SELF-ADAPTING IMPROVEMENT FOR AUTOMATED DRIVING WITH MAP AND CAMERA PROBLEMS

The system autonomously corrects discrepancies in map and camera data by detecting curvature mismatches and adjusting confidence levels, improving the accuracy and efficiency of autonomous vehicle navigation.

DE102023129288A1Pending Publication Date: 2024-12-05GM GLOBAL TECHNOLOGY OPERATIONS LLC
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE102023129288
Authority / Receiving Office
DE · DE
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-06-01
Filing Date
2023-10-25
Publication Date
2024-12-05

AI Technical Summary

Technical Problem

Automated vehicle systems rely on map and camera data for navigation, but discrepancies between these data sources require manual intervention by technical personnel, which is time-consuming.

Method used

A system and method to automatically detect and correct mismatches in map and camera data by determining road curvatures, generating case reports, and adjusting confidence levels based on rationality checks and environmental conditions.

Benefits of technology

Automatically identifies and corrects data errors, reducing the need for manual intervention and enhancing the accuracy and efficiency of autonomous vehicle navigation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 00000000_0000_ABST
    Figure 00000000_0000_ABST
Patent Text Reader

Abstract

A vehicle comprises a system and a procedure for operating the vehicle. The system includes a camera, a map database, and a processor. The camera collects camera data about the location on a road traveled by the vehicle. The map database provides map data about the location on the road. The processor determines a first curvature of the road at that location from the camera data, determines a second curvature of the road at that location from the map data, identifies a mismatch between the first and second curvatures at that location, generates a case report for the map data or the camera data at that location upon occurrence of the mismatch, and adjusts the map data or a confidence level in the camera data for that location based on the case report.
Need to check novelty before this filing date? Find Prior Art

Description

INTRODUCTION

[0001] The present disclosure relates to autonomous vehicles and, more particularly, to a system and method for correcting map data and / or camera data used to guide the autonomous vehicles along a section of road.

[0002] Automated driving of a vehicle relies on map data and / or camera data to detect the direction of a road and road curvature. If the camera data and map data do not match at a particular location, technical personnel are called in to manually determine the causes of the discrepancy and propose solutions to improve data integrity. Such reliance on technical personnel is time-consuming. Therefore, it is desirable to provide a system that determines which of the map and camera data are inaccurate and makes the necessary corrections. SUMMARY

[0003] In an exemplary embodiment, a method for operating a vehicle is disclosed. Camera data of a location on a road is acquired and used by the vehicle to travel through the location. Map data of the location on the road is obtained and used by the vehicle to travel through the location. A first curvature of the road at the location is determined from the camera data. A second curvature of the road at the location is determined from the map data. A discrepancy or mismatch is detected between the first curvature and the second curvature at the road location. When the discrepancy occurs, a case report is created for the map data or the camera data of the location. The map data and a confidence level in the camera data are adjusted for the location based on the case report.

[0004] In addition to one or more of the features described herein, the method further comprises determining a rationality of the map data or the camera data by determining a deflection in the first curvature or the second curvature or an oscillation in the first curvature or the second curvature. The method further comprises determining the occurrence of an escalation event when either the first curvature or the second curvature is determined to satisfy a rationality condition. The method further comprises comparing a trajectory of a human-driven vehicle with the trajectory of the camera data or the map data at the location. The method further comprises generating the case report for each time the vehicle passes through the location. The method further comprises generating the case report for multiple vehicles passing through the location at different times.The method further includes excluding a case report for the camera data based on an environmental condition.

[0005] In another exemplary embodiment, a system for operating a vehicle is disclosed. The system includes a camera, a map database, and a processor. The camera is configured to receive camera data from a location on a road the vehicle is traveling on. The map database is configured to provide map data about the location on the road. The processor is configured to determine a first curvature of the road at the location from the camera data, determine a second curvature of the road at the location from the map data, identify a mismatch between the first curvature and the second curvature at the road location, generate a case report for the map data or the camera data at the location upon the occurrence of the mismatch, and adjust the map data or a confidence level in the camera data for the location based on the case report.

