Semantic map construction method, construction device, processor and vehicle system
By updating obstacle information in the geospatial database and building a semantic map of the construction area, the problem of driving failures of autonomous vehicles in the construction area is solved, and accurate identification and safe driving of obstacle information are achieved.
Patent Information
- Application Number
- CN202011555698.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2020-12-24
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2040-12-24
AI Technical Summary
The 3D point cloud image in the prior art cannot display the object information corresponding to the point cloud, resulting in a failure of autonomous driving vehicles in the construction area, lack of semantic level obstacle recognition, affecting driving strategy adjustment.
By receiving obstacle information sent by the vehicle, including location, shape and status information, update object information in the geospatial database, build a semantic map of the construction area, ensure that the map matches the current construction area, and achieve accurate identification of obstacle information.
Improve the accuracy of obstacle identification in the vehicle in the construction area, avoid driving failures, and ensure safe driving.
Smart Images

Figure CN112579724B_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of autonomous driving, and more specifically, to a semantic map construction method, construction device, computer-readable storage medium, processor, and vehicle system. Background Art
[0002] The 3D point cloud images in the existing technology cannot show what the object corresponds to, that is, they do not distinguish the objects, and the semantic level is low, which is not conducive to the vehicle to accurately and efficiently identify the construction area, nor is it conducive to the subsequent accurate and efficient adjustment of driving strategies based on the information of the construction area. Therefore, autonomous driving vehicles often have driving failures in construction areas.
[0003] Therefore, there is an urgent need for a method that enables vehicles to drive safely in construction areas.
[0004] The above information disclosed in the background technology section is only used to enhance the understanding of the background technology of the technology described in this article. Therefore, the background technology may contain certain information that does not form the prior art known in this country for those skilled in the art. Summary of the invention
[0005] The main purpose of this application is to provide a semantic map construction method, construction device, computer-readable storage medium, processor and vehicle system to solve the problem of frequent driving failures of autonomous driving vehicles in construction areas in the prior art.
[0006] According to one aspect of an embodiment of the present invention, a method for constructing a semantic map is provided, comprising: receiving first obstacle information sent by a vehicle, the first obstacle information comprising location information, shape information and state information of the first obstacle, the first obstacle being an obstacle detected by the vehicle or an obstacle in an initial semantic map, the state information comprising first state information, second state information and third state information, wherein the first state information is information characterizing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information characterizing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information characterizing that the obstacle detected by the vehicle is reduced compared to the initial semantic map; updating a geospatial database according to the first obstacle information, the geospatial database comprising object information, the object information comprising an ID, shape information and location information of the object; and constructing a semantic map of the construction area according to the updated geospatial database.
[0007] Optionally, receiving the first obstacle information sent by a vehicle includes: receiving the first obstacle information respectively sent by a plurality of the vehicles.
[0008] Optionally, updating the geospatial database according to the first obstacle information includes: determining a first predetermined obstacle according to multiple first obstacle information, the first predetermined obstacle being the same first obstacle among the multiple first obstacles, deleting the first obstacle information corresponding to redundant first predetermined obstacles, and retaining only one first obstacle information corresponding to the first predetermined obstacle, to obtain second obstacle information; and updating the geospatial database according to the second obstacle information.
[0009] Optionally, determining the first predetermined obstacle based on the multiple first obstacle information includes: determining the area intersection-and-union ratio of any two of the first obstacles in the multiple first obstacle information, the area intersection-and-union ratio being the ratio of the area of the intersecting part to the total area, the total area being the sum of the areas minus the area of the intersecting part; when the area intersection-and-union ratio is greater than the intersection-and-union ratio threshold, determining that the two first obstacles are the first predetermined obstacles.
[0010] Optionally, the geospatial database is updated according to the second obstacle information, including: determining whether there is a second predetermined obstacle among the objects corresponding to the object information, the second predetermined obstacle being an object that is the same as the first obstacle corresponding to the second state information and is at a distance less than a predetermined distance; in case the second predetermined obstacle exists, updating the predetermined position information to the position information of the second predetermined obstacle, the predetermined position information being the position information of the first obstacle that is the same as the second predetermined obstacle.
[0011] Optionally, updating the above-mentioned geographic spatial database according to the above-mentioned second obstacle information includes: in the absence of the above-mentioned second predetermined obstacle, adding the above-mentioned first obstacle information including the above-mentioned second status information to the above-mentioned geographic spatial database.
[0012] Optionally, determining whether there is a second predetermined obstacle among the objects corresponding to the above-mentioned object information includes: determining whether there is a prepared object among the objects corresponding to the above-mentioned object information, the prepared object being an object whose distance from the above-mentioned first obstacle corresponding to the above-mentioned second state information is less than the above-mentioned predetermined distance; in the case where the above-mentioned prepared object exists, determining whether the above-mentioned prepared object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information; in the case where the above-mentioned prepared object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information, determining that the corresponding above-mentioned prepared object is the above-mentioned second predetermined obstacle.
[0013] Optionally, determining whether there is a preliminary object in the object corresponding to the above-mentioned object information includes: determining a first border, the first border being the smallest rectangle including the polygonal outline of the object corresponding to the above-mentioned object information; determining a second border, the second border being the smallest rectangle including the polygonal outline of the above-mentioned first obstacle corresponding to the above-mentioned second state information; determining a preliminary border based on the position of the above-mentioned first border and the position of the above-mentioned second border, the above-mentioned preliminary border being the first border whose distance with the above-mentioned second border is less than the above-mentioned predetermined distance; calculating the distance between the above-mentioned polygonal outline corresponding to the above-mentioned preliminary border and the polygonal outline corresponding to the above-mentioned second state information to obtain multiple preliminary distances; determining the above-mentioned object whose corresponding above-mentioned preliminary distance is less than the above-mentioned predetermined distance is the above-mentioned preliminary object.
[0014] Optionally, determining whether the prepared object is the same as the first obstacle corresponding to the second state information includes: determining an area intersection-and-union ratio of the prepared object and the first obstacle corresponding to the second state information, the area intersection-and-union ratio being a ratio of an area of an intersecting part to a total area; when the area intersection-and-union ratio is greater than an intersection-and-union ratio threshold, the prepared object is the same as the first obstacle corresponding to the second state information; when the area intersection-and-union ratio is less than or equal to the intersection-and-union ratio threshold, the prepared object is not the same as the first obstacle corresponding to the second state information.
