vehicles with accurate warning information. This puts forward new technical requirements for the processing of map data, the correlation analysis between traffic elements and vehicle positions, and real-time calculation. However, in the existing technologies, when processing collision warnings between vehicles and surrounding traffic participants, there are problems such as large map data volume, slow loading, low matching degree, and high data latency requirements, which cannot meet the needs of intelligent networked vehicles for efficient and accurate data processing.

[0022] Based on the above purpose, the first aspect of the embodiment of the present invention provides a collision warning method. Figure 1 As shown, the collision warning method includes the following steps: Step S100, obtaining a high-precision map and performing segmentation processing on it to obtain a preset map database; Step S200, determining a first range according to the real-time position of the target vehicle, and acquiring a plurality of first map slices within the first range based on a preset map database; Step S300, obtaining geographic feature labels corresponding to each first map slice in a preset geographic feature library; Step S400, determining a collision warning range based on the geographic feature tags corresponding to all first map slices; Step S500, obtaining corresponding target traffic element data within the collision warning range, and determining the collision state of the target vehicle based on all target traffic element data; Step S600: sending collision warning information to the target vehicle based on the collision status.

[0023] In some embodiments, based on the latitude and longitude origin of the high-precision map, the high-precision map is divided according to the fixed spacing rule to obtain multiple map slices to complete the slice processing of the high-precision map, and then the preset map database is constructed using the slice description information corresponding to the multiple map slices obtained by the division. The first range is the initial collision warning range preset by the user according to actual business needs. For example, the first range can be a circular range with a radius of 300 meters drawn with the real-time position of the target vehicle as the center. The first range may include one or more map slices. Using geometric intersection calculation, a set of map slices intersecting with the first range is found in the preset map database, that is, a number of first map slices are obtained. The preset map database stores multiple map slices and their corresponding slice description information, and the slice description information includes a map slice ID (identification), a map sheet number, a map slice graphic, and an associated perception group number. Among them, the map slice ID is a unique identifier of the map slice, which is convenient for fast indexing and querying. The map sheet number is a grid position calculated according to the latitude and longitude coordinates of the corresponding map slice, recorded in the (X, Y) coordinate format, X can be longitude, and Y can be latitude. The map slice graph is used to record the geometric shape of the map slice, which can be a closed geometric figure composed of multiple vertices. The associated perception group number is the roadside perception device associated with the map slice, such as a sensor or camera, which can determine the map slice associated with the roadside perception device based on its latitude and longitude.

[0024] In some embodiments, the preset geographic feature library includes the correspondence between each map slice and the geographic feature tag, and the geographic feature tag corresponding to the map slice can be quickly obtained through the map slice ID. Geographic feature tags include Link, Lane, Junction, Stopline, etc. It should be noted that each map slice can correspond to one or more geographic feature tags, and the preset geographic feature library allows one-to-many records. According to the collision warning scenario, the geographic feature tags are queried on demand to obtain the corresponding map slice and its traffic element data. In one example, if map slice 1 has two geographic feature tags of "ramp" and "road" at the same time, and map slice 2 has two geographic feature tags of "junction" and "road" at the same time, when the collision warning calculation is performed in the ramp scenario, the traffic element data in map slice 1 will be obtained, and when the collision warning calculation is performed in the road scenario, the traffic element data in map slice 1 and map slice 2 will be obtained. Among them, the collision warning scenario is determined according to the geographic feature tag corresponding to the first map slice.

[0025] In some implementations, the data loading range is dynamically adjusted in combination with the geographic feature tag corresponding to the first map slice to update the first range to obtain the final collision warning range, so as to achieve on-demand loading of map slices and their target traffic element data according to the collision warning scenario, thereby avoiding full data loading and unnecessary consumption of computing resources, and reducing the loading time of traffic data around the vehicle. Furthermore, the acquired target traffic element data is subjected to traffic participant type filtering, distance threshold condition calculation, relative position relationship calculation, traffic flow and traffic accident status judgment, etc., to obtain the collision status of the target vehicle, thereby achieving real-time delivery of collision warning information to the target vehicle.