[0006] In addition to one or more of the features described herein, the processor is further configured to determine the rationality of the map data or the camera data by determining an excursion of the first curvature or the second curvature, or an oscillation of the first curvature or the second curvature. The processor is further configured to determine the occurrence of an escalation event when either the first curvature or the second curvature satisfies a rationality condition. The processor is further configured to compare a trajectory of a human-driven vehicle with the trajectory of the camera data or the map data at the location. The processor is further configured to generate the case report for each time the vehicle passes through the location. The processor is further configured to generate the case report for multiple vehicles passing through the location at different times.The processor is also configured to exclude a fall report for the camera data based on an environmental condition.

[0007] In another exemplary embodiment, a vehicle is disclosed. The vehicle has a camera, a map database, and a processor. The camera is configured to receive camera data from a location on a road that the vehicle is traveling on. The map database is configured to provide map data about the location on the road. The processor is configured to determine a first curvature of the road at the location from the camera data, determine a second curvature of the road at the location from the map data, identify a mismatch between the first curvature and the second curvature at the road location, generate a case report for the map data or the camera data at the location upon the occurrence of the mismatch, and adjust the map data or a confidence level in the camera data for the location based on the case report.

[0008] In addition to one or more of the features described herein, the processor is further configured to determine the rationality of the map data or the camera data by determining an excursion of the first curvature or the second curvature, or an oscillation of the first curvature or the second curvature. The processor is further configured to determine the occurrence of an escalation event when either the first curvature or the second curvature satisfies a rationality condition. The processor is further configured to compare a trajectory of a human-driven vehicle with the trajectory of the camera data or the map data at the location. The processor is further configured to generate the case report for each time the vehicle passes through the location. The processor is further configured to generate the case report for multiple vehicles passing through the location at different times.

[0009] The above features and advantages, as well as other features and advantages of the disclosure, will be readily apparent from the following detailed description when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Further features, advantages and details are listed only as examples in the following detailed description, which refers to the drawings in which: Fig. 1 shows a vehicle that can be operated in an autonomous mode or in a cruise control mode according to an exemplary embodiment; Fig. 2 shows a flowchart of a method for improving automated driving in one embodiment; Fig. 3 shows a flowchart of a method for identifying a problem for improved use of map and camera data in enhanced driving; Fig. 4 shows a graphical representation of a timeline of a trajectory of a vehicle following a selected data set, in one embodiment; Fig. 5 shows a graphical representation of a timeline of the trajectory of a vehicle following a selected data set in another embodiment; Fig. 6 shows a flowchart of a method for determining the rationality of a curvature excursion of a trajectory; Fig. 7 shows a flowchart 700 of a method for determining a rationality of a curvature oscillation; Fig. Figure 8 shows a first curvature for a trajectory of a human-driven vehicle and a second curvature for a trajectory based on a data set; and Fig. Figure 9 shows a flowchart illustrating a method for location analysis. DETAILED DESCRIPTION

[0011] The following description is merely exemplary in nature and is not intended to limit the disclosure, its application, or uses. It should be understood that throughout the drawings, corresponding reference numerals indicate like or corresponding parts and features.

[0012] According to an exemplary embodiment, Fig. 1 shows a vehicle 100 that can be operated in an autonomous mode or in a cruise control mode. The vehicle 100 has a driving system 102 that controls the movement of the vehicle over a section of road. The driving system 102 may include a camera 104 that captures information about the surroundings of the vehicle 100, including the road on which the vehicle is traveling. The driving system 102 also includes a map database 108 that provides a map of the section of road traveled by the vehicle. The vehicle 100 may receive both map data and camera data. The vehicle has a controller 106 that performs various calculations for operating the vehicle using the map data and the camera data.

[0013] The controller 106 may include processing circuitry, which may include an application-specific integrated circuit (ASIC), an electronic circuit, a processor (collectively, dedicated, or as a group), and memory executing one or more software or firmware programs, combinational logic circuitry, and / or other suitable components that provide the described functionality. The controller 106 may include a non-transitory computer-readable medium storing instructions that, when processed by one or more processors of the controller 106, implement a method for determining a difference between map data and camera data and adjusting a parameter related to either the map data or the camera data for later use.The method includes selecting map data or camera data to create a trajectory for the vehicle and maneuvering the vehicle along the trajectory.

[0014] The vehicle 100 may also include a communication unit 110 that enables the controller 106 to communicate with a map server 120. In an alternative embodiment, the map data may be received from the map database 108, and the results of the controller 106's calculations may be transmitted to the map database for implementation.

