Vehicle-end light map real-time updating method, system and device and storage medium

Abstract

The invention discloses a vehicle end light map real-time updating method, system and device and a storage medium. The method comprises the steps of obtaining an SD map and camera perception recognition data within a preset range according to a vehicle position and a road scene; superposing the camera perception identification data to an SD map; extracting lane feature point data corresponding to a map scene from the fused perception map, and associating the lane feature point data to a road Link; and according to the Link ID, the lane feature point data is stored in a KFD database for differential updating, and the vehicle end light map is updated in real time based on the KFD database after differential updating and the map scene. Compared with the prior art in which a map is difficult to update in time in case of a temporary road condition, the method has the advantages that vehicle camera sensing data are fused to construct a map scene and extract lane feature point information, a more accurate lane-level map is generated by real-time dynamic mapping at the vehicle end during each running, and real-time updating of the light map at the vehicle end is realized.

Description

Technical Field

[0001] The present invention relates to the field of map updating technology, and in particular to a method, system, device and storage medium for real-time updating of a vehicle-side light map. Background Art

[0002] Lane-level maps (including lane center lines, lane boundary lines, road boundary lines, stop lines, traffic lights, and other information) are lightweight high-precision maps that provide reliable prior knowledge for autonomous driving / assisted driving.

[0003] At present, there are two main ways to update high-precision maps: 1. Traditional update: Professional surveying and mapping fleets are equipped with high-precision sensors (such as lidar, high-definition cameras, IMU, etc.), but traditional methods are usually updated periodically, which makes it difficult to reflect the dynamic changes of roads in real time, such as temporary construction and traffic accidents. In the face of emergencies and temporary changes, the response speed is slow, and it is difficult to update map information in time. The data volume is large, the processing time is long, and the cost is high. 2. Crowdsourcing update: Using a large number of vehicles to collect data, the coverage is wide, and it is suitable for large-scale map updates. Vehicles upload data in real time during driving, which can more quickly reflect the dynamic changes of roads, such as temporary conditions such as traffic accidents and road construction. However, the quality and calibration of sensors of different vehicles are different, which may lead to low accuracy and consistency of collected data. The location information and sensor data uploaded by the vehicle may involve user privacy and require strict privacy protection measures. Data may face security risks during transmission and storage, and it is necessary to ensure the security and integrity of the data. In the crowdsourcing mode, the amount of data is huge, requiring powerful data processing and storage capabilities. Therefore, how to achieve real-time update of vehicle-side light maps has become an urgent problem to be solved.

[0004] The above contents are only used to assist in understanding the technical solution of the present invention and do not constitute an admission that the above contents are prior art. Summary of the invention

[0005] The main purpose of the present invention is to provide a method, system, device and storage medium for real-time updating of vehicle-side light maps, aiming to solve the technical problem of how to achieve real-time updating of vehicle-side light maps.

[0006] To achieve the above object, the present invention provides a vehicle-side light map real-time update method, the vehicle-side light map real-time update method comprising:

[0007] Acquire an SD map and camera perception recognition data within a preset range according to the vehicle position and road scene corresponding to the current driving vehicle, wherein the SD map includes a road Link and a Link ID corresponding to the road Link;

[0008] Overlaying the camera perception recognition data onto the SD map to obtain a fused perception map;

[0009] Extracting lane feature point data corresponding to the map scene from the fused perception map, and associating the lane feature point data with the road Link;

[0010] The lane feature point data is stored in the KFD database according to the Link ID for differential update, and the vehicle-side light map is updated in real time based on the differentially updated KFD database and the map scene.

[0011] Optionally, the step of superimposing the camera perception recognition data onto the SD map to obtain a fused perception map includes:

[0012] Determining the perception confidence corresponding to each perception element in the camera perception recognition data respectively;

[0013] Selecting at least one target perception element corresponding to the perception confidence greater than a preset confidence threshold from the multiple perception elements;

[0014] The at least one target perception element is superimposed on the SD map to obtain a fused perception map.

[0015] Optionally, the step of extracting lane feature point data corresponding to the map scene from the fused perception map includes:

[0016] Constructing a map scene according to the fused perception map;

[0017] Acquire multiple to-be-processed perception elements at the same position at different time stamps within the preset range based on the map scene;

[0018] Determine the element type and element position corresponding to each perception element to be processed;

[0019] Lane scene feature point data corresponding to the map scene is extracted from the fused perception map according to the element type and the element position.