[0015] Optionally, the first obstacle information further includes an ID of the first obstacle, and the object information further includes an ID of the object.
[0016] Optionally, updating the geospatial database according to the second obstacle information includes: updating the state information in the second obstacle information including the first state information to the object information with the same ID.
[0017] Optionally, updating the geospatial database according to the second obstacle information includes: updating the status information in the second obstacle information including the third status information to the object information with the same ID.
[0018] Optionally, constructing a semantic map of the construction area according to the updated geospatial database includes: updating the outline of the construction area in the initial semantic map according to the updated geospatial database to obtain the semantic map.
[0019] According to another aspect of an embodiment of the present invention, a semantic map construction device is also provided, including a receiving unit, an updating unit and a construction unit, wherein the receiving unit is used to receive first obstacle information sent by a vehicle, the first obstacle information includes location information, shape information and state information of the first obstacle, the first obstacle is an obstacle detected by the vehicle or an obstacle in the initial semantic map, the state information includes first state information, second state information and third state information, wherein the first state information is information representing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information representing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information representing that the obstacle detected by the vehicle is reduced compared to the initial semantic map; the updating unit is used to update a geospatial database according to the first obstacle information, the geospatial database includes object information, and the object information includes ID, shape information and location information of the object; the construction unit is used to construct a semantic map of the construction area according to the updated geospatial database.
[0020] According to another aspect of the embodiments of the present invention, a computer-readable storage medium is provided, wherein the storage medium includes a stored program, wherein the program executes any one of the above methods.
[0021] According to yet another aspect of the embodiments of the present invention, a processor is provided, and the processor is used to run a program, wherein the program executes any one of the above methods when running.
[0022] According to another aspect of an embodiment of the present invention, a vehicle system is also provided, comprising: a vehicle, one or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include means for executing any one of the methods described above.
[0023] The method for constructing a semantic map of the present application firstly receives the first obstacle information sent by the vehicle, the first obstacle information includes the position information, shape information and state information of the first obstacle, the first obstacle is the obstacle detected by the vehicle or the obstacle in the initial semantic map, the state information includes the first state information, the second state information and the third state information; then, the geospatial database is updated according to the first obstacle information, the geospatial database includes object information, the object information includes the ID, shape information and position information of the object, and finally, the semantic map of the construction area is constructed according to the updated geospatial database. In the method, by receiving the position information, shape information and state information of the first obstacle in the construction area, the object information such as the ID, shape information and position information of the object in the geospatial database is updated in real time, and the semantic map of the construction area is constructed according to the updated geospatial database, so as to ensure that the constructed semantic map basically matches the current construction area, and ensure that the constructed semantic map is relatively accurate, so as to ensure that the vehicle can accurately and efficiently identify the obstacle information in the construction area according to the constructed semantic map, avoid the problem of driving failure of the vehicle in the construction area, and ensure the safe driving of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS
[0024] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:
[0025] Figure 1 A schematic diagram showing a flow chart of a method for constructing a semantic map according to an embodiment of the present application;
[0026] Figure 2 A schematic diagram of a semantic map construction device according to an embodiment of the present application is shown. DETAILED DESCRIPTION
[0027] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.
[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.
[0030] As mentioned in the background technology, autonomous driving vehicles in the prior art often experience driving failures in construction areas. In order to solve the above problems, in a typical embodiment of the present application, a semantic map construction method, construction device, computer-readable storage medium, processor and vehicle system are provided.
[0031] According to an embodiment of the present application, a method for constructing a semantic map is provided.
[0032] Figure 1 FIG. 1 is a flow chart of a method for constructing a semantic map according to an embodiment of the present application. Figure 1 As shown, the method comprises the following steps:
[0033] Step S101, receiving first obstacle information sent by a vehicle, the first obstacle information including position information, shape information and state information of the first obstacle, the first obstacle being an obstacle detected by the vehicle or an obstacle in the initial semantic map, the state information including first state information, second state information and third state information, wherein the first state information is information representing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information representing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information representing that the obstacle detected by the vehicle is reduced compared to the initial semantic map;
[0034] Step S102, updating a geospatial database according to the first obstacle information, the geospatial database including object information, the object information including object ID, shape information and location information;
[0035] Step S103: constructing a semantic map of the construction area based on the updated geospatial database.
[0036] The method for constructing the semantic map, first, receives the first obstacle information sent by the vehicle, the first obstacle information includes the position information, shape information and state information of the first obstacle, the first obstacle is the obstacle detected by the vehicle or the obstacle in the initial semantic map, the state information includes the first state information, the second state information and the third state information; then, updates the geospatial database according to the first obstacle information, the geospatial database includes object information, the object information includes the ID, shape information and position information of the object, and finally constructs the semantic map of the construction area according to the updated geospatial database. In the method, by receiving the position information, shape information and state information of the first obstacle in the construction area, the object information such as the ID, shape information and position information of the object in the geospatial database is updated in real time, and the semantic map of the construction area is constructed according to the updated geospatial database, so as to ensure that the constructed semantic map basically matches the current construction area, and ensure that the constructed semantic map is relatively accurate, thereby ensuring that the vehicle can accurately and efficiently identify the obstacle information in the construction area according to the constructed semantic map, avoid the problem of driving failure of the vehicle in the construction area, and ensure the safe driving of the vehicle.
[0037] In actual application, the type of the first obstacle may include one or more of cone barrels, water barriers and boards. Of course, the type of the first obstacle may also be other obstacles.
[0038] In order to further ensure that the constructed semantic map is relatively accurate, and further ensure that the vehicle can accurately and efficiently identify obstacle information in the construction area based on the constructed semantic map to ensure driving safety, according to a specific embodiment of the present application, receiving the first obstacle information sent by the vehicle includes: receiving the first obstacle information sent by multiple vehicles. By receiving the first obstacle information sent by multiple vehicles, and constructing a semantic map of the construction area based on the first obstacle information of multiple vehicles, it is further ensured that the constructed semantic map basically matches the current construction area.
