Adaptive justifier for robust autonomous driving control oriented vehicle perception system

Abstract

The present invention relates to an adaptive justifier for a vehicle perception system oriented towards robust autonomous driving control. A system includes a computing device. The computing device includes a processor and a memory including instructions causing the processor to be configured to: convert vehicle dynamics data to corresponding road geometry data; calculate a justifier value based on a difference between the road geometry data and map-based road geometry data; and determine whether to modify a vehicle state based on the justifier value.

Description

TECHNICAL FIELD

[0001] The technical field is generally related to perception systems, and more particularly, to systems and methods for determining a rationalization value representing a discrepancy between sensor data and corresponding map data. BACKGROUND

[0002] Motor vehicles are increasingly incorporating advanced driver assistance systems (ADAS) designed to automate and / or augment the driving process and improve overall safety of the vehicle during operation. For example, ADAS can include a “horizon-based” driver assistance system that utilizes map data and external sensor data to predict a path that the vehicle can take as it travels along a roadway. SUMMARY

[0003] A system includes a computing device. The computing device includes a processor and a memory including instructions causing the processor to be configured to: convert vehicle dynamics data to corresponding road geometry data; compute a rationalization value based on a discrepancy between the road geometry data and map-based road geometry data; and determine whether to modify a vehicle state based on the rationalization value.

[0004] In other features, the processor is further configured to compute a rate of change of the rationalization value.

[0005] In other features, the processor is further configured to determine whether to modify the vehicle state based on the rationalization value and the rate of change of the rationalization value.

[0006] In other features, the processor is further configured to flag map data corresponding to a road segment, wherein the vehicle dynamics data corresponds to the road segment.

[0007] In other features, the processor is further configured to compute the rationalization value according to the equation:

[0008]

[0009] wherein, includes a curvature estimation error, includes a bank angle estimation error, represents a grade angle estimation error, includes a map-based curvature estimation, includes a map-based bank angle estimation, includes a map-based grade angle estimation, includes a vehicle curvature estimation, includes a vehicle bank angle estimation, and including vehicle slope angle estimation.

[0010] In other features, the processor is further configured to calculate the rationalization value based on wherein, , R is the rationalization value, is E a transpose of and Q is adjusted based on a lane centering error tracking.

[0011] In other features, the processor is further configured to transition a vehicle state from an autonomous state to at least one of a semi-autonomous state or a non-autonomous state based on the rationalization value.

[0012] In other features, the processor is further configured to generate an alert based on the rationalization value.

[0013] In other features, the processor is further configured to at least one of generate an alert based on the rationalization value or reduce control authority in uncertain environments.

[0014] In other features, the processor is further configured to evaluate local lane likelihoods based on the rationalization value with respect to the identified lane lines and correct lane level positioning.

[0015] In other features, the processor is further configured to adjust at least one of availability or control of autonomous or semi-autonomous features based on an evaluation of rationalization error over a prediction horizon.

[0016] In other features, the processor is further configured to combine positioning and perception data over a rolling horizon with historical vehicle dynamics measurements to isolate sources of potential discrepancies to enable robust lane following control.

[0017] A vehicle including a system is disclosed. The system includes a controller including a processor and a memory. The memory includes instructions causing the processor to be configured to: convert vehicle dynamics data to corresponding road geometry data; calculate a rationalization value based on a difference between the road geometry data and map-based road geometry data; and determine whether to modify a vehicle state based on the rationalization value.

[0018] In other features, the processor is further configured to calculate a rate of change of the rationalization value.

[0019] In other features, the processor is further configured to determine whether to modify the vehicle state based on the rationalization value and the rate of change of the rationalization value.

[0020] In other features, the processor is further configured to flag map data corresponding to a road segment, wherein the vehicle dynamics data corresponds to the road segment.

[0021] In other features, the processor is further configured to calculate the rationalization value according to the equation:

[0022]

[0023] wherein, includes a curvature estimation error, includes a bank angle estimation error, represents a grade angle estimation error, includes a map-based curvature estimation, includes a map-based bank angle estimation, includes a map-based grade angle estimation, includes a vehicle curvature estimation, includes a vehicle bank angle estimation, and includes a vehicle grade angle estimation.