[0020] Optionally, the step of acquiring a plurality of to-be-processed perception elements at the same position at different timestamps within the preset range based on the map scene includes:

[0021] When the map scene is an intersection scene, the multiple perception elements to be processed at the same position at different time stamps within the preset range include traffic lights, stop lines, ground arrows, traffic signs, lane lines and stop lines.

[0022] Optionally, the step of acquiring a plurality of to-be-processed perception elements at the same position at different timestamps based on the road scene further includes:

[0023] When the map scene is a lane change scene, the multiple perception elements to be processed at the same position at different time stamps within the preset range include ground arrows, traffic signs and lane lines.

[0024] Optionally, the step of acquiring a plurality of to-be-processed perception elements at the same position at different timestamps based on the road scene further includes:

[0025] When the map scene is a divergence and merging scene, the multiple perception elements to be processed at the same position at different time stamps within the preset range include ground arrows, traffic signs, lane lines and guide lines.

[0026] In addition, to achieve the above purpose, the present invention also proposes a vehicle-side light map real-time update system, the vehicle-side light map real-time update system comprising:

[0027] An acquisition module, used to acquire an SD map and camera perception recognition data within a preset range according to a vehicle position and a road scene corresponding to a currently driven vehicle, wherein the SD map includes a road Link and a Link ID corresponding to the road Link;

[0028] A fusion module, used for superimposing the camera perception recognition data onto the SD map to obtain a fused perception map;

[0029] An extraction module, used to extract lane feature point data corresponding to the map scene from the fused perception map, and associate the lane feature point data with the road Link;

[0030] An updating module is used to store the lane feature point data in the KFD database for differential updating according to the Link ID, and to update the vehicle-side light map in real time based on the differentially updated KFD database and the map scene.

[0031] In addition, to achieve the above-mentioned purpose, the present invention also proposes a vehicle-side light map real-time update device, which includes: a memory, a processor, and a vehicle-side light map real-time update program stored in the memory and executable on the processor, and the vehicle-side light map real-time update program is configured to implement the steps of the vehicle-side light map real-time update method as described above.

[0032] In addition, to achieve the above-mentioned purpose, the present invention also proposes a storage medium, on which a vehicle-side light map real-time update program is stored. When the vehicle-side light map real-time update program is executed by a processor, the steps of the vehicle-side light map real-time update method as described above are implemented.

[0033] The present invention first obtains the SD map and camera perception recognition data within a preset range according to the vehicle position and road scene corresponding to the current driving vehicle. The SD map includes the road Link and the Link ID corresponding to the road Link. Then the camera perception recognition data is superimposed on the SD map to obtain a fused perception map. Then the lane feature point data corresponding to the map scene is extracted from the fused perception map, and the lane feature point data is associated with the road Link. Finally, the lane feature point data is stored in the KFD database according to the Link ID for differential update, and the vehicle-side light map is updated in real time based on the differentially updated KFD database and map scene. Compared with the traditional update and crowdsourcing update used in the prior art, it is difficult to update the map in time if a temporary road condition is encountered. The present invention obtains the SD map of a certain range ahead based on the vehicle position, integrates the vehicle camera perception data to construct the map scene and extracts the lane feature point information. After multiple runs, the extracted lane feature point data has covered all the information inconsistent with reality. Each run is real-time dynamic mapping on the vehicle side to generate a more accurate lane-level map for automatic driving / assisted driving, so as to realize the real-time update of the vehicle-side light map. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a structural diagram of a vehicle-side light map real-time updating device in a hardware operating environment involved in an embodiment of the present invention;

[0035] Figure 2 It is a flowchart of a first embodiment of a method for real-time updating of a vehicle-side light map according to the present invention;

[0036] Figure 3 This is a flow chart of the real-time update of the vehicle-side light map according to the first embodiment of the real-time update method of the vehicle-side light map of the present invention;

[0037] Figure 4 The camera perception image of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0038] Figure 5 It is a fused perception map image of the intersection scene of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0039] Figure 6 The lane feature point image of the intersection scene of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0040] Figure 7 This is a rendering of a road intersection scene according to the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0041] Figure 8 It is a fused perception map image of the lane change scene of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0042] Fig. 9 The lane feature point image of the lane change scene of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0043] Fig.10 This is a rendering of a lane change scene according to the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0044] Fig.11 It is a fused perception map image of the divergence and merging scene of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0045] Fig.12 The lane feature point image of the divergence and merging scene of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0046] Fig.13 This is a rendering of a diverging and merging scene according to the first embodiment of the vehicle-side light map real-time updating method of the present invention;

[0047] Fig.14 This is a lane feature point differential update effect diagram of the first embodiment of the vehicle-side light map real-time update method of the present invention;

[0048] Fig.15 This is a structural block diagram of the first embodiment of the vehicle-side light map real-time update system of the present invention.