[0039] In actual application, there are multiple vehicles passing through the same area at the same time and reporting the same obstacle. Since the information of obstacles reported by different vehicles may be different, it is necessary to deduplicate the information reported by different vehicles. According to another specific embodiment of the present application, the geospatial database is updated according to the first obstacle information, including: determining a first predetermined obstacle according to the plurality of first obstacle information, the first predetermined obstacle being the same first obstacle among the plurality of first obstacles, deleting the first obstacle information corresponding to the redundant first predetermined obstacles, and retaining only the first obstacle information corresponding to one of the first predetermined obstacles, to obtain the second obstacle information; and updating the geospatial database according to the second obstacle information. By deleting the first obstacle information corresponding to the redundant first predetermined obstacles and retaining only the first obstacle information corresponding to one of the first predetermined obstacles, it is ensured that the duplicate information in the first obstacle information is removed, and the obtained second obstacle information is relatively concise, which is convenient for the subsequent updating of the geospatial database according to the second obstacle information and the construction of the semantic map.
[0040] In order to ensure that the first predetermined obstacle is determined accurately, avoid mistaken deletion of different first obstacle information, and further ensure that the constructed semantic map is more accurate, in actual application, the first predetermined obstacle is determined according to the multiple first obstacle information, including: determining the area intersection-and-union ratio of any two of the first obstacles in the multiple first obstacle information, the area intersection-and-union ratio is the ratio of the area of the intersection part to the total area, and the total area is the sum of the areas minus the area of the intersection part; when the area intersection-and-union ratio is greater than the intersection-and-union ratio threshold, determining the two first obstacles as the first predetermined obstacles.
[0041] According to another specific embodiment of the present application, the above-mentioned geospatial database is updated according to the above-mentioned second obstacle information, including: determining whether there is a second predetermined obstacle among the objects corresponding to the above-mentioned object information, the above-mentioned second predetermined obstacle is an object that is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information and the distance is less than the predetermined distance; in the case where the above-mentioned second predetermined obstacle exists, the predetermined position information is updated to the position information of the second predetermined obstacle, and the predetermined position information is the position information of the first obstacle that is the same as the second predetermined obstacle. In the above-mentioned method, in the case where the above-mentioned second predetermined obstacle exists, that is, the above-mentioned second predetermined obstacle information exists in the above-mentioned geospatial database, the position information of the above-mentioned first obstacle that is the same as the above-mentioned second predetermined obstacle is updated to the position information of the above-mentioned second predetermined obstacle, thereby ensuring that the updated above-mentioned geospatial database is more accurate, and further ensuring that the semantic map of the construction area subsequently constructed according to the above-mentioned geospatial database is more accurate.
[0042] In another specific embodiment of the present application, the geospatial database is updated according to the second obstacle information, including: in the absence of the second predetermined obstacle, the first obstacle information including the second state information is added to the geospatial database. In the above method, in the absence of the second predetermined obstacle, the first obstacle information including the second state information is added to the geospatial database, that is, the second obstacle information obtained after the update adds the information of the obstacles newly added compared to the initial semantic map, further ensuring that the second obstacle information obtained after the update is basically matched with the obstacle information of the current construction area, and further ensuring that the semantic map constructed subsequently is relatively accurate.
[0043] In order to further ensure that the result of determining whether the above-mentioned second predetermined obstacle exists is relatively accurate, and thus ensure that the subsequently constructed map is relatively accurate, according to another specific embodiment of the present application, determining whether there is a second predetermined obstacle among the objects corresponding to the above-mentioned object information includes: determining whether there is a prepared object among the objects corresponding to the above-mentioned object information, the prepared object being an object whose distance from the above-mentioned first obstacle corresponding to the above-mentioned second state information is less than the above-mentioned predetermined distance; in the case where the above-mentioned prepared object exists, determining whether the above-mentioned prepared object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information; in the case where the above-mentioned prepared object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information, determining that the corresponding above-mentioned prepared object is the above-mentioned second predetermined obstacle.
[0044] In actual application, when representing the distance between irregular obstacles, since the coordinates of obstacles in a construction area in the direction perpendicular to the ground are relatively close, the obstacle is projected onto the ground and approximately regarded as a polygonal obstacle on a two-dimensional plane. The coordinates of the vertices and the center point of the polygonal obstacle are determined on the two-dimensional plane. The above coordinates are relative to the origin of the custom semantic map to calculate the distance between the obstacles. Finally, the coordinates in the direction perpendicular to the ground are filtered to prevent the vertical coordinates between the obstacles from being too different.
[0045] According to another specific embodiment of the present application, determining whether there is a preliminary object in the object corresponding to the above-mentioned object information includes: determining a first border, the above-mentioned first border is the smallest rectangle including the polygonal outline of the object corresponding to the above-mentioned object information; determining a second border, the above-mentioned second border is the smallest rectangle including the polygonal outline of the above-mentioned first obstacle corresponding to the above-mentioned second state information; determining a preliminary border based on the position of the above-mentioned first border and the position of the above-mentioned second border, the above-mentioned preliminary border is a first border whose distance from the above-mentioned second border is less than the above-mentioned predetermined distance; calculating the distance between the above-mentioned polygonal outline corresponding to the above-mentioned preliminary border and the polygonal outline corresponding to the above-mentioned second state information to obtain multiple preliminary distances; determining the above-mentioned object whose corresponding above-mentioned preliminary distance is less than the above-mentioned predetermined distance is the above-mentioned preliminary object. The above method determines the first frame and the second frame, and determines the preliminary frame according to the position of the first frame and the position of the second frame, calculates multiple preliminary distances between the polygonal outline corresponding to the preliminary frame and the polygonal outline corresponding to the second state information, and determines whether the corresponding preliminary distance is less than the predetermined distance. When the preliminary distance is less than the predetermined distance, the corresponding physical object is the preliminary object. This further ensures that the determination result of whether there is a preliminary object is relatively accurate, thereby further ensuring that the subsequent construction of the semantic map is relatively accurate, and further ensuring the driving safety of the vehicle.
