Crowdsourcing High-Precision Map Quality Evaluation Method, System and Storage Medium

Abstract

The present invention discloses a crowdsourcing high-precision map quality evaluation method, system and storage medium, including: Step 1. Obtain the vehicle information, positioning information and traveling direction of each crowdsourcing vehicle; Step 2. Send high-precision map data to the crowdsourcing vehicle; Step 3. Obtain road information and record the traveled trajectory information; Step 4. Collect road information, query the corresponding elements in the high-precision map data and compare them, and record the comparison results; Use the recorded trajectory information to make a judgment in the high-precision map data to determine whether there is a situation inconsistent with the trajectory, and give an evaluation result; Step 5. Upload the driving range, comparison results, trajectory information and evaluation results to the cloud; Step 6. The cloud conducts a comprehensive analysis of the corresponding high-precision map data to obtain the quality evaluation result of this time; Step 7. The cloud conducts an overall evaluation of the high-precision map data according to the historical quality evaluation results. The present invention avoids the uncertainty error of manual measurement.

Description

Technical Field

[0001] The present invention belongs to the technical field of high-precision maps, and particularly relates to a crowdsourcing high-precision map quality evaluation method, system and storage medium. Background Art

[0002] Under the information technology revolution, the automotive industry is gradually moving towards personalization, digitalization, and intelligence. Countries around the world have also introduced a series of policies and regulations to promote the rapid implementation of the autonomous driving industry. In intelligent driving, high-precision maps are an important component. Compared with traditional electronic maps, high-precision maps have richer and more complete road network information, including data such as lane slope, curvature, heading, elevation, and roll, and detailed information on road surface elements can provide a basis for automatic navigation, planning decisions, etc., playing a very important role in intelligent driving.

[0003] Since high-precision maps directly affect the driving routes of autonomous vehicles, the quality evaluation of high-precision map data is particularly important. Traditional map quality evaluation methods usually rely on manual inspection, using professional teams and high-precision equipment to collect ground truth data, and then manually comparing the ground truth data with map data to obtain the quality of map data. Since the existing method relies on manual collection, it takes a long time, has a high labor cost, and low evaluation efficiency, which is no longer a suitable practice for high-precision maps.

[0004] For example, a high-precision map test and evaluation method, device, vehicle and medium disclosed in patent document CN114440929A, the method includes: according to the high-precision map pre-deployed in the vehicle, real-time obtaining high-precision map data within a preset range of the vehicle; according to the high-precision map data, controlling the vehicle to perform autonomous driving, and real-time obtaining real road images within the preset range of the vehicle during the autonomous driving of the vehicle; simultaneously displaying the high-precision map data and the real road images to the user through a visualization interface, so that the user can evaluate the performance of the high-precision map according to the display results of the visualization interface. This method conducts actual autonomous driving tests and evaluates on a visualization interface. The test data needs to be pre-deployed on the vehicle, and at the same time, a preset test range is required, and the scope of the test evaluation is limited. The evaluation process requires manual confirmation on the interface, and the test process involves manual participation throughout. When conducting a large number of high-precision map data test evaluations, a large amount of labor costs are also required, it takes a long time, and large-scale high-precision map data evaluation tests are also difficult to carry out.

[0005] Therefore, it is necessary to develop a crowdsourcing high-precision map quality evaluation method, storage medium, device and vehicle. Summary of the Invention

[0006] The object of the present invention is to provide a crowdsourcing high-precision map quality assessment method, system and storage medium, so as to avoid the uncertainty error of manual measurement, improve the detection efficiency and quality, and reduce the detection cost at the same time.

[0007] In the first aspect, a crowdsourcing high-precision map quality assessment method according to the present invention includes the following steps:

[0008] Step 1. The cloud obtains the vehicle information, positioning information and traveling direction of each crowdsourcing vehicle.

[0009] Step 2. The cloud issues high-precision map data within a preset range to the crowdsourcing vehicle according to the vehicle information, positioning information and traveling direction of the crowdsourcing vehicle.