[0049] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0050] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0051] Reference Figure 1 , Figure 1 This is a schematic diagram of the structure of a vehicle-side light map real-time update device in the hardware operating environment involved in an embodiment of the present invention.

[0052] like Figure 1As shown, the vehicle-side light map real-time update device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage system independent of the aforementioned processor 1001.

[0053] Those skilled in the art will understand that Figure 1 The structure shown in does not constitute a limitation on the vehicle-side light map real-time update device, and may include more or fewer components than shown in the figure, or a combination of certain components, or a different arrangement of components.

[0054] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a vehicle-side light map real-time update program.

[0055] exist Figure 1 In the vehicle-side light map real-time update device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the vehicle-side light map real-time update device of the present invention can be set in the vehicle-side light map real-time update device, and the vehicle-side light map real-time update device calls the vehicle-side light map real-time update program stored in the memory 1005 through the processor 1001, and executes the vehicle-side light map real-time update method provided by the embodiment of the present invention.

[0056] The embodiment of the present invention provides a vehicle-side light map real-time update method, referring to Figure 2 , Figure 2 It is a flow chart of the first embodiment of the vehicle-side light map real-time update method of the present invention.

[0057] In this embodiment, the vehicle-side light map real-time update method includes the following steps:

[0058] Step S10: Acquire an SD map and camera perception recognition data within a preset range according to the vehicle position and road scene corresponding to the current driving vehicle, wherein the SD map includes a road Link and a Link ID corresponding to the road Link.

[0059] It is easy to understand that the execution subject of this embodiment can be a vehicle-side light map real-time update system with functions such as data processing, network communication and program running, or it can be other computer equipment with similar functions, etc. This embodiment is not limited.

[0060] In this embodiment, reference Figure 3 , Figure 3 This is a vehicle-side real-time update flowchart of the first embodiment of the vehicle-side light map real-time update method of the present invention. Every time the vehicle moves, updated information will be extracted, and it will be reflected in the light map in real time when the vehicle-side dynamic map is built, so that the update frequency is high; and real-time updates are performed based on the sensor data of the vehicle, so that the update cost is low; updates are performed based on the navigation road Link, and only the information involved in the changes is updated, which occupies a small amount of memory and takes a short time.

[0061] It should also be noted that, during the first movement of the ego vehicle, since the lane feature point data (kotei lane featuredata KFD) database is empty, the lane feature point data can be determined based on the camera perception data obtained during the first movement of the ego vehicle. After that, the lane feature point data and the SD map are directly dynamically mapped, and a light map display can be implemented on the ego vehicle side.

[0062] The second self-driving vehicle movement can obtain the lane feature point data stored in the KFD database for the first movement. During dynamic mapping, the lane feature point data in the KFD data and the lane feature point data of this time can be differentially updated, and the lane feature point data after the differential update can replace the lane feature point data of the first movement. The self-driving vehicle light map construction is completed based on the lane feature point data after the differential update.

[0063] It should be understood that the Nth vehicle movement mapping can be realized based on the second movement mapping process, and the first movement to the Nth movement are the same road movement.

[0064] In the specific implementation, the vehicle position corresponding to the current driving vehicle is obtained in real time through GNSS and IMU, and the road scene corresponding to the current vehicle position is identified. The preset range is determined according to the vehicle position and the road scene, and then the SD map and camera perception recognition data within the preset range are obtained.

[0065] Due to different road scenes, the preset range is also different. The preset range can be a straight-ahead range of the road, a circle range with the vehicle as the origin, etc. The preset range can also be set according to user needs, etc., and this embodiment does not limit it.

[0066] The road scene is the scene of the current road identified by the sensor, which can be an intersection scene, a lane change scene, a divergence and merging scene, etc.

[0067] In this embodiment, based on the timestamp and vehicle location information, the camera perception and recognition data (traffic lights, stop lines, traffic signs, ground arrows, lane lines, etc.) of the vehicle passing through the preset Link range (i.e., the preset range) is obtained, and reference is made to Figure 4 , Figure 4 The camera of the first embodiment of the vehicle-side light map real-time update method of the present invention senses the image and obtains the SD map within the preset Link range. The SD map includes the road Link and the Link ID corresponding to the road Link.