[0026] The collision warning method of the present invention improves the real-time and efficiency of collision warning calculation of target traffic participants by combining map slicing technology, geographic feature labeling and traffic element data binding, meets the needs of smart networked vehicles for efficient and accurate data processing, realizes real-time collision warning, provides reliable basic support for smart driving, and reduces the incidence of traffic accidents.

[0027] In some embodiments, the step of determining the collision warning range based on the geographic feature tags corresponding to all first map slices includes: judging whether the geographic feature tags corresponding to each first map are first type tags; if they are first type tags, using the first map slice as the target map slice; if they are not first type tags, recursively searching for other map slices that are adjacent to the first map slice and have the same corresponding geographic feature tags, and using the first map slice and the other map slices as target map slices respectively; and determining the collision warning range based on all target map slices.

[0028] In some embodiments, the first type of label includes an intersection, and the second type of label includes a road, a ramp, and a roundabout, etc. In one example, when the geographic feature label corresponding to the first map slice is an intersection, and the collision warning scene is an intersection, only the traffic element data cached in the first map slice is obtained, that is, only the first map slice is used as the target map slice, and no other map slices need to be associated. When the geographic feature label corresponding to the first map slice is a road, and the collision warning scene is a road, the traffic element data cached in the first map slice and the surrounding map slices with the same geographic feature label are obtained at the same time, that is, the first map slice and the surrounding map slices with the same geographic feature label are respectively used as a target map slice.

[0029] In some embodiments, if there are multiple first map slices, the geographic feature labels corresponding to each first map slice are analyzed respectively to determine the target map slice to be obtained, and all target map slices obtained by analyzing the multiple first map slices are deduplicated to obtain the final collision warning range. For example, the first map slice includes map slice 1 and map slice 2. If the geographic feature labels corresponding to map slice 1 include intersections and roads, and the geographic feature labels corresponding to map slice 2 include roads. For intersection collision warning scenarios, map slice 1 is used as the target map slice. For road collision warning scenarios, all adjacent map slices centered on map slice 1 and map slice 2 with the geographic feature label of "road" are traversed respectively. If all adjacent map slices centered on map slice 1 and with the geographic feature label "road" are traversed, including map slice 3 and map slice 4, and all adjacent map slices centered on map slice 2 and with the geographic feature label "road" are traversed, including map slice 4 and map slice 5, then the target map slices in the collision warning range are finally obtained as map slice 1, map slice 2, map slice 3, map slice 4 and map slice 5 respectively.

[0030] In some embodiments, the acquisition of traffic element data cached in the first map slice and the map slices with the same geographic feature labels in its surrounding areas is described. With the first map slice as the center, recursively search for map slices with the same map feature labels in the four directions of up, down, left, and right until there are no matching map slices. In one example, assuming that the map slice number of the first map slice is (1,1), the map slice numbers of the map slices in the four directions of up, down, left, and right adjacent to it are (1,2), (1,0), (0,1), and (2,1), respectively. If the map feature label corresponding to the first map slice is road, and the map feature labels corresponding to the adjacent map slices with sheet numbers (1,2) and (0,1) are also roads, then the map slices with sheet numbers (1,1), (1,2) and (0,1) including the first map slice are taken as target map slices respectively, and the map slices with sheet numbers (1,2) and (0,1) are recursively searched for map slices with the same map feature labels in the four directions of up, down, left and right, with the map slices with sheet numbers (1,2) and (0,1) as the center, until no map slices with the same map feature labels are found.

[0031] The collision warning method of the present invention, in combination with the geographic feature tag corresponding to the first map slice, dynamically adjusts the data loading range to obtain the collision warning range, realizes on-demand loading of map slices and their target traffic element data according to the collision warning scenario, avoids full data loading and unnecessary consumption of computing resources, reduces the loading time of vehicle surrounding traffic data, realizes efficient spatial indexing and map slice dynamic update mechanism, and ensures the relevance and integrity of the required data.

[0032] In some embodiments, the step of obtaining a number of target traffic element data corresponding to the collision warning range includes: based on each traffic sensing device, collecting first traffic element data in real time; judging whether there is a target map segment matching the first traffic element data in the collision warning range according to the latitude and longitude coordinates and the influence range corresponding to each first traffic element data; in response to the existence of a target map segment matching the first traffic element data, taking a traffic element data as the target traffic element data.

