Road information updating method and system, electronic equipment and storage medium

Abstract

The invention provides a road information updating method and system, electronic equipment and a storage medium. The method comprises the following steps: determining an abnormal driving event of a reported vehicle in a driving process based on a target map; determining a labeling result of the abnormal driving event, wherein the labeling result at least comprises abnormal position information of an abnormal position where the abnormal driving event occurs and object description information of an abnormal road object associated with the abnormal driving event; and issuing a labeling result to a to-be-updated vehicle, so that the to-be-updated vehicle generates a driving strategy at the abnormal position according to the object description information and the target map.

Description

Technical Field

[0001] The present specification relates to the field of autonomous driving technology, and in particular to methods, systems, electronic devices and storage media for updating road information. Background Art

[0002] Autonomous driving is a technology that enables vehicles to drive automatically without human intervention. The development goal of autonomous driving is to achieve a transition from basic driving assistance to fully autonomous driving, so that vehicles can perceive, make decisions and control autonomously, so as to improve traffic safety, traffic efficiency and free up drivers' energy. High-precision maps play a vital role in autonomous driving. High-definition maps (HD Maps) are map data that is more detailed and accurate than traditional navigation maps, and are used to support environmental perception, positioning and path planning for autonomous driving. For example, HD maps contain lane-level road structures, which provide detailed road layout information for autonomous driving and help the system make more accurate path planning. Especially in complex urban streets, intersections, ramps and other scenes, the detailed data of the map enables vehicles to better cope with changing road environments. For example, HD maps can provide information about road conditions ahead in advance, such as upcoming curves, slopes, speed limits, etc., to help the autonomous driving system adjust speed and steering in advance to ensure smooth and safe driving of the vehicle. However, maintaining the timeliness and accuracy of HD maps is a huge challenge. Road infrastructure often changes, such as construction, new roads, and replacement of traffic signs, which will affect the accuracy of HD maps. If the map is not updated in a timely manner, it may cause perception deviations in the actual driving process of the autonomous driving vehicle, thus affecting driving safety.

[0003] At present, the update of high-precision maps mainly relies on professional surveying and mapping vehicles, which are equipped with high-precision lidar, cameras and GPS devices, and generate and update high-precision maps by collecting a large amount of road data. Although this method has high accuracy, it is costly and has a long data collection and processing cycle, resulting in untimely updates of road information in high-precision maps, making it difficult for vehicles that rely on high-precision maps for navigation to cope with rapidly changing road environments. Summary of the invention

[0004] In order to overcome the problems existing in the related art, this specification provides a method, system, electronic device and storage medium for updating road information.

[0005] According to a first aspect of an embodiment of this specification, a method for updating road information is provided, the method comprising:

[0006] Determine and report abnormal driving events that occur when the vehicle is driving based on the target map;

[0007] Determine a labeling result of the abnormal driving event, wherein the labeling result at least includes abnormal position information of an abnormal position where the abnormal driving event occurs and object description information of an abnormal road object associated with the abnormal driving event;

[0008] The annotation result is sent to the vehicle to be updated, so that the vehicle to be updated generates a driving strategy at the abnormal position according to the object description information and the target map.

[0009] According to a second aspect of an embodiment of this specification, another method for updating road information is provided, the method comprising:

[0010] Receive the annotation result of the abnormal driving event sent by the server; the annotation result at least includes abnormal position information of the abnormal position where the abnormal driving event occurs and object description information of the abnormal road object associated with the abnormal driving event; the abnormal driving event is a driving event that occurs during the driving process of the reporting vehicle based on the target map;

[0011] A driving strategy at the abnormal location is generated according to the object description information and the target map.

[0012] According to a third aspect of an embodiment of this specification, a road information updating system is provided, the system comprising a service end, a reporting vehicle and a vehicle to be updated, wherein:

[0013] The server determines an abnormal driving event that occurs during the driving process of the reported vehicle based on the target map; determines a labeling result of the abnormal driving event and sends the labeling result to the vehicle to be updated, wherein the labeling result at least includes abnormal position information of the abnormal position where the abnormal driving event occurs and object description information of an abnormal road object associated with the abnormal driving event;

[0014] The to-be-updated vehicle generates a driving strategy at the abnormal position according to the annotation result and the target map.

[0015] According to the fourth aspect of the embodiments of this specification, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method described in the first aspect and / or the second aspect are implemented.

[0016] According to a fifth aspect of the embodiments of this specification, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of the method described in the first aspect and / or the second aspect are implemented.