[0015] Fig. Figure 2 shows a flowchart 200 of a method for improving automated driving in one embodiment. In block 202, data is received at vehicle 100 for use in automated driving of the vehicle. The data includes camera data from camera 104 and map data from map database 108 or map server 120.

[0016] In field 204, a first curvature of the road at a selected location on the road is determined from the map data and a second curvature of the road at the selected location is determined from the camera data.

[0017] In block 206, the first curvature and the second curvature are compared to determine if there is a discrepancy or mismatch between them. If the first curvature and the second curvature differ by an amount less than a match threshold (i.e., the curvatures match), the method may return to block 202, where new records are received from the map database 108 and the camera 104. If the first curvature and the second curvature differ by an amount greater than a match threshold (i.e., the curvatures do not match), the method continues to block 208.

[0018] In field 208, both the first and second curvatures can be tested for rationality. The rationality of a curvature is determined by converting a deflection or oscillation of the road curvature into a value and then comparing the value to a threshold. If the value of the first curvature (i.e., the map data) or the second curvature (i.e., the camera data) is greater than the threshold, their curvatures are classified as irrational. If the value is greater than the threshold, the process continues with field 210. Otherwise, if the value is less than or equal to the threshold, the process continues with field 212.

[0019] First, a case report is created for an irrational data set in field 210. A case report can be created each time the location is passed through. This can happen each time the same vehicle passes through the location, or multiple vehicles pass through the location, or both. In field 222, the case reports associated with the location or road section are collected and counted. From field 222, the process continues in either field 224 or field 226. If there are a large number of case reports for the map data (i.e., greater than a threshold), the process continues with field 224. If there are a large number of case reports for the camera data, the process continues with field 226.

[0020] In field 224, the map is updated with a corrected map for the location or section of road. In field 226, the external environmental conditions that may affect the data are reviewed. For example, nighttime shots, rainy weather, etc., can result in poor camera data, even from a camera in good condition. This data is related to the environmental conditions. In field 228, the camera's confidence level for the road section is reduced or adjusted based on the environmental conditions.

[0021] In field 212, the camera data is reviewed to determine whether an escalation event occurred concurrently with the curvature deviation. A typical example of an escalation event is a lane departure. If an escalation event did not occur concurrently with the curvature deviation, the method proceeds to field 214. Otherwise, the method proceeds to field 218 for location analysis.

[0022] In field 214, the trajectory of a human-driven vehicle is compared with the map data and the camera data to determine whether the human-driven vehicle is following or generally aligned with the map data or the camera data. In field 216, the data set that the human-driven vehicle is not following is marked as potentially false data. The method proceeds from field 216 to field 222 for location analysis.

[0023] Field 218 determines whether a curvature deviation occurs after an escalation event. If a deviation occurs after an escalation event, the process continues with field 214. Otherwise, the process continues with field 220. Field 220 determines the number of map data cases and the number of camera data cases. From field 220, the process continues to field 222.

[0024] Fig. 3 shows a flowchart 300 of a method for identifying a problem for improved use of the map data and the camera data. The method begins in box 302 when a discrepancy has occurred between a first curvature at a road location based on the map data at a road location and a second curvature at the road location based on the camera data. In box 304, both the first curvature and the second curvature are compared to various rationality thresholds to determine whether the curvatures are rational or irrational. In box 306, a decision is made based on the results of the calculations performed in box 304. If the first curvature (map data) is irrational, the method proceeds to box 308. If the second curvature (camera data) is irrational, the method proceeds to box 310.If both the first curvature and the second curvature are rational, the method proceeds to box 320.

[0025] In field 308 (first curvature is irrational), a case report is created indicating a possible map problem. In field 310 (second curvature is irrational), a case report is created indicating a possible camera problem. In field 312, a location with a large number of case reports is identified. From field 312, the method proceeds to field 314 if a potential map problem exists (i.e., a large number of case reports for the map data). In field 314, the map database for the location is updated. From field 314, the method continues to field 330, where the method ends.

[0026] Referring again to box 312, if a potential camera problem exists (i.e., a large number of case reports for the camera data), the method continues to box 316. In box 316, the influence of weather conditions on the camera data is considered. In box 318, a confidence level for the camera data is adjusted (or reduced) based on the camera data problems and the impact of weather conditions on the camera data. From box 318, the method continues to box 330, where the method ends.