[0068] It should be understood that each road Link has a corresponding Link ID.

[0069] Step S20: superimposing the camera perception recognition data onto the SD map to obtain a fused perception map.

[0070] Furthermore, the perception confidence corresponding to each perception element in the camera perception recognition data is determined respectively; at least one target perception element corresponding to a perception confidence greater than a preset confidence threshold is selected from multiple perception elements; and at least one target perception element is superimposed on the SD map to obtain a fused perception map.

[0071] In this embodiment, based on the data perceived and recognized by the camera (traffic lights, stop lines, traffic signs, ground arrows, lane lines, etc.), the data with a confidence level less than or equal to a preset confidence threshold (80%) is filtered out (generally the confidence level is greater than 80%, the location information is correct, and the type information is relatively correct), and the data is fused and superimposed on the SD map to obtain a fused perception map.

[0072] In a specific implementation, the fused perception map can be a fused perception map of an intersection scene, a fused perception map of a lane change scene, or a fused perception map of a divergence and merging scene, etc. Figure 5 , Figure 8 and Fig.11 , Figure 5 The fused perception map image of the intersection scene of the first embodiment of the vehicle-side light map real-time update method of the present invention is Figure 8 It is a fused perception map image of a lane change scene in the first embodiment of the vehicle-side light map real-time update method of the present invention, Fig.11This is a fused perceived map image of a divergent and merging scene in the first embodiment of the vehicle-side light map real-time update method of the present invention.

[0073] Step S30: extracting lane feature point data corresponding to the map scene from the fused perception map, and associating the lane feature point data to the road Link.

[0074] Furthermore, a processing method for extracting lane feature point data corresponding to a map scene from a fused perception map is to construct a map scene based on the fused perception map; based on the map scene, obtain multiple perception elements to be processed at the same position at different time stamps within a preset range; determine the element type and element position corresponding to each perception element to be processed; and extract lane scene feature point data corresponding to the map scene from the fused perception map according to the element type and element position.

[0075] It should also be noted that the map scene is the scene that needs to be constructed by fusing the camera perception and recognition data onto the SD map.

[0076] In this embodiment, the type and position of the perceived element are obtained: data with too few identification times of the same element ID are filtered out, elements at the same position at different timestamps are obtained, the type with the largest proportion is set as the type of the current element, and the average position point of the element position of this type is calculated as the position point of the element.

[0077] When the map scene is an intersection scene, multiple perception elements to be processed at the same position at different time stamps within a preset range include traffic lights, stop lines, ground arrows, traffic signs, lane lines, and stop lines.

[0078] In the specific implementation, refer to Figure 6 , Figure 6This is the lane feature point image of the intersection scene of the first embodiment of the real-time update method of the vehicle-side light map of the present invention. The traffic lights, stop lines, and ground arrows in the light map are all generated based on logical algorithms, and there will be inconsistencies with reality. The traffic lights, stop lines, and ground arrows in the perception data are associated with the road Link, and lane feature points of the types of traffic lights, stop lines, and ground arrows are generated respectively; the light map does not store traffic signs, and the traffic signs in the perception data are associated with the road Link to generate lane feature points of the type of traffic signs; based on the perceived and recognized lane lines, the number of lanes, lane width, and lane line attributes (color, virtuality, etc.) of the road Link are extracted, and the starting point of the road Link is used as the lane feature point, and the number of lanes, lane width, lane lines and other attributes are set; based on the stop line position in the perception data, the intersection entry Link is cut, and the intersection entry point is set; based on the perception and recognition of the stop line of the opposite road, the intersection exit Link is cut, and the intersection exit point is set. The intersection entry point and the intersection exit point are associated with the road Link, and lane feature points of the types of intersection entry point and road opening exit point are generated respectively. Extract the arrow information of ground arrows and traffic signboards, and generate the lane topology relationship between the intersection entry point and the intersection exit point.

[0079] When the map scene is a lane change scene, multiple perception elements to be processed at the same position at different time stamps within a preset range include ground arrows, traffic signs, and lane lines.

[0080] In the specific implementation, refer to Fig. 9 , Fig. 9 This is an image of lane feature points of a lane change scenario in the first embodiment of the vehicle-side light map real-time update method of the present invention. In a lane change scenario, ground arrows, traffic signs, and lane line information can be obtained based on camera perception data. The feature point extraction method for ground arrows, traffic signs, and lane lines is consistent with the feature point extraction method for intersection scenes. Based on lane lines, ground arrows, etc., the start and end points where the number of lanes changes are identified, the lane change start and end points are set, and the lane change start and end points are associated with the road Link, and lane feature points of the type of lane change start and end points are generated respectively. The topological relationship between the lane change start and end points is generated based on the ground arrow and traffic sign information.