[0033] In some embodiments, the traffic sensing device includes a roadside sensing device, and the first traffic element data is the traffic element data collected in real time by the roadside sensing device, including the type of traffic participants, the location information of traffic participants, the traffic flow status, etc. In order to collect traffic element data in real time, the time interval of collection can be set to a smaller value according to business requirements, such as 0.2ms. According to the latitude and longitude coordinates and the impact range (such as a circular range or a polygonal range) of the first traffic element data, a geometric intersection calculation with the map slice is performed, so that the first traffic element data is cached in the corresponding map slice. When the result of the geometric intersection calculation is that the first traffic element data is completely contained in the map slice, the first traffic element data is cached separately in the map slice, and the first traffic element data matches a single map slice. When the result of the geometric intersection calculation is that the first traffic element data is located in multiple map slices, the first traffic element data is cached in all relevant map slices, and the first traffic element data matches multiple map slices. For example, the impact range of a traffic accident is a circular area with a radius of 100m. If the circular area covers two map slices, the accident data needs to be cached in both map slices at the same time. Therefore, when the geometric intersection calculation result of the target map slice and the corresponding influence range of the first traffic element data is included or intersected, the target map slice matches the first traffic element data. When the geometric intersection calculation result is that the first traffic element data does not intersect with any map slice, it means that the first traffic element data is not within the valid map range and is not cached.

[0034] The collision warning method of the present invention caches the collected traffic element data into corresponding map slices through geographic coordinate mapping and geometric intersection calculation. The traffic element data cached in each map slice supports real-time dynamic update, which solves the problem of high data latency in the collision warning process in the prior art, improves the real-time and efficiency of collision warning calculation, realizes real-time collision warning, and provides reliable basic support for intelligent driving.

[0035] In some embodiments, the step of determining the collision state of the target vehicle based on all target traffic element data includes: determining the traffic flow state based on all target traffic element data; screening out the target traffic participant and its location information from all target traffic element data, and determining the distance and orientation between the target traffic participant and the target vehicle based on the location information; determining the collision state of the target vehicle based on the distance, orientation, and traffic flow state. In one embodiment, this may include: determining the collision state as a first state in response to the distance and orientation meeting the collision threshold condition and the traffic flow state being a preset traffic state; determining the collision state as a second state in response to the distance and orientation not meeting the collision threshold condition or the traffic flow state not being a preset traffic state.

[0036] In some embodiments, based on the target traffic element data, the potential collision risk of the target vehicle is calculated. The traffic flow state includes the traffic flow state and / or the accident state, the target traffic participant is a vulnerable traffic participant, and the location information includes the latitude and longitude coordinates of the vulnerable traffic participant. The collision threshold condition is that the distance between the vulnerable traffic participant and the target vehicle is less than or equal to the preset distance and the direction is in the direction of travel of the target vehicle. The preset distance can be set according to actual business needs, for example, 80m. The preset traffic state is that the traffic flow state is a non-congested state and the accident state is a traffic accident-free state. When the distance between the vulnerable traffic participant and the target vehicle is less than or equal to the preset distance, the direction is in the direction of travel of the target vehicle, the traffic flow state is a non-congested state and a traffic accident-free state, the collision state is a first state. When the distance between the vulnerable traffic participant and the target vehicle is greater than the preset distance, the direction is not in the direction of travel of the target vehicle, the traffic flow state is a congested state and / or a traffic accident state, the collision state is a second state.

[0037] The collision warning method of the present invention combines the target map segment and the cached traffic element data to calculate the distance and orientation relationship between the vulnerable traffic participants and the target vehicle in real time, and comprehensively judges the collision risk through the traffic flow status, thereby ensuring the accuracy and reliability of the collision status detection results, realizing real-time detection and warning of vulnerable traffic participants, and reducing the incidence of traffic accidents.