[0046] In order to further ensure that whether the above-mentioned prepared object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information is determined more accurately, in the actual application process, determining whether the above-mentioned prepared object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information includes: determining the area intersection and union ratio of the above-mentioned prepared object and the above-mentioned first obstacle corresponding to the second state information, the above-mentioned area intersection and union ratio is the ratio of the area of the intersection part to the total area, and the total area is the sum of the areas minus the area of the intersection part; when the above-mentioned area intersection and union ratio is greater than the intersection and union ratio threshold, the above-mentioned prepared object is the same as the above-mentioned first obstacle corresponding to the second state information; when the above-mentioned area intersection and union ratio is less than or equal to the intersection and union ratio threshold, the above-mentioned prepared object is not the same as the above-mentioned first obstacle corresponding to the second state information.
[0047] In another specific embodiment of the present application, the first obstacle information further includes an ID of the first obstacle, and the object information further includes an ID of the object.
[0048] Specifically, the obstacle ID is the self-incrementing ID (i.e., serial number) of the obstacle inserted into the database. The obstacle ID will be included in the initial semantic map distributed to the vehicle, as well as in the first state information and the third state information. The first state information represents objects that the vehicle sees in the real world and are similar in position, shape, and height to the corresponding ID in the semantic map. The third state information represents that the vehicle cannot see similar objects near the corresponding ID position in the semantic map in the real world. For newly added objects, the ID is not included in the information reported by the vehicle, and is assigned by the server.
[0049] The server will first search the geospatial database for the same object (similar in position, shape and height) based on the position and shape information. If so, the position and shape information of the same object in the geospatial database will be updated to the latest observation information (i.e., the data reported by the vehicle). If it is found that there are multiple objects in the geospatial database with the same obstacle information as the obstacle reported by the vehicle (the position and shape are highly similar - in a high concurrency scenario, multiple vehicles report the same new object at the same time), only the original obstacle will be retained, and other duplicate obstacle information in the database will be deleted, and the position, outline and shape information of the retained obstacle will be updated (i.e., the ID will be kept as the smallest ID among the duplicate obstacles, the obstacle ID will be kept as stable as possible, and the information will be the latest information). If it is found that there is no identical obstacle in the database, the obstacle information will be inserted into the database, and the newly inserted obstacle information will automatically get an ID according to its insertion order.
[0050] In another specific embodiment of the present application, the geospatial database is updated according to the second obstacle information, including: updating the status information in the second obstacle information including the first status information to the object information of the same ID.
[0051] In actual application, the above object information also includes update time. The above method checks the update time of obstacles in the above geospatial database through a scheduled task. If the obstacle has not been updated for a long time, it is considered that the obstacle has been removed. If all obstacles in the construction area have been completely removed, it is considered that the entire construction area has been removed.
[0052] In the actual application process, the geospatial database is updated according to the second obstacle information, including: updating the state information in the second obstacle information including the third state information to the object information with the same ID. The method further ensures that the updated geospatial database is more accurate by updating the state information in the second obstacle information including the third state information to the object information with the same ID. The semantic map of the construction area constructed according to the updated geospatial database basically matches the actual construction area, further ensuring the safe driving of the vehicle in the construction area.
[0053] In order to further ensure that the above-mentioned semantic map basically matches the current construction area, and to further ensure that the constructed semantic map is relatively accurate, in another specific embodiment of the present application, a semantic map of the construction area is constructed according to the updated above-mentioned geospatial database, including: updating the outline of the above-mentioned construction area in the above-mentioned initial semantic map according to the updated above-mentioned geospatial database to obtain the above-mentioned semantic map.
[0054] It should be noted that the outline of the above-mentioned construction area is the convex hull of all obstacles in the construction area. The construction area in this application is the area between obstacles whose distance is within a predetermined threshold, that is, the distance between any two adjacent obstacles in a construction area is less than the predetermined threshold. For multiple obstacles, it may constitute one construction area or multiple construction areas, and the specific situation should be determined and divided according to the distance between adjacent obstacles.
[0055] In the actual application process, in order to further ensure that the constructed semantic map is more accurate, it is necessary to merge the construction areas. For each construction area, the ID of the obstacles it includes is recorded. For each obstacle, the ID of the construction area to which it belongs is recorded. The specific steps for merging the construction areas include: according to the received first obstacle information, the state information in the above-mentioned second obstacle information including the above-mentioned first state information is updated to the above-mentioned object information with the same ID, and the state information in the above-mentioned second obstacle information including the above-mentioned third state information is updated to the above-mentioned object information with the same ID, and then the outline of the construction area is updated, and the obstacles near the construction area are determined according to the polygonal obstacles on the two-dimensional plane to merge adjacent construction areas.
[0056] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.
[0057] The embodiment of the present application also provides a device for constructing a semantic map. It should be noted that the device for constructing a semantic map in the embodiment of the present application can be used to execute the method for constructing a semantic map provided in the embodiment of the present application. The following introduces the device for constructing a semantic map provided in the embodiment of the present application.
[0058] Figure 2 Schematic diagram of a device for constructing a semantic map according to an embodiment of the present application. Figure 2 As shown, the device includes a receiving unit 10, an updating unit 20 and a constructing unit 30, wherein the receiving unit 10 is used to receive first obstacle information sent by a vehicle, the first obstacle information includes location information, shape information and state information of the first obstacle, the first obstacle is an obstacle detected by the vehicle or an obstacle in the initial semantic map, the state information includes first state information, second state information and third state information, wherein the first state information is information representing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information representing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information representing that the obstacle detected by the vehicle is reduced compared to the initial semantic map; the updating unit 20 is used to update the geospatial database according to the first obstacle information, the geospatial database includes object information, and the object information includes ID, shape information and location information of the object; the constructing unit 30 is used to construct a semantic map of the construction area according to the updated geospatial database.
[0059] The device for constructing the semantic map receives the first obstacle information sent by the vehicle through the receiving unit, the first obstacle information includes the position information, shape information and state information of the first obstacle, the first obstacle is the obstacle detected by the vehicle or the obstacle in the initial semantic map, the state information includes the first state information, the second state information and the third state information, according to the first obstacle information, the updating unit updates the geospatial database, the geospatial database includes object information, the object information includes the ID, shape information and position information of the object, and according to the updated geospatial database, the semantic map of the construction area is constructed by the construction unit. In the device, by receiving the position information, shape information and state information of the first obstacle in the construction area, the object information such as the ID, shape information and position information of the object in the geospatial database is updated in real time, and the semantic map of the construction area is constructed according to the updated geospatial database, so as to ensure that the constructed semantic map basically matches the current construction area, and ensure that the constructed semantic map is relatively accurate, thereby ensuring that the vehicle can accurately and efficiently identify the obstacle information in the construction area according to the constructed semantic map, avoid the problem of driving failure of the vehicle in the construction area, and ensure the safe driving of the vehicle.
