A double high-precision map verification method and system

Abstract

The application provides a double high-precision map verification method, and belongs to the field of map surveying and mapping. The method comprises the following steps: obtaining first high-precision map data and second high-precision map data; comparing map elements in the first high-precision map data and the second high-precision map data, and extracting difference elements with differences in all map elements; classifying and analyzing and verifying all difference elements to calculate the correctness, redundancy rate, omission rate and / or error rate of the first high-precision map data and / or the second high-precision map data; and outputting a verification report of the first high-precision map data and the second high-precision map data according to the calculation result. The different high-precision map data is verified by being obtained, the difference result of the high-precision map data is analyzed after verification, and then targeted verification and correction are performed according to the difference result, so that the verification period and verification cost are effectively reduced.

Description

Technical Field

[0001] This invention relates to the field of map surveying, and in particular to a method and system for verifying high-precision maps. Background Technology

[0002] The current high-precision map field verification process adopts a combination of machine recognition and manual processing. It typically requires multiple personnel for verification. Personnel A is responsible for driving the vehicle and collecting data; the data collection operation is a one-click start, requiring no human intervention. Personnel B listens to the voice broadcast of the high-precision map on the vehicle's infotainment system, observes the real-world scene, and reports any discrepancies between the high-precision map and reality to the recording personnel. Personnel C is responsible for recording the problems, including the type, time of occurrence, and a detailed description. After verification, the records are transmitted to the server for storage. Finally, professionals analyze the data collected during verification, combining it with the high-precision map data, to confirm the problems. After all problems are analyzed, a high-precision map verification report and problem analysis and improvement measures are output.

[0003] However, due to the varying accuracy of the equipment used by map providers, the high-precision maps they produce may contain erroneous elements or be inaccurate in their depiction of the real world, such as the curvature and slope of lanes. Consequently, the accuracy may also deviate significantly, with both horizontal and vertical accuracy deviations.

[0004] Furthermore, existing solutions involve high-precision map verification in the field, which is costly and time-consuming; simulation verification requires building simulation scenarios to simulate vehicle driving conditions, which is technically complex and costly. Moreover, these solutions are all based on their own map verification and cannot obtain third-party verification.

[0005] Therefore, there is a need to provide a high-precision map verification method to solve the problems of high verification costs and low map accuracy caused by long verification cycles, the need to build simulation scenarios and simulate vehicle driving conditions in existing technologies. Summary of the Invention

[0006] One objective of this invention is to provide a verification method for dual high-precision maps, in order to solve the problems of high verification costs caused by long verification cycles, the need to build simulation scenarios, and the need to simulate vehicle driving conditions in the prior art.

[0007] A further objective of this invention is to address the problem that high-precision maps from map providers cannot be verified by third parties.

[0008] Specifically, this invention provides a method for verifying dual high-precision maps, comprising the following steps:

[0009] Acquire the first and second high-precision map data;

[0010] Compare the map elements in the first high-precision map data and the second high-precision map data, and extract the difference elements among all map elements;

[0011] All discrepancies are classified and analyzed for verification in order to calculate the accuracy, redundancy, omission rate and / or error rate of the first high-precision map data and / or the second high-precision map data;

[0012] Based on the calculation results, output a verification report for the first high-precision map data and the second high-precision map data.

[0013] Furthermore, the step of classifying and analyzing all discrepancies to calculate the accuracy, redundancy, omission rate, and / or error rate of the first high-precision map data and / or the second high-precision map data includes the following steps:

[0014] For all differing elements in the same category, off-site verification and / or on-site verification are used to determine the total number of differing elements, the number of correct elements, and the number of redundant elements in the first high-precision map data and / or the second high-precision map data;

[0015] The accuracy of the first high-precision map data and / or the second high-precision map data is calculated according to the following formula: Accuracy = 1 - (number of errors + number of omissions + number of redundancies) / (total number of different elements + number of omissions) * 100%;

[0016] The redundancy rate of the first high-precision map data and / or the second high-precision map data is calculated using the following formula: Redundancy rate = Number of redundancies / (Total number of differential elements - Number of redundancies + Number of omissions) * 100%.

