High-precision map updating method and device, storage medium and vehicle

Abstract

The high-precision map updating method, apparatus, storage medium, and vehicle provided in this application include: In the current cycle, if the vehicle is in motion, collecting road data of the road the vehicle is currently traveling on, and extracting the actual coordinates of the lane markings from the road data corresponding to the current cycle; obtaining the historical coordinates of the lane markings from the high-precision map based on the vehicle's position in the current cycle; calculating the lane marking deviation value based on the actual lane marking coordinates and the historical lane marking coordinates corresponding to the current cycle, and obtaining the relationship between the lane marking deviation value and a preset deviation threshold; if it is determined that the high-precision map needs to be updated based on the relationship corresponding to the current cycle, uploading the road data corresponding to the current cycle to the server, and proceeding to the next cycle; the road data corresponding to the current cycle is used to update the high-precision map. This ensures the timeliness of high-precision map updates.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, and particularly relates to a high-precision map updating method and device, a storage medium and a vehicle. BACKGROUND

[0002] In recent years, with the rapid development of automatic driving technology, automatic driving vehicles play an increasingly important role in the field of transportation. The smooth operation and safety of automatic driving vehicles cannot be separated from the support of high-precision maps. A high-precision map is a map containing road geometric information, traffic signs and lane-level details, which can provide accurate position information and surrounding environment perception ability for automatic driving vehicles. A high-precision map usually includes an image map and a semantic map. The image map provides physical information of the road, and the semantic map provides annotation information of the road.

[0003] However, there are some significant problems in the high-precision map used by the current automatic driving vehicle. When the road changes, is newly built or is repaired, etc., the road conditions cannot be fed back in the high-precision map in time, resulting in poor timeliness of high-precision map updating. SUMMARY

[0004] The purpose of the present application is to at least solve one of the above technical defects, in particular, the technical defect of poor timeliness of high-precision map updating in the prior art.

[0005] In a first aspect, the present application provides a high-precision map updating method, which comprises:

[0006] In the current round, if the vehicle is in a driving state, the road data of the current driving road of the vehicle is collected, and the actual marking line coordinates of the driving lane marking line are extracted from the road data corresponding to the current round;

[0007] According to the vehicle position of the vehicle in the current round, the historical marking line coordinates of the driving lane marking line are obtained from the high-precision map;

[0008] According to the actual marking line coordinates corresponding to the current round and the historical marking line coordinates corresponding to the current round, the marking line deviation value is calculated, and the size relationship between the marking line deviation value corresponding to the current round and the preset deviation threshold value is obtained.

[0009] If it is determined according to the size relationship corresponding to the current round that the high-precision map needs to be updated, the road data corresponding to the current round is uploaded to the server, and the next round is entered; wherein the road data corresponding to the current round is used to instruct the server to update the high-precision map.

[0010] In one embodiment, the step of uploading road data corresponding to the current round to the server if it is determined that the high-precision map needs to be updated based on the size relationship corresponding to the current round includes:

[0011] If the size relationship corresponding to the current round is that the deviation value of the datum line corresponding to the current round is greater than the preset deviation threshold, then the number of rounds to be selected N is determined, and based on the size relationship corresponding to the previous N rounds, it is determined whether the high-precision map needs to be updated.

[0012] When it is determined that the high-precision map needs to be updated, the road data corresponding to the current cycle is uploaded to the server.

[0013] In one embodiment, the step of determining the number of rounds to be selected includes:

[0014] The previous round of the current round is taken as the round to be evaluated, and the number of rounds selected is initialized to zero;

[0015] Calculate the vehicle distance between the vehicle position in the round to be evaluated and the vehicle position in the current round. If the vehicle distance is less than or equal to a preset distance threshold, increment the number of rounds selected by 1, update the round to be evaluated to the previous round, and compare the vehicle distance corresponding to the updated round to be evaluated with the preset distance threshold until the vehicle distance corresponding to the updated round to be evaluated is greater than the preset distance threshold.

[0016] In one embodiment, the step of determining whether the high-precision map needs to be updated based on the size relationship corresponding to the previous N rounds includes:

[0017] In the first N rounds, if the corresponding deviation value of the datum line in each round is greater than the preset deviation threshold, then it is determined that the high-precision map needs to be updated.

[0018] In one embodiment, the method further includes:

[0019] If the vehicle uploaded road data in the previous round of the current round, the uploaded road length corresponding to the previous round is determined; wherein, when the vehicle did not upload road data in the two rounds prior to the current round, the uploaded road length corresponding to the previous round is the road length corresponding to the road data collected by the vehicle in the previous round; when the vehicle uploaded road data in the two rounds prior to the current round, the uploaded road length corresponding to the previous round is the sum of the road lengths corresponding to the road data collected by the vehicle in multiple consecutive rounds, wherein the multiple consecutive rounds include the round preceding the current round;

[0020] If the length of the road already uploaded in the previous round is less than the preset upload length, then the road data corresponding to the current round is uploaded to the server.

[0021] In one embodiment, the method further includes:

[0022] If the length of the road that has been uploaded in the previous round is greater than or equal to the preset upload length, then when there is at least one round in the multiple consecutive rounds whose marking deviation value is greater than the preset deviation threshold, the road data corresponding to the current round is uploaded to the server.

[0023] In one embodiment, the method further includes:

[0024] In response to the map update command sent by the server, update lane marking data is obtained from the map update command;

[0025] The high-precision map stored in the vehicle is updated based on the updated lane marking data.