[0010] Step 3. The crowdsourcing vehicle receives the high-precision map data, and uses its own sensor module to obtain road information during driving, and records the driving trajectory information at the same time; wherein, the road information includes the position, geometric shape and type of the elements.

[0011] Step 4. Using the road information collected by the sensor module, query the corresponding elements in the received high-precision map data and make a comparison, and record the comparison result; at the same time, use the recorded trajectory information to make a judgment in the high-precision map data to determine whether there is a situation that does not match the trajectory, and give an evaluation result.

[0012] Step 5. After stopping driving or driving out of the current high-precision map range under comparison, upload the driving range, comparison result, trajectory information and evaluation result to the cloud.

[0013] Step 6. The cloud comprehensively analyzes the corresponding high-precision map data according to the driving range, comparison result, trajectory information and evaluation result to obtain the quality assessment result of this time.

[0014] Step 7. After accumulating a preset number of multi-vehicle evaluation times, the cloud makes an overall evaluation of the high-precision map data according to the historical quality assessment results to obtain the quality assessment result.

[0015] Optionally, in the above step 4, using the road information collected by the sensor module, query the corresponding elements in the received high-precision map data and make a comparison, wherein the comparison content includes whether the element exists, whether the type of the element matches, and whether the geometric shape of the element is consistent.

[0016] Optionally, determining whether there is a situation that does not match the trajectory specifically includes:

[0017] Whether the trajectory continuously stays within the lane during driving;

[0018] Whether the trajectory route appears at a position where it should not appear in the high-precision map;

[0019] Whether the trajectory direction is correct;

[0020] And based on the above judgment results, give the evaluation result of the current road.

[0021] Optionally, the evaluation result of the current road includes the overall quality score, whether serious errors occur, the number of ordinary errors, and the number of serious errors.

[0022] Optionally, in step 5, when uploading data, the unique vehicle identifier and vehicle evaluation weight are uploaded simultaneously.

[0023] Optionally, in step 6, the quality evaluation result includes the quality score, the range corresponding to the quality score, error information, the corresponding vehicle information, and the vehicle evaluation weight.

[0024] Optionally, in step 1, the vehicle information includes the unique identifier of the crowdsourcing vehicle in the cloud, the type of the crowdsourcing vehicle, the self-configuration of the crowdsourcing vehicle, and the vehicle evaluation weight.

[0025] Optionally, in step 2, the cloud sends high-precision map data within a range of 5 kilometers to 10 kilometers to the vehicle terminal according to the vehicle's positioning information and traveling direction.

[0026] In a second aspect, a crowdsourcing high-precision map quality evaluation system according to the present invention includes at least one processor and at least one memory; wherein, one or more computer-readable programs are stored in the memory, and when the computer-readable programs are called by the processor, they can execute the steps of the crowdsourcing high-precision map quality evaluation method as described in the present invention.

[0027] In a third aspect, a storage medium according to the present invention stores a computer-readable program, and when the computer-readable program is called, it can execute the steps of the crowdsourcing high-precision map quality evaluation method as described in the present invention.

[0028] The present invention has the following advantages: By connecting the crowdsourcing vehicle to the cloud and having the cloud aggregate and analyze the data, the present invention avoids the uncertainty error of manual measurement, improves the detection efficiency and quality, and reduces the detection cost at the same time. The present invention is analyzed based on big data technology. Due to reasons such as different vehicle configurations, different driving habits of drivers, and different current road conditions, the evaluation results of a single time are not the same and cannot completely evaluate all data. Therefore, after accumulating enough evaluation tests, the cloud determines that the historical data is sufficient and all data has been evaluated and tested enough times, and then comprehensively analyzes the historical data. As the number of times increases, the accuracy of the evaluation results will continuously improve. Brief Description of the Drawings

[0029] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention, and thus should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 is the flowchart of this embodiment. Detailed Embodiment

[0031] The present invention will be described in detail below with reference to the drawings.