[0017] Furthermore, the step of classifying and analyzing all discrepancies to calculate the accuracy, redundancy, omission rate, and / or error rate of the first high-precision map data and / or the second high-precision map data also includes the following steps:

[0018] For all differing elements in the same category, determine the number of omissions and errors in the first high-precision map data and / or the second high-precision map data using off-site verification and / or on-site verification methods;

[0019] The omission rate of the first high-precision map data and / or the second high-precision map data shall be calculated according to the following formula: Omission rate = Number of omissions / (Total number of differential elements - Number of redundancies + Number of omissions) * 100%;

[0020] The error rate of the first high-precision map data and / or the second high-precision map data shall be calculated according to the following formula: Error rate = (number of errors + number of redundancies) / (total number of differential elements - number of redundancies + number of omissions) * 100%.

[0021] Furthermore, the step of determining the number of correct, redundant, missing, and erroneous elements in the first high-precision map data and / or the second high-precision map data for all differing elements within the same category using off-site verification and / or on-site verification methods includes the following steps:

[0022] For a given difference element, query the collection time of that difference element in the first high-precision map data and the second high-precision map data;

[0023] Newer high-precision map data was used as reference data.

[0024] Furthermore, in the step of comparing map elements in the first high-precision map data and the second high-precision map data, and extracting the difference elements among all map elements, the same regional specification is used to extract the differences in the maps.

[0025] Furthermore, the map elements include the geometric attributes and topological connections of roads, as well as the geometric and attribute information of traffic equipment.

[0026] Furthermore, it also includes improving and correcting the first high-precision map data and / or the second high-precision map data based on the output verification report.

[0027] The present invention also discloses a verification system for dual high-precision maps, including a control device. The control device includes a memory and a processor. The memory stores a control program, which, when executed by the processor, is used to implement the aforementioned verification method for dual high-precision maps.

[0028] This invention acquires high-precision map data from different manufacturers, performs regional analysis on these different high-precision map data, and then verifies the differences of each road element within the analyzed area. It calculates the accuracy, redundancy, omission rate, and error rate of each type of difference element in the high-precision map and outputs a verification report. Finally, based on the verification report, the high-precision map data is specifically verified and corrected, thereby effectively reducing the verification cycle and cost. This solves the problems of long verification cycles, the need to build simulation scenarios, and the simulation of vehicle driving states that cause high verification costs in existing technologies.

[0029] Furthermore, this invention verifies different high-precision map data by acquiring them from different sources. After verification, the differences between the high-precision map data are analyzed, and then targeted verification and correction are carried out based on these differences. This effectively reduces the verification cycle and verification cost. In addition, by acquiring high-precision map data from different manufacturers, it also avoids system errors caused by equipment limitations of the same map manufacturer, thus solving the problem that the verification of high-precision maps from map vendors cannot be obtained from third-party verification.

[0030] The above and other objects, advantages and features of the present invention will become more apparent to those skilled in the art from the following detailed description of specific embodiments of the invention in conjunction with the accompanying drawings. Attached Figure Description

[0031] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings:

[0032] Figure 1 This is a flowchart illustrating a high-precision map verification method according to an embodiment of the present invention;

[0033] Figure 2 This is a schematic diagram of high-precision map verification according to an embodiment of the present invention;

[0034] Figure 3 This is a schematic diagram of the structure of a high-precision map verification system according to an embodiment of the present invention;

[0035] In the diagram: 10 - Data acquisition module, 11 - Data parsing module, 12 - Comparison and verification module, 13 - Problem analysis module, 14 - Statistical analysis module, 15 - Data improvement module. Detailed Implementation

[0036] Figure 1 This is a flowchart illustrating a high-precision map verification method according to an embodiment of the present invention. In one embodiment, such as... Figure 1 As shown, the verification method for verifying dual high-precision maps includes the following steps:

[0037] S1. Obtain the first high-precision map data and the second high-precision map data;

[0038] S2. Compare the map elements in the first high-precision map data and the second high-precision map data, and extract the difference elements among all map elements;

[0039] S3. Classify and analyze all the differences to calculate the accuracy, redundancy, omission rate and / or error rate of the first high-precision map data and / or the second high-precision map data.

[0040] S4. Output a verification report of the first high-precision map data and the second high-precision map data based on the calculation results;

[0041] S5. Based on the output verification report, improve and correct the first high-precision map data and / or the second high-precision map data.

[0042] In other words, see Figure 1 First, high-precision map data is obtained from both the first and second map providers. This data can be obtained from the map providers' cloud servers or through offline transmission. Next, since different map providers use different map data specifications, it is necessary to obtain the specifications from both providers separately, parse the data, set up the parsing environment, and store the parsed data.

[0043] It should be noted that the map elements that must be parsed mainly include: the geometric and attribute information of regular roads and lanes, the topological connections of roads and lanes, the geometric and attribute information of ground traffic signs, the geometric and attribute information of roadside traffic signs, and the geometric and attribute information of traffic facilities such as gantries, streetlights, overhead tunnels, and turnstiles.