[0027] If both the map data and the camera data are determined to be rational, the processor determines in box 320 whether an escalation event has occurred. If no escalation event has occurred, the method continues to box 322. In box 322, the data indicative of the path of a human-driven vehicle is compared with the map data and the camera data to determine whether a human-driven vehicle is following the map data or the camera data. If the path of the human-driven vehicle is aligned with the camera data, the method continues to box 308. If the path of the human-driven vehicle is aligned with the map data, the method continues to box 310. If the path of the human-driven vehicle is not aligned with either map or camera data, the method continues to box 328 where further analysis is requested.

[0028] If an escalation event has occurred according to field 320, the method continues to field 324. In field 324, the data is reviewed to determine if a curvature deviation exists after the escalation event. If a curvature deviation exists after the escalation, the method continues to field 322. If there is no curvature deviation after the escalation event, the method continues to field 326. In field 326, the map data and the camera data are reviewed to determine if a change in curvature has occurred in either the map data or the camera data. If a change in curvature occurs in the map data, the method continues to field 308. If a change in curvature occurs in the camera data, the method continues to field 310. If no change occurs in either the camera data or the map data, the method continues to field 328. In field 328, further analysis is requested and / or performed.From field 328, the procedure continues with field 330, where the procedure ends.

[0029] Fig. Figure 4 shows a diagram 400 of a time axis of a trajectory 402 of a vehicle following a selected data set. Time in seconds (s) is shown on the abscissa, and amplitude (A) is shown on the ordinate. A section of the trajectory 402 can be isolated using a sliding window 404, and the curvature within the sliding window can be analyzed. In one embodiment, the deflection of the curvature can be checked for rationality.

[0030] To analyze the deflection, three curvature amplitudes are determined. A first curvature amplitude (Curv A ) is the curvature at the back of the sliding window. A second curvature amplitude (Curv B ) is the curvature that has the highest value within the sliding window. A third curvature amplitude (Curv C) is the curvature at the front of the sliding window. The rationality of the deflection can be determined by comparing the curvatures under different conditions (see Eqs. (1) - (4)) using the amplitudes. The change in curvature between any two curvatures can be compared with a relative threshold, as shown in Eq. (1): |Curv1−Curv2|>max|Curv|∗Threl where the indices (1,2) can be (A,B) or (B,C), Curv is the data set of all curvature values ​​and Th rel is a relative threshold that can be specified. The change in curvature can also be compared to an absolute threshold, as shown in Eqs. (2) and (3): |CurvA−CurvB|>Thabs |CurvB−CurvC|>Thabs

[0031] In addition, a product of the curvature changes is checked to see whether it is positive or negative, as shown in Eq. (4): (CurvA−CurvB)∗(CurvB−CurvC)<0

[0032] If all equations (1) - (4) are satisfied for a data set, the curvature of the data set is considered irrational.

[0033] Fig. Figure 5 shows a diagram 500 of a time axis of the trajectory 502 of a vehicle following a selected data set, in another embodiment. Time in seconds (s) is shown on the abscissa, and amplitude (A) is shown on the ordinate. A section of the trajectory 502 can be isolated by the sliding window 404. The oscillation of the curvature can be calculated.

[0034] To analyze the oscillation, three curvature amplitudes are determined. A first curvature amplitude (Curv A ) is the point with the locally lowest curvature value within the sliding window. A second curvature amplitude (Curv B) is the curvature that has the highest curvature value within the sliding window. A third curvature amplitude (Curv C ) is the curvature at an inflection point of the curvature after the second curvature amplitude. The rationality of the oscillation can be determined by comparing the curvatures under different conditions (shown in Eqs. (5)-(8)) based on the amplitudes. The change in curvature between any two curvatures can be compared with a relative threshold, as shown in Eq. (5): |Curv1−Curv2|>max|Curv|∗Threl

[0035] The change in curvature can also be compared to an absolute threshold value, as shown in Eqs. (6) and (7): |CurvA−CurvB|>Thabs / 2 |CurvB−CurvC|>Thabs / 2

[0036] In addition, a product of the curvature changes is checked to see whether it is positive or negative, as shown in Eq. (8): (CurvA−CurvB)∗(CurvB−CurvC)<0

[0037] In addition, the trajectory is checked to determine whether the same pattern occurs within a 10-second interval.