[0026] Secondly, this application provides a high-precision map updating device, the device comprising:

[0027] The actual lane marking coordinate determination module is used to collect road data of the current driving road of the vehicle in the current round if the vehicle is in a driving state, and extract the actual lane marking coordinates of the driving lane from the road data corresponding to the current round.

[0028] The historical lane marking coordinate determination module is used to obtain the historical lane marking coordinates from a high-precision map based on the vehicle's position in the current round.

[0029] The marking deviation value determination module is used to calculate the marking deviation value based on the actual marking coordinates corresponding to the current round and the historical marking coordinates corresponding to the current round, and to obtain the relationship between the marking deviation value corresponding to the current round and the preset deviation threshold.

[0030] The high-precision map update module is used to upload the road data corresponding to the current round to the server and proceed to the next round if it is determined that the high-precision map needs to be updated based on the size relationship corresponding to the current round; wherein, the road data corresponding to the current round is used to instruct the server to update the high-precision map.

[0031] Thirdly, this application provides a storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the high-precision map update method described in any of the above embodiments.

[0032] Fourthly, this application provides a vehicle, including: one or more processors, and a memory;

[0033] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the high-precision map update method described in any of the above embodiments.

[0034] As can be seen from the above technical solutions, the embodiments of this application have the following advantages:

[0035] In the high-precision map updating method, apparatus, storage medium, and vehicle provided in this application, if the vehicle is in a driving state in the current cycle, road data of the road on which the vehicle is currently driving can be collected, and the actual coordinates of the driving lane markings can be obtained from the road data. Based on the vehicle's position in the current cycle, the historical coordinates of the driving lane markings can be obtained from the high-precision map. The marking deviation value is calculated based on the actual marking coordinates in the current cycle and the historical marking coordinates, thereby obtaining the relationship between the marking deviation value corresponding to the current cycle and a preset deviation threshold. Based on the relationship corresponding to the current cycle, if it is necessary to update the high-precision map, the road data corresponding to the current cycle can be uploaded to the server to update the high-precision map, and the next cycle can begin. In this way, road data can be collected in real time while the vehicle is in motion, and the system can automatically determine whether the high-precision map needs to be updated based on the deviation between the actual and historical coordinates of the lane markings. When the high-precision map needs to be updated, the system can automatically upload the real-time road data to the server so that the server can update the high-precision map based on the road data. This ensures the timeliness of the high-precision map updates and guarantees that the road information in the high-precision map is up-to-date. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is an application environment diagram of a high-precision map update method provided in one embodiment of this application;

[0038] Figure 2 A flowchart illustrating a high-precision map update method provided in one embodiment of this application;

[0039] Figure 3 This is a schematic diagram of the structure of a high-precision map updating device provided in one embodiment of this application;

[0040] Figure 4 This is a schematic diagram of the internal structure of a computer device provided in one embodiment of this application. Detailed Implementation

[0041] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0042] In some embodiments, the high-precision map update method provided in this application can be applied to Figure 1 In the illustrated application environment, each vehicle 101 can communicate with a remote server 102 to achieve data interaction. Each vehicle 101 can be an autonomous driving vehicle, capable of collecting real-time road data and data from traffic participants, automatically planning its driving trajectory based on the collected data, and performing autonomous driving based on the planned trajectory. Each vehicle 101 may be equipped with computer equipment for data processing and a LiDAR system for collecting road data.

[0043] The remote server 102 can be a non-vehicle-mounted control center, such as a remote control center, which may include one or more servers. The remote server 102 is used to further process the data reported by each vehicle 101 and / or send data to each vehicle 101 to control the vehicle 101 to perform corresponding functions. The remote server 102 can update the high-precision map using the road data uploaded by each vehicle 101. After technicians or drivers confirm that the high-precision map update is correct in the background of the remote server 102, they can send a high-precision map update command to the vehicle 101 through the remote server 102, thereby remotely controlling the vehicle 101 to update the high-precision map.

[0044] In one embodiment, this application provides a high-precision map update method, which can be applied to... Figure 1 The vehicle shown. (As shown) Figure 2 As shown, the high-precision map update method of this application may include the following steps:

[0045] S201: In the current round, if the vehicle is in motion, collect the road data of the road the vehicle is currently traveling on, and extract the actual coordinates of the lane markings from the road data corresponding to the current round.

[0046] Road data refers to information and data related to the road on which a vehicle is traveling. This includes, but is not limited to, the topology of the road the vehicle is currently traveling on, traffic signs, lane markings, road conditions, and environmental information. Lane markings are the lines drawn on the road to indicate the direction of travel, lane boundaries, and traffic rules. Types of lane markings include, but are not limited to, solid lines, dashed lines, double solid lines, domino-style markings, and zebra crossings.

[0047] The vehicle processes the collected road data in a cyclical manner and determines whether to upload the road data to update the high-precision map based on the processing results. Specifically, in each cycle, that is, within the current cycle, the vehicle can determine whether it is in motion. If so, it can send a command to the lidar unit installed on the vehicle, instructing it to collect road data for the route the vehicle is currently traveling. Furthermore, if the vehicle is not in motion, it is not necessary to collect road data for the route the vehicle is currently traveling on. For example, when the vehicle is not running, or has started but is not moving, there is no need to collect road data.