[0024] In other features, the processor is further configured to calculate the rationalization value according to wherein, , R is the rationalization value, is E is a transpose of, and Q is adjusted based on a lane centering error tracking.

[0025] In other features, the processor is further configured to transition a vehicle state from an autonomous state to at least one of a semi-autonomous state or a non-autonomous state based on the rationalization value.

[0026] In other features, the processor is further configured to generate an alert based on the rationalization value.

[0027] The disclosure also includes the following technical solutions.

[0028] Solution 1. A system comprising a computing device, the computing device comprising a processor and a memory, the memory comprising instructions causing the processor to be configured to:

[0029] convert vehicle dynamics data to corresponding road geometry data;

[0030] calculate a rationalization value based on a difference between the road geometry data and map-based road geometry data; and

[0031] determine whether to modify a vehicle state based on the rationalization value.

[0032] Scheme 2. The system of scheme 1, wherein the processor is further configured to compute a rate of change of the rationalized value.

[0033] Scheme 3. The system of scheme 2, wherein the processor is further configured to determine whether to modify the vehicle state based on the rationalized value and the rate of change of the rationalized value.

[0034] Scheme 4. The system of scheme 1, wherein the processor is further configured to flag map data corresponding to a road segment and correct a map database in the segment, wherein the vehicle dynamics data corresponds to the road segment.

[0035] Scheme 5. The system of scheme 1, wherein the processor is further configured to compute the rationalized value according to the equation:

[0036]

[0037] wherein, includes a curvature estimation error, includes a bank angle estimation error, represents a grade angle estimation error, includes a map-based curvature estimation, includes a map-based bank angle estimation, includes a map-based grade angle estimation, includes a vehicle curvature estimation, includes a vehicle bank angle estimation, and includes a vehicle grade angle estimation.

[0038] Scheme 6. The system of scheme 5, wherein the processor is further configured to compute the rationalized value according to wherein, R is the rationalized value, is a transpose of E, and Q adjusts based on lane centering error tracking.

[0039] Scheme 7. The system of scheme 1, wherein the processor is further configured to transition a vehicle state from an autonomous state to at least one of a semi-autonomous state or a non-autonomous state based on the rationalized value.

[0040] Scheme 8. The system of scheme 1, wherein the processor is further configured to at least one of generate an alert based on the rationalized value or reduce control authority in an uncertain environment.

[0041] Scheme 9. The system of Scheme 1, wherein the processor is further configured to evaluate a local lane likelihood based on the rationalization value with respect to the identified lane line, and correct a lane level localization.

[0042] Scheme 10. The system of Scheme 1, wherein the processor is further configured to adjust at least one of availability or control of autonomous or semi-autonomous features based on an evaluation of rationalization error over a prediction horizon.

[0043] Scheme 11. The system of Scheme 1, wherein the processor is further configured to combine localization and perception data over a rolling horizon with historical vehicle dynamics measurements to isolate sources of potential discrepancies to enable robust lane following control.

[0044] Scheme 12. A vehicle comprising a system, the system comprising a controller comprising a processor and a memory, the memory comprising instructions causing the processor to be configured to:

[0045] convert vehicle dynamics data to corresponding road geometry data;

[0046] calculate a rationalization value based on a difference between the road geometry data and map-based road geometry data; and

[0047] determine whether to modify a vehicle state based on the rationalization value.

[0048] Scheme 13. The vehicle of Scheme 12, wherein the processor is further configured to calculate a rate of change of the rationalization value.

[0049] Scheme 14. The vehicle of Scheme 13, wherein the processor is further configured to determine whether to modify the vehicle state based on the rationalization value and the rate of change of the rationalization value.

[0050] Scheme 15. The vehicle of Scheme 12, wherein the processor is further configured to flag map data corresponding to a road segment, wherein the vehicle dynamics data corresponds to the road segment.

[0051] Scheme 16. The vehicle of Scheme 12, wherein the processor is further configured to calculate the rationalization value according to the equation:

[0052]

[0053] wherein, comprises a curvature estimation error, comprises a bank angle estimation error, denotes a grade angle estimation error, including a map-based curvature estimate, including a map-based bank angle estimate, including a map-based grade angle estimate, including a vehicle curvature estimate, including a vehicle bank angle estimate, and including a vehicle grade angle estimate.