[0038] Fig. 6 shows a flowchart 600 of a method for determining the rationality of a curvature excursion. The method begins in block 602. In block 604, a sliding window is identified. If a sliding window cannot be identified within a preselected time period, the method continues to block 610. In an exemplary embodiment, the preselected time period is 3 seconds. In block 610, an output is generated indicating that no curvature excursion is present. The method then continues to block 614, where it ends.

[0039] Return to field 604: If a sliding window is identified, curvature points can be located and tested while the method continues with field 606. Otherwise, the method returns to field 604. In field 606, the first curvature (Curv A ) and the second curvature Curv B ) is determined. If Eq. (2) and Eq. (1) are satisfied for the first curvature and the second curvature, the method continues with box 608. Otherwise, the method returns to box 604. In box 608, the third curvature (Curv C ). If Eq. (3) and Eq. (1) are satisfied for the second curvature and the third curvature, and Eq. (4) is satisfied, the method continues to box 612. Otherwise, the method returns to box 604. In box 612, an output is generated indicating that a curvature excursion is present.

[0040] Fig. Figure 7 shows a flowchart 700 of a method for determining the rationality of a curvature oscillation. The method begins in block 702. In block 704, a sliding window is identified. If a sliding window with a preselected time duration cannot be determined, the method proceeds to block 712. In block 712, an output is generated indicating that no curvature oscillation is present. The method then proceeds to block 716, where it ends.

[0041] Returning to field 704, if a sliding window is identified, curvature points can be located and tested while the method continues with field 706. In field 706, the first curvature (Curv A ) and the second curvature Curv B ) is determined. If Eq. (6) and Eq. (6) are satisfied for the first curvature and the second curvature, the method continues with box 708. Otherwise, the method returns to box 704. In box 708, the third curvature (Curv C). If Eq. (7) and Eq. (5) are satisfied for the second curvature and the third curvature, and Eq. (8) is satisfied, the method continues with box 710. Otherwise, the method returns to box 704. In 710, it is checked whether these conditions are satisfied for another portion of the time axis within a 10-second window. If the conditions are satisfied within the 10-second window, the method continues in box 714. Otherwise, the method returns to box 704. In box 714, an output is generated indicating that a curvature oscillation is present.

[0042] Fig. 8 shows a first curvature 802 for a trajectory of a human-driven vehicle and a second curvature 804 for a trajectory based on a data set (either map data or camera data). Time (t) is plotted on the abscissa and curvature (C) on the ordinate axis. These curvatures can be compared to determine whether the vehicle has followed the data set. With respect to the first curvature 802, the curvature values ​​CV0, CV1, ..., CVn are calculated at intervals over a sampling period ΔS. Similar curvature values ​​(second curvature 804) include the curvature values ​​CM0, CM1, ..., CMn for map data and / or CC0, CC1, ..., CCn for camera data, which are calculated at the same intervals as for the human-driven vehicle. An area sum for the vehicle data is shown in Eq. (9) shown: SV=(CV+CV1+⋯+CVn)∗ΔS

[0043] Similarly, an area sum for the map data is shown in Eq. (10): SM=(CM0+CM1+⋯+CMn)∗ΔS and an area sum for the camera data is shown in Eq. (11): SC=(CC0+CC1+⋯+CCn)∗ΔS

[0044] The difference between the vehicle curvature and the curvature of the map data is calculated as follows: D1=SV−SM and the difference between the vehicle curvature and the curvature of the camera data is calculated by: D2=SV−SC

[0045] If: D2−D1≥ThCurv where Th Curv is a calibratable threshold that can be determined at design time, then the vehicle follows the map data. Otherwise, if: D1−D2≥ThCurv then the vehicle follows the camera data.

[0046] Fig. 9 shows a flowchart 900 illustrating a method for location analysis (as shown in field 222 of Fig. 2 or in field 312 of Fig. 3). The process begins in field 902. In field 904, the cases are organized so that they are arranged in a case sequence. In field 906, the cases are reviewed to determine if there are any pairs of cases that are not processed. If there are no pending cases, the process continues with field 916, where it ends.

[0047] If there are any pending cases, the process continues to field 908. In field 908, the distance is compared to a distance threshold. If the distance is less than the distance threshold, the process returns to field 906. Otherwise, the process continues with field 910. The distance can be determined as follows: distance=2r sin−1(sin2(φ2−φ12)+cos φ1⋅cos φ2⋅sin2(λ2−λ12))

[0048] Where φ1 and φ2 are the latitude of point 1 and the latitude of point 2, respectively, and λ1 and λ2 are the longitude of point 1 and the longitude of point 2, respectively.