[0060] In actual application, the type of the first obstacle may include one or more of cone barrels, water barriers and boards. Of course, the type of the first obstacle may also be other obstacles.
[0061] In order to further ensure that the constructed semantic map is relatively accurate, and further ensure that the vehicle can accurately and efficiently identify obstacle information in the construction area based on the constructed semantic map to ensure driving safety, according to a specific embodiment of the present application, the above-mentioned receiving unit includes a receiving subunit, and the above-mentioned receiving subunit is used to receive the above-mentioned first obstacle information sent by the above-mentioned multiple vehicles. By receiving the above-mentioned first obstacle information sent by the above-mentioned multiple vehicles respectively, and constructing a semantic map of the construction area based on the above-mentioned first obstacle information of the multiple vehicles, it is further ensured that the constructed semantic map basically matches the current construction area.
[0062] In actual application, there are multiple vehicles passing through the same area at the same time and reporting the same obstacle. Since the information of obstacles reported by different vehicles may be different, it is necessary to deduplicate the information reported by different vehicles. According to another specific embodiment of the present application, the updating unit includes a determining subunit and a first updating subunit, wherein the determining subunit is used to determine a first predetermined obstacle according to the plurality of first obstacle information, the first predetermined obstacle is the same first obstacle among the plurality of first obstacles, delete the first obstacle information corresponding to the redundant first predetermined obstacles, and only retain the first obstacle information corresponding to one of the first predetermined obstacles, to obtain the second obstacle information; the first updating subunit is used to update the geospatial database according to the second obstacle information. By deleting the first obstacle information corresponding to the redundant first predetermined obstacles and retaining the first obstacle information corresponding to one of the first predetermined obstacles, it is ensured that the duplicate information in the first obstacle information is removed, and the obtained second obstacle information is relatively concise, which is convenient for the subsequent updating of the geospatial database according to the second obstacle information and the construction of the semantic map.
[0063] In order to ensure that the determined first predetermined obstacle is relatively accurate, avoid mistaken deletion of different first obstacle information, and further ensure that the constructed semantic map is relatively accurate, in the actual application process, the determination subunit includes a first determination module and a second determination module, wherein the first determination module is used to determine the area intersection and union ratio of any two of the first obstacles in the plurality of the first obstacle information, the area intersection and union ratio is the ratio of the area of the intersection to the total area, the total area is the sum of the areas minus the area of the intersection, for example, the area of the intersection of the two first obstacles is 1cm 2 , the sum of the areas of the two first obstacles is 4cm 2 , so the total area of the two is 3cm 2 , the area intersection-and-union ratio is 1 / 3; the second determination module is used to determine that the two first obstacles are the first predetermined obstacles when the area intersection-and-union ratio is greater than the intersection-and-union ratio threshold.
[0064] According to another specific embodiment of the present application, the first updating subunit includes a third determining module and a first updating module, wherein the third determining module is used to determine whether there is a second predetermined obstacle in the object corresponding to the object information, the second predetermined obstacle being an object that is the same as the first obstacle corresponding to the second state information and is at a distance less than a predetermined distance; the first updating module is used to update the predetermined position information to the position information of the second predetermined obstacle when the second predetermined obstacle exists, the predetermined position information being the position information of the first obstacle that is the same as the second predetermined obstacle. The device, when the second predetermined obstacle exists, that is, the second predetermined obstacle information exists in the geospatial database, updates the position information of the first obstacle that is the same as the second predetermined obstacle to the position information of the second predetermined obstacle, thereby ensuring that the updated geospatial database is more accurate, and further ensuring that the semantic map of the construction area subsequently constructed based on the geospatial database is more accurate.
[0065] In another specific embodiment of the present application, the first updating subunit further includes an adding module, and the adding module is used to add the first obstacle information including the second state information to the geospatial database when the second predetermined obstacle does not exist. The device adds the first obstacle information including the second state information to the geospatial database when the second predetermined obstacle does not exist, that is, the second obstacle information obtained after the update adds the information of the obstacles newly added compared to the initial semantic map, further ensuring that the second obstacle information obtained after the update basically matches the obstacle information of the current construction area, and further ensuring that the semantic map constructed subsequently is more accurate.
[0066] In order to further ensure that the result of determining whether the above-mentioned second predetermined obstacle exists is relatively accurate, and thus ensure that the semantic map constructed subsequently is relatively accurate, according to another specific embodiment of the present application, the above-mentioned third determination module includes a first determination submodule, a second determination submodule and a third determination submodule, wherein the above-mentioned first determination submodule is used to determine whether there is a preparatory object among the objects corresponding to the above-mentioned object information, and the above-mentioned preparatory object is an object whose distance with the above-mentioned first obstacle corresponding to the above-mentioned second state information is less than the above-mentioned predetermined distance; the above-mentioned second determination submodule is used to determine whether the above-mentioned preparatory object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information when the above-mentioned preparatory object exists; and the above-mentioned third determination submodule is used to determine that the corresponding above-mentioned preparatory object is the above-mentioned second predetermined obstacle when the above-mentioned preparatory object is the same as the above-mentioned first obstacle corresponding to the above-mentioned second state information.
[0067] In actual application, when representing the distance between irregular obstacles, since the coordinates of obstacles in a construction area in the direction perpendicular to the ground are relatively close, the obstacle is projected onto the ground and approximately regarded as a polygonal obstacle on a two-dimensional plane. The coordinates of the vertices and the center point of the polygonal obstacle are determined on the two-dimensional plane. The above coordinates are relative to the origin of the custom semantic map to calculate the distance between the obstacles. Finally, the coordinates in the direction perpendicular to the ground are filtered to prevent the vertical coordinates between the obstacles from being too different.
