Map fusion method, electronic device and program product
Through multiple acquisition and minimum spanning tree fusion technology, the problem of low map accuracy in crowdsourcing acquisition and map construction is solved, and a higher-precision map fusion and production is achieved.
Patent Information
- Application Number
- CN202411535280.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-30
- Publication Date
- 2025-02-11
AI Technical Summary
The existing crowdsourcing acquisition and mapping method has low accuracy due to incomplete sensor accuracy and single-pass acquisition data, which affects the subsequent map production accuracy.
By obtaining map data collected by multiple vehicles through multiple trips, dividing it into sub-map data, and establishing a minimum spanning tree based on the sub-map data, determining the maximum matching sub-map data, and using the relative pose matrix to fusion map data to improve the map fusion accuracy.
Through the integration of multiple collection data and the establishment of the minimum spanning tree, the integrity and accuracy of map data are improved and the accuracy of subsequent map production is improved.
Smart Images
Figure CN119413191A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of map updating, and in particular to a map fusion method, electronic equipment and program product. Background Art
[0002] In the field of autonomous driving, high-precision maps are indispensable data for vehicle positioning and path planning. Compared with professional map collection and construction methods, crowdsourcing map collection and construction with high update frequency and low computational complexity is becoming more and more popular.
[0003] Crowdsourcing data collection can be understood as users collecting road data through the sensors of autonomous vehicles themselves or other low-cost sensors and transmitting them to the cloud for data fusion. This fusion method can be used to improve data accuracy and complete the production of crowdsourced high-precision maps or semantic maps. Among them, visual crowdsourcing data collection solutions are currently the main collection solutions.
[0004] However, the current crowdsourcing mapping method is affected by the accuracy of sensors and the incompleteness of the data collected in a single trip, resulting in low accuracy of the collected maps, which in turn affects the accuracy of subsequent map production. Summary of the invention
[0005] The embodiments of the present disclosure at least provide a map fusion method, electronic device, and program product, which can improve the accuracy of map fusion.
[0006] The present disclosure provides a map fusion method, including:
[0007] Acquire multiple map data for the same target road section collected by crowdsourcing, wherein the multiple map data are collected by multiple vehicles in multiple trips; wherein each map data corresponds to data collected by one vehicle in one trip;
[0008] For each map data, the map data is divided into a plurality of sub-map data, and a minimum spanning tree is established based on the plurality of sub-map data;
[0009] Performing association matching on each sub-graph data corresponding to the target road section to obtain a matching relationship, and determining the maximum matching sub-graph data corresponding to each map data based on the minimum spanning tree;
[0010] Based on the associated matching relationship, a relative pose matrix between any two associated maximum matching sub-map data is determined, and based on each relative pose matrix, the multiple map data are fused to obtain a target fused map.
[0011] In a possible implementation, establishing a minimum spanning tree based on the plurality of subgraph data includes:
[0012] The plurality of subgraph data are converted into a weighted undirected graph using a preset algorithm, and the minimum spanning tree is established based on the undirected graph.
[0013] In a possible implementation, each map data includes a semantic map element; and the associative matching of each sub-map data corresponding to the target road section to obtain a matching relationship includes:
[0014] Based on the position and direction of each semantic map element in each sub-map data, association matching is performed on each semantic map element in each sub-map data to obtain the matching relationship.
[0015] In a possible implementation manner, each map data includes a semantic map element; and determining the maximum matching subgraph data corresponding to each map data based on the minimum spanning tree includes:
[0016] For each target subgraph corresponding to each map data, based on the connection relationship indicated by the minimum spanning tree and the similarity between each target subgraph, each target subgraph is matched and spliced to obtain the maximum matching subgraph data corresponding to the map data.
[0017] In a possible implementation, each map data includes a semantic map element; and fusing the multiple map data based on the relative pose matrices to obtain a target fused map includes:
[0018] For any two associated maximum matching sub-graph data, based on the relative pose matrix, the relative poses between the associated semantic map elements are optimized;
[0019] Based on the optimized relative positions between the semantic map elements, the map data are fused to obtain the target fused map.