[0081] When the map scene is a divergence and merging scene, multiple perception elements to be processed at the same position at different time stamps within a preset range include ground arrows, traffic signs, lane lines, and guide lines.

[0082] In the specific implementation, refer to Fig.12 , Fig.12This is an image of lane feature points of a divergence and merging scene in the first embodiment of the real-time update method of a vehicle-side light map of the present invention. In a divergence and merging scene, ground arrows, traffic signs, and lane line information can be obtained based on camera perception data. The feature point extraction method for ground arrows, traffic signs, and lane lines is consistent with the feature point extraction method for intersection scenes. The lane line is extracted as a guide line attribute, and the guide line starting point is used as the divergence starting point / merging ending point. The divergence ending point / merging starting point is calculated based on the speed of the road. The divergence starting point, divergence ending point, merging starting point, and merging ending point are associated with the road Link, and lane feature points of the types of divergence starting point, divergence ending point, merging starting point, and merging ending point are generated respectively. The topological relationship of the start and end points of the lane is generated based on the ground arrows and traffic signs.

[0083] It should also be noted that if the map scene is other complex scenes, it can be expressed by the combination of the above three scenes, and the lane feature points are extracted according to the above three scene methods. In the straight road scene (the road link is not a link within the scope of the intersection, lane change, divergence / merging scene), only the number of lanes, lane width, lane line attributes (color, virtuality, etc.) of the link are updated.

[0084] Step S40: storing the lane feature point data in the KFD database for differential update according to the Link ID, and updating the vehicle-side light map in real time based on the differentially updated KFD database and the map scene.

[0085] It should also be noted that there is a corresponding relationship between Link ID and lane feature point data in the KFD database. When performing dynamic mapping, the corresponding lane feature point data can be directly obtained based on the Link ID.

[0086] In the specific implementation, refer to Figure 7 , Fig.10 , Fig.13 and Fig.14 , Figure 7 This is a scene effect diagram of an intersection according to the first embodiment of the vehicle-side light map real-time update method of the present invention. Fig.10 This is a lane change scene effect diagram of the first embodiment of the vehicle-side light map real-time update method of the present invention. Fig.13 This is a rendering of a divergence and merging scene according to the first embodiment of the vehicle-side light map real-time update method of the present invention. Fig.14 This is a lane feature point differential update effect diagram of the first embodiment of the vehicle-side light map real-time update method of the present invention. The KFD data will be updated every time the run is run, and the lane-level map constructed by this run is more consistent with reality than the previous run. The KFD data generated after multiple runs has covered all the inconsistencies with reality, realizing the real-time update of the vehicle-side light map during self-vehicle commuting.

[0087] In this implementation, firstly, the SD map and camera perception recognition data within a preset range are obtained according to the vehicle position and road scene corresponding to the current driving vehicle. The SD map includes the road link and the link ID corresponding to the road link. Then, the camera perception recognition data is superimposed on the SD map to obtain a fused perception map. Then, the lane feature point data corresponding to the map scene is extracted from the fused perception map, and the lane feature point data is associated with the road link. Finally, the lane feature point data is stored in the KFD database according to the link ID for differential update, and the vehicle-side light map is updated in real time based on the differentially updated KFD database and map scene. Compared with the traditional update and crowdsourcing update used in the prior art, it is difficult to update the map in time if a temporary road condition is encountered. In this implementation, the SD map of a certain range ahead is obtained based on the vehicle position, and the map scene is constructed by integrating the vehicle camera perception data and extracting the lane feature point information. After multiple runs, the extracted lane feature point data has covered all the information inconsistent with reality. Each run is real-time dynamic mapping on the vehicle side to generate a more accurate lane-level map for automatic driving / assisted driving, so as to realize the real-time update of the vehicle-side light map.

[0088] Reference Fig.15 , Fig.15 This is a structural block diagram of the first embodiment of the vehicle-side light map real-time update system of the present invention.