[0032] As Figure 1 shown, in this embodiment, a crowdsourcing high-precision map quality assessment method includes the following steps:

[0033] Step 1. The cloud obtains the vehicle information, positioning information, and traveling direction of each crowdsourcing vehicle. Among them, the vehicle information includes the unique identifier of the crowdsourcing vehicle in the cloud, the type of the crowdsourcing vehicle, the self-configuration of the crowdsourcing vehicle, and the vehicle evaluation weight.

[0034] A positioning device is deployed on the crowdsourcing vehicle. The positioning device can be a GNSS device, and the positioning information and traveling direction are reported to the cloud through the GNSS device.

[0035] Step 2. The cloud sends the high-precision map data within a preset range to the crowdsourcing vehicle according to the vehicle information, positioning information, and traveling direction of the crowdsourcing vehicle.

[0036] In this step, the preset range is any value within the range of 5 kilometers to 10 kilometers. The data of the high-precision map can be sliced in the way of tiles, and 2 - 5 tile range data is selected when sending.

[0037] Step 3. The crowdsourcing vehicle receives the high-precision map data, and uses its own sensor module to obtain road information during driving, and at the same time records the driving trajectory information; among them, the road information includes the position, geometric shape, and type of the elements.

[0038] In this embodiment, the sensor module includes a lidar, a camera, etc., and the lidar and camera are used to identify the current road information. The identified road information includes various elements, such as arrows, markings, signs, etc.

[0039] Step 4. Query the corresponding features in the received high-precision map data based on the road information collected by the sensor module, compare them, and record the comparison results. Among them, the comparison content includes whether the features exist, whether the types of the features match, and whether the geometric shapes of the features are consistent.

[0040] At the same time, use the recorded trajectory information to judge in the high-precision map data, judge whether there is a situation that does not match the trajectory, and give an evaluation result.

[0041] In this embodiment, judging whether there is a situation that does not match the trajectory specifically includes:

[0042] Whether the trajectory continuously maintains driving within the lane;

[0043] Whether the trajectory route appears at a position where it should not appear in the high-precision map;

[0044] Whether the trajectory direction is correct;

[0045] And give an evaluation result of the current road based on the above judgment results.

[0046] In this embodiment, the evaluation result of the current road includes the overall quality score, whether there are serious errors, the number of ordinary errors, and the number of serious errors; and mark the error points.

[0047] Step 5. After stopping driving or driving out of the current high-precision map range for comparison, upload the driving range, comparison results, trajectory information, and evaluation results to the cloud; when uploading data, upload the vehicle unique identifier and vehicle evaluation weight at the same time.

[0048] Step 6. The cloud performs a comprehensive analysis on the corresponding high-precision map data based on the driving range, comparison results, trajectory information, and evaluation results, obtains the quality evaluation result of this time, and stores the quality evaluation result; among them, the quality evaluation result includes the quality score, the range corresponding to the quality score, error information, the corresponding vehicle information, and the vehicle evaluation weight.

[0049] Step 7. After accumulating a preset number of multi-vehicle evaluation times, the cloud performs an overall evaluation on the high-precision map data based on the historical quality evaluation results, obtains the quality evaluation result. At the same time, judge whether this section of data meets the release standard.

[0050] In step 7 here, the cloud will aggregate and statistically analyze based on big data technology. Due to reasons such as different vehicle configurations, different driving habits of drivers, and different current road conditions, the single evaluation results are different and cannot comprehensively evaluate all data. Therefore, after accumulating a sufficient number of evaluation tests, the cloud determines that the historical data is sufficient and all data has been evaluated enough times, and then comprehensively analyzes the historical data. As the number of times increases, the accuracy of the evaluation results will increase accordingly.

[0051] In this embodiment, a crowdsourcing high-precision map quality evaluation system includes at least one processor and at least one memory; wherein, one or more computer-readable programs are stored in the memory, and when the computer-readable programs are called by the processor, they can execute the steps of the crowdsourcing high-precision map quality evaluation method described in this embodiment.

[0052] In this embodiment, a storage medium stores a computer-readable program, and when the computer-readable program is called, it can execute the steps of the crowdsourcing high-precision map quality evaluation method described in this embodiment.