[0038] In some embodiments, a high-precision map is obtained and sliced ​​to obtain a preset map database, including: obtaining a high-precision map, starting from the longitude and latitude origin in the high-precision map, and gridding it according to a preset spacing to obtain a number of second map slices; obtaining slice description information corresponding to each second map slice, and constructing a preset map database based on all slice description information; wherein the slice description information includes a map slice identifier, a map sheet number, a map slice graphic, and an associated perception group number.

[0039] In some embodiments, the longitude and latitude origin can be represented as (0,0), and the preset spacing can be 300m*300m. Based on the longitude and latitude origin of the high-precision map, the high-precision map is divided according to the rule of fixed spacing to obtain multiple map slices, namely the second map slice. A unique ID is assigned to each divided map slice, and the sheet number of each map slice is calculated based on the longitude and latitude origin and the grid offset step. In one example, assuming that the sheet number of the map slice 1 including the longitude and latitude origin is represented by (0,0), the sheet number of the map slice 2 obtained by offsetting the preset spacing in the direction of the positive axis of longitude (i.e., to the right) on the basis of map slice 1 can be represented as (1,0). The map slice figure is a geometric figure composed of the boundary points of the map slice, usually a rectangle determined by the longitude and latitude coordinates of the four vertices of the corresponding grid. The associated perception group number is the roadside perception device associated with the map slice, such as the roadside perception device located in the map slice.

[0040] The collision warning method of the present invention obtains map slices by gridding the high-precision map and constructing a preset map database to achieve efficient spatial indexing, which is conducive to realizing a subsequent dynamic update mechanism for the traffic element data in the map slices.

[0041] In some embodiments, the steps of constructing a preset geographic feature library of the present invention include: traversing all geographic attribute information in the high-precision map; based on each geographic attribute information, determining whether there is a third map slice matching it in the preset map database; in response to the existence of a third map slice matching it, using the geographic attribute information as the geographic feature label of the third map slice; and constructing a preset geographic feature library based on the third map slice and its corresponding geographic feature label.

[0042] In some embodiments, the high-precision map includes multiple geographic attribute information such as ramps, lanes, intersections, stop lines, and roundabouts. According to the latitude and longitude coordinates and range of each geographic attribute information, the intersection relationship between the geographic attribute information and the geometric shape of the map slice is determined by geometric intersection calculation. If the two intersect, the geographic attribute information matches the map slice. If they do not intersect, the geographic attribute information does not match the map slice. Thus, one or more map slices matching the geographic attribute information are found in the preset map database, that is, the third map slice. Based on the third map slice and its matching one or more geographic feature tags, a preset geographic feature library is constructed.

[0043] The collision warning method of the present invention assigns a geographic feature label to each map slice based on high-precision map attributes and forms a preset geographic feature library, which is conducive to the subsequent rapid retrieval of associated map slices and traffic element data around the target vehicle based on the geographic feature labels, and realizes the accurate issuance of collision warning information to the target vehicle in a specific collision warning scenario.

[0044] In some embodiments, based on a preset map database, the step of obtaining a plurality of first map slices within a first range includes: based on geometric intersection calculation, querying the preset map database for all map slices intersecting with the first range and taking them as first map slices.

[0045] In some embodiments, the step of screening out target traffic participants and their location information from all target traffic element data includes: determining whether the type of traffic participants in each target traffic element data is a vulnerable traffic participant; in response to the traffic participant type being a vulnerable traffic participant, taking the vulnerable traffic participant as a target traffic participant, and obtaining the location information of the vulnerable traffic participant.

[0046] The collision warning method of the present invention combines geographic slicing technology, geographic feature labeling and traffic element data binding to quickly load the location information of vulnerable traffic participants, aiming at the characteristics of vulnerable traffic participants such as poor protection ability, strong uncertainty of dynamic behavior and visual blind spots that make it difficult for drivers to discover them in time, so as to calculate the distance and direction relationship between the vulnerable traffic participants and the target vehicle in real time, and respond quickly to the vulnerable traffic participants in a targeted manner, which is conducive to promoting the development of intelligent transportation systems to assist drivers in perceiving the status information of surrounding vulnerable traffic participants and vehicles and reducing the incidence of traffic accidents.