[0020] In a possible implementation, the semantic map element includes at least one of the following:
[0021] Lane markings, zebra crossings, road sign arrows, road edges, stop lines.
[0022] In a possible implementation manner, the map data is acquired by a visual sensor of the vehicle, and the method further includes:
[0023] The target fusion maps corresponding to the multiple target road sections are spliced and fused to obtain a target map, and the target map is sent to the multiple target vehicles.
[0024] In a possible implementation, the map data is obtained by preprocessing image data collected by a visual sensor of the vehicle.
[0025] An embodiment of the present disclosure provides an electronic device, including: a processor, a memory and a bus, wherein the memory stores machine-readable instructions executable by the processor, and when the electronic device is running, the processor and the memory communicate via the bus, and when the machine-readable instructions are executed by the processor, the map fusion method described in any possible implementation manner described above is performed.
[0026] An embodiment of the present disclosure provides a computer program product, characterized in that it includes a computer program. When the computer program is executed by a computer device or a processor, the computer device or the processor executes the map fusion method described in any possible implementation manner described above.
[0027] The map fusion method, electronic device, and program product provided by the embodiments of the present disclosure can improve the integrity of the collected data compared to the data collected in a single trip, because the multiple map data are collected by multiple vehicles in multiple trips, which is beneficial to improve the accuracy of subsequent map fusion. Secondly, in the map fusion process, each map data is first divided into multiple sub-graph data, and a minimum spanning tree is established based on the multiple sub-graph data to facilitate the maintenance of the sub-graph data, and then based on the minimum spanning tree, the maximum matching sub-graph data corresponding to each map data is determined, and then based on the relative pose matrix between any two associated maximum matching sub-graphs, the multiple map data are fused to obtain a target fused map, which can further improve the accuracy of map fusion.
[0028] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.
[0030] Figure 1 A flow chart of a map fusion method provided by an embodiment of the present disclosure is shown;
[0031] Figure 2 A schematic diagram showing a plurality of map data provided by an embodiment of the present disclosure;
[0032] Figure 3 A schematic diagram of obtaining a target fused map after performing map fusion provided by an embodiment of the present disclosure is shown;
[0033] Figure 4 Another schematic diagram of obtaining a target fused map after performing map fusion provided by an embodiment of the present disclosure is shown;
[0034] Figure 5 A functional module diagram of a map fusion device provided by an embodiment of the present disclosure is shown;
[0035] Figure 6 A schematic structural diagram of an electronic device provided by an embodiment of the present disclosure is shown. DETAILED DESCRIPTION
[0036] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure provided in the drawings is not intended to limit the scope of the present disclosure for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.
[0037] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.
[0038] The term "and / or" herein only describes an association relationship, indicating that three relationships may exist. For example, A and / or B may represent the following three situations: A exists alone, A and B exist at the same time, and B exists alone. In addition, the term "at least one" herein represents any combination of at least two of any one or more of a plurality of. For example, including at least one of A, B, and C may represent including any one or more elements selected from the set consisting of A, B, and C.
[0039] In the following description of this application, unless otherwise clearly specified and limited, the term "connection" should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integrated connection; it can be a direct connection or an indirect connection through an intermediate medium.
[0040] Crowdsourcing data collection is gradually being used to assist in the production and update of high-precision map data due to its low price and large data volume. Crowdsourcing collection vehicles use consumer-grade global navigation satellite systems (GNSS) and cameras, but do not have active lidars. They need to rely on the temporal fusion of visual information to complete the production of crowdsourcing high-precision maps or semantic maps.
[0041] However, research has found that the current crowdsourcing mapping method is affected by sensor accuracy and incomplete data collected in a single trip, resulting in low accuracy of the collected maps, which in turn affects the accuracy of subsequent map production.