[0068] According to another specific embodiment of the present application, the above-mentioned first determination submodule is also used to determine a first border, the above-mentioned first border is the smallest rectangle including the polygonal outline of the object corresponding to the above-mentioned object information; determine a second border, the above-mentioned second border is the smallest rectangle including the polygonal outline of the above-mentioned first obstacle corresponding to the above-mentioned second state information; determine a preliminary border based on the position of the above-mentioned first border and the position of the above-mentioned second border, the above-mentioned preliminary border is the first border whose distance with the above-mentioned second border is less than the above-mentioned predetermined distance; calculate the distance between the above-mentioned polygonal outline corresponding to the above-mentioned preliminary border and the polygonal outline corresponding to the above-mentioned second state information to obtain multiple preliminary distances; determine the above-mentioned object whose corresponding above-mentioned preliminary distance is less than the above-mentioned predetermined distance as the above-mentioned preliminary object. The above-mentioned device determines the first frame and the second frame, and determines the preliminary frame according to the position of the above-mentioned first frame and the position of the above-mentioned second frame, calculates multiple preliminary distances between the above-mentioned polygonal outline corresponding to the above-mentioned preliminary frame and the polygonal outline corresponding to the above-mentioned second state information, and determines whether the corresponding above-mentioned preliminary distance is less than the above-mentioned predetermined distance. When the above-mentioned preliminary distance is less than the above-mentioned predetermined distance, the corresponding above-mentioned physical object is the above-mentioned preliminary object, which further ensures that the determination result of whether there is a preliminary object is relatively accurate, thereby further ensuring that the subsequent construction of the semantic map is relatively accurate, and further ensuring the driving safety of the vehicle.
[0069] In order to further ensure a relatively accurate determination of whether the prepared object is the same as the first obstacle corresponding to the second state information, in actual application, the second determination submodule is further used to determine the area intersection-and-union ratio of the prepared object and the first obstacle corresponding to the second state information, where the area intersection-and-union ratio is the ratio of the area of the intersecting part to the total area; when the area intersection-and-union ratio is greater than the intersection-and-union ratio threshold, the prepared object is the same as the first obstacle corresponding to the second state information; when the area intersection-and-union ratio is less than or equal to the intersection-and-union ratio threshold, the prepared object is not the same as the first obstacle corresponding to the second state information.
[0070] In another specific embodiment of the present application, the first obstacle information further includes an ID of the first obstacle, and the object information further includes an ID of the object.
[0071] In another specific embodiment of the present application, the first updating subunit further includes a second updating module, and the second updating module is used to update the state information in the second obstacle information including the first state information to the object information of the same ID.
[0072] In actual application, the above-mentioned object information also includes update time. The above-mentioned device checks the update time of the obstacles in the above-mentioned geographic spatial database through a scheduled task. If the obstacle has not been updated for a long time, it is considered that the obstacle has been removed. If all obstacles in the construction area have been completely removed, it is considered that the entire construction area has been removed.
[0073] In the actual application process, the first updating subunit further includes a third updating module, and the third updating module is used to update the state information in the second obstacle information including the third state information to the object information of the same ID. The device further ensures that the updated geospatial database is more accurate by updating the state information in the second obstacle information including the third state information to the object information of the same ID, and the semantic map of the construction area constructed according to the updated geospatial database basically matches the actual construction area, further ensuring the safe driving of the vehicle in the construction area.
[0074] In order to further ensure that the above-mentioned semantic map basically matches the current construction area, and further ensure that the constructed semantic map is more accurate, in another specific embodiment of the present application, the above-mentioned construction unit includes a second updating sub-unit, and the above-mentioned second updating sub-unit is used to update the outline of the above-mentioned construction area in the above-mentioned initial semantic map according to the updated above-mentioned geographic spatial database to obtain the above-mentioned semantic map.
[0075] It should be noted that the outline of the above-mentioned construction area is the convex hull containing all obstacles in the construction area.
[0076] In the actual application process, in order to further ensure that the constructed semantic map is more accurate, it is necessary to merge the construction areas. For each construction area, the ID of the obstacles it includes is recorded. For each obstacle, the ID of the construction area to which it belongs is recorded. The specific steps for merging the construction areas include: according to the received first obstacle information, the state information in the above-mentioned second obstacle information including the above-mentioned first state information is updated to the above-mentioned object information with the same ID, and the state information in the above-mentioned second obstacle information including the above-mentioned third state information is updated to the above-mentioned object information with the same ID, and then the outline of the construction area is updated, and the obstacles near the construction area are determined according to the polygonal obstacles on the two-dimensional plane to merge adjacent construction areas.
[0077] The apparatus for constructing the semantic map comprises a processor and a memory. The receiving unit, updating unit and constructing unit are all stored in the memory as program units. The processor executes the program units stored in the memory to implement corresponding functions.
[0078] The processor includes a kernel, which retrieves the corresponding program unit from the memory. One or more kernels can be set, and the problem of frequent driving failures of autonomous driving vehicles in construction areas in the prior art can be solved by adjusting kernel parameters.
[0079] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.
[0080] An embodiment of the present invention provides a storage medium on which a program is stored. When the program is executed by a processor, the method for constructing the semantic map is implemented.
[0081] An embodiment of the present invention provides a processor, which is used to run a program, wherein the method for constructing a semantic map is executed when the program is running.
[0082] An embodiment of the present invention provides a device, the device including a processor, a memory, and a program stored in the memory and executable on the processor, and when the processor executes the program, at least the following steps are implemented:
[0083] Step S101, receiving first obstacle information sent by a vehicle, the first obstacle information including position information, shape information and state information of the first obstacle, the first obstacle being an obstacle detected by the vehicle or an obstacle in the initial semantic map, the state information including first state information, second state information and third state information, wherein the first state information is information representing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information representing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information representing that the obstacle detected by the vehicle is reduced compared to the initial semantic map;
[0084] Step S102, updating a geospatial database according to the first obstacle information, the geospatial database including object information, the object information including object ID, shape information and location information;
[0085] Step S103: constructing a semantic map of the construction area based on the updated geospatial database.
[0086] The devices in this article can be servers, PCs, PADs, mobile phones, etc.