[0044] Then, all the discrepancies are categorized, analyzed, and verified. Assuming the high-precision map we need to verify is the first high-precision map data provided by the first map provider, we can use the second high-precision map data provided by the second map provider as a reference for verification. Input the coordinates of a specific path or global path (X1, Y1; X2, Y2; X3, Y3; X4, Y4; X5, Y5; ...), and then extract the map data of the first and second map providers within the same range according to the input coordinates. After obtaining the data, start comparing and verifying the map elements in the area. If the difference between the two is greater than a certain threshold, it is listed as a problem, and the accuracy, redundancy, omission rate, and error rate of the first high-precision map in the area are calculated. A verification report is output based on the calculation results.

[0045] Finally, based on the output verification report, determine whether the first high-precision map needs improvement or correction, and the direction of improvement or correction.

[0046] In this embodiment, high-precision map data from different manufacturers is acquired, and regional analysis is performed on the different high-precision map data. Then, the differences of each road element within the analyzed area are verified, and the accuracy, redundancy, omission rate, and error rate of each type of difference element in the high-precision map are calculated. A verification report is then output, and finally, the high-precision map data is specifically verified and corrected based on the verification report. This effectively reduces the verification cycle and verification cost. It solves the problems of high verification costs caused by long verification cycles, the need to build simulation scenarios, and the simulation of vehicle driving states in existing technologies.

[0047] Furthermore, this embodiment verifies the data by acquiring different high-precision map data. After verification, the differences between the high-precision map data are analyzed, and then targeted verification and correction are carried out based on the differences. This effectively reduces the verification cycle and verification cost. In addition, by acquiring high-precision map data from different manufacturers, system errors caused by equipment limitations of the same map manufacturer are avoided, thus solving the problem that the verification of high-precision maps from map vendors cannot be verified by a third party.

[0048] Furthermore, this invention utilizes a dual high-precision map verification method, eliminating the need for specialized equipment and reducing the cost of purchasing such equipment. It also eliminates the need for professional testing personnel to manually test and record data on-site, reducing errors and lowering labor, travel, and vehicle costs. Moreover, this dual high-precision map verification method ensures the accuracy of map providers' data is verified by a third party, making the verification results more authoritative. Additionally, this method can be performed simultaneously by multiple computers, resulting in high efficiency and significantly shortening the verification cycle.

[0049] According to one embodiment of the present invention, in step S2, in order to extract the difference elements among all map elements, the collected data during the high-precision map production process can be referenced for comparison and verification. For example, if the map marker positions of the first map provider and the second map provider are inconsistent at the same location, the latest collected data shall prevail. If the collected data of the first map provider is the most recent, then the first map provider's collected data shall be referenced, and the comparison shall be made according to the latest collected data. If they match, it is considered correct; otherwise, an error is recorded. There may also be cases where the first map provider has a certain map element at the same location, while the second map provider does not. Similarly, the collected data shall be referenced for judgment. According to the latest collected data, if the element exists in the collected data, the first map provider records it as correct, and the second map provider records it as an omission; if the element does not exist in the collected data, the first map provider marks it as redundant, and the second map provider records it as correct.

[0050] In this embodiment, by comparing the collection times of the first high-precision map data and the second high-precision map data, and using the latest map data as reference data, the timeliness of the map data can be guaranteed, changes in various map elements in the map data can be detected in a timely manner, and the map data can be corrected in a timely manner, thus ensuring the user experience of the map.

[0051] According to an embodiment of the present invention, in step S3, when calculating the accuracy, redundancy, omission rate and / or error rate of the first high-precision map data and / or the second high-precision map data, the problems are first statistically analyzed and classified according to the type of map elements, such as roads, lanes, ground markings, signs, traffic facilities, etc., wherein each type of map element includes the accuracy of geometry and attributes, and the accuracy rate of each type is calculated.

[0052] Then, for all discrepancies within the same category, off-site verification and / or on-site verification are used to determine the number of correct, redundant, missing, and incorrect elements in the first and / or second high-precision map data. The accuracy rate of the first and / or second high-precision map data is calculated according to the following formulas: Accuracy rate = 1 - (number of errors + number of omissions + number of redundancies) / (total number of discrepancies + number of omissions) * 100%; Redundancy rate = number of redundancies / (total number of discrepancies - number of redundancies + number of omissions) * 100%; Omission rate = number of omissions / (total number of discrepancies - number of redundancies + number of omissions) * 100%; Error rate = (number of errors + number of redundancies) / (total number of discrepancies - number of redundancies + number of omissions) * 100%.