[0049] In field 910, the cases are checked to see if they are linked to the same VIN (vehicle identification number) and date. If they are linked, the process returns to field 906. Otherwise, the process continues with field 912. In field 912, the cases are compared to determine if they originate from the same side of the road. If they do not originate from the same side of the road, the process returns to field 906. Otherwise, the process continues with field 914. In field 914, the two cases are placed in the same group. If neither of the existing groups can accommodate the two cases, a new group is created.

[0050] The terms "a" and "an" do not imply a limitation of quantity, but denote the presence of at least one of the mentioned items. The term "or" means "and / or" unless the context clearly indicates otherwise. Whenever "an aspect" is mentioned throughout the description, this means that a particular element (e.g. a feature, structure, step or property) described in connection with the aspect is included in at least one of the aspects described therein and may or may not be present in other aspects. It is understood that the described elements in the various aspects can be combined in any suitable way.

[0051] When an element such as a layer, film, sheet, region, or substrate is described as lying "on" another element, it may lie directly on top of the other element, or there may be intervening elements. In contrast, when an element is described as lying "directly on" another element, there are no intervening elements.

[0052] Unless otherwise indicated herein, all examination standards are the most recent standard in effect on the filing date of this application or, if priority is claimed, on the filing date of the earliest priority application in which the examination standard appears.

[0053] Unless otherwise defined, technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0054] While the above disclosure has been described with reference to exemplary embodiments, those skilled in the art will understand that various changes may be made and equivalents may be substituted for elements thereof without departing from the scope of the disclosure. Furthermore, many modifications may be made to adapt a particular situation or material to the teachings of the disclosure without departing from the essential scope of the disclosure. Therefore, the present disclosure is not intended to be limited to the particular embodiments disclosed, but is intended to include all embodiments falling within its scope.

[0055] Figure labeling of the Fig. 2, Fig. 3 , Fig. 6, Fig. 7: Y Yes N No

Claims

[1] A method of operating a vehicle, comprising: Receiving camera data of a location on a road, the camera data being used by the vehicle to drive through the location; Receiving map data of the location on the road, the map data being used by the vehicle to drive through the location; determining a first curvature of the road at the location from the camera data; determining a second curvature of the road at the location from the map data; identifying a mismatch between the first curvature and the second curvature at the location on the road; Generating an incident report for the map data or camera data at the location where the mismatch occurs; and Adjusting the map data or a confidence level in the camera data for the location based on the case report. [2] The method of claim 1, further comprising determining the rationality of the map data and / or the camera data by determining: (i) a deflection in the first curvature or the second curvature; or (ii) an oscillation of the first curvature or the second curvature. [3] The method of claim 1, further comprising determining the occurrence of an escalation event when either the first curvature or the second curvature is determined to satisfy a rationality condition. [4] The method of claim 1, further comprising comparing a trajectory of a human-driven vehicle with the trajectory of the camera data or the map data at the location. [5] The method of claim 1, further comprising creating the case report for one of the following cases: (i) each time the vehicle passes through the location; and (ii) multiple vehicles passing through the location at different times. [6] A system for operating a vehicle, comprising: a camera configured to receive camera data of a location on a road traversed by the vehicle; a map database configured to provide map data about the location on the street; a processor configured to: determining a first curvature of the road at the location from the camera data; determining a second curvature of the road at the location from the map data; Detecting a mismatch between the first curvature and the second curvature at the location on the road; Generating an incident report for the map data or camera data at the location where the mismatch occurs; and Adjusting either the map data or a confidence level in the camera data for the location based on the case report. [7] The system of claim 6, wherein the processor is further configured to determine the rationality of the map data or the camera data by determining one of: (i) a deflection in the first curvature or the second curvature; or (ii) an oscillation of the first curvature or the second curvature. [8] The system of claim 6, wherein the processor is further configured to determine the occurrence of an escalation event when it is determined that either the first curvature or the second curvature satisfies a rationality condition. [9] The system of claim 6, wherein the processor is further configured to compare a trajectory of a human-driven vehicle with the trajectory of the camera data or the map data at the location. [10] The system of claim 6, wherein the processor is further configured to generate the case report for one of the following cases: (i) each time the vehicle passes through the location; and (ii) multiple vehicles passing through the location at different times.