[0048] In one embodiment, when collecting road data, the vehicle can use a lidar mounted on the vehicle to collect point cloud data of the current road, and identify lane markings based on the collected point cloud data to obtain the actual coordinates of the lane markings. In one example, the point cloud data may include reflection intensity information and coordinate information of traffic facilities. The vehicle can identify the specific shape of the lane markings from the reflection intensity information, and thus identify the lane markings in the point cloud data. After identifying the lane markings, the actual coordinates of the lane markings can be determined using the coordinate information of multiple traffic facilities in the point cloud data. Traffic facilities refer to facilities installed to ensure the safe and normal operation of the road traffic system, and include, but are not limited to, road traffic signs, lane markings, and guardrails.

[0049] Furthermore, the current cycle can be the moment each vehicle travels, for example, collecting road data every 0.1 seconds or every 1 millisecond. The time interval between any two adjacent cycles on the time axis can be determined based on actual conditions, such as the scanning frequency of the LiDAR. The actual lane marking coordinates can be three-dimensional or two-dimensional, depending on the specific circumstances. Road data can include road shape data, lane marking data, traffic light data, and road obstacle data.

[0050] S202: Based on the vehicle's position in the current round, obtain the historical coordinates of the driving lane markings from the high-precision map.

[0051] Specifically, high-precision maps can store the coordinate information of each lane marking. By matching the vehicle's position in the current cycle with the coordinate information stored in the high-precision map, the coordinates of the lane marking corresponding to the current cycle in the high-precision map can be determined, and the historical coordinates of the lane marking can be obtained accordingly. The coordinate type corresponding to the coordinate information stored in the high-precision map can be determined according to the coordinate system used by the high-precision map. The coordinate system used by the high-precision map can be a geocentric coordinate system, a geodetic coordinate system (i.e., a latitude, longitude, and altitude coordinate system), or a station-centered coordinate system. Before matching the vehicle position with the coordinate information stored in the high-precision map, the position information type of the vehicle and the coordinate type stored in the high-precision map need to be unified to facilitate matching operations within the same coordinate system.

[0052] In one example, a vehicle can use coordinate information determined from a high-precision map as historical paving coordinates. In another example, for coordinate information extracted from a high-precision map, the vehicle can calibrate the coordinate information based on its position and driving status in the current cycle, and use the calibrated coordinates as historical paving coordinates, thus obtaining more accurate historical paving coordinates. It is understood that historical paving coordinates and actual paving coordinates have the same dimension; that is, both historical and actual paving coordinates can be three-dimensional or two-dimensional.

[0053] In one example, the vehicle's location can be latitude and longitude coordinates, which can be obtained through the GPS (Global Positioning System) installed in the vehicle. Matching the vehicle's location in the current cycle with the road network in a high-precision map determines the road the vehicle is currently traveling on. Based on the road and the vehicle's direction of travel, the lane the vehicle is currently in can be determined. Historical lane marking data for that lane can be retrieved from the high-precision map's database. Parsing this historical data allows the extraction of historical lane marking coordinates, which in turn provides the historical lane marking coordinates corresponding to the vehicle's current cycle.

[0054] S203: Calculate the marking deviation value based on the actual marking coordinates corresponding to the current round and the historical marking coordinates corresponding to the current round, and obtain the relationship between the marking deviation value corresponding to the current round and the preset deviation threshold.

[0055] The lane marking deviation value refers to the degree of deviation between the historical lane marking coordinates recorded on the high-precision map and the actual lane marking coordinates collected. The preset deviation threshold can be determined according to the actual situation, and this application does not make specific requirements for it; for example, it can be 0.2 meters or 0.3 meters.

[0056] In this step, the method for calculating the lane marking deviation value based on the actual lane marking coordinates and the historical lane marking coordinates for the current cycle can be determined according to the actual situation. For example, the Euclidean distance between the actual and historical lane marking coordinates can be used as the lane marking deviation value, or the least squares method can be used to determine the lane marking deviation value between the actual and historical lane markings. The calculated lane marking deviation value is compared with a preset deviation threshold to determine the relationship between the two values. This relationship can be used to assess whether the lane marking deviation value exceeds an acceptable range, i.e., whether it exceeds the preset deviation threshold. If the lane marking deviation value for the current cycle is less than or equal to the preset deviation threshold, it can be determined that the deviation between the historical and actual lane marking coordinates recorded for the current cycle in the high-precision map is small and meets the requirements. If the lane marking deviation value for the current cycle is greater than the preset deviation threshold, it can be determined that the deviation between the historical and actual lane marking coordinates recorded for the current cycle in the high-precision map is large.

[0057] S204: If it is determined that the high-precision map needs to be updated based on the size relationship corresponding to the current round, the road data corresponding to the current round is uploaded to the server, and the next round is started; wherein, the road data corresponding to the current round is used to instruct the server to update the high-precision map.

[0058] For example, the computer device can determine whether a high-precision map needs to be updated based solely on the size relationship corresponding to the current cycle. If the size relationship corresponding to the current cycle indicates a lane marking deviation value greater than a preset deviation threshold—that is, if the historical lane marking coordinates recorded for the current cycle in the high-precision map deviate significantly from the actual lane marking coordinates—it can be determined that the high-precision map needs to be updated. Alternatively, the computer device can determine whether a high-precision map needs to be updated based on both the size relationship corresponding to the current cycle and the size relationships corresponding to historical cycles. For example, it could statistically analyze the size relationships corresponding to the current cycle and historical cycles, and determine whether a high-precision map update is necessary based on the statistical results.