[0017] The technical solutions provided by the embodiments of this specification may have the following beneficial effects:

[0018] In the embodiment of this specification, the server determines the abnormal driving event that occurred in the reporting vehicle, locates the abnormal road object that caused the abnormal driving event, and marks the abnormal road object, and then sends the marking result to the vehicle to be updated, so that the vehicle to be updated generates a driving strategy at the abnormal position according to the marking result. It can be seen that this solution directly sends the marking result of the latest road condition containing the abnormal road object to the vehicle to be updated, so that the vehicle to be updated can also determine the driving strategy at the abnormal position in combination with the unupdated high-precision map data and the marking result without receiving the high-precision map data with the latest road condition updated with the abnormal road object, thereby solving the problem of driving strategy decision errors caused by the lag in data update in the high-precision map.

[0019] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the specification and, together with the description, serve to explain the principles of the specification.

[0021] Figure 1 It is an architecture diagram of a road information updating system shown in this specification according to an exemplary embodiment.

[0022] Figure 2 The flowchart of a method for updating road information is shown in this specification according to an exemplary embodiment.

[0023] Figure 3 It is a software architecture diagram of a labeling platform shown in this specification according to an exemplary embodiment.

[0024] Figure 4 It is a schematic diagram of a front-end function page of a tag management module 31 according to an exemplary embodiment of this specification.

[0025] Figure 5 It is a schematic diagram of a front-end function page of a branch management module 32 according to an exemplary embodiment of this specification.

[0026] Figure 6 It is a schematic diagram of a front-end function page of a labeling process according to an exemplary embodiment of this specification.

[0027] Figure 7 It is a schematic diagram of a front-end function page of a label management module 33 according to an exemplary embodiment of this specification.

[0028] Figure 8It is a schematic diagram of a front-end function page for querying annotation results according to an exemplary embodiment of this specification.

[0029] Fig. 9 It is a flowchart of another method for updating road information according to an exemplary embodiment of the present specification.

[0030] Fig.10 It is a schematic diagram of updating a high-precision map according to an exemplary embodiment of this specification.

[0031] Fig.11 It is a structural schematic diagram of an electronic device according to an exemplary embodiment of the present specification.

[0032] Fig.12 It is a block diagram of a device for updating road information according to an exemplary embodiment of the present specification.

[0033] Fig.13 It is a block diagram of another device for updating road information according to an exemplary embodiment of the present specification. DETAILED DESCRIPTION

[0034] Autonomous driving is a technology that enables vehicles to drive automatically without human intervention. The development goal of autonomous driving is to achieve a transition from basic driving assistance to fully autonomous driving, so that vehicles can perceive, make decisions and control autonomously, so as to improve traffic safety, traffic efficiency and free up drivers' energy. High-precision maps play a vital role in autonomous driving. High-definition maps (HD Maps) are map data that is more detailed and accurate than traditional navigation maps, and are used to support environmental perception, positioning and path planning for autonomous driving. For example, HD maps contain lane-level road structures, which provide detailed road layout information for autonomous driving and help the system make more accurate path planning. Especially in complex urban streets, intersections, ramps and other scenes, the detailed data of the map enables vehicles to better cope with changing road environments. For example, HD maps can provide information about road conditions ahead in advance, such as upcoming curves, slopes, speed limits, etc., to help the autonomous driving system adjust speed and steering in advance to ensure smooth and safe driving of the vehicle. However, maintaining the timeliness and accuracy of HD maps is a huge challenge. Road infrastructure often changes, such as construction, new roads, and replacement of traffic signs, which will affect the accuracy of HD maps. If the map is not updated in a timely manner, it may cause perception deviations in the actual driving process of the autonomous driving vehicle, thus affecting driving safety.

[0035] At present, the update of high-precision maps mainly relies on professional surveying and mapping vehicles, which are equipped with high-precision lidar, cameras and GPS devices, and generate and update high-precision maps by collecting a large amount of road data. Although this method has high accuracy, it is costly and has a long data collection and processing cycle, resulting in untimely updates of road information in high-precision maps, and vehicles that rely on high-precision maps for navigation find it difficult to cope with rapidly changing road environments.

[0036] In response to the above problems, this specification provides a method for updating road information, by directly sending the labeling results of abnormal road objects to the vehicle to be updated, so that the vehicle to be updated generates a driving strategy for the abnormal location of the abnormal road object based on the labeling results, thereby reducing the dependence on the update data of traditional high-precision maps, thereby improving the timeliness of updating road information.

[0037] like Figure 1 As shown, Figure 1 It is an architecture diagram of a road information update system shown in this specification according to an exemplary embodiment. The road information update system 1 includes a server 10 and a vehicle 13, and the vehicle 13 includes a to-be-updated vehicle 13B and a reporting vehicle 13A. Among them, the server 10 is used to determine the abnormal driving event 11 that occurred in the reporting vehicle 13A; determine the abnormal road object that caused the abnormal driving event 11 to occur, and mark the abnormal road object, and the corresponding marking result includes the abnormal position information of the abnormal position where the abnormal driving event is located and the object description information for updating the abnormal road object; and send the marking result to the to-be-updated vehicle 13B. The to-be-updated vehicle 13B is used to generate a driving strategy at the abnormal position according to the marking result.