[0087] The present application also provides a computer program product, which, when executed on a data processing device, is suitable for executing a program for initializing at least the following method steps:
[0088] Step S101, receiving first obstacle information sent by a vehicle, the first obstacle information including position information, shape information and state information of the first obstacle, the first obstacle being an obstacle detected by the vehicle or an obstacle in the initial semantic map, the state information including first state information, second state information and third state information, wherein the first state information is information representing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information representing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information representing that the obstacle detected by the vehicle is reduced compared to the initial semantic map;
[0089] Step S102, updating a geospatial database according to the first obstacle information, the geospatial database including object information, the object information including object ID, shape information and location information;
[0090] Step S103: constructing a semantic map of the construction area based on the updated geospatial database.
[0091] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.
[0092] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the above-mentioned units can be 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 interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.
[0093] The units described above 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 units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.
[0094] In addition, each functional unit in each embodiment of the present invention 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. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0095] According to another typical embodiment of the present application, a vehicle system is also provided, comprising: a vehicle, one or more processors, a memory and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include means for executing any one of the methods described above.
[0096] The above-mentioned vehicle system includes a car, one or more processors, a memory and one or more programs, wherein the one or more programs include a method for executing any of the above-mentioned methods. The above-mentioned vehicle system, through the program therein, executes any of the above-mentioned methods, thereby realizing real-time updating of object information such as ID, shape information and location information of objects in the geospatial database, and constructs a semantic map of the construction area based on the updated geospatial database, thereby ensuring that the constructed semantic map basically matches the current construction area, and ensuring that the constructed semantic map is relatively accurate, thereby ensuring that the vehicle can accurately and efficiently identify obstacle information in the construction area based on the constructed semantic map, thereby avoiding driving failures of the vehicle in the construction area and ensuring safe driving of the vehicle.
[0097] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of 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 above-mentioned methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.
[0098] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:
[0099] 1) The method for constructing a semantic map of the present application, first, receives the first obstacle information sent by the vehicle, the first obstacle information includes the position information, shape information and state information of the first obstacle, the first obstacle is the obstacle detected by the vehicle or the obstacle in the initial semantic map, the state information includes the first state information, the second state information and the third state information; then, updates the geospatial database according to the first obstacle information, the geospatial database includes object information, the object information includes the ID, shape information and position information of the object, and finally constructs the semantic map of the construction area according to the updated geospatial database. In this method, by receiving the position information, shape information and state information of the first obstacle in the construction area, the object information such as the ID, shape information and position information of the object in the geospatial database is updated in real time, and the semantic map of the construction area is constructed according to the updated geospatial database, so as to ensure that the constructed semantic map basically matches the current construction area, and ensure that the constructed semantic map is relatively accurate, so as to ensure that the vehicle can accurately and efficiently identify the obstacle information in the construction area according to the constructed semantic map, avoid the problem of driving failure of the vehicle in the construction area, and ensure the safe driving of the vehicle.
[0100] 2) The semantic map construction device of the present application receives the first obstacle information sent by the vehicle through the above-mentioned receiving unit, the above-mentioned first obstacle information includes the position information, shape information and state information of the first obstacle, the above-mentioned first obstacle is an obstacle detected by the above-mentioned vehicle or an obstacle in the initial semantic map, the above-mentioned state information includes first state information, second state information and third state information, according to the above-mentioned first obstacle information, the above-mentioned updating unit updates the geospatial database, the above-mentioned geospatial database includes object information, the above-mentioned object information includes the ID, shape information and position information of the object, according to the updated above-mentioned geospatial database, the above-mentioned construction unit constructs a semantic map of the construction area. In the device, by receiving the position information, shape information and status information of the first obstacle in the construction area, real-time updating of object information such as the ID, shape information and position information of the object in the geospatial database is achieved. According to the updated geospatial database, a semantic map of the construction area is constructed, which ensures that the constructed semantic map basically matches the current construction area and that the constructed semantic map is relatively accurate, thereby ensuring that the vehicle can accurately and efficiently identify obstacle information in the construction area based on the constructed semantic map, avoiding driving failures of the vehicle in the construction area and ensuring safe driving of the vehicle.
[0101] 3) The vehicle system of the present application includes a car, one or more processors, a memory and one or more programs, wherein the one or more programs include a method for executing any of the above methods. The above vehicle system, through the program therein, executes any of the above methods, thereby realizing real-time update of object information such as ID, shape information and location information of objects in the geospatial database, and constructs a semantic map of the construction area based on the updated geospatial database, thereby ensuring that the constructed semantic map basically matches the current construction area, and ensuring that the constructed semantic map is relatively accurate, thereby ensuring that the vehicle can accurately and efficiently identify obstacle information in the construction area based on the constructed semantic map, thereby avoiding driving failures of the vehicle in the construction area and ensuring safe driving of the vehicle.
[0102] The above are only preferred embodiments of the present application and are not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A method for constructing a semantic map, characterized in that: include: receiving first obstacle information sent by a vehicle, the first obstacle information including position information, shape information and state information of the first obstacle, the first obstacle being an obstacle detected by the vehicle or an obstacle in an initial semantic map, the state information including first state information, second state information and third state information, wherein the first state information is information representing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information representing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information representing that the obstacle detected by the vehicle is reduced compared to the initial semantic map; updating a geospatial database according to the first obstacle information, wherein the geospatial database includes object information, and the object information includes an ID, shape information, and location information of the object; Based on the updated geospatial database, a semantic map of the construction area is constructed. Based on the updated geospatial database, a semantic map of the construction area is constructed, including: According to the updated geospatial database, the outline of the construction area in the initial semantic map is updated to obtain the semantic map, Receiving the first obstacle information sent by the vehicle, including: receiving the first obstacle information respectively sent by a plurality of the vehicles, Updating the geospatial database according to the first obstacle information includes: determining a first predetermined obstacle according to the plurality of first obstacle information, where the first predetermined obstacle is the same first obstacle among the plurality of first obstacles, deleting the first obstacle information corresponding to redundant first predetermined obstacles, and retaining only the first obstacle information corresponding to one first predetermined obstacle, to obtain second obstacle information; updating the geospatial database according to the second obstacle information, Determining a first predetermined obstacle according to the plurality of first obstacle information includes: Determine an area intersection-and-combination ratio of any two of the first obstacles in the plurality of first obstacle information, where the area intersection-and-combination ratio is a ratio of an area of an intersecting portion to a total area, and the total area is a sum of areas minus an area of the intersecting portion; When the area intersection-and-union ratio is greater than the intersection-and-union ratio threshold, determining the two first obstacles as the first predetermined obstacles, Updating the geospatial database according to the second obstacle information includes: determining whether there is a second predetermined obstacle among the objects corresponding to the object information, where the second predetermined obstacle is an object that is the same as the first obstacle corresponding to the second state information and is less than a predetermined distance away; In the case where the second predetermined obstacle exists, the predetermined position information is updated to the position information of the second predetermined obstacle, and the predetermined position information is the position information of the first obstacle which is the same as the second predetermined obstacle.