[0053] In a specific embodiment, such as Figure 1 and Figure 2 As shown, to verify a road within a certain area of ​​the first high-precision map, we first obtain the geometric coordinates of the same elements, such as... Figure 2 The ground markings shown are as follows: the coordinates of the seven corner points of the first map are (XA1, YA1, ZA1; XA2, YA2, ZA2; XA3, YA3, ZA3; XA4, YA4, ZA4; XA5, YA5, ZA5; XA6, YA6, ZA6; XA7, YA7, ZA7), and the coordinates of the seven corner points of the second map are (XB1, YB1, ZB1; XB2, YB2, ZB2; XB3, YB3, ZB3; XB4, YB4, ZB4; XB5, YB5, ZB5; XB6, YB6, ZB6; XB7, YB7, ZB7). Next, the difference between the first and second map quotients for each corner point of the ground marker is calculated: Δ1 = (|XB1-XA1| + |YB1-YA1| + |ZB1-ZA|), and so on, to obtain the values ​​of Δ2, Δ3, and so on. Finally, the differences are summed to obtain the comparison result of the first and second map quotients for the ground marker. If the result is greater than a preset threshold, it is included in the problem.

[0054] The elements included in the problem are categorized, such as roads, lanes, ground markings, signs, traffic facilities, etc. Each category of map elements includes the accuracy of geometry and attributes, and the accuracy rate of each category is calculated.

[0055] Then, for all discrepancies within the same category, off-site verification and / or on-site verification are used to determine the total number of discrepancies, correct numbers, redundant numbers, omissions, and errors in the first and / or second high-precision map data. The accuracy rate of the first and / or second high-precision map data is calculated according to the following formulas: Accuracy rate = 1 - (number of errors + number of omissions + number of redundancies) / (total number of discrepancies + number of omissions) * 100%; Redundancy rate = number of redundancies / (total number of discrepancies - number of redundancies + number of omissions) * 100%; Omission rate = number of omissions / (total number of discrepancies - number of redundancies + number of omissions) * 100%; Error rate = (number of errors + number of redundancies) / (total number of discrepancies - number of redundancies + number of omissions) * 100%.

[0056] In this embodiment, a specific example is used to illustrate the working process of the present invention, but it is not limited to the situation described in this embodiment. It is also applicable when the road element in this embodiment is replaced with other elements.

[0057] This invention eliminates the need for specialized equipment, reducing the cost of purchasing such equipment. It also eliminates the need for professional testing personnel to conduct on-site manual testing and recording, reducing errors in manual recording and lowering labor, travel, and vehicle costs. Furthermore, through the dual high-precision map verification method, the accuracy of the map provider's own data is verified by a third party, making the verification results more authoritative.

[0058] Furthermore, the present invention also discloses a verification system for dual high-precision maps. The system includes a control device, which includes a memory and a processor. The memory stores a control program, and when the control program is executed by the processor, it is used to implement the aforementioned verification method for dual high-precision maps.

[0059] Specifically, such as Figure 3 As shown, the control device includes: a data acquisition module 10, a data parsing module 11, a comparison and verification module 12, a problem analysis module 13, a statistical analysis module 14, and a data improvement module 15. The data acquisition module 10 acquires a first high-precision map and a second high-precision map. The data parsing module 11, connected to the data acquisition module 10, parses the acquired first and second high-precision maps. The comparison and verification module 12, connected to the data parsing module 11, compares and verifies the differences between the first and second high-precision maps. The problem analysis module 13, connected to the comparison and verification module 12, calculates and analyzes the differences, records elements with differences exceeding a preset threshold, and outputs the analysis results. The statistical analysis module 14, connected to the problem analysis module 13, performs statistical calculations on the analysis results, calculating the accuracy rate, error rate, omission rate, and redundancy rate for each type of problem, and then sends the calculation results to the data improvement module 15 for improvement or data correction of the first or second high-precision map.

[0060] Furthermore, the problem analysis module 13 includes a problem classification unit and a problem integration unit. The problem classification unit calculates and analyzes the difference elements in the verification module and determines whether the difference value of the difference element is greater than a preset threshold. If so, the problem is recorded. The problem integration unit classifies and integrates the recorded problems according to the elements of the map, including geometric problems, attribute problems, etc.