[0047] In some embodiments, the step of determining the collision state of the target vehicle based on the distance and direction and the traffic flow state includes: in response to the distance and direction meeting the collision threshold condition and the traffic flow state being a preset traffic state, determining the collision state as a first state in which a collision warning message should be sent to the target vehicle; in response to the distance and direction not meeting the collision threshold condition or the traffic flow state being not a preset traffic state, determining the collision state as a second state in which a collision warning message is not sent to the target vehicle.

[0048] In some embodiments, the real-time collision status is obtained through the above steps. When the collision status is the first state, that is, the real-time calculation shows that the distance between the target traffic participant and the target vehicle is less than the preset distance, the direction is in the forward direction of the target vehicle, and the traffic flow state is the preset traffic state, a collision warning information between the target vehicle and the target traffic participant is sent to the target vehicle.

[0049] In some embodiments, Figure 2 Take as an example to illustrate the execution process of the collision warning method: Step 1: After the target vehicle equipment is started, determine whether the target vehicle equipment system has loaded high-precision map data; Step 2: If high-precision map data is loaded, the high-precision map is sliced ​​to form a map database; if high-precision map data is not loaded, the process ends directly; Step 3: Set geographic feature labels (such as intersections, roads, and ramps) for each map slice in the map database to form a geographic feature library; Step 4: Bind the traffic element data collected in real time by the roadside sensing device to each map slice in the map database to form an association between the map slice and the traffic element data; Step 5: According to the real-time position and geographic feature library of the target vehicle, the target map slices related to the collision warning range of the target vehicle are matched from the map database; Step 6: For each target map slice, load the traffic element data cached in the target map slice, including the type of traffic participant, the location of the traffic participant, and the traffic flow status; Step 7, after entering the vulnerable traffic participant collision warning calculation program, determine whether the traffic participant type in the target map slice is a vulnerable traffic participant; Step 8: If it is a vulnerable traffic participant, calculate the distance and direction between the vulnerable traffic participant and the target vehicle; if it is not a vulnerable traffic participant, end the process; Step 9: determine whether the distance and direction meet the collision threshold condition, that is, determine whether the distance between the vulnerable traffic participant and the target vehicle is less than or equal to a preset distance and whether the direction is in the direction of travel of the target vehicle, the preset distance is, for example, 80m; Step 10: If the collision threshold condition is met, determine whether the traffic flow state is a preset traffic state, that is, determine whether the traffic flow is in a non-congested state and there are no traffic accidents in the surrounding area; if the collision threshold condition is not met, end the process; Step 11: If the traffic flow state is the preset traffic state, determine that there is a potential collision risk between the vulnerable traffic participant and the target vehicle, generate high-precision collision information and send it to the target vehicle.

[0050] The collision warning method of the present invention divides a high-precision map into multiple map slices, each map slice is associated with a geographic feature tag, and can dynamically adjust the data loading range to achieve on-demand loading of map slices and their target traffic element data in specific scenarios, avoiding unnecessary consumption of computing resources, and being beneficial to improving data management and retrieval efficiency; caching traffic element data to map slices is convenient for rapid data acquisition and updating, and when using map slices for collision warning calculations, real-time collision warning calculations can be performed based on the latest traffic element data, thereby ensuring the accuracy and timeliness of collision warnings.

[0051] Based on the same inventive concept, according to another aspect of the present invention, Figure 3 As shown, an embodiment of the present invention further provides a collision warning system, comprising: A map slicing processing unit 110 is configured to obtain a high-precision map and perform slicing processing on it to obtain a preset map database; The initial slice acquisition unit 120 is configured to determine a first range according to the real-time position of the target vehicle, and acquire a plurality of first map slices located within the first range based on a preset map database; The geographic feature determination unit 130 is configured to obtain the geographic feature labels corresponding to each of the first map slices in a preset geographic feature library; A warning range determination unit 140, configured to determine a collision warning range based on the geographic feature tags corresponding to all the first map slices; A collision state determination unit 150 is configured to obtain a plurality of target traffic element data corresponding to the collision warning range, and determine the collision state of the target vehicle based on all the target traffic element data; The collision warning unit 160 is configured to send collision warning information to the target vehicle based on the collision state.