2. The method according to claim 1, characterized in that Updating the geospatial database according to the second obstacle information includes: In the absence of the second predetermined obstacle, the first obstacle information including the second state information is added to the geospatial database.
3. The method according to claim 1, characterized in that Determining whether there is a second predetermined obstacle in the object corresponding to the object information includes: determining whether there is a prepared object among the objects corresponding to the object information, the prepared object being an object whose distance from the first obstacle corresponding to the second state information is less than the predetermined distance; In the case where the prepared object exists, determining whether the prepared object is the same as the first obstacle corresponding to the second state information; In a case where the prepared object is identical to the first obstacle corresponding to the second state information, the corresponding prepared object is determined to be the second predetermined obstacle.
4. The method according to claim 3, characterized in that Determining whether there is a prepared object among the objects corresponding to the object information includes: Determine a first frame, where the first frame is a minimum rectangle including a polygonal outline of the object corresponding to the object information; Determine a second frame, where the second frame is a minimum rectangle including a polygonal outline of the first obstacle corresponding to the second state information; Determine a preliminary frame according to the position of the first frame and the position of the second frame, wherein the preliminary frame is a first frame that is less than the predetermined distance from the second frame; calculating the distance between the polygonal outline corresponding to the preliminary frame and the polygonal outline corresponding to the second state information to obtain a plurality of preliminary distances; The object corresponding to the preparation distance less than the predetermined distance is determined as the preparation object.
5. The method according to claim 3, characterized in that: Determining whether the prepared object is the same as the first obstacle corresponding to the second state information includes: Determine an area intersection-and-merge ratio of the preparation object and the first obstacle corresponding to the second state information, where the area intersection-and-merge ratio is a ratio of an area of the intersection to a total area, and the total area is a sum of areas minus an area of the intersection; When the area intersection-over-union ratio is greater than the intersection-over-union ratio threshold, the prepared object is the same as the first obstacle corresponding to the second state information; When the area intersection-over-union ratio is less than or equal to an intersection-over-union ratio threshold, the prepared object is different from the first obstacle corresponding to the second state information.
6. The method according to any one of claims 1 to 5, characterized in that The first obstacle information further includes an ID of the first obstacle, and the object information further includes an ID of the object.
7. The method according to claim 6, characterized in that Updating the geospatial database according to the second obstacle information includes: The state information and position information in the second obstacle information including the first state information are updated to the object information with the same ID.
8. The method according to claim 6, characterized in that Updating the geospatial database according to the second obstacle information includes: The state information and position information in the second obstacle information including the third state information are updated to the object information with the same ID.
9. A device for constructing a semantic map, characterized in that: include: a receiving unit, configured to receive first obstacle information sent by a vehicle, the first obstacle information including position information, shape information and state information of the first obstacle, the first obstacle being an obstacle detected by the vehicle or an obstacle in an initial semantic map, the state information including first state information, second state information and third state information, wherein the first state information is information representing that the obstacle detected by the vehicle matches the obstacle in the initial semantic map, the second state information is information representing that the obstacle detected by the vehicle is newly added compared to the initial semantic map, and the third state information is information representing that the obstacle detected by the vehicle is reduced compared to the initial semantic map; An updating unit, configured to update a geospatial database according to the first obstacle information, wherein the geospatial database includes object information, and the object information includes an ID, shape information, and location information of the object; A construction unit is used to construct a semantic map of the construction area according to the updated geospatial database, The construction unit includes a second updating subunit, and the second updating subunit is used to update the outline of the construction area in the initial semantic map according to the updated geospatial database to obtain the semantic map. The receiving unit includes a receiving subunit, and the receiving subunit is used to receive the first obstacle information respectively sent by a plurality of the vehicles. The updating unit includes a determining subunit and a first updating subunit, wherein the determining subunit is used to determine a first predetermined obstacle according to a plurality of first obstacle information, the first predetermined obstacle being the same first obstacle among the plurality of first obstacles, deleting the first obstacle information corresponding to redundant first predetermined obstacles, and retaining only the first obstacle information corresponding to one first predetermined obstacle, to obtain second obstacle information; the first updating subunit is used to update the geospatial database according to the second obstacle information, The determination subunit includes a first determination module and a second determination module, wherein the first determination module is used to determine an area intersection-and-union ratio of any two of the first obstacles in the plurality of first obstacle information, the area intersection-and-union ratio being a ratio of an area of an intersecting part to a total area, the total area being a sum of areas minus an area of the intersecting part; and the second determination module is used to determine that the two first obstacles are the first predetermined obstacles when the area intersection-and-union ratio is greater than an intersection-and-union ratio threshold, The first updating subunit includes a third determination module and a first updating module, wherein the third determination module is used to determine whether there is a second predetermined obstacle in the object corresponding to the object information, and the second predetermined obstacle is an object that is the same as the first obstacle corresponding to the second state information and the distance is less than a predetermined distance; the first updating module is used to update the predetermined position information to the position information of the second predetermined obstacle when the second predetermined obstacle exists, and the predetermined position information is the position information of the first obstacle that is the same as the second predetermined obstacle.
10. A computer-readable storage medium, characterized in that: The storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 8.
11. A processor, characterized in that: The processor is used to run a program, wherein the program executes the method according to any one of claims 1 to 8 when running.
12. A vehicle system, characterized in that: include: An automobile, one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include methods for executing any one of claims 1 to 8.
Citation Information
Patent Citations
Method, device, apparatus for updating electronic map and storage medium
CN109284348A
Dangerous road condition reminding method and device, vehicle-mounted terminal, server and medium
CN109785633A
Method and device for positioning obstacles in semantic maps, computer equipment and storage medium
CN110032181A
High definition map updates
CN110832474A
Obstacle tracking method, device and equipment and storage medium
CN110927742A