[0059] When it is determined that a high-precision map update is needed, the computer device uploads the road data corresponding to the current cycle to the server. The road data corresponding to the current cycle refers to the road data collected by the vehicle in the current cycle. The computer device can establish a network communication connection with the server using communication protocols such as HTTP (Hypertext Transfer Protocol) before uploading the road data, ensuring the integrity and security of the road data during communication. After receiving the uploaded road data, the server can update the high-precision map database, update high-precision map markers, and optimize route planning algorithms based on the latest road data to ensure that the latest road information is updated in the high-precision map.

[0060] It is understandable that computer devices can preprocess road data before uploading it to the server. Alternatively, the computer devices can upload raw road data to the server, which then preprocesses it.

[0061] In the above embodiments, if the vehicle is in motion during the current cycle, road data of the current road can be collected, and the actual coordinates of the lane markings can be obtained from this road data. Based on the vehicle's position in the current cycle, historical coordinates of the lane markings can be obtained from a high-precision map. The lane marking deviation value is calculated based on the actual and historical coordinates of the lane markings in the current cycle, thus revealing the relationship between the lane marking deviation value for the current cycle and a preset deviation threshold. If the high-precision map needs updating, the road data for the current cycle can be uploaded to the server to update the high-precision map, and the cycle can proceed to the next cycle. In this way, while the vehicle is in motion, road data can be collected in real time, and the high-precision map can be automatically determined based on the lane marking deviation value between the actual and historical coordinates of the lane markings. When an update is needed, the real-time collected road data can be automatically uploaded to the server, allowing the server to update the high-precision map based on the road data. This ensures the timeliness of the high-precision map updates and guarantees that the road information in the high-precision map is up-to-date.

[0062] In one embodiment, the step of uploading road data corresponding to the current round to the server if it is determined that the high-precision map needs to be updated based on the size relationship corresponding to the current round includes:

[0063] If the size relationship corresponding to the current round is that the deviation value of the datum line corresponding to the current round is greater than the preset deviation threshold, then the number of rounds to be selected N is determined, and based on the size relationship corresponding to the previous N rounds, it is determined whether the high-precision map needs to be updated.

[0064] When it is determined that the high-precision map needs to be updated, the road data corresponding to the current cycle is uploaded to the server.

[0065] Specifically, if the deviation value of the datum line corresponding to the current round is greater than a preset deviation threshold, then the size relationships of the N rounds closest to the current round can be used to comprehensively determine whether the high-precision map needs to be updated. Here, N is a positive integer.

[0066] For example, assuming the current round is round X, we can determine whether a high-precision map needs to be updated based on the size relationships of rounds (X-1), (X-2), (X-3), ..., (XN). The number of rounds selected, N, can be determined based on the difference between the line deviation value corresponding to the current round and a preset deviation threshold, or by using a machine learning model to predict the value of the number of rounds selected, N.

[0067] Furthermore, one way to determine whether a high-precision map needs updating is to count the number of rounds in the first N rounds where the map deviation value exceeds a preset deviation threshold. If this number exceeds the preset threshold, then it can be determined that the high-precision map needs updating. The preset threshold can be selected based on the actual situation. Alternatively, the magnitude relationship between the first N rounds can be input into a pre-trained update prediction model to obtain the high-precision map update probability output by the update prediction model, and then the decision on whether to update the high-precision map can be made based on this probability.

[0068] In this embodiment, by determining the number of rounds N and judging whether the high-precision map needs to be updated based on the size relationship of the first N rounds, false alarms caused by a single abnormal data can be reduced, invalid operations caused by temporary deviations can be reduced, and the current road conditions can be better reflected, so that the high-precision map can be updated in a timely manner according to the actual situation.

[0069] In one embodiment, the step of determining the number of rounds to be selected includes:

[0070] The previous round of the current round is taken as the round to be evaluated, and the number of rounds selected is initialized to zero;

[0071] Calculate the vehicle distance between the vehicle position in the round to be evaluated and the vehicle position in the current round. If the vehicle distance is less than or equal to a preset distance threshold, increment the number of rounds selected by 1, update the round to be evaluated to the previous round, and compare the vehicle distance corresponding to the updated round to be evaluated with the preset distance threshold until the vehicle distance corresponding to the updated round to be evaluated is greater than the preset distance threshold.

[0072] Specifically, the vehicle's position in each round can be determined from the real-time vehicle position recorded during the vehicle's movement. When determining the number of rounds to select, the rounds to be evaluated can be initialized to the round preceding the current round, and the number of rounds to select can be initialized to zero.

[0073] Based on the vehicle's position in the previous round and its position in the current round, the vehicle spacing is calculated and compared to a preset distance threshold. If the vehicle spacing is less than or equal to the preset distance threshold, the number of rounds selected is incremented by 1, and the round to be evaluated is updated to the round preceding it, i.e., the two rounds before the current round. Based on the vehicle's position in the updated round to be evaluated and its position in the current round, the vehicle spacing is recalculated and compared to the preset distance threshold. If the vehicle spacing is greater than the preset distance threshold, the round to be evaluated is no longer updated, and the final number of rounds selected is determined. If the vehicle spacing is less than or equal to the preset distance threshold, the round to be evaluated continues to be updated to the round preceding it, i.e., the three rounds before the current round, until the vehicle spacing corresponding to the updated round to be evaluated is greater than the preset distance threshold, thus obtaining the number of rounds selected, N.

