Rule-based lane contact analysis for automated driving
By using sensors to obtain lane markings and vehicle parameters in autonomous vehicles, using processors to detect fault domains, and automatically diagnose the causes of lane contact events, solving the problem of manual diagnosis in the prior art, improving the safety and efficiency of the system , reducing the occurrence of lane contact incidents.
Patent Information
- Application Number
- CN202311382188.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2023-06-01
- Filing Date
- 2023-10-23
- Publication Date
- 2024-12-03
AI Technical Summary
In the prior art, when an autonomous vehicle or vehicle cruise control system occurs, the engineering team needs to manually check the data to determine the cause, which is time-consuming and subjective, and it is difficult to effectively reduce or eliminate the occurrence of unexpected lane contact events.
During the operation of the vehicle's autonomous driving program, the first sensor acquires the external parameters of the lane mark and the second sensor obtains the internal parameters of the vehicle, and the processor detects the lane contact event, determines the fault domain that causes the lane contact event, and adjusts the relevant drivers System programming to identify and improve the boot system.
Automatic lane contact event fault identification and diagnosis is realized, reducing the time and subjectivity of manual inspections, improving the safety and efficiency of the system, and helping to design and improve the guidance system to reduce the occurrence of future lane contact events.
Smart Images

Figure CN119058730A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the operation of an autonomous vehicle or vehicle cruise control system and, more particularly, to systems and methods for analyzing the occurrence of a lane contact event to identify a fault domain of a vehicle guidance system that caused the lane contact. Background Art
[0002] Autonomous vehicles, as well as vehicles operating in super cruise control mode, can operate a guidance system that allows the vehicle to move safely along a stretch of road. The guidance system keeps the vehicle within a lane indicated by lane markings. A lane contact event occurs when the vehicle touches or crosses a lane marking or veers outside of the lane. For safety reasons, it is desirable to minimize or eliminate the occurrence of unintended lane contact events. When a lane contact event occurs during autonomous operation of the vehicle, it is desirable to determine its cause so that the guidance system can be upgraded or redesigned. Currently, determining the cause of a lane contact requires engineering teams to manually review data, which is time consuming and subjective. Therefore, it is desirable to provide a system that can determine the cause of a lane contact event in order to aid in the design and improvement of guidance systems. Summary of the invention
[0003] In an exemplary embodiment, a method of operating a drive system for a vehicle is disclosed. During a time period when an autonomous driving program is running at the vehicle, an external parameter indicating a lane marking is obtained using a first sensor. During the time period, an internal parameter of the vehicle is obtained using a second sensor. A lane contact event occurring during the time period is detected. An error occurring at the drive system that caused the lane contact event is determined at a processor based on the internal parameter and the external parameter. A fault domain of the drive system in which the error occurred is determined at the processor. A design of a program of the drive system associated with the fault domain is changed.
[0004] In addition to one or more features described herein, each of the internal parameters and the external parameters includes time series data obtained within a time interval. The fault domain is at least one of a vehicle sensing domain, a path planning domain, a vehicle control domain, and a reporting domain. Errors in the vehicle sensing domain include errors in detecting lane markings. The method also includes determining an error in the path planning domain when lane contact occurs by comparing a difference between a planned path of the vehicle and an actual path of the vehicle to a threshold. The method also includes determining a lane change problem caused by one of a delayed abort of a lane change, oversteering during a lane change, and overadjustment of a lane change. The method also includes determining the responsiveness of the drive system to a lane contact event.
[0005] In another exemplary embodiment, a drive system for a vehicle is disclosed. The drive system includes a first sensor, a second sensor, and a processor. The first sensor is configured to obtain an external parameter indicating a lane marking during a time period when an autonomous driving program is running at the vehicle. The second sensor is configured to obtain an internal parameter of the vehicle during the time period. The processor is configured to detect a lane contact event that occurs during the time period, determine an error that occurs in the drive system that causes the lane contact event based on the internal parameter and the external parameter, identify a fault domain of the drive system where the error occurs, and change the design of the program of the drive system related to the fault domain.
[0006] In addition to one or more features described herein, each of the internal parameters and the external parameters includes time series data obtained within a time interval. The fault domain is at least one of a vehicle sensing domain, a path planning domain, a vehicle control domain, and a reporting domain. The error in the vehicle sensing domain includes an error in detecting lane markings. The processor is also configured to determine an error in the path planning domain when lane contact occurs by comparing the difference between the planned path of the vehicle and the actual path of the vehicle to a threshold. The processor is also configured to determine a lane change problem caused by one of a delayed abort of a lane change, oversteering during a lane change, and overadjustment of a lane change. The processor is also configured to determine the responsiveness of the drive system to a lane contact event.
[0007] In yet another exemplary embodiment, a system for designing a drive system of a vehicle is disclosed. The system includes a first sensor, a second sensor, and a processor. The first sensor is configured to obtain an external parameter indicating a lane marking during a time period when an autonomous driving program is running at the vehicle. The second sensor is configured to obtain an internal parameter of the vehicle during the time period. The processor is configured to detect a lane contact event that occurs during the time period, determine an error that occurs in the drive system that causes the lane contact event based on the internal parameter and the external parameter, identify a fault domain of the drive system where the error occurs, and change the design of the program of the drive system related to the fault domain.
[0008] In addition to one or more features described herein, the fault domain is at least one of a vehicle sensing domain, a path planning domain, a vehicle control domain, and a reporting domain. The error in the vehicle sensing domain includes an error in detecting lane markings. The processor is also configured to determine an error in the path planning domain when lane contact occurs by comparing a difference between a planned path of the vehicle and an actual path of the vehicle to a threshold. The processor is also configured to determine a lane change problem caused by one of a delayed abort of a lane change, oversteering during a lane change, and overadjustment of a lane change. The processor is also configured to determine a responsiveness of the drive system to a lane contact event.
[0009] The above features and advantages and other features and advantages of the present disclosure will become apparent from the following detailed description when taken in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] Other features, advantages and details appear, by way of example only, in the following detailed description, which refers to the accompanying drawings, in which:
[0011] Figure 1 A vehicle capable of operating in an autonomous mode or a cruise control mode is shown according to an exemplary embodiment;
[0012] Figure 2 is a schematic diagram showing an operation flow for analyzing and improving a driving system with respect to detecting lane markings and with respect to lane marking movement;
[0013] Figure 3 A schematic diagram illustrating various routines executed at a drive system to maneuver a vehicle along a trajectory in an illustrative embodiment is shown;
[0014] Figure 4 A schematic diagram showing a procedure for identifying fault domains of a drive system;
[0015] Figure 5 shows a plurality of time series that may be analyzed at a remote processor to identify fault domains;
[0016] Figure 6 A flow chart showing a method for identifying the cause of lane contact occurrence in an illustrative embodiment;
[0017] Figure 7 shows a flow chart for identifying a lane marking detection problem in an illustrative embodiment;
[0018] Figure 8 A flow chart showing a method for identifying a path planning problem;
[0019] Fig. 9 A flow chart showing a method for identifying a lane change problem; and
[0020] Fig.10 A flow chart of a method for identifying straight road problems is shown. DETAILED DESCRIPTION
[0021] The following description is merely exemplary in nature and is not intended to limit the present disclosure, its application, or uses. It should be understood that throughout the drawings, corresponding reference numerals indicate the same or corresponding parts and features.
[0022] According to an exemplary embodiment, Figure 1A vehicle 100 is shown that can operate in an autonomous mode or a cruise control mode. The vehicle 100 includes a drive system 102 that controls movement of the vehicle along a stretch of road.
[0023] The drive system 102 may include a sensor system 104 for obtaining information about the surroundings or environment of the vehicle 100 or the vehicle itself, a controller 106 for calculating possible actions of the autonomous vehicle based on the obtained information and for implementing one or more possible actions, and an actuator system 108 for controlling vehicle navigation based on the actions calculated at the controller 106, such as a steering wheel actuator, a brake actuator, an acceleration actuator, etc.
[0024] The sensor system 104 may include external sensors (first sensors) that measure various external parameters of the environment, road, lane marking position, etc., such as cameras, Lidar, radar, GPS, etc. The sensor system 104 may also include internal sensors (second sensors) that measure internal parameters of the vehicle, such as accelerometers, angle measuring devices, etc. The data obtained for a given parameter (internal or external) can be obtained as a time series. Exemplary parameters include road curvature, the distance between the vehicle and the lane and the right lane or the left lane and the right lane marking, vehicle speed, vehicle yaw rate, vehicle lateral acceleration, vehicle longitudinal acceleration, vehicle position, steering wheel angle (SWA), steering torque, LCOD (Lane Change on Demand) state and / or LCC (Lane Centering Control) state, etc.
[0025] The controller 106 may include processing circuitry that may include an application specific integrated circuit (ASIC), electronic circuitry, a processor (shared, dedicated, or group) executing one or more software or firmware programs and memory, combinatorial 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 of analyzing parameters obtained by the sensor system 104, determining the occurrence of a lane contact event, determining a root cause or fault domain of the lane contact event, and suggesting possible improvements to mitigate the occurrence of future lane contact events.
[0026] The vehicle 100 also includes a communication unit 110 capable of communicating with a remote processor 120. In various embodiments, data can be transmitted from the vehicle 100 to the remote processor 120 for off-site processing. Data or software can also be transmitted from the remote processor 120 to the vehicle 100.
[0027] Figure 2A schematic diagram 200 illustrating an operational flow for analyzing and improving the driver system 102 with respect to detecting lane markings and moving relative to lane markings is shown. The operational flow includes calculations performed on the vehicle 100 using a processor of the vehicle 100 (e.g., at the controller 106), or calculations performed off-vehicle using a remote processor 120. The fleet 202 is driven in various environments to obtain data in the form of a plurality of time series data. The plurality of time series data is sent to the remote processor 120, which analyzes the time series data to determine when lane contact occurs and the cause of the lane contact. In particular, the remote processor 120 determines the domain in which the error that caused the lane contact occurred. Lane contact events may occur due to faults in multiple domains. Exemplary errors include, but are not limited to, lane perception errors, path planning errors, steering errors, etc. When a lane contact event occurs, the remote processor 120 determines in which domain (the "fault domain") the error that caused the lane contact event occurred. The analysis results in a review of the fault domain and / or a recommendation for design improvements that can be implemented and used in the fleet 202. If the analysis fails to identify a fault domain, a new (unknown category) domain can be created and used in subsequent analysis.
[0028] Figure 3 A schematic diagram 300 is shown illustrating various programs that are executed at the drive system 102 in an illustrative embodiment to maneuver the vehicle along a trajectory. These programs include a perception model 302, a path planning model 304, a vehicle tracking controller 306, a steering controller 308, and a vehicle position sensing / estimation program 312. The perception model 302 determines the presence and location of lane markings based on data obtained by the sensing system 104. The data may be related to weather conditions, road conditions, wind conditions, road slope, etc. The path planning model 304 is an algorithm or model for planning a vehicle trajectory based on the location of lane markings. The trajectory may include straight-line motion, turning, changing lanes, merging, exiting a highway, turning at an intersection, etc. The tracking controller 306 receives input indicating that the vehicle path deviates from the planned trajectory and determines actions that can be taken at the vehicle to make corrections to keep the vehicle moving along the planned trajectory. The tracking controller 306 sends a signal to the steering controller 308, which determines the steering torque or steering angle to be applied to the steering wheel of the vehicle 100 through the actuator 108.
[0029] The vehicle position sensing / estimation program 312 tracks the position of the vehicle 100 in response to the turn signal applied at the vehicle. The vehicle position can be the lateral position of the vehicle within the lane or relative to the lane markings. The vehicle position is sent to the difference engine 314, which determines the path deviation based on the vehicle position and the planned trajectory. The path deviation can then be sent to the tracking controller 306 to create a feedback loop.
[0030] Lane contact event monitor 316 obtains data from the program of drive system 102 and analyzes the data to determine when a lane contact event occurs. Further analysis can determine the root cause of the lane contact event, which can lead to improvements and redesign.
[0031] Figure 4 A schematic diagram 400 is shown identifying fault domains of a program of the drive system 102. The schematic diagram 400 includes a vehicle sensing domain 402, a path planning domain 404, a vehicle control domain 406, and a reporting domain 408. The vehicle sensing domain 402 includes a perception model 302 and a vehicle position sensing / estimation program 312. The path planning domain 404 includes the path planning model 304. The vehicle control domain 406 includes a tracking control program 306 and a steering control program 308. The reporting domain includes a lane contact event monitor 316.
[0032] Errors occurring in the vehicle sensing domain 402 may be due to lane marking issues, such as blurred lane markings, old and new lane markings, changes in road color, road separations, or road merges. Errors may also occur due to global positioning satellite (GPS) errors or map errors, poor perception due to weather, time of day, surroundings, etc., or sensor failures.
[0033] Errors occurring in the path planning domain 404 may be due to road segmentation or road merging, the presence of construction zones, the presence of other objects (i.e., vehicles, pedestrians) in the road that are near or cut off from the field of view, etc. The errors may also be the result of phantom object detection or misidentified object locations.
[0034] Errors occurring in the vehicle control domain 406 may occur in the vehicle tracking controller 306 and / or the steering controller 308. The error may be caused by a control change error, a weather-related control error, or a vehicle degradation error. A control change error occurs at a transition point between manual control and vehicle control. A control change error may be caused by operation of a lane change on demand (LCOD) program or a manual lane change. Errors are likely to occur in places such as sharp turns, zigzag roads, path changes, etc. Weather-related control errors may be errors caused by crosswinds, rain, snow, slippery roads, bumpy roads, potholes, roads, road slopes, bank angles, etc. Vehicle degradation errors may include errors caused by tire pressure, wheel misalignment, suspension, uneven braking, uneven wheel propulsion, differential propulsion, steering sensor errors, actuator failures, etc.
[0035] The error that occurs in the reporting domain 408 is due to a malfunction of the lane contact event monitor 316. Such errors may be due to data collection problems, onboard memory problems, and electronic control unit problems, misattribution of lane contact to vehicle control (rather than manual control), and other false positives due to erroneous lane marking detection.
[0036] Figure 5 A plurality of time series 500 that can be analyzed at the remote processor 120 to identify fault domains are shown. For each time series, time is shown in seconds along the abscissa. A first time series 502 includes data related to the occurrence of Super Cruise 509 and lane contact events 511 of the participant vehicle 100. A second time series 504 includes data related to steering and road curves. A third time series 506 includes data related to lane deviation. A fourth time series 508 includes data related to the operation of the lane contact event monitor 316. In the first time series 502, the Super Cruise curve 510 indicates that Super Cruise was engaged at approximately t=176.5 seconds, and the Lane Event curve 512 indicates that lane contact occurred at approximately t=178.3 seconds. In the second time series 504, the Steering Angle (SWA) curve 514 indicates that the steering wheel briefly turned to the left at approximately t=177.0 seconds. In the third time series 506, the Distance to Target Lane curve 516 indicates that the vehicle moved quickly to the left at approximately t=176.5 seconds. The occurrence of lane contact event 511 is also shown.
[0037] Figure 6 A flow chart 600 of a method for identifying the cause of a lane contact occurrence in an illustrative embodiment is shown. The method begins at box 602, where a lane contact report is received. At box 604, the lane contact report is cross-checked against the left lane and right lane offsets to determine the completeness of the report. At box 606, a decision is made based on the results of the cross-check in box 604. If the report is not an accurate report, the method proceeds to box 608. At box 608, an output is generated indicating that the report is erroneous. The method then proceeds to box 636, where the method ends.
[0038] Returning to box 606, if the report is an accurate report, the method proceeds to box 610. At box 610, a check is made to determine if there is a problem determining lane markings. At box 612, a decision is made based on the results of the check in box 610. If there is a lane marking detection problem, the method proceeds to box 614. At box 614, the method checks for relevant conditions and outputs a lane detection problem. Figure 7 Details of the checks performed at block 614 are discussed. From block 614, the method proceeds to block 636 where the method ends.
[0039] Returning to box 612, if there is no lane marking detection problem, the method proceeds to box 616. At box 616, a check is made to determine if there are any path planning problems. At box 618, a decision is made based on the results of box 616. If there is a path planning problem, the method proceeds to box 620. At box 620, the path planning problem is identified and output. Figure 8 Details of the method performed at block 620 are discussed. From block 620, the method proceeds to block 636 where the method ends.
[0040] Returning to box 618, if there is no path planning problem, the method proceeds to box 622. At box 622, it is determined whether the error occurred within a selected time range shortly after the cruise control was enabled. Such an error may be related to the transition from manual control of the vehicle to vehicle control of the vehicle. If the error occurred within the selected time range, the method proceeds to box 624. At box 624, the recorded conditions are checked just before the cruise control is activated, such as whether the vehicle is close to a lane marking when the transition is made, whether the vehicle is on the road, etc. These conditions are provided as output. The method then proceeds to box 636, where the method ends.
[0041] Returning to block 622, if the error occurred after the selected time frame, the method proceeds to block 626. At block 626, the log is checked to determine if the vehicle was on a curved road. If the vehicle was on a curved road when the lane contact occurred, the method proceeds to block 628. At block 628, the log is checked for a road entry or road exit condition or a curve condition, outputting a curve control problem. The method then proceeds to block 636, where the method ends.
[0042] Returning to box 626, if the vehicle is not on a curved road, the method proceeds to box 630. At box 630, the log is checked to determine if the vehicle acted in response to a lane change on demand (LCOD) command. If the vehicle responded to the LCOD, the method proceeds to box 632. At box 632, the log is checked for relevant LCOD conditions. The control problem associated with the LCOD is output. The method then proceeds to box 636, where the method ends.
[0043] Returning to box 630, if the vehicle does not respond to the LCOD, the method proceeds to box 634. In box 634, the log is checked for any relevant reaction conditions. These possible relevant actions are output. The method then proceeds to box 636, where the method ends.
[0044] Figure 7A flow chart 700 is shown for identifying a lane marking detection problem in an illustrative embodiment. The method begins at box 702. At box 704, variables for lane marking determination prior to lane contact are input and checked. These variables include, but are not limited to, warning_lane_uncertainty, left lane quality indication, right lane quality indication, map quality, etc.
[0045] At box 706, a decision is made based on the alert_lane_uncertainty variable. If alert_lane_uncertainty is "true" (i.e., if the lane detection system indicates that its detection of the lane marking is uncertain), the method proceeds to box 712. Otherwise, the method proceeds to box 708. At box 708, a sliding window is used to calculate a moving average of the lane quality for the left lane marking and the right lane marking, and is based on data from the camera and the map.
[0046] At box 710, a decision is made based on the lane quality. If the lane quality is low or poor, the method proceeds to box 712. At box 712, the record is checked for relevant conditions, such as lane merging or lane splitting, the occurrence of road construction, curved roads, lane changes, etc. At box 714, the lane detection problem is output along with the relevant conditions. The method then proceeds to box 718, where the method ends. Returning to box 710, if the lane quality is good, the method proceeds to box 716. At box 716, an output is generated to indicate that there is no lane marking problem.
[0047] Figure 8 A flow chart 800 of a method for identifying a path planning problem is shown. The method begins at box 802. At box 804, the values of path planning variables and thus the values obtained at a selected time before the lane contact event are examined. Such path planning variables include, but are not limited to, distance to target path (t). In the illustrative embodiment, the time period includes the time period from three seconds before lane contact to one second before lane contact. A negative value of distance to target path > 0 indicates that the vehicle is moving to the left, and a positive value indicates that the vehicle is moving to the right.
[0048] At block 806 , a minimum value of the distance to the target path during the selected time is determined, and a maximum value of the distance to the target path during the selected time is determined.
[0049] At box 808, a path planning error condition is checked. As an example, the condition is whether the minimum is within 40 centimeters of the left lane and left lane contact is received, or whether the maximum is within 40 centimeters of the right lane marking and right lane contact is received, and the road does not bend and no lane change is specified. When this is true, the method proceeds to box 810. At box 810, the path planning problem is output. The method proceeds to box 814, where the method ends.
[0050] Returning to block 808, if the condition is false, the method proceeds to block 812. At block 812, an output is generated to indicate that there is no path planning problem. From block 812, the method proceeds to block 814, where the method ends.
[0051] Fig. 9 A flow chart 900 of a method for identifying lane change problems is shown. The method begins at box 902. At box 904, lane change variable values obtained within a selected time prior to a lane contact event are input. Such lane change variables include, but are not limited to, lane change status and lane offset.
[0052] At block 906, the log is checked to determine if the lane change was aborted before the lane contact event occurred. If the lane change was aborted before the lane contact occurred, the method proceeds to block 908. At block 908, an output is generated to indicate that a delayed abort of the lane change occurred. The method proceeds to block 920, where the method ends.
[0053] Returning to block 906, if the passage is not aborted or aborted after the lane contact event, the method proceeds to block 910. At block 910, the record is checked to determine if another lane marker is contacted, such as the next lane marker of the next lane. If the next lane marker is contacted, the method proceeds to block 912. At block 912, an output is generated to indicate oversteering during the lane change. The method proceeds to block 920, where the method ends.
[0054] Returning to block 910, if the next lane marker has not been contacted, the method proceeds to block 914. At block 914, the log is checked to determine if the nearest lane has been contacted again after the lane contact event. If the nearest lane has been contacted again, the method proceeds to block 916. At block 916, an output is generated to indicate that the lane change was over-adjusted. The method then proceeds to block 920, where the method ends.
[0055] Returning to box 914, if the nearest lane has not been contacted after the original lane contact event, the method proceeds to box 918. At box 918, an output is generated to indicate that there is no lane change issue. From box 918, the method proceeds to box 920, where the method ends.
[0056] Fig.10 A flow chart 1000 of a method for identifying straight road problems is shown. The method begins at box 1002. At box 1004, steering variable values occurring within a time window prior to a lane contact event are input. Such steering variable values may include, but are not limited to, steering angle, lane offset, etc. In an illustrative embodiment, the time window may include from about 2 seconds prior to a lane contact event to about 1 second prior to a lane contact event.
[0057] At box 1006, an average steering angle value A is calculated within the time window. At box 1008, the average steering angle is compared to a threshold steering angle value. If the absolute value of the average steering angle is less than or equal to the threshold, the method proceeds to box 1010. At box 1010, an output is generated to indicate that the steering is not responsive to the lane contact event. From box 1010, the method proceeds to box 1018, where the method ends.
[0058] Returning to box 1008, if the absolute value of the average steering angle is greater than the threshold, the method proceeds to box 1012. In box 1012, the record is checked to determine whether the steering angle is directing the vehicle toward the contacted lane. If the steering is toward the contacted lane, the method proceeds to box 1014. Otherwise, the method proceeds to box 1016.
[0059] At box 1014, an output is generated to indicate that the steering performed an incorrect reaction to the lane contact. At box 1016, an output is generated to indicate that the steering was too late or too little to react to the lane contact event. From either box 1014 or box 1016, the method proceeds to box 1018, where the method ends.
[0060] The terms "one" and "an" do not indicate a limitation of quantity, but rather indicate the presence of at least one of the referenced items. The term "or" means "and / or", unless the context clearly indicates otherwise. References to "an aspect" throughout the specification mean that a particular element (e.g., feature, structure, step, or characteristic) described in conjunction with that aspect is included in at least one aspect described herein, and may or may not be present in other aspects. In addition, it should be understood that the described elements may be combined in any suitable manner in various aspects.
[0061] When an element such as a layer, film, region, or substrate is referred to as being “on” another element, it can be directly on the other element or intervening elements may also be present. In contrast, when an element is referred to as being “directly on” another element, there are no intervening elements present.
[0062] Unless otherwise indicated herein, all test standards are the most recent standards in effect as of the filing date of the present application or, if priority is claimed, the filing date of the earliest priority application in which the test standards appear.
[0063] Unless defined otherwise, 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 belongs.
[0064] Although the above disclosure has been described with reference to exemplary embodiments, it will be appreciated by those skilled in the art that various changes may be made and equivalents may be substituted for its elements without departing from the scope thereof. In addition, many modifications may be made to adapt specific situations or materials to the teachings of the disclosure without departing from the essential scope of the disclosure. Therefore, it is intended that the disclosure is not limited to the specific embodiments disclosed, but will include all embodiments falling within its scope.
Claims
1. A method of operating a drive system for a vehicle, comprising: obtaining, during a period of time when the automated driving program is running at the vehicle, an external parameter indicative of a lane marking using a first sensor; obtaining an internal parameter of the vehicle using a second sensor during the time period; detecting a lane contact event occurring during the time period; determining, at a processor, an error occurring at a drive system that caused the lane contact event based on the internal parameters and the external parameters; identifying, at a processor, a fault domain of a drive system in which an error occurred; as well as Change the design of the program that drives the system in relation to the fault domain.
2. The method according to claim 1, wherein: Each of the internal parameters and the external parameters includes time series data obtained within a time interval.
3. The method according to claim 1, wherein: The fault domain is at least one of: (i) a vehicle sensing domain; (ii) a path planning domain; (iii) a vehicle control domain; and (iv) a reporting domain.
4. The method according to claim 3, wherein: The errors in the vehicle sensing domain include errors in detecting lane markings.
5. The method of claim 3 further comprising determining an error in the path planning domain when lane contact occurs by comparing a difference between the planned path of the vehicle and the actual path of the vehicle to a threshold.
6. A drive system for a vehicle, comprising: a first sensor configured to obtain an external parameter indicative of a lane marking during a period of time when an automated driving program is being executed at the vehicle; a second sensor configured to obtain an internal parameter of the vehicle during the time period; Processor, which is configured as: detecting a lane contact event occurring during the time period; determining an error occurring in the drive system that caused the lane contact event based on the internal parameters and the external parameters; Identify the fault domain of the drive system where the error occurred; as well as Change the design of the program that drives the system in relation to the fault domain.
7. The drive system according to claim 6, wherein: Each of the internal parameters and the external parameters includes time series data obtained within a time interval.
8. The drive system according to claim 6, wherein: The fault domain is at least one of: (i) a vehicle sensing domain; (ii) a path planning domain; (iii) a vehicle control domain; and (iv) a reporting domain.
9. The drive system according to claim 8, wherein: The errors in the vehicle sensing domain include errors in detecting lane markings.
10. The drive system according to claim 8, wherein: The processor is also configured to determine an error in the path planning domain when lane contact occurs by comparing a difference between the planned path of the vehicle and the actual path of the vehicle to a threshold.