[0052] Based on the real-time position of the target vehicle, the first map slice that is around the target vehicle and within a fixed range is quickly loaded, and combined with the geographic feature label corresponding to the first map slice, the data loading range is dynamically adjusted to obtain the collision warning range, so as to realize on-demand loading of map slices and their target traffic element data according to the collision warning scenario, avoid full data loading and unnecessary consumption of computing resources, reduce the loading time of traffic data around the vehicle, and realize efficient spatial indexing and dynamic update mechanism of map slices. The collision warning system of the present invention improves the real-time and efficiency of collision warning calculation of target traffic participants by combining map slicing technology, geographic feature labeling and traffic element data binding, meets the needs of intelligent networked vehicles for efficient and accurate data processing, realizes real-time collision warning, provides reliable basic support for intelligent driving, and reduces the incidence of traffic accidents.

[0053] Based on the same inventive concept, according to another aspect of the present invention, Figure 4 As shown, an embodiment of the present invention further provides a computer device 30, which includes a processor 310 and a memory 320. The memory 320 stores a computer program 321 that can be run on the processor. When the processor 310 executes the program, the steps of the above method are performed.

[0054] Finally, it should be noted that a person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium of the program can be a disk, an optical disk, a read-only storage memory (ROM) or a random access memory (RAM), etc. The above-mentioned computer program embodiments can achieve the same or similar effects as the corresponding above-mentioned any method embodiments.

[0055] It will also be appreciated by those skilled in the art that various exemplary logic blocks, modules, circuits and algorithm steps described in conjunction with the disclosure herein can be implemented as electronic hardware, computer software or a combination of the two. In order to clearly illustrate this interchangeability of hardware and software, a general description has been given to the functions of various schematic components, blocks, modules, circuits and steps. Whether this function is implemented as software or hardware depends on specific applications and the design constraints imposed on the entire system. Those skilled in the art can implement the function in various ways for each specific application, but this implementation decision should not be interpreted as causing a departure from the disclosed scope of the embodiments of the present invention.

[0056] The above are exemplary embodiments disclosed in the present invention, but it should be noted that various changes and modifications may be made without departing from the scope of the embodiments disclosed in the present invention as defined in the claims. The functions, steps and / or actions of the method claims according to the disclosed embodiments described herein do not need to be performed in any particular order. The serial numbers of the embodiments disclosed in the above embodiments of the present invention are for description only and do not represent the advantages and disadvantages of the embodiments. In addition, although the elements disclosed in the embodiments of the present invention may be described or required in individual form, they may also be understood as multiple unless explicitly limited to the singular.

[0057] It should be understood that, as used herein, the singular forms "a", "an" are intended to include the plural forms as well, unless the context clearly supports an exception. It should also be understood that, as used herein, "and / or" refers to any and all possible combinations including one or more of the associated listed items.

[0058] A person skilled in the art should understand that the discussion of any of the above embodiments is only exemplary and is not intended to imply that the scope of the disclosure of the embodiments of the present invention (including the claims) is limited to these examples; under the concept of the embodiments of the present invention, the technical features in the above embodiments or different embodiments can also be combined, and there are many other changes in different aspects of the above embodiments of the present invention, which are not provided in detail for the sake of simplicity. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present invention should be included in the protection scope of the embodiments of the present invention.

Claims

1. A collision warning method, characterized in that: include: Obtain a high-precision map and slice it to obtain a preset map database; Determine a first range according to the real-time position of the target vehicle, and acquire a plurality of first map slices located within the first range based on the preset map database; In a preset geographic feature library, obtain the geographic feature labels corresponding to each of the first map slices; Determining a collision warning range based on the geographic feature tags corresponding to all the first map slices; Acquire a plurality of target traffic element data corresponding to the collision warning range, and determine the collision state of the target vehicle based on all the target traffic element data; Based on the collision state, collision warning information is sent to the target vehicle.