[0074] For example, suppose the current round is round Y, and the previous round is round (Y-1). Round (Y-1) is selected as the round to be evaluated, and the number of rounds selected is initialized to zero. The distance between the vehicle's position in round (Y-1) and its position in round Y is calculated. If the distance is less than or equal to a preset distance threshold, the number of rounds selected is incremented by 1 (updated to 1), and the round to be evaluated is updated to round (Y-2). The distance between the vehicle's position in round (Y-2) and its position in round Y is calculated. If the distance is less than or equal to a preset distance threshold, the number of rounds selected is incremented by 1 (updated to 2), and so on, until the distance between the updated rounds to be evaluated is greater than the preset distance threshold, resulting in the number of rounds selected, N.

[0075] It is understandable that the preset distance threshold can be determined according to the actual situation. This application does not make specific requirements on this. For example, it can be 1 meter or 2 meters.

[0076] Furthermore, the method for calculating vehicle spacing can be selected based on the actual situation. For example, it can be the absolute distance method, which calculates the straight-line distance between the vehicle's position in the round to be evaluated and its position in the current round. Alternatively, it can be the relative distance method, which calculates the difference between the vehicle's position in the round to be evaluated and its position in the current round, and then standardizes this difference based on the vehicle's length to obtain the relative vehicle spacing.

[0077] In this embodiment, the number of rounds selected is determined by the vehicle spacing and a preset distance threshold. Only rounds that meet the conditions are selected and further processed, avoiding unnecessary calculations for processing all rounds and thus improving calculation efficiency.

[0078] In one embodiment, the step of determining whether the high-precision map needs to be updated based on the size relationship corresponding to the previous N rounds includes:

[0079] In the first N rounds, if the deviation value of the datum line corresponding to each round is greater than the preset deviation threshold, then it is determined that the high-precision map needs to be updated.

[0080] Specifically, the map deviation value for each of the first N rounds is obtained, and the deviation value for each round is compared with a preset deviation threshold. If the deviation value for each of the first N rounds is greater than the preset deviation threshold, then the high-precision map needs to be updated. Furthermore, if at least one round in the first N rounds has a deviation value less than or equal to the preset deviation threshold, then the high-precision map does not need to be updated.

[0081] In this embodiment, the computer device can determine that the map needs to be updated when the lane marking deviation value in each round is greater than a preset deviation threshold. Thus, by judging through the first N rounds, misjudgments caused by a single round can be avoided, accurately identifying the deviation of lane markings in the high-precision map. Updating the high-precision map based on the deviation ensures the accuracy of the high-precision map information. Restricting the high-precision map update operation to only when certain conditions are met avoids unnecessary high-precision map updates, saving related resources and costs.

[0082] In one embodiment, the high-precision map update method of this application further includes:

[0083] If the vehicle uploaded road data in the previous round of the current round, the uploaded road length corresponding to the previous round is determined; wherein, when the vehicle did not upload road data in the two rounds prior to the current round, the uploaded road length corresponding to the previous round is the road length corresponding to the road data collected by the vehicle in the previous round; when the vehicle uploaded road data in the two rounds prior to the current round, the uploaded road length corresponding to the previous round is the sum of the road lengths corresponding to the road data collected by the vehicle in multiple consecutive rounds, wherein the multiple consecutive rounds include the round preceding the current round;

[0084] If the length of the road already uploaded in the previous round is less than the preset upload length, then the road data corresponding to the current round is uploaded to the server.

[0085] The uploaded road length refers to the road length recorded in the road data uploaded by the vehicle. For example, if the vehicle has continuously collected 100 meters of road data, the road length recorded in the road data is 100 meters. If the road data that has been continuously collected for 100 meters is uploaded, the uploaded road length will be 100 meters.

[0086] Specifically, vehicles typically upload road data over longer distances to ensure data integrity. This allows for updates to road planning based on complete road data during high-precision map updates, further guaranteeing the integrity of road information in the high-precision map. If the uploaded road data is discontinuous, it may result in missing or incomplete road information in the high-precision map.

[0087] To ensure the server receives complete road data and thus guarantees the integrity of road information in high-precision maps, vehicles can determine whether they uploaded road data in the previous round. If so, the vehicle can further determine whether the uploaded road data provides sufficiently complete road information. If the uploaded road data does not provide sufficiently complete road information, then the vehicle needs to continue uploading road data for the current round. Furthermore, if the uploaded road data provides sufficiently complete road information, then uploading road data for the current round is not necessary.

[0088] Specifically, if a vehicle uploaded road data in the previous round, it is determined whether the vehicle uploaded road data in the two rounds prior to the current round. If the vehicle did not upload road data in the two rounds prior to the current round, it indicates that the road data upload was interrupted in those rounds. To ensure the continuity and integrity of the road data, the road data from the two rounds prior to that round is unavailable. Therefore, the uploaded road length corresponding to the previous round can be determined as the road length corresponding to the road data collected by the vehicle in the previous round. If the vehicle uploaded road data in the two rounds prior to the current round, it indicates that the vehicle uploaded road data in at least two consecutive rounds. To ensure the continuity and integrity of the road data, the road data uploaded in the two rounds prior to this round can be used. Therefore, the uploaded road length corresponding to the previous round can be determined as the road length corresponding to the road data collected by the vehicle in multiple consecutive rounds, where multiple consecutive rounds include the previous round.

[0089] If the length of the road already uploaded in the current round is less than the preset upload length, then the road data for the current round is uploaded to the server. The preset upload length can be determined based on actual conditions; this application does not impose specific restrictions on it, for example, it could be 300 meters or 400 meters.