[0042] Based on the above research, an embodiment of the present disclosure provides a map fusion method, which first obtains multiple map data for the same target road section collected by crowdsourcing, and the multiple map data are collected by multiple vehicles in multiple trips; wherein each map data corresponds to data collected by one vehicle in one trip; then, for each map data, the map data is divided into multiple sub-graph data, and a minimum spanning tree is established based on the multiple sub-graph data; then, association matching is performed on the sub-graph data corresponding to the target road section to obtain a matching relationship, and based on the minimum spanning tree, the maximum matching sub-graph data corresponding to each map data is determined; finally, based on the association matching relationship, the relative pose matrix between any two associated maximum matching sub-graphs is determined, and the multiple map data are fused based on the respective relative pose matrices to obtain a target fused map.
[0043] In the disclosed embodiment, firstly, since the multiple map data are collected by multiple vehicles in multiple trips, the integrity of the collected data can be improved compared to the data collected in a single trip, which is beneficial to improving the accuracy of subsequent map fusion. Secondly, in the map fusion process, each map data is first divided into multiple sub-graph data, and a minimum spanning tree is established based on the multiple sub-graph data to facilitate the maintenance of the sub-graph data, and then based on the minimum spanning tree, the maximum matching sub-graph data corresponding to each map data is determined, and then based on the relative pose matrix between any two associated maximum matching sub-graphs, the multiple map data are fused to obtain a target fused map, which can further improve the accuracy of map fusion.
[0044] To facilitate understanding of this embodiment, the execution subject of a map fusion method disclosed in the embodiment of the present disclosure is first introduced in detail. The execution subject of the map fusion method may also be an electronic device with certain computing capabilities, wherein the electronic device may include a server or other processing device. The server may be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud storage, big data and artificial intelligence platforms. Of course, the electronic device may also include a terminal device, which may be a user terminal, a handheld device, a computing device, a vehicle-mounted device, a wearable device, etc.
[0045] In some embodiments, the map fusion method may also be implemented by a processor calling computer-readable instructions stored in a memory.
[0046] The map fusion method provided by the embodiment of the present disclosure is described below with reference to the accompanying drawings. Figure 1 As shown, it is a flowchart of a map fusion method provided by an embodiment of the present disclosure, such as Figure 1 As shown, the method may include the following S101 to S104:
[0047] S101, obtaining a plurality of map data for the same target road section collected by crowdsourcing, wherein the plurality of map data are collected by a plurality of vehicles in a plurality of trips; wherein each map data corresponds to data collected by a vehicle in a single trip.
[0048] Among them, crowdsourcing is a data collection method that assigns work tasks to a large number of objects to complete. The core of this method is to transform the data collection work that may originally require a lot of professional resources or time to complete into small tasks that can be completed by many non-professional objects in fragmented time through decentralized task allocation. For example, in the process of updating high-precision maps, the task of data collection originally performed by professional collection vehicles is transformed into a task of data collection performed by non-professional ordinary vehicles.
[0049] For example, the non-professional ordinary vehicle can be an autonomous driving vehicle with an autonomous driving function. The autonomous driving vehicle can rely on artificial intelligence, visual computing, radar, monitoring devices and global positioning systems to work together, so that the vehicle computer can automatically and safely operate the motor vehicle without any human active operation. Of course, it can also be other vehicles, as long as they are equipped with components such as visual sensors and global positioning systems.
[0050] The visual sensor may be a camera for collecting image data. Exemplarily, the camera includes but is not limited to a monocular camera, a binocular camera, a fisheye camera, etc. The radar is used to collect point cloud data.
[0051] It should be noted that the map data in the embodiments of the present disclosure is obtained by preprocessing the image data collected by the visual sensor of the vehicle. For example, the image data can be collected by the visual sensor of the vehicle, and then the image data can be inferred and predicted by a deep learning neural network to obtain the inference prediction result, and then the inference result can be preprocessed to obtain the map data, wherein the preprocessing can include lane line parameter curve equation fitting, zebra crossing convex hull fitting, etc.
[0052] For example, see Figure 2 FIG. 1 is a schematic diagram of a plurality of map data provided by an embodiment of the present disclosure, wherein each map data includes a semantic map element. The semantic map element includes at least one of a lane line, a zebra crossing, a road surface indication arrow, a road edge, and a stop line.