[0054] Scheme 17. The vehicle of Scheme 16, wherein the processor is further configured to calculate the rationalization value according to wherein, R is the rationalization value, is a transpose of E, and Q adjusts based on lane centering error tracking.

[0055] Scheme 18. The vehicle of Scheme 12, wherein the processor is further configured to transition a vehicle state from an autonomous state to at least one of a semi-autonomous state or a non-autonomous state based on the rationalization value.

[0056] Scheme 19. The vehicle of Scheme 12, wherein the processor is further configured to generate an alert based on the rationalization value.

[0057] Scheme 20. The vehicle of Scheme 12, evaluating local lane likelihoods based on the rationalization value with respect to the identified lane lines, and correcting lane-level localization. BRIEF DESCRIPTION OF DRAWINGS

[0058] Exemplary embodiments will hereinafter be described in conjunction with the following drawings, wherein like numerals denote like elements, and wherein:

[0059] Figure 1 depicts an exemplary environment with a vehicle traversing a roadway, wherein sensor data representing a roadway geometry captured by one or more vehicle systems differs relative to corresponding map data representing the roadway geometry;

[0060] Figure 2 depicts a system for calculating a rationalization value;

[0061] Figure 3 depicts an exemplary communication system;

[0062] Figure 4 depicts an exemplary vehicle including one or more sensor devices, a camera, and a computing device;

[0063] Figure 5 depicts an exemplary computing device;

[0064] Figure 6An exemplary road segment including multiple lanes is depicted; and

[0065] Figure 7 FIG. 1 is a flow diagram illustrating an exemplary process for computing a rationalized value. DETAILED DESCRIPTION

[0066] Embodiments of the disclosure can be described herein in terms of functional and / or logical block components and various processing steps. It should be appreciated that such block components can be realized by any number of hardware, software, and / or firmware components configured to perform the specified functions. For example, an embodiment of the disclosure can employ various integrated circuit components, e.g., memory elements, digital signal processing elements, logic elements, look-up tables, or the like, which can carry out a variety of functions under the control of one or more microprocessors or other control devices. Similarly, the software implemented aspects of an embodiment of the disclosure are not described with reference to a particular computer system in terms of its hardware architecture. Rather, computers useful for

[0067] Autonomous and semi-autonomous vehicles are capable of sensing their environment and navigating based on the sensed environment. Such vehicles use multiple types of sensing devices, such as radar, lidar, image sensors, to sense their environment.

[0068] The present disclosure relates to systems and processes that combine localization and perception data over a receding horizon with historical vehicle dynamic data to determine potential discrepancies between the two data sources, such as map data, sensor data, and / or camera data.

[0069] Figure 1 An exemplary environment including a vehicle 20 traversing a road surface 10 is illustrated. As shown, the road surface 10 includes lane markings 12, 14. The lane markings 12 can be used to define the road surface 10 into at least two lanes, and the lane markings 14 can be used to define a perimeter of the road surface 10. Although depicted as solid lines, it is to be understood that the lane markings 12 can include intermittent markings.

[0070] The vehicle 20 can include a driver assistance system (DAS) and / or an advanced driver assistance system (ADAS) that uses data provided by one or more sensors within the vehicle 20 to assist a driver in controlling the vehicle 20. The DAS can include, but is not limited to, an electronic stability control system, an anti-lock braking system, and a traction control system. The ADAS can include, but is not limited to, a lane keep assist (LKA) system and an adaptive cruise control (ACC) system.

[0071] As vehicle 20 travels along roadway 10, the vehicle's road geometry sensing and estimation system, i.e., vehicle dynamic state, mapping and / or localization, camera sensors, can be subject to errors and / or uncertainties. For example, as shown in FIG. 1, predicted vehicle path trajectory 25 and predicted map vehicle path 30 through which vehicle 20 travels can initially overlap. However, as the distance between vehicle 20 and a perceived object increases, predicted map vehicle path 30 and predicted vehicle trajectory 25 can diverge, as shown by the increasing vertical lines. Figure 1

[0072] Figure 2 An exemplary control system 100 for operating a vehicle, such as vehicle 20, is illustrated. Control system 100 includes one or more sensor devices 105, one or more cameras 140, and a computing device 205 that includes a vehicle state and path estimation module 110, an adaptive rationalizer module 115, an alert module 120, a vehicle control module 125, a localization module 130, a digital map database 135, and a fusion module 145.