[0090] Furthermore, if a vehicle did not upload road data in the previous round, the need to update the high-precision map is determined based on the size relationship corresponding to the current round. It can be understood that determining the road length corresponding to the previous round is to determine whether the uploaded continuous road data can provide complete road information to the high-precision map. The server updates the high-precision map based on the uploaded road data. If a vehicle did not upload road data in the two rounds prior to the current round, the road data from the previous two rounds cannot be used. To ensure the continuity of road data, when the road data from the previous two rounds cannot be used, the road length corresponding to the road data collected by the vehicle in the previous round can be determined as the uploaded road length from the previous round.

[0091] For example, assuming the current round is round N, if a vehicle uploads road data in round (N-1), then determine whether the vehicle uploads road data in round (N-2). If the vehicle does not upload road data in round (N-2), the uploaded road length is determined based on the road data collected in round (N-1), i.e., the distance of road data collected in round (N-1). If the vehicle uploads road data in round (N-2), the road lengths corresponding to the road data collected in consecutive rounds including round (N-1) are determined. That is, if the vehicle uploads road data in round (N-3) but not in round (N-4), the road lengths corresponding to the road data collected in rounds (N-1), (N-2), and (N-3) are determined, and the sum of the road lengths corresponding to rounds (N-1), (N-2), and (N-3) is the uploaded road length. If this uploaded road length is less than the preset upload length, the road data collected by the vehicle in round N is uploaded to the server.

[0092] In this embodiment, by recording and judging the data upload status between rounds, it can be ensured that the uploaded road data is as long as possible, thereby maximizing the collection and storage of road data collected by vehicles; by considering road data from multiple consecutive rounds, the continuity of road data can be improved, the discontinuity of road data can be reduced, and more complete and accurate road information can be provided, ultimately ensuring the accuracy of high-precision maps.

[0093] In one embodiment, the high-precision map update method of this application further includes:

[0094] If the length of the road that has been uploaded in the previous round is greater than or equal to the preset upload length, then when there is at least one round in the multiple consecutive rounds whose marking deviation value is greater than the preset deviation threshold, the road data corresponding to the current round is uploaded to the server.

[0095] It's understandable that in real-world scenarios, there might be long stretches of lane marking maintenance and renovation, resulting in high-precision maps failing to update in a timely manner and retaining a significant amount of raw data. In such cases, even after a vehicle uploads road data of a length greater than or equal to a preset upload length, the updated high-precision map still cannot fully and comprehensively update the lane markings to ensure the entire lane marking is up-to-date. Therefore, to ensure a complete and comprehensive update of lane markings in the high-precision map, when the uploaded road length from the previous round is greater than or equal to the preset upload length, the vehicle can also determine if at least one round's lane marking deviation value exceeds a preset deviation threshold. If so, it continues to upload road data for the current round to the server, ensuring the high-precision map can update lane markings based on more road data. If not, it can choose not to upload road data for the current round to the server.

[0096] Specifically, if the uploaded road length from the previous round is greater than or equal to a preset upload length, the road marking deviation value for each round in a series of consecutive rounds is determined. If the road marking deviation value for at least one round is greater than a preset deviation threshold, the road data for the current round is uploaded to the server. If the road marking deviation values ​​for all rounds are less than or equal to the preset deviation threshold, it can be determined whether road data needs to be uploaded based on the relative values ​​for the current round.

[0097] For example, suppose the sum of the road lengths corresponding to the road data uploaded by the vehicle in multiple consecutive cycles reaches the preset upload length, but there is a deviation value of the road markings in at least one cycle that is greater than the preset deviation threshold. This indicates that the historical road data corresponding to the current driving road stored in the high-precision map deviates significantly from the actual road data. Therefore, it is necessary to continue uploading the road data collected by the vehicle in the current cycle.

[0098] In this embodiment, when the deviation of the road markings is large, even if enough road data has been uploaded, road data for the current cycle can be uploaded to provide more accurate road information, thereby further ensuring the accuracy of the high-precision map.

[0099] In one embodiment, the high-precision map update method of this application further includes:

[0100] In response to the map update command sent by the server, update lane marking data is obtained from the map update command;

[0101] The high-precision map stored in the vehicle is updated based on the updated lane marking data.

[0102] Specifically, under normal circumstances, the server can communicate with multiple vehicles and update lane marking information for multiple map areas in the high-precision map based on road data uploaded by each vehicle. Vehicles can receive map update commands sent by the server and extract updated lane marking data from these commands. This updated lane marking data refers to the data extracted from the map update command used to update the high-precision map stored in the vehicle. In one example, the updated lane marking data may include information such as the location, type, or length of the lane markings. The extracted updated lane marking data is used to update the high-precision map stored in the vehicle; for example, the lane marking information can be updated based on the location and type of the lane markings in the updated lane marking data.

[0103] Thus, for any vehicle, even if it is not driving in other areas, it can still update the lane marking information in other areas through map update commands sent by the server, thereby further improving the timeliness of high-precision map updates.