[0053] For example, the target road section can be set according to actual needs. For example, the target road section can be a road section with a length of 5 kilometers or 10 kilometers within the preset position range. Each vehicle travels on the target road section once, and a corresponding map data can be obtained.
[0054] S102: for each map data, divide the map data into a plurality of sub-map data, and establish a minimum spanning tree based on the plurality of sub-map data.
[0055] Specifically, for each map data, the map data can be divided into multiple sub-map data according to preset rules. For example, the map data can be divided into multiple sub-map data according to a preset length (such as every 5 meters) in the extension direction of the map data, or the map part with a similarity greater than a preset threshold can be divided into one sub-map data according to the similarity. For example, if the target road section includes an intersection area, a first road part before the intersection area, and a second road part after the intersection area, then the map data corresponding to the first road part can be divided into one sub-map data, the map data corresponding to the intersection area part can be divided into another sub-map data, and the map data corresponding to the second road part can be divided into another sub-map data.
[0056] In some embodiments, when establishing a minimum spanning tree based on the multiple sub-graph data, it may include: using a preset algorithm to generate a weighted undirected graph from the multiple sub-graph data, and establishing the minimum spanning tree based on the undirected graph.
[0057] The preset algorithm includes Kruskal's algorithm or Prim's algorithm. In the disclosed embodiment, the Kruskal algorithm is used to generate a weighted undirected graph from multiple subgraph data to facilitate subsequent subgraph maintenance, and then a minimum spanning tree can be established based on the undirected graph.
[0058] S103 , performing association matching on each sub-graph data corresponding to the target road section to obtain a matching relationship, and determining the maximum matching sub-graph data corresponding to each map data based on the minimum spanning tree.
[0059] Exemplarily, after obtaining multiple sub-map data corresponding to each map data, all sub-map data corresponding to the target road section can be obtained, and then similarity matching can be performed on each sub-map data according to the type of data to obtain a matching relationship, which is used to indicate the association relationship between different sub-maps at the same location. For example, after matching, it is determined that sub-map data No. 1 and sub-map data No. 2 are map data corresponding to the same location (such as XX intersection), that is, through association matching, the sub-map data corresponding to the same location can be found and associated.
[0060] Specifically, since each map data includes semantic map elements, in order to improve the accuracy of association matching, when performing association matching on each sub-map data corresponding to the target road section to obtain a matching relationship, it can include: based on the position and direction of each semantic map element in each sub-map data, performing association matching on each semantic map element in the sub-map data to obtain the matching relationship.
[0061] Exemplarily, if the sub-map data includes semantic map elements such as lane lines and road surface indication arrows, and each sub-map data has a corresponding data identifier (such as a number), lane 1 in sub-map 1 can be associated with lane 1 in sub-map 2 through association matching, and the road surface indication arrow in sub-map 1 can be associated with the road surface indication arrow in sub-map 2.
[0062] Optionally, the matching relationship can be obtained by ∑<lm_i|ln_j> Indicates that lm_i is a semantic map element, lm_j is the semantic map element associated with lm_i, i and j are respectively used to indicate the data identifiers of different sub-map data. For example, if i is 1 and j is 3, it indicates that the semantic map element in sub-map data No. 1 is associated and matched with the semantic map element in sub-map data No. 3.
[0063] It can be understood that since each map data is divided in step S102, one map data is split into multiple sub-map data. For the multiple sub-map data of each map data, the consistency between adjacent sub-map data may be high. In order to improve the efficiency and accuracy of subsequent map fusion, the maximum matching sub-map data corresponding to each map data can be determined based on the minimum spanning tree.
[0064] Exemplarily, in order to improve the accuracy of determining the maximum matching sub-graph data, when determining the maximum matching sub-graph data corresponding to each map data based on the minimum spanning tree, it may include: for each target sub-graph data corresponding to each map data, based on the connection relationship indicated by the minimum spanning tree and the similarity between each target sub-graph data, matching and splicing each target sub-graph data to obtain the maximum matching sub-graph data corresponding to the map data. That is, in the process of determining the maximum matching sub-graph data, the previous and next frame data of the same map data are associated and matched to obtain a maximum matching sub-graph data corresponding to the map data.