[0038] It should be noted that the vehicle 13 may include ordinary vehicles used by the public. Since there are a large number of ordinary vehicles, which are spread across various regions and roads, changes in roads can be discovered more promptly. Therefore, the server 10 can quickly obtain the latest road change information based on the driving data of ordinary vehicles. This instant feedback mechanism can greatly improve the timeliness of obtaining the latest road information. Any vehicle in the vehicle 13 can be a reporting vehicle 13A and a vehicle to be updated 13B at the same time, and all vehicles in the vehicle 13 can be a reporting vehicle 13A, that is, they can report driving data to the server 10. In addition, all vehicles in the vehicle 13 can be vehicles to be updated 13B, receive the annotation results 12 issued by the server 10, and update the driving strategy at the abnormal position according to the annotation results 12. Since any vehicle in the vehicle 13 can be a reporting vehicle 13A, different vehicles travel at different times and places, and can capture various abnormal road conditions, which greatly broadens the source of driving data collection. Once an abnormal road condition is discovered and marked by the server 10, other vehicles can receive update information in the shortest time and adjust their driving strategies.

[0039] like Figure 2 As shown, Figure 2 This is a flowchart of a method for updating road information according to an exemplary embodiment of the present specification. The method can be applied to Figure 1 The server 10 includes steps 201-203:

[0040] Step 201: Determine and report abnormal driving events that occur when the vehicle is driving based on a target map.

[0041] In the example autonomous driving scenario, the autonomous driving system of the reporting vehicle can plan the standard driving strategy of the route in detail based on the location of the destination and the road information in the target map. For example, if the target map shows that the road ahead is a turning intersection, the reporting vehicle can automatically adjust the speed according to the curvature and speed limit requirements of the curve recorded in the target map to ensure that the reporting vehicle can pass safely.

[0042] However, the reporting vehicle does not always drive according to the standard driving strategy on the planned driving route, and various abnormal driving events that do not conform to the standard driving strategy may occur. Abnormal driving events include detection driving events that do not match the standard driving strategy set by the reporting vehicle for the target road. The detection driving event is a driving event obtained by event recognition of the driving log of the reporting vehicle on the target road. For example, due to frequent changes in road conditions, the traffic signs at a certain intersection were replaced some time ago, but the old sign information is still displayed in the target map, which may cause the reporting vehicle to make wrong driving decisions at the intersection. At this time, the reporting vehicle can also issue voice prompts and visual warnings after detecting that the target map is invalid to remind the driver to manually control the vehicle in time. In addition, if there are temporary construction areas that are not marked by the target map, road collapses, or part of the road surface is occupied, the reporting vehicle may deviate from the normal driving route. For example, on urban roads, due to emergency construction on a certain section of the road, temporary roadblocks are set up. After detecting these changes, the reporting vehicle may deviate from the originally planned route in order to avoid the construction area. The above situations can be regarded as abnormal driving events that occurred in the reporting vehicle.

[0043] In an illustrated embodiment, the server can obtain the driving log uploaded by the reporting vehicle on the target road, and perform event recognition on the driving log. If the detected driving event does not match the standard driving strategy set by the reporting vehicle for the target road, the detected driving event is determined to be an abnormal driving event. For example, assuming that the target map shows that the speed limit of a certain road is 60km / h, the standard driving strategy can drive according to the speed limit requirement. However, the driving log shows that the speed of the reporting vehicle suddenly drops to 30km / h when driving on the road, and it lasts for a period of time, then it can be identified from the driving log that the reporting vehicle has an abnormal driving event on the road. For another example, assuming that the standard driving strategy requires stable driving in the middle lane of a certain road according to the instructions of the target map, but the driving log shows that the reporting vehicle frequently switches between different lanes, and there is no obvious overtaking or avoiding other vehicles, which can also be identified as an abnormal driving event. In order to avoid misjudgment caused by inaccurate or incomplete driving log data of a single vehicle, the server can obtain driving logs uploaded by multiple reporting vehicles on the same road, and identify common abnormal driving events from each driving log. The driving log may include basic driving data of the vehicle, road images taken by the on-board camera, etc.

[0044] In another illustrated embodiment, the reporting vehicle can determine an abnormal driving event based on the driving log of a certain road, and make a report of the abnormal driving event, and then upload the report to the server. For example, when the reporting vehicle is driving on a certain road, it is monitored that the vehicle's driving behavior does not conform to the standard driving strategy generated for the road. The reporting vehicle can make a report of the abnormal driving event based on the driving log on the road. The report can include vehicle identification, time of occurrence, location of occurrence, description information of the abnormal driving event, image data captured by the on-board camera, etc. The reporting vehicle uploads the report of the abnormal driving event to the server to reduce the pressure on the server to handle abnormal driving events.