2. The collision warning method according to claim 1, characterized in that: The step of determining the collision warning range based on the geographic feature tags corresponding to all the first map slices includes: Based on the geographic feature labels corresponding to the first maps, determining whether they are first type labels; If it is a first type tag, the first map slice is used as the target map slice; If it is not a first type of label, recursively search for other map slices that are adjacent to the first map slice and have the same corresponding geographic feature label, and use the first map slice and the other map slices as target map slices respectively; Based on all the target map slices, the collision warning range is determined.

3. The collision warning method according to claim 2, characterized in that: The step of obtaining data of a plurality of target traffic elements corresponding to the collision warning range includes: Based on each traffic sensing device, real-time collection of first traffic element data; According to the latitude and longitude coordinates and the impact range corresponding to each of the first traffic element data, determining whether there is the target map segment matching the first traffic element data in the collision warning range; In response to the existence of the target map slice matching the first traffic element data, the traffic element data is used as the target traffic element data.

4. The collision warning method according to claim 1, characterized in that: The step of determining the collision state of the target vehicle based on all the target traffic element data comprises: Determining a traffic flow state based on all of the target traffic element data; Filtering out target traffic participants and their location information from all the target traffic element data, and determining the distance and direction between the target traffic participants and the target vehicle based on the location information; A collision state of the target vehicle is determined based on the distance, the orientation, and the traffic flow state.

5. The collision warning method according to claim 1, characterized in that: The step of obtaining a high-precision map and slicing it to obtain a preset map database includes: The high-precision map is obtained, and starting from the latitude and longitude origin in the high-precision map, grid division is performed according to a preset interval to obtain a plurality of second map slices; Obtaining the tile description information corresponding to each of the second map tiles, and constructing the preset map database based on all the tile description information; The tile description information includes a map tile identifier, a map sheet number, a map tile graphic, and an associated perception group number.

6. The collision warning method according to claim 5, characterized in that: The steps of constructing the preset geographic feature library include: Traversing all geographic attribute information in the high-precision map; Based on each of the geographic attribute information, determining whether there is a third map slice matching it in the preset map database; In response to the existence of a third map slice matching the third map slice, using the geographic attribute information as a geographic feature tag of the third map slice; The preset geographic feature library is constructed based on the third map slice and the corresponding geographic feature tag.

7. The collision warning method according to claim 1, characterized in that: The step of acquiring a plurality of first map slices located within the first range based on the preset map database includes: Based on geometric intersection calculation, all map slices intersecting with the first range are searched in the preset map database and are used as the first map slices.

8. The collision warning method according to claim 4, characterized in that: The step of screening out target traffic participants and their location information from all the target traffic element data includes: Determining whether the type of traffic participant in each of the target traffic element data is a vulnerable traffic participant; In response to the type of the traffic participant being a vulnerable traffic participant, the vulnerable traffic participant is taken as the target traffic participant, and the location information of the vulnerable traffic participant is acquired.

9. The collision warning method according to claim 4, characterized in that: The step of determining the collision state of the target vehicle based on the distance and the orientation and the traffic flow state comprises: In response to the distance and the orientation meeting a collision threshold condition and the traffic flow state being a preset traffic state, determining that the collision state is a first state in which a collision warning message should be sent to the target vehicle; In response to the distance and the orientation not meeting a collision threshold condition or the traffic flow state not being a preset traffic state, determining the collision state as a second state in which collision warning information is not sent to the target vehicle.

10. A collision warning system, characterized in that: include: A map slicing processing unit configured to obtain a high-precision map and perform slicing processing on it to obtain a preset map database; An initial slice acquisition unit, configured to determine a first range according to a real-time position of the target vehicle, and acquire a plurality of first map slices located within the first range based on a preset map database; A geographic feature determination unit, configured to obtain, from a preset geographic feature library, geographic feature labels corresponding to each of the first map slices; a warning range determining unit, configured to determine a collision warning range based on the geographic feature tags corresponding to all the first map slices; A collision state determination unit configured to obtain a plurality of target traffic element data corresponding to the collision warning range, and determine the collision state of the target vehicle based on all the target traffic element data; The collision warning unit is configured to send collision warning information to the target vehicle based on the collision state.

11. A computer device comprising: at least one processor; as well as A memory storing a computer program executable on the processor, wherein the processor executes the steps of the method according to any one of claims 1 to 9 when executing the program.

Images