[0089] like Fig.15 As shown, the vehicle-side light map real-time update system proposed in the embodiment of the present invention includes:

[0090] The acquisition module 1501 is used to acquire the SD map and camera perception recognition data within a preset range according to the vehicle position and road scene corresponding to the current driving vehicle, wherein the SD map includes a road Link and a LinkID corresponding to the road Link;

[0091] A fusion module 1502 is used to superimpose the camera perception recognition data onto the SD map to obtain a fused perception map;

[0092] An extraction module 1503 is used to extract lane feature point data corresponding to the map scene from the fused perception map, and associate the lane feature point data with the road Link;

[0093] The updating module 1504 is used to store the lane feature point data in the KFD database for differential update according to the Link ID, and to update the vehicle-side light map in real time based on the differentially updated KFD database and the map scene.

[0094] Other embodiments or specific implementation methods of the vehicle-side light map real-time update system of the present invention can refer to the above-mentioned method embodiments and will not be repeated here.

[0095] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0096] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0097] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory / random access memory, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods described in each embodiment of the present invention.

[0098] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A vehicle-side light map real-time update method, characterized in that: The vehicle-side light map real-time update method comprises the following steps: Acquire an SD map and camera perception recognition data within a preset range according to the vehicle position and road scene corresponding to the current driving vehicle, wherein the SD map includes a road Link and a Link ID corresponding to the road Link; Overlaying the camera perception recognition data onto the SD map to obtain a fused perception map; Extracting lane feature point data corresponding to the map scene from the fused perception map, and associating the lane feature point data with the road Link; The lane feature point data is stored in the KFD database according to the Link ID for differential update, and the vehicle-side light map is updated in real time based on the differentially updated KFD database and the map scene.

2. The method according to claim 1, characterized in that The step of superimposing the camera perception recognition data onto the SD map to obtain a fused perception map includes: Determining the perception confidence corresponding to each perception element in the camera perception recognition data respectively; Selecting at least one target perception element corresponding to the perception confidence greater than a preset confidence threshold from the multiple perception elements; The at least one target perception element is superimposed on the SD map to obtain a fused perception map.

3. The method according to claim 2, characterized in that The step of extracting lane feature point data corresponding to the map scene from the fused perception map includes: Constructing a map scene according to the fused perception map; Acquire multiple to-be-processed perception elements at the same position at different time stamps within the preset range based on the map scene; Determine the element type and element position corresponding to each perception element to be processed; Lane scene feature point data corresponding to the map scene is extracted from the fused perception map according to the element type and the element position.

4. The method according to claim 3, characterized in that The step of acquiring a plurality of to-be-processed perception elements at the same position at different time stamps within the preset range based on the map scene comprises: When the map scene is an intersection scene, the multiple perception elements to be processed at the same position at different time stamps within the preset range include traffic lights, stop lines, ground arrows, traffic signs, lane lines and stop lines.

5. The method according to claim 3, characterized in that The step of acquiring a plurality of to-be-processed perception elements at the same position at different timestamps based on the road scene further includes: When the map scene is a lane change scene, the multiple perception elements to be processed at the same position at different time stamps within the preset range include ground arrows, traffic signs and lane lines.

6. The method according to claim 3, characterized in that The step of acquiring a plurality of to-be-processed perception elements at the same position at different timestamps based on the road scene further includes: When the map scene is a divergence and merging scene, the multiple perception elements to be processed at the same position at different time stamps within the preset range include ground arrows, traffic signs, lane lines and guide lines.

7. A vehicle-side light map real-time update system, characterized in that: The vehicle-side light map real-time update system includes: An acquisition module, used to acquire an SD map and camera perception recognition data within a preset range according to a vehicle position and a road scene corresponding to a currently driven vehicle, wherein the SD map includes a road Link and a Link ID corresponding to the road Link; A fusion module, used for superimposing the camera perception recognition data onto the SD map to obtain a fused perception map; An extraction module, used to extract lane feature point data corresponding to the map scene from the fused perception map, and associate the lane feature point data with the road Link; An updating module is used to store the lane feature point data in the KFD database for differential updating according to the Link ID, and to update the vehicle-side light map in real time based on the differentially updated KFD database and the map scene.

8. A vehicle-side light map real-time update device, characterized in that: The device includes: a memory, a processor, and a vehicle-side light map real-time update program stored in the memory and executable on the processor, wherein the vehicle-side light map real-time update program is configured to implement the steps of the vehicle-side light map real-time update method as described in any one of claims 1 to 6.

9. A storage medium, characterized in that: The storage medium stores a vehicle-side light map real-time update program, and when the vehicle-side light map real-time update program is executed by the processor, the steps of the vehicle-side light map real-time update method as described in any one of claims 1 to 6 are implemented.

Images