[0104] To facilitate understanding of the solution presented in this application, a specific example is provided below. The high-precision map update method provided in this example may include the following steps:

[0105] Assuming the vehicle executes the high-precision map update method of this application in 1ms cycles, in the 10ms, when the vehicle is in motion, the vehicle's lane marking recognition mode is activated. This involves collecting road data of the road the vehicle is traveling on in the 10ms and extracting the actual 3D coordinates of the lane markings from this data. Based on the vehicle's position in the 10ms, the historical 3D coordinates of the lane markings are obtained from the high-precision map, and the deviation between the actual and historical 3D coordinates is calculated. If this deviation is greater than a 0.2m deviation threshold, it is determined whether a continuous 1m deviation greater than the 0.2m threshold has occurred based on the road data before the 10ms. If so, the high-precision map update mode is activated, and at least 300m of road data is continuously uploaded. Starting from the 10ms, the collected road data is uploaded to the server, with at least 300m of road data uploaded. During the continuous uploading of 300m of road data, the system continues to check whether the lane marking deviation value corresponding to each 1ms is greater than the 0.2m deviation threshold. If there is a consecutive 1m lane marking deviation value less than the 0.2m deviation threshold in the uploaded 300m of road data, the uploading of road data can be paused; otherwise, the uploading of road data will continue after the initial 300m of road data is uploaded until there is a consecutive 1m lane marking deviation value less than the 0.2m deviation threshold in the uploaded road data. After receiving the road data uploaded by the vehicle, the server annotates the high-precision map according to the road data. After the annotation is completed, a high-precision map update confirmation message is sent to the cartographic engineer. After the cartographic engineer confirms the high-precision map update, the high-precision map update command is sent to each vehicle through the server, so that the vehicle updates its onboard high-precision map according to the high-precision map update command.

[0106] The high-precision map updating apparatus provided in the embodiments of this application is described below. The high-precision map updating apparatus described below can be referred to in correspondence with the high-precision map updating method described above. Figure 3 As shown, this application provides a high-precision map updating device, which may include the following structure:

[0107] The actual lane marking coordinate determination module 301 is used to collect road data of the current driving road of the vehicle in the current round if the vehicle is in a driving state, and extract the actual lane marking coordinates of the driving lane from the road data corresponding to the current round.

[0108] The historical lane marking coordinate determination module 302 is used to obtain the historical lane marking coordinates of the driving lane from a high-precision map based on the vehicle's position in the current round.

[0109] The marking deviation value determination module 303 is used to calculate the marking deviation value based on the actual marking coordinates corresponding to the current round and the historical marking coordinates corresponding to the current round, and to obtain the relationship between the marking deviation value corresponding to the current round and the preset deviation threshold.

[0110] The high-precision map update module 304 is used to upload the road data corresponding to the current round to the server and proceed to the next round if it is determined that the high-precision map needs to be updated based on the size relationship corresponding to the current round; wherein, the road data corresponding to the current round is used to instruct the server to update the high-precision map.

[0111] In one embodiment, the high-precision map update module 304 includes:

[0112] The round selection quantity determination submodule is used to determine the round selection quantity N if the size relationship corresponding to the current round is that the line deviation value corresponding to the current round is greater than the preset deviation threshold, and to determine whether the high-precision map needs to be updated based on the size relationship corresponding to the previous N rounds.

[0113] The road data upload submodule is used to upload the road data corresponding to the current round to the server when it is determined that the high-precision map needs to be updated.

[0114] In one embodiment, the submodule for determining the number of rounds selected includes:

[0115] The round to be evaluated determination unit is used to take the previous round of the current round as the round to be evaluated and initialize the number of rounds selected to zero.

[0116] The round selection quantity determination unit is used to calculate the vehicle distance between the vehicle position in the round to be evaluated and the vehicle position in the current round. If the vehicle distance is less than or equal to a preset distance threshold, the round selection quantity is incremented by 1, and the round to be evaluated is updated to the previous round. The vehicle distance corresponding to the updated round to be evaluated is compared with the preset distance threshold until the vehicle distance corresponding to the updated round to be evaluated is greater than the preset distance threshold.

[0117] In one embodiment, the submodule for determining the number of rounds selected includes:

[0118] The high-precision map update determination unit is used to determine that the high-precision map needs to be updated if, in the first N rounds, the corresponding deviation value of the datum line in each round is greater than the preset deviation threshold.

[0119] In one embodiment, the apparatus further includes:

[0120] The uploaded road length determination module is used to determine the uploaded road length corresponding to the previous round if the vehicle uploaded road data in the round preceding the current round. Specifically, when the vehicle did not upload road data in the two rounds preceding the current round, the uploaded road length corresponding to the previous round is the road length corresponding to the road data collected by the vehicle in the previous round. When the vehicle uploaded road data in the two rounds preceding the current round, the uploaded road length corresponding to the previous round is the sum of the road lengths corresponding to the road data collected by the vehicle in multiple consecutive rounds, including the round preceding the current round.

[0121] The first road data upload module is used to upload the road data corresponding to the current round to the server if the length of the road uploaded in the previous round is less than the preset upload length.

[0122] In one embodiment, the apparatus further includes:

[0123] The second road data upload module is used to upload the road data corresponding to the current round to the server if the length of the uploaded road in the previous round is greater than or equal to the preset upload length, and if the marking deviation value of at least one round in the multiple consecutive rounds is greater than the preset deviation threshold.

[0124] In one embodiment, the apparatus further includes:

[0125] The lane marking data acquisition module is used to acquire updated lane marking data from the map update command sent by the server in response to the map update command.

[0126] The vehicle high-precision map update module is used to update the high-precision map stored in the vehicle based on the updated lane marking data.

[0127] In one embodiment, this application also provides a storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the high-precision map update method as described in any of the above embodiments.

[0128] In one embodiment, this application also provides a vehicle storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the high-precision map update method as described in any of the above embodiments.

[0129] In one embodiment, the vehicle may include computer equipment for data processing and lidar for acquiring road data. (Illustratively, as shown...) Figure 4 As shown,Figure 4 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. (Refer to...) Figure 4 The computer device 400 includes a processing component 402, which further includes one or more processors, and memory resources represented by memory 401 for storing instructions, such as application programs, that can be executed by the processing component 402. The application programs stored in memory 401 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 402 is configured to execute instructions to perform method steps executed by the computer device in any of the above embodiments.