[0065] It can be understood that the number of maximum matching sub-graph data corresponding to each map data may be one or more, which needs to be determined according to the actual matching results.
[0066] S104, based on the associated matching relationship, determining the relative pose matrix between any two associated maximum matching subgraphs, and fusing the plurality of map data based on the respective relative pose matrices to obtain a target fused map.
[0067] For example, based on the association matching relationship, the relative pose matrix between two arbitrarily associated maximum matching subgraphs can be solved using the Iterative Closest Point (ICP) algorithm, wherein the ICP algorithm is an algorithm based on data registration method and the closest point search method to solve the problem based on free-form surfaces.
[0068] In some embodiments, when the plurality of map data are fused based on the relative pose matrices to obtain a target fused map, the following steps (I) to (II) may be included:
[0069] (I) for any two associated maximum matching sub-graph data, optimizing the relative postures between the associated semantic map elements based on the relative posture matrix;
[0070] (II) Based on the optimized relative positions between the semantic map elements, the map data are fused to obtain the target fused map.
[0071] Specifically, the relative pose change matrix of two arbitrarily associated maximum matching sub-graph data can be jointly optimized and the corresponding map element P lm , the optimized result can get the optimal global consistent map. It can be expressed by the following expression:
[0072]
[0073] Among them, source and target are the associated subgraph data (which can be the maximum matching subgraph data), P lm It is a semantic map element. neighbor constrian represents the pose constraint of the adjacent sub-graph, which can be specifically expressed by the following formula (1); where neighbor contains multiple adjacent sub-graph pairs, landmark constrain It represents the element constraint on the matching within the sub-graph, which can be specifically expressed by the following formula (2), where landmark represents a map element, such as a lane line, a curb, etc. In this way, the optimized map is restored and aligned to obtain the target fused map.
[0074] neightbor constrian =(source_connect_P lm -target_connect_P lm ) (1)
[0075] landmark constrain =(R*source_P lm +t)-target_P lm (2)
[0076] Among them, R represents the rotation deviation, t represents the displacement deviation, source_connect_P lm 、target_connect_P lm They represent two adjacent subgraphs in neighbor, where neighbor refers to subgraphs at two adjacent positions in the data collected by a single vehicle at a single time. source_P lm Represents the semantic map element in a subgraph data of the associated subgraph pair, target_P lm It represents the corresponding map element in another sub-map data in the associated sub-map pair. The associated sub-map pair refers to the sub-map at the same position of the data collected by multiple vehicles multiple times.
[0077] See also Figure 3 and Figure 4As shown, it is a schematic diagram of obtaining a target fused map by performing map fusion using the map fusion method of the disclosed embodiment. In the disclosed embodiment, since multiple map data are collected by multiple vehicles in multiple trips, the integrity of the collected data can be improved compared to the data collected in a single trip, which is beneficial to improving the accuracy of subsequent map fusion. Secondly, in the map fusion process, each map data is first divided into multiple sub-graph data, and a minimum spanning tree is established based on the multiple sub-graph data to facilitate the maintenance of the sub-graph data, and then based on the minimum spanning tree, the maximum matching sub-graph data corresponding to each map data is determined, and then based on the relative pose matrix between any two associated maximum matching sub-graphs, the multiple map data are fused to obtain a target fused map, which can further improve the accuracy of map fusion.
[0078] In some embodiments, after obtaining the target fusion map corresponding to the target road section, the target fusion maps corresponding to multiple target road sections can be spliced and fused to obtain a target map, and the target map is sent to multiple target vehicles. In this way, the target vehicle can use the target map to implement corresponding functions according to actual needs, for example, the automatic driving function can be implemented based on the target map.
[0079] Those skilled in the art will appreciate that, in the above method of specific implementation, the order in which the steps are written does not imply a strict execution order and does not constitute any limitation on the implementation process. The specific execution order of the steps should be determined by their functions and possible internal logic.