[0061] After the problem integration unit sends the processed data to the statistical analysis module 14, the statistical analysis module 14 calculates the total number of elements, error rate, omission rate, and redundancy rate for this type of problem based on the following formulas: Accuracy rate = 1 - (number of errors + number of omissions + number of redundancies) / (total number of differing elements + number of omissions) * 100%; Redundancy rate = number of redundancies / (total number of differing elements - number of redundancies + number of omissions) * 100%; Omission rate = number of omissions / (total number of differing elements - number of redundancies + number of omissions) * 100%; Error rate = (number of errors + number of redundancies) / (total number of differing elements - number of redundancies + number of omissions) * 100%;

[0062] This invention employs a dual high-precision map verification method, performing mutual verification to simultaneously obtain the accuracy of both high-precision maps. Furthermore, by comparing and verifying any element in the high-precision maps, this invention allows for timely improvement of the high-precision maps based on the verified data. Moreover, it identifies key issues based on problem classification and provides targeted improvements, achieving continuous improvement of the high-precision maps. In addition, it offers advantages such as low cost and short development cycle.

[0063] Therefore, those skilled in the art should recognize that although numerous exemplary embodiments of the present invention have been shown and described in detail herein, many other variations or modifications conforming to the principles of the present invention can be directly determined or derived from the disclosure of the present invention without departing from the spirit and scope of the invention. Thus, the scope of the present invention should be understood and construed as covering all such other variations or modifications.

Claims

1. A method for verifying dual high-precision maps, characterized in that, Includes the following steps: Acquire the first and second high-precision map data; Compare the map elements in the first high-precision map data and the second high-precision map data, and extract the difference elements among all map elements; All discrepancies are classified and analyzed for verification in order to calculate the accuracy, redundancy, omission rate and / or error rate of the first high-precision map data and / or the second high-precision map data; Output a verification report for the first high-precision map data and the second high-precision map data based on the calculation results; The step of classifying and analyzing all discrepancies to calculate the accuracy, redundancy, omission rate, and / or error rate of the first high-precision map data and / or the second high-precision map data includes the following steps: For all differing elements in the same category, off-site verification and / or on-site verification are used to determine the total number of differing elements, the number of correct elements, and the number of redundant elements in the first high-precision map data and / or the second high-precision map data; The accuracy of the first high-precision map data and / or the second high-precision map data is calculated using the following formula: Accuracy = 1 - (Number of errors + Number of omissions + Number of redundancies) / (Total number of differing elements + Number of omissions) 100%; The redundancy rate of the first high-precision map data and / or the second high-precision map data is calculated using the following formula: Redundancy rate = Number of redundancies / (Total number of differing elements - Number of redundancies + Number of omissions) 100%.

2. The verification method according to claim 1, characterized in that, The step of classifying and analyzing all discrepancies to calculate the accuracy, redundancy, omission rate, and / or error rate of the first high-precision map data and / or the second high-precision map data further includes the following steps: For all differing elements in the same category, determine the number of omissions and errors in the first high-precision map data and / or the second high-precision map data using off-site verification and / or on-site verification methods; The omission rate of the first high-precision map data and / or the second high-precision map data is calculated using the following formula: Omission rate = Number of omissions / (Total number of differing elements - Number of redundancies + Number of omissions) 100%; The error rate of the first high-precision map data and / or the second high-precision map data shall be calculated according to the following formula: Error rate = (Number of errors + Number of redundancies) / (Total number of differing elements - Number of redundancies + Number of omissions) 100%.

3. The verification method according to claim 2, characterized in that, The steps for determining the number of correct, redundant, missing, and erroneous elements in the first and / or second high-precision map data for all differing elements within the same category using off-site and / or on-site verification methods include the following steps: For a given difference element, query the collection time of that difference element in the first high-precision map data and the second high-precision map data; Newer high-precision map data was used as reference data.

4. The verification method according to any one of claims 1-3, characterized in that, In the step of comparing map elements in the first high-precision map data and the second high-precision map data and extracting the differences among all map elements, the same regional specification is used to extract the differences from the maps.

5. The verification method according to claim 4, characterized in that, The map elements include the geometric attributes and topological connections of roads, as well as the geometric and attribute information of traffic equipment.

6. The verification method according to claim 5, characterized in that, It also includes improving and correcting the first high-precision map data and / or the second high-precision map data based on the output verification report.

7. A verification system for dual high-precision maps, characterized in that, The system includes a control device, which comprises a memory and a processor. The memory stores a control program, which, when executed by the processor, is used to implement the verification method for dual high-precision maps according to any one of claims 1-6.

Images