[0045] Step 202: Determine a labeling result of the abnormal driving event, wherein the labeling result at least includes abnormal position information of an abnormal position where the abnormal driving event occurs and object description information of an abnormal road object associated with the abnormal driving event.

[0046] The abnormal road object may be various road objects on the road, for example, the abnormal road object may be road markings on the abnormal road, such as lane markings, crosswalk markings and other road markings; may be traffic signs on the abnormal road, such as speed limit signs, intersection signs, etc.; may be obstacles on the road surface, such as electric poles, billboards, etc. Of course, the above is only an exemplary description, and the abnormal road object may also include other objects related to the road information of the abnormal road.

[0047] After determining the abnormal driving event in step 201, the abnormal road where the abnormal driving event occurs can be determined first, and then the abnormal road object that causes the abnormal driving event on the abnormal road can be determined. For example, assuming that the abnormal driving event that occurs on a certain abnormal road is that the reported vehicle does not travel at the prescribed speed, the cause of the abnormal driving event may be that the speed limit on the abnormal road has changed, or the road section is in an area where pedestrians can easily pass through, or the road surface of the road is damaged, etc. The abnormal road object is a reason related to the road condition, that is, the cause of the abnormal driving event is caused by a road object on the abnormal road. For example, if the cause of the abnormal driving event is that the speed limit on the abnormal road has changed, the abnormal road object can be a speed limit sign. For another example, if the cause of the abnormal driving event is that the road section is in an area where pedestrians can easily pass through or the road surface of the road is damaged, etc., the abnormal road object can be the road surface of the road.

[0048] In an illustrated embodiment, the abnormal location where the abnormal driving event occurs can be determined first, and the current object information of each road object at the abnormal location can be determined. The current object information of each road object is compared with the historical object information of the corresponding road object in the target map, and the road object whose current object information is different from the historical object information is determined as an abnormal road object.

[0049] Exemplarily, the server obtains the detection results of each road object at the abnormal position by the reporting vehicle, such as the current object information such as the speed limit value of the current speed limit sign, the number and distribution of the current lane lines, and the current location of the electric pole. Then, the server calls the target map data and obtains the historical object information of the road object at the corresponding abnormal position from the target map, which includes the speed limit value of the historical speed limit sign, the number and distribution of the historical lane lines, the location of the historical electric poles and other historical object information. The current object information is compared with the historical object information item by item. For the speed limit sign, compare whether the speed limit value is the same; for the lane line, compare whether the number and distribution are consistent; for fixed objects such as electric poles, compare whether the position has changed. For example, if it is found that the current speed limit sign shows a speed limit of 60km / h, and the speed limit at this location in the target map is 80km / h, the speed limit sign is determined to be an abnormal road object. In addition to the numerical difference, some indirect changes can also be considered, such as the color change of the lane line, the damage of the traffic sign, etc., which may also be regarded as abnormal.

[0050] After the abnormal road object is determined, its specific location information at the abnormal location can be marked. For example, for a speed limit sign, its latitude and longitude coordinates and altitude information can be marked; for a lane line, the road sign of the road where it is located can be marked. According to the specific situation of the abnormal road object, object description information for updating the abnormal road object is generated. For a speed limit sign, the marked object description information can be the new speed limit value (such as a speed limit of 60km / h); for a lane line, the marked object description information can be the latest distribution of the lane line (such as the number, shape and direction of the lane line).

[0051] It is not difficult to understand that the cause of abnormal driving events is not necessarily that the road objects in the target map have changed. Even if the individual road objects in the road have not changed, abnormal driving events may still occur. For example, some roads are in peak traffic during off-get off work hours, resulting in the reported vehicles being unable to pass at the speed set by the standard driving strategy. This reason is difficult to obtain by comparing the current object information of the road object with the historical object information in the target map. For some complex scenes, such as less obvious potholes on the road, it is also difficult for image recognition models to detect the problem. For another example, for a partially obscured traffic sign, the machine may not be able to accurately identify its content, but human annotators can infer the type of sign and possible abnormalities by observing factors such as the shape, color, and surrounding environment of the sign.