[0080] See also Figure 5 FIG. 5 is a functional module diagram of a map fusion device provided by an embodiment of the present disclosure. The map fusion device 500 includes:
[0081] The map data acquisition module 501 is used to acquire a plurality of map data collected by crowdsourcing for the same target road section, wherein the plurality of map data are collected by a plurality of vehicles in a plurality of trips; wherein each map data corresponds to data collected by a vehicle in a single trip;
[0082] A map data partitioning module 502 is used to partition each map data into a plurality of sub-map data and establish a minimum spanning tree based on the plurality of sub-map data;
[0083] The data association matching module 503 is used to perform association matching on each sub-graph data corresponding to the target road section to obtain a matching relationship, and determine the maximum matching sub-graph data corresponding to each map data based on the minimum spanning tree;
[0084] The map data fusion module 504 is used to determine the relative pose matrix between any two maximum matching sub-graphs based on the associated matching relationship, and fuse the multiple map data based on each relative pose matrix to obtain a target fused map.
[0085] In a possible implementation, the map data division module 502 is specifically used for:
[0086] The plurality of subgraph data are converted into a weighted undirected graph using a preset algorithm, and the minimum spanning tree is established based on the undirected graph.
[0087] In a possible implementation, each map data includes a semantic map element; the data association matching module 503 is specifically used for:
[0088] Based on the position and direction of each semantic map element in each sub-map data, association matching is performed on each semantic map element in each sub-map data to obtain the matching relationship.
[0089] In a possible implementation, each map data includes a semantic map element; the data association matching module 503 is specifically used for:
[0090] For each target subgraph corresponding to each map data, based on the connection relationship indicated by the minimum spanning tree and the similarity between each target subgraph, each target subgraph is matched and spliced to obtain the maximum matching subgraph data corresponding to the map data.
[0091] In a possible implementation, each map data includes a semantic map element; the map data fusion module 504 is specifically used for:
[0092] For any two target maximum matching subgraphs, based on the relative pose matrix, the relative poses between the associated semantic map elements are optimized;
[0093] Based on the optimized relative positions between the semantic map elements, the map data are fused to obtain the target fused map.
[0094] In a possible implementation, the semantic map element includes at least one of the following:
[0095] Lane markings, zebra crossings, road sign arrows, road edges, stop lines.
[0096] In a possible implementation manner, the map data is acquired by a visual sensor of the vehicle, and the map data fusion module 504 is further used for:
[0097] The target fusion maps corresponding to the multiple target road sections are spliced and fused to obtain a target map, and the target map is sent to the multiple target vehicles.
[0098] Based on the same technical concept, the embodiment of the present disclosure also provides an electronic device. Figure 6 , which is a schematic diagram of the structure of an electronic device 600 provided in an embodiment of the present disclosure, includes a processor 601, a memory 602, and a bus 603. The memory 602 is used to store execution instructions.
[0099] In the embodiment of the present disclosure, the memory 602 is specifically used to store the application code for executing the solution of the present disclosure, and the execution is controlled by the processor 601. That is, when the electronic device 600 is running, the processor 601 communicates with the memory 602 through the bus 603, so that the processor 601 executes the application code stored in the memory 602, and then executes the method described in any of the above embodiments.
[0100] Among them, the memory 602 can be, but is not limited to, random access memory (Random Access Memory, RAM), read only memory (Read Only Memory, ROM), programmable read-only memory (Programmable Read-Only Memory, PROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, EPROM), electrically erasable read-only memory (Electric Erasable Programmable Read-Only Memory, EEPROM), etc.
[0101] Processor 601 may be an integrated circuit chip with signal processing capabilities. The above-mentioned processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present invention may be implemented or executed. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.
[0102] It is to be understood that the structure illustrated in the embodiment of the present application does not constitute a specific limitation on the electronic device 600. In other embodiments of the present application, the electronic device 600 may include more or fewer components than shown in the figure, or combine some components, or split some components, or arrange the components differently. The components shown in the figure may be implemented in hardware, software, or a combination of software and hardware.
[0103] The embodiment of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the map fusion method in the above method embodiment are executed.
[0104] The embodiments of the present disclosure also provide a computer program product, including a computer program / instruction. When the computer program / instruction processor is executed, the map fusion method provided in the embodiments of the present disclosure is implemented. For details, please refer to the above method embodiments, which will not be repeated here.
[0105] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is implemented as a computer storage medium. In another optional embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).
[0106] The method in the embodiments of the present disclosure may be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instruction is downloaded and executed on a computer, the process or function described in this application is executed in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user device, a core network device, an OAM or other programmable device.
[0107] The computer program or instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer program or instructions may be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired or wireless means. The computer-readable storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium may be a magnetic medium, such as a floppy disk, a hard disk, or a magnetic tape; it may also be an optical medium, such as a digital video disk; it may also be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both volatile and non-volatile types of storage media.
[0108] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.
[0109] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0110] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
[0111] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.
[0112] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure should be based on the protection scope of the claims.
Claims
1. A map fusion method, characterized in that: include: Acquire multiple map data for the same target road section collected by crowdsourcing, wherein the multiple map data are collected by multiple vehicles in multiple trips; wherein each map data corresponds to data collected by one vehicle in one trip; For each map data, the map data is divided into a plurality of sub-map data, and a minimum spanning tree is established based on the plurality of sub-map data; Performing association matching on each sub-graph data corresponding to the target road section to obtain a matching relationship, and determining the maximum matching sub-graph data corresponding to each map data based on the minimum spanning tree; Based on the associated matching relationship, a relative pose matrix between any two associated maximum matching sub-map data is determined, and based on each relative pose matrix, the multiple map data are fused to obtain a target fused map.
2. The method according to claim 1, characterized in that The step of establishing a minimum spanning tree based on the plurality of subgraph data comprises: The plurality of subgraph data are converted into a weighted undirected graph using a preset algorithm, and the minimum spanning tree is established based on the undirected graph.
3. The method according to claim 1, characterized in that Each map data includes a semantic map element; and the associative matching of each sub-map data corresponding to the target road section to obtain a matching relationship includes: Based on the position and direction of each semantic map element in each sub-map data, association matching is performed on each semantic map element in each sub-map data to obtain the matching relationship.
4. The method according to claim 1, characterized in that: Each map data includes a semantic map element; and determining the maximum matching subgraph data corresponding to each map data based on the minimum spanning tree includes: For each target subgraph data corresponding to each map data, based on the connection relationship indicated by the minimum spanning tree and the similarity between each target subgraph data, each target subgraph data is matched and spliced to obtain the maximum matching subgraph data corresponding to the map data.
5. The method according to claim 1, characterized in that Each map data includes a semantic map element; and fusing the plurality of map data based on the relative pose matrices to obtain a target fused map includes: For any two associated maximum matching sub-graph data, based on the relative pose matrix, the relative poses between the associated semantic map elements are optimized; Based on the optimized relative positions between the semantic map elements, the map data are fused to obtain the target fused map.
6. The method according to any one of claims 3 to 5, characterized in that: The semantic map element includes at least one of the following: Lane markings, zebra crossings, road sign arrows, road edges, stop lines.
7. The method according to any one of claims 1 to 5, characterized in that The map data is collected by a visual sensor of the vehicle, and the method further includes: The target fusion maps corresponding to the multiple target road sections are spliced and fused to obtain a target map, and the target map is sent to the multiple target vehicles.
8. The method according to any one of claims 1 to 5, characterized in that The map data is obtained by preprocessing image data collected by a visual sensor of the vehicle.
9. An electronic device, characterized in that: It includes a processor, a memory and a bus, the memory stores machine-readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and when the machine-readable instructions are executed by the processor, the map fusion method as described in any one of claims 1 to 8 is performed.
10. A computer program product, characterized in that The computer program product stores a computer program, and when the computer program is executed by a processor, the map fusion method according to any one of claims 1 to 8 is executed.