[0053] It should be noted that the storage medium shown in this embodiment can be a computer-readable signal medium, a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted by any suitable medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0054] The above embodiments are preferred embodiments of the present invention, but the embodiments of the present invention are not limited to the above embodiments. Any other changes, modifications, substitutions, combinations, or simplifications made without departing from the spirit and principle of the present invention shall be equivalent replacement methods and are all included in the protection scope of the present invention.

Claims

1. A crowdsourcing high-precision map quality assessment method, characterized in that, It includes the following steps: Step 1. The cloud obtains the vehicle information, positioning information, and traveling direction of each crowdsourcing vehicle; Step 2. The cloud issues high-precision map data within a preset range to the crowdsourcing vehicle according to the vehicle information, positioning information, and traveling direction of the crowdsourcing vehicle; Step 3. The crowdsourcing vehicle receives the high-precision map data, and uses its own sensor module to obtain road information during driving, and at the same time records the driving trajectory information; wherein, the road information includes the position, geometry, and type of the elements; Step 4. Use the road information collected by the sensor module to query and compare the corresponding elements in the received high-precision map data, and record the comparison result; at the same time, use the recorded trajectory information to make a judgment in the high-precision map data to determine whether there is a situation that does not match the trajectory, and give an evaluation result; Step 5. After stopping driving or driving out of the current compared high-precision map range, upload the driving range, comparison result, trajectory information, and evaluation result to the cloud; Step 6. The cloud comprehensively analyzes the corresponding high-precision map data according to the driving range, comparison result, trajectory information, and evaluation result to obtain the quality evaluation result of this time; the quality evaluation result includes a quality score, the range corresponding to the quality score, error information, the corresponding vehicle information, and the vehicle evaluation weight; Step 7. After accumulating a preset number of multi-vehicle evaluation times, the cloud conducts an overall evaluation of the high-precision map data according to the historical quality evaluation results to obtain the quality evaluation result.

2. The crowdsourcing high-precision map quality evaluation method according to claim 1, wherein: In the said Step 4, use the road information collected by the sensor module to query and compare the corresponding elements in the received high-precision map data, wherein the comparison content includes whether the element exists, whether the type of the element matches, and whether the geometry of the element is consistent.

3. The crowdsourcing high-precision map quality assessment method according to claim 1 or 2, characterized in that: In the said Step 4, judging whether there is a situation that does not match the trajectory specifically includes: Whether the trajectory continuously keeps driving within the lane; Whether the trajectory route appears at a position that should not appear in the high-precision map; Whether the trajectory direction is correct; And based on the above judgment results, give the evaluation result of the current road.

4. The crowdsourcing high-precision map quality assessment method according to claim 3, wherein: The evaluation result of the current road includes the overall quality score, whether serious errors occur, the number of ordinary errors that occur, and the number of serious errors that occur.

5. The crowdsourcing high-precision map quality assessment method according to claim 4, wherein: In the said Step 5, when uploading data, upload the vehicle unique identifier and the vehicle evaluation weight at the same time.

6. The crowdsourcing high-precision map quality evaluation method according to claim 1 or 2 or 4 or 5, characterized in that: In the said Step 1, the vehicle information includes the unique identifier of the crowdsourcing vehicle in the cloud, the type of the crowdsourcing vehicle, the self-configuration of the crowdsourcing vehicle, and the vehicle evaluation weight.

7. The crowdsourcing high-precision map quality assessment method according to claim 6, characterized in that: In the said Step 2, the cloud issues high-precision map data within a range of 5 kilometers to 10 kilometers to the vehicle end according to the positioning information and traveling direction of the vehicle.

8. A crowdsourcing high-precision map quality assessment system, characterized in that: It includes at least one processor and at least one memory; wherein, one or more computer-readable programs are stored in the memory, and when the computer-readable program is called by the processor, it can execute the steps of the crowdsourcing high-precision map quality evaluation method as described in any one of claims 1 to 7.

9. A storage medium, characterized in that: It stores a computer-readable program, and when the computer-readable program is called, it can execute the steps of the crowdsourcing high-precision map quality evaluation method as described in any one of claims 1 to 7.

Images