[0052] Therefore, this manual provides a annotation platform, such as Figure 3 As shown, the annotation platform 3 can process Figure 1 The server 10 collects abnormal driving events reported by the vehicle, and the abnormal road objects that cause the abnormal driving events are manually determined by the annotation personnel, and the abnormal road objects are annotated to improve the accuracy of abnormal road event processing. The annotation platform 3 includes a label management module 31, a branch management module 32 and an annotation management module 33. Among them, the label management module 31 is used to predefine labels representing object description information of abnormal road objects. The label management module 31 includes a label registration submodule 311, a label query submodule 312 and a label status management submodule 313. The branch management module 32 is used to manage the annotation results under different branch versions. The branch management module 32 includes a branch registration submodule 321, a branch query submodule 322 and a branch status management submodule 323. The annotation management module 33 is used to annotate abnormal road objects using the labels defined by the label management module 31. The annotation management module 33 includes a annotation submodule 331, an annotation result query submodule 332 and an annotation result sending submodule 333.

[0053] Specifically, Figure 4 As shown, Figure 4This is a schematic diagram of a front-end function page of a tag management module 31 shown in this specification according to an exemplary embodiment. The labeling personnel can register tags through the control of tag registration 407. The tag status of each tag is managed by the control of operation 408, and the tag status includes two states: disabled and enabled. A tag registration list is displayed on the tag management page, and the header of each registered tag data includes a tag identifier 401, a tag name 402, a tag description 403, a number of annotation records 404, a creation time 405, a creator 406, and a tag status managed by the operation control 408. For example, the specific content of a registered tag data is: the tag identifier is 2, the tag name is "speed reduction section (20km / h)", the tag description is "it is recommended to reduce the speed to 20km / h", the number of annotation records is 4, indicating that the tag is used 4 times, the creation time is XX year XX month XX day, indicating the creation time of the tag, and the creator is "XX". In this embodiment, by pre-defining labels, when labeling abnormal road objects, labelers do not need to manually enter label content every time, but can directly select appropriate labels from the pre-registered label list to save labeling time. In addition, the label description information also helps to provide accurate labeling guidance for labelers, so that labelers can quickly understand the meaning and scope of application of the label according to the label description, so as to label more accurately.

[0054] like Figure 5 As shown, Figure 5 It is a schematic diagram of a front-end function page of a branch management module 32 shown in this specification according to an exemplary embodiment. The annotation personnel can set branches for different target map versions and different areas (such as counties, cities, districts, etc.) under different map versions to manage different annotation results under different branch versions. Among them, the two controls of importing intelligence 507 and adding branches 508 are used to register branches. The header of each branch data after registration includes a branch identifier 501, a branch name 502, an area identifier 503, a map version 504, an intelligence option 505, a number of annotation records 405, a creation time 506, a creator 509, and a visualization option under the operation control 510. Among them, the area identifier 503 can be the area identifier of different sub-areas in the target map. The map version 504 can be the map version of the target map. By clicking the visualization option, you can see the distribution of all abnormal road objects under the branch in the target map.

[0055] like Figure 6 As shown, Figure 6This is a schematic diagram of the front-end page of a labeling process shown in this specification according to an exemplary embodiment. For any abnormal driving event that occurs to a reported vehicle, the labeler can determine the abnormal road object that caused the abnormal driving event, locate the abnormal position of the abnormal road object 6011 in the target map 601, and label the abnormal road object in the labeling result editing page 602. For example, the report of the abnormal driving event records the abnormal problem that the reported vehicle deviates from the dotted line when merging into the main road lane, and the report also includes the time and location of the reported vehicle and the video file taken by the on-board camera. The labeler can then determine the abnormal road object 6011 that caused the abnormal driving event based on the abnormal problem and the video file taken by the on-board camera. For example, if the abnormal road object 6011 is a change in the dotted line direction of the road, the labeler can do as follows Figure 6 As shown, the abnormal area where the abnormal road object 6011 is located in the target map 601, and then the abnormal location information 6021 and object description information 6022 of the abnormal area are recorded in the annotation result editing page 602, and submitted after the annotation is completed. Among them, the annotation platform can locate the abnormal area where the abnormal road object is located in the target map, and generate the regional coordinate information of the abnormal area, and use the regional coordinate information to represent the abnormal location information 6021 where the abnormal road object is located. Of course, the annotation platform can also locate the abnormal road where the abnormal road object is located in the target map, and generate the abnormal road mark of the abnormal road, and use the abnormal road mark to represent the abnormal location information 6021 where the abnormal road object is located.

[0056] like Figure 7 As shown, Figure 7 1 is a schematic diagram of a front-end function page of a label management module 33 according to an exemplary embodiment of the present specification. Figure 6 After marking the abnormal road objects in the manner shown, a Figure 7 A labeling result in the front-end page of the annotation result list. The header of the labeling result list may include a labeling result identifier 701, a performance tag 702, a reason tag 703, a branch identifier 501, a branch name 502, a note 704, an associated road and lane 705, and a creation time 706. Among them, the object description information used to update the abnormal road object in the annotation result may be a performance tag 702 and a reason tag 703. For example, the performance tag 702 may be "wrong data" and the reason tag may be "easy to degrade". The abnormal position information of the abnormal position of the abnormal road object in the annotation result can be represented by the associated road and lane 705. The labeling result sent to the vehicle to be updated may include a performance tag 702, a reason tag 703, and an associated road and lane 705.