[0073] Sensor devices 105 can include one or more of a radar device, a LIDAR device, an ultrasonic device, an inertial measurement unit (IMU), a wheel encoder, a power steering device, a global positioning system (GPS) device, or other similar devices for collecting data about the vehicle, the vehicle's environment, and the behavior of other vehicles on the roadway. Cameras 140 can capture a series of images related to the environment near and in the vehicle's path, including but not limited to images of the roadway, images of lane markings, images of potential obstacles near the vehicle, images of vehicles around the vehicle, and other images of related information for controlling the vehicle.

[0074] Vehicle state and path estimation module 110 receives sensor data and / or camera data from sensor devices 105 and / or cameras 140. Vehicle state and path estimation module 110 can determine the vehicle dynamic state of vehicle 20, i.e., position, heading, steering angle, and vehicle path to be traversed, using suitable sensor fusion techniques and / or vehicle path determination techniques. Vehicle state and path estimation module 110 can also include a road geometry approximation module 150 that generates road geometry approximation data based on the vehicle dynamic data, which is described in more detail below.

[0075] ​The adaptive rationalizer module 115 can receive vehicle localization data from the localization module 130 and road geometry approximation data from the road geometry approximation module 150. The adaptive rationalizer module 115 generates a rationalization value over a prediction horizon of the vehicle 20. The rationalization value can be defined as a difference between map data provided by the digital map database 135 and camera data provided by the camera 140, as explained in more detail below.

[0076] The alert module 120 can generate an alert based on the rationalization value. For example, the alert module 120 can generate a visual and / or audible alert to alert an operator of the vehicle 20 of a potential discrepancy. The alert can be displayed at a human-machine interface within the vehicle 20, such as a display screen. Based on the alert, the operator can take action, such as operating the vehicle 20, etc.

[0077] The vehicle control module 125 receives data from the vehicle state and path estimation module 110 to cause the vehicle 20 to traverse the determined path. For example, based on vehicle path data determined by the vehicle state and path estimation module 110, the vehicle control module 125 can operate the vehicle 20 according to a determined desired vehicle speed and desired vehicle trajectory. For example, the vehicle control module 125 can control steering, braking, acceleration, lateral and longitudinal movement of the vehicle 20.

[0078] The localization module 130 receives map data from the digital map database 135 and localizes the vehicle 20 based on the map data and / or sensor data. In an example embodiment, based on a GPS location of the vehicle 20, the localization module 130 determines a vehicle position relative to the map data. For example, the localization module 130 can determine a relative position of the vehicle 20 on a road, such as which lane the vehicle 20 is traversing.

[0079] The digital map database 135 includes data about an area near the vehicle 20, including: historically recorded road geometry; lanes within a road; synthetic data, such as vehicle-to-vehicle or infrastructure-to-vehicle data about road geometry; and other information that can be monitored and stored about a particular area that the vehicle 20 can travel. Road geometry data can include, but is not limited to, road curvature estimates, bank angle estimates, and / or grade angle estimates.

[0080] Fusion module 145 can receive camera data from camera 140 and apply suitable fusion techniques to determine lane-level offset, lane-level curvature, lane-level tilt angle, and / or lane-level grade angle data. For example, fusion module 145 can use suitable object classification and / or detection techniques to determine lane-level offset, lane-level curvature, lane-level tilt angle, and / or lane-level grade angle data based on camera images captured by camera 140.

[0081] Figure 3 An exemplary data communication system 200 within a controlled vehicle is illustrated. Data communication system 200 includes sensors 110, camera 140, and computing device 205 communicatively connected via vehicle data bus 240. Vehicle data bus 240 includes a communication network capable of quickly passing data back and forth between various connected devices and modules. Data can be collected from each of sensor device 105, camera 140, and / or digital map database 120. As described in greater detail herein, computing device 205 can use sensor data to determine whether there is a discrepancy in map data obtained from digital map database 120.

[0082] Figure 4 An exemplary vehicle 20 that can be navigated using control system 100 is illustrated. As shown, vehicle 20 is traveling on road surface 10 that includes lane markings 14. Vehicle 20 can include computing device 205, sensor device 105, and camera 140. Camera 140 includes field of view 112 and is positioned to capture images of road surface 10 and other objects and obstacles in the vicinity of vehicle 20. Sensor device 105 can additionally provide data about objects in the vicinity of vehicle 20. Computing device 205 receives data from sensor device 105 and / or camera 140.

[0083] Figure 5 An exemplary computing device 205 is illustrated. Computing device 205 includes processor 410, data input / output module 420, communication module 430, and memory storage 440. Note that computing device 205 can include other components, and some of these components are not present in certain embodiments.

[0084] Processor 410 can include memory, such as read-only memory (ROM) and random access memory (RAM), that stores processor-executable instructions, and one or more processors that execute the processor-executable instructions. In embodiments where processor 410 includes two or more processors, the processors can operate in parallel or distributed fashion. Processor 410 can execute an operating system of computing device 205. Processor 410 can include one or more modules that execute programming code or computerized processes or methods that include executable steps. The illustrated modules can include a single physical device or functionality that spans multiple physical devices.

[0085] The data input / output module 420 is an apparatus operable to acquire data collected from sensors and devices throughout the vehicle and process the data into a format that is readily usable by the processor 410. The data input / output module 420 is also operable to process output from the processor 410 and enable the output to be used by other devices or modules throughout the vehicle. For example, the data input / output module 420 can include a human-machine interface to display alerts to an operator of the vehicle 20.

[0086] The communication module 430 can include communication / data connections with bus apparatus configured to transmit data to different components of the system and can include one or more wireless transceivers for performing wireless communication.

[0087] The memory storage device 440 is an apparatus that stores data generated or received by the computing apparatus 205. The memory storage device 440 can include, but is not limited to, a hard disk drive, an optical disk drive, and / or a flash drive.

[0088] The computing apparatus 205 can determine whether there is a discrepancy between the map data, the sensor data, and / or the camera data. In some embodiments, if there is a discrepancy, the computing apparatus 205 can generate an alert and / or modify the vehicle state. The computing apparatus 205 includes a road geometry approximation module 150. The road geometry approximation module 150 converts vehicle dynamic state data into corresponding road geometry data. The road geometry approximation module 150 can transform vehicle dynamic state data into road geometry data according to Equations 1-12, which assume constant longitudinal velocity (by incorporating non-zero , the same theory applies for variable velocity):

[0089] Equation 1

[0090] Equation 2

[0091] Equation 3

[0092] Equation 4

[0093] Equation 5

[0094] Equation 6

[0095] Equation 7

[0096] Equation 8

[0097] Equation 9

[0098] Equation 10

[0099] Equation 11

[0100] Equation 12

[0101] where, is the vehicle and wheel angle, is the vehicle lateral velocity, is the longitudinal velocity, is the yaw rate, is the CG to vehicle rear axle distance, is the CG to vehicle front axle distance, is the host vehicle to lane marker relative heading, is the host vehicle lateral offset from the desired trajectory, is the estimated vehicle curvature, is the estimated road bank angle, is the estimated road pitch angle, is the lateral acceleration, is the estimated vehicle heading, is the front tire cornering stiffness, is the rear tire cornering stiffness, m is the vehicle mass, is the vehicle moment of inertia, is the weighted average of the design parameters.

[0102] The adaptive rationalizer module 115 computes various error values defined in Equations 13-15:

[0103] Equation 13

[0104] Equation 14

[0105] Equation 15

[0106] where, includes the curvature estimation error, includes the bank angle estimation error, denotes the pitch angle estimation error, includes the map-based curvature estimate, includes the map-based bank angle estimate, includes the map-based pitch angle estimate, includes the vehicle curvature estimate, includes the vehicle bank angle estimate, and including vehicle grade angle estimation.

[0107] The adaptive justifier module 115 can then compute a justification value that indicates a difference between the map-based road estimate and the sensor-based road estimate. The adaptive justifier module 115 can compute the justification value according to equation 16:

[0108] Equation 16

[0109] wherein, , R is the justification value, is E transpose of and Q is adjusted in real-time based on the SC / UC tracking control error wherein includes lane centering error tracking, and includes derivative of to represent rate of change of .

[0110] The adaptive justifier