[0130] The computer device 400 may also include a power supply component 403 configured to perform power management of the computer device 400, a wired or wireless network interface 404 configured to connect the computer device 400 to a network, and an input / output (I / O) interface 405. The computer device 400 may operate on an operating system stored in memory 401, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0131] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0132] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element. In this document, "a," "an," "the," "the," and "its" may also include plural forms unless the context clearly indicates otherwise. "Multiple" refers to at least two, such as 2, 3, 5, or 8, etc. "And / or" includes any and all combinations of the related listed items.

[0133] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0134] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A high-precision map update method, characterized in that, The method includes: In the current round, if the vehicle is in motion, the road data of the road on which the vehicle is currently traveling is collected, and the actual coordinates of the lane markings are extracted from the road data corresponding to the current round. Based on the vehicle's position in the current round, the historical coordinates of the driving lane markings are obtained from a high-precision map; Based on the actual line coordinates corresponding to the current round and the historical line coordinates corresponding to the current round, calculate the line deviation value and obtain the relationship between the line deviation value corresponding to the current round and the preset deviation threshold. If the deviation value of the lane markings corresponding to the current round is greater than the preset deviation threshold, then the previous round is selected as the round to be evaluated, and the number of rounds selected, N, is initialized to zero. The vehicle distance between the vehicle's position in the round to be evaluated and its position in the current round is calculated. If the vehicle distance is less than or equal to a preset distance threshold, the number of rounds selected, N, is incremented by 1, and the round to be evaluated is updated to the previous round. The vehicle distance corresponding to the updated round to be evaluated is compared with the preset distance threshold until the vehicle distance corresponding to the updated round to be evaluated is greater than the preset distance threshold. Based on the size relationships corresponding to the previous N rounds, it is determined whether the high-precision map needs to be updated. When it is determined that the high-precision map needs to be updated, the road data corresponding to the current round is uploaded to the server, and the next round begins. The road data corresponding to the current round is used to instruct the server to update the high-precision map.

2. The high-precision map updating method according to claim 1, characterized in that, The step of determining whether the high-precision map needs to be updated based on the size relationship corresponding to the previous N rounds includes: In the first N rounds, if the corresponding deviation value of the datum line in each round is greater than the preset deviation threshold, then it is determined that the high-precision map needs to be updated.

3. The high-precision map updating method according to claim 1, characterized in that, The method further includes: If the vehicle uploaded road data in the previous round of the current round, the uploaded road length corresponding to the previous round is determined; wherein, when the vehicle did not upload road data in the two rounds prior to the current round, the uploaded road length corresponding to the previous round is the road length corresponding to the road data collected by the vehicle in the previous round; when the vehicle uploaded road data in the two rounds prior to the current round, the uploaded road length corresponding to the previous round is the sum of the road lengths corresponding to the road data collected by the vehicle in multiple consecutive rounds, wherein the multiple consecutive rounds include the round preceding the current round; If the length of the road already uploaded in the previous round is less than the preset upload length, then the road data corresponding to the current round is uploaded to the server.

4. The high-precision map updating method according to claim 3, characterized in that, The method further includes: If the length of the road that has been uploaded in the previous round is greater than or equal to the preset upload length, then when there is at least one round in the multiple consecutive rounds whose marking deviation value is greater than the preset deviation threshold, the road data corresponding to the current round is uploaded to the server.

5. The high-precision map updating method according to any one of claims 1 to 4, characterized in that, The method further includes: In response to the map update command sent by the server, update lane marking data is obtained from the map update command; The high-precision map stored in the vehicle is updated based on the updated lane marking data.

6. A high-precision map updating device, characterized in that, The device includes: The actual lane marking coordinate determination module is used to collect road data of the current driving road of the vehicle in the current round if the vehicle is in a driving state, and extract the actual lane marking coordinates of the driving lane from the road data corresponding to the current round. The historical lane marking coordinate determination module is used to obtain the historical lane marking coordinates from a high-precision map based on the vehicle's position in the current round. The marking deviation value determination module is used to calculate the marking deviation value based on the actual marking coordinates corresponding to the current round and the historical marking coordinates corresponding to the current round, and to obtain the relationship between the marking deviation value corresponding to the current round and the preset deviation threshold. The high-precision map update module is used to: if the datum deviation value corresponding to the current round is greater than the preset deviation threshold, then the previous round is selected as the round to be evaluated, and the number of rounds selected, N, is initialized to zero. The module calculates the vehicle distance between the vehicle position in the round to be evaluated and the vehicle position in the current round. If the vehicle distance is less than or equal to a preset distance threshold, the number of rounds selected, N, is incremented by 1, and the round to be evaluated is updated to the previous round. The vehicle distance corresponding to the updated round to be evaluated is compared with the preset distance threshold until the vehicle distance corresponding to the updated round to be evaluated is greater than the preset distance threshold. Based on the size relationships corresponding to the previous N rounds, the module determines whether the high-precision map needs to be updated. When it is determined that the high-precision map needs to be updated, the module uploads the road data corresponding to the current round to the server and proceeds to the next round. The road data corresponding to the current round is used to instruct the server to update the high-precision map.

7. A storage medium, characterized in that: The storage medium stores computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the high-precision map update method as described in any one of claims 1 to 5.

8. A vehicle, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the high-precision map update method as described in any one of claims 1 to 5.

Images