[0057] like Figure 8 As shown, Figure 8 It is a front-end function page schematic diagram of a query annotation result shown in this specification according to an exemplary embodiment. The annotation personnel can query the annotation of all abnormal road objects within the area according to the specified coordinate area range, and can be limited to a specific area and / or a specific map version for query. For example, the annotation personnel can enter the coordinate information in the coordinate input box 801, and enter the radius value in the radius input box 802, so that the coordinate area range can be determined, and can choose to enter the map sub-area identification in the input box of the area identification 503 and / or enter the map version information in the input box of the map version 504, so as to limit the query within the specific sub-area and / or specific map version range. Of course, the annotation personnel can also query the annotation of all abnormal road objects within the road range within the specified road range. For example, the annotation personnel can enter the road identification in the input box of the road identification 803, and can also choose to enter the map sub-area identification in the input box of the area identification 503 and / or enter the map version information in the map version input box, so as to limit the query within the specific map sub-area and / or specific map version range. By clicking the control for obtaining the annotation results 804, the annotation results of all abnormal road objects within different coordinate regions or road ranges are obtained, so that these annotation results can be sent to the vehicles to be updated. In this embodiment, the annotation platform can also perform in-depth mining and analysis on the annotation results of all abnormal road objects within different coordinate regions or road ranges, so as to discover the laws and trends of road environment changes, so as to further optimize and update the high-precision map data.

[0058] Step 203: Send the annotation result to the vehicle to be updated, so that the vehicle to be updated generates a driving strategy at the abnormal position according to the object description information and the target map.

[0059] The annotation result may include the abnormal position information of the abnormal position of the abnormal driving event and the object description information of the abnormal road object associated with the abnormal driving event. The object description information may be a description of the latest state of the abnormal road object. For example, for a speed limit sign, the annotated object description information may be a new speed limit value (such as a speed limit of 60km / h); for a lane line, the annotated object description information may be the latest distribution of the lane line (such as the number, shape and direction of the lane line). It may be represented by a label in a unified format so that the vehicle to be updated can be accurately identified. When the vehicle to be updated generates a driving strategy at an abnormal position according to the target map, the driving strategy may be generated according to the latest object description information of the abnormal road object at the corresponding abnormal position in the target map. For example, object description information may be added to the abnormal road object at the corresponding abnormal position in the target map. When the vehicle to be updated reads the target map data, the state information of the abnormal road object at the corresponding abnormal position in the target map may be adjusted according to the object description information.

[0060] In addition, after determining the labeling results of the abnormal driving events, the object description information of the abnormal road objects in the target map can also be updated based on the labeling results, and the updated target map can be sent to the vehicle to be updated, so that the vehicle to be updated can generate a driving strategy at the abnormal location according to the labeling results and the updated target map, or the vehicle to be updated can generate a driving strategy at the abnormal location according to the updated target map.

[0061] like Fig. 9 As shown, Fig. 9 is a flowchart of another method for updating road information according to an exemplary embodiment of the present specification. The method can be applied to Figure 1 The vehicle to be updated 13B in the embodiment includes steps 901-902:

[0062] Step 901: Receive the annotation results of abnormal driving events sent by the server; the annotation results at least include abnormal position information of the abnormal position where the abnormal driving event occurs and object description information of the abnormal road object associated with the abnormal driving event; the abnormal driving event is a driving event that occurs during the driving process of the reporting vehicle based on the target map.

[0063] Step 902: Generate a driving strategy at the abnormal location according to the object description information and the target map.

[0064] In the existing technology, the update of high-precision maps mainly relies on professional mapping vehicles equipped with high-precision lidar, cameras and GPS devices, which generate and update high-precision maps by collecting a large amount of road data. Although this method has high accuracy, it is costly, has a long data collection and processing cycle, and is difficult to cope with the rapidly changing road environment.

[0065] Therefore, since the update speed of high-precision maps cannot keep up with the changing speed of the road environment, if the vehicle to be updated still uses a high-precision map that is not updated in time as the target map, it may cause the autonomous driving system to make incorrect driving strategies.

[0066] In an illustrated embodiment, this solution can combine the current version of the high-precision map as the target map and the object description information of the abnormal road object to generate a driving strategy at the abnormal location. For example, the current version of the high-precision map has not updated the road conditions of a certain section of road, and the object description information of the abnormal road object represents the road conditions of the section of road, such as the object description information of the abnormal road object can be "road damage". The driving strategy of the vehicle can be to slow down in advance when the vehicle to be updated drives near the abnormal location in the target map. This solution allows the vehicle to be updated to determine the driving strategy at the abnormal location in combination with the unupdated target map data and the annotation results without receiving the target map data with the latest road conditions updated for the abnormal road objects, thereby solving the problem of driving strategy decision errors caused by the lag in data updates in the target map.

[0067] In an illustrated embodiment, the vehicle to be updated can determine a navigation route set through a target map, and when the navigation route passes through an abnormal road object, a driving strategy at an abnormal position corresponding to the navigation route is generated based on object description information of the abnormal road object.

[0068] In one illustrated embodiment, Fig.10 As shown, the server 10 can update the object description information of the abnormal road object in the current version of the target map 14 based on the annotation result 12, and send the updated map update data 15 to the vehicle to be updated 13B, so as to update the target map 14 currently used by the vehicle to be updated 13B. The vehicle to be updated 13B can generate a driving strategy at the abnormal position according to the latest version of the target map. In this embodiment, the annotation results in this solution can also be used to update the map data in the target map. Compared with collecting road information through professional surveying and mapping vehicles, this solution can update the map data in the target map at a lower cost.

[0069] Corresponding to the embodiments of the aforementioned method, this specification also provides embodiments of a device and a terminal to which it is applied.

[0070] like Fig.11 As shown, Fig.11It is a structural diagram of an electronic device shown in this specification according to an exemplary embodiment. At the hardware level, the electronic device 1100 includes a processor 1102, an internal bus 1104, a network interface 1106, a memory 1108 and a non-volatile memory 1110, and of course may also include hardware required for other services. One or more embodiments of this specification can be implemented based on software, such as the processor 1102 reading the corresponding computer program from the non-volatile memory 1110 to the memory 1108 and then running it. Of course, in addition to the software implementation, one or more embodiments of this specification do not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is, the execution subject of the following processing flow is not limited to each logic module, but can also be hardware or logic devices.

[0071] like Fig.12 As shown, Fig.12 is a block diagram of a road information updating device according to an exemplary embodiment of the present specification. The device can be applied to Fig.11 The electronic device 1100 shown in the figure is used to implement the technical solution of this specification. The device includes:

[0072] The abnormal driving event determination module 1202 is used to determine the abnormal driving event that occurs when the reporting vehicle is driving based on the target map;

[0073] A labeling module 1204 is used to determine a labeling result of the abnormal driving event, wherein the labeling result at least includes abnormal position information of an abnormal position where the abnormal driving event occurs and object description information of an abnormal road object associated with the abnormal driving event;

[0074] The sending module 1206 is used to send the annotation result to the vehicle to be updated, so that the vehicle to be updated generates a driving strategy at the abnormal position according to the object description information and the target map.

[0075] Optionally, the abnormal driving event determination module 1202, the abnormal driving event includes a detected driving event that does not match the standard driving strategy set for the target road by the reporting vehicle, and the detected driving event is a driving event obtained by performing event identification on the driving log of the reporting vehicle on the target road; it is specifically used to perform event identification on the driving log of the reporting vehicle on the target road, if the identified detected driving event does not match the standard driving strategy set for the target road by the reporting vehicle, then the identified detected driving event is determined as the abnormal driving event; and / or, receiving the abnormal driving event uploaded by the reporting vehicle.

[0076] Optionally, the labeling module 1204 is specifically used to determine the abnormal location where the abnormal driving event occurs, and determine the current object information of each road object at the abnormal location; compare the current object information of each road object with the historical object information in the target map, and determine the road objects whose current object information is different from the historical object information as abnormal road objects, and determine the object description information of each abnormal road object.

[0077] Optionally, the labeling module 1204 is specifically used to perform the following operations for each abnormal road object: determine the regional coordinate information of the area where the abnormal road object is located in the target map, and determine the regional coordinate information as the abnormal position information; and / or determine the road sign of the road where the abnormal road object is located in the target map, and determine the road sign as the abnormal position information.

[0078] Optionally, a map update module 1208 is also included, which is used to update the object description information of abnormal road objects in the current version of the high-precision map based on the annotation results, and send the updated latest version of the high-precision map to the vehicle to be updated.

[0079] like Fig.13 As shown, Fig.13 is a block diagram of another road information updating device according to an exemplary embodiment of the present specification. The device can be applied to Fig.11 The electronic device 1100 shown in the figure is used to implement the technical solution of this specification. The device includes:

[0080] The annotation result receiving module 1302 is used to receive the annotation result of the abnormal driving event sent by the server; the annotation result at least includes the abnormal position information of the abnormal position where the abnormal driving event occurs and the object description information of the abnormal road object associated with the abnormal driving event; the abnormal driving event is a driving event that occurs during the driving process of the reporting vehicle based on the target map;

[0081] The driving strategy generating module 1304 is used to generate a driving strategy at the abnormal location according to the object description information and the target map.

[0082] Optionally, the driving strategy generation module 1304 is specifically used to generate a driving strategy at an abnormal location by combining a current version of a high-precision map and object description information of abnormal road objects.

[0083] Optionally, the driving strategy generation module 1304 is specifically used to determine the navigation route set through the target map; when the current navigation route passes through the abnormal road object, generate the driving strategy at the abnormal position corresponding to the navigation route according to the object description information of the abnormal road object.

[0084] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, which will not be repeated here.

[0085] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this specification. Ordinary technicians in this field can understand and implement it without paying creative work.

[0086] The present specification also provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of any of the aforementioned road information updating methods provided in the present application are implemented.

[0087] Specifically, computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including, for example, semiconductor memory devices (such as EPROM, EEPROM and flash memory devices), magnetic disks (such as internal hard disks or removable disks), magneto-optical disks, and CD ROM and DVD-ROM disks.

Claims

1. A method for updating road information, characterized in that: The method comprises: Determine and report abnormal driving events that occur when the vehicle is driving based on the target map; Determine a labeling result of the abnormal driving event, wherein the labeling result at least includes abnormal position information of an abnormal position where the abnormal driving event occurs and object description information of an abnormal road object associated with the abnormal driving event; The annotation result is sent to the vehicle to be updated, so that the vehicle to be updated generates a driving strategy at the abnormal position according to the object description information and the target map.

2. The method according to claim 1, characterized in that The abnormal driving event includes a detected driving event that does not match the standard driving strategy set by the reporting vehicle for the target road, and the detected driving event is a driving event obtained by performing event recognition on the driving log of the reporting vehicle on the target road; and / or The determining of the abnormal driving event that occurs during the driving process of the reporting vehicle based on the target map includes: performing event recognition on the driving log of the reporting vehicle on the target road, and if the recognized detected driving event does not match the standard driving strategy set by the reporting vehicle for the target road, then determining the recognized detected driving event as the abnormal driving event; And / or, receiving the abnormal driving event uploaded by the reporting vehicle.

3. The method according to claim 1, characterized in that: The determining of the labeling result of the abnormal driving event includes: Determining an abnormal location where the abnormal driving event occurs, and determining current object information of each road object at the abnormal location; The current object information of each road object is compared with the historical object information in the target map, and the road objects whose current object information is different from the historical object information are determined as abnormal road objects, and the object description information of each abnormal road object is determined.

4. The method according to claim 1, characterized in that: The determining of the labeling result of the abnormal driving event includes: The following operations are performed for each abnormal road object: Determine the regional coordinate information of the region where the abnormal road object is located in the target map, and determine the regional coordinate information as the abnormal position information; and / or, A road sign of a road where the abnormal road object is located in the target map is determined, and the road sign is determined as the abnormal position information.

5. The method according to claim 1, characterized in that After determining the labeling result of the abnormal driving event, the method further includes: The object description information of the abnormal road object in the target map is updated based on the annotation result, and the updated target map is sent to the vehicle to be updated, so that the vehicle to be updated generates a driving strategy at the abnormal position according to the annotation result and the updated target map, or the vehicle to be updated generates a driving strategy at the abnormal position according to the updated target map.

6. A method for updating road information, characterized in that: The method comprises: Receive the annotation result of the abnormal driving event sent by the server; the annotation result at least includes abnormal position information of the abnormal position where the abnormal driving event occurs and object description information of the abnormal road object associated with the abnormal driving event; the abnormal driving event is a driving event that occurs during the driving process of the reporting vehicle based on the target map; A driving strategy at the abnormal location is generated according to the object description information and the target map.

7. The method according to claim 6, characterized in that Generating a driving strategy at the abnormal position according to the annotation result includes: Determining a navigation route set by the target map; In a case where the current navigation route passes through the abnormal road object, a driving strategy at the abnormal position corresponding to the navigation route is generated according to the object description information of the abnormal road object.

8. A road information updating system, characterized in that: The system includes a server, a reporting vehicle and a vehicle to be updated, wherein: The server determines an abnormal driving event that occurs during the driving process of the reported vehicle based on the target map; determines a labeling result of the abnormal driving event and sends the labeling result to the vehicle to be updated, wherein the labeling result at least includes abnormal position information of the abnormal position where the abnormal driving event occurs and object description information of an abnormal road object associated with the abnormal driving event; The to-be-updated vehicle generates a driving strategy at the abnormal position according to the annotation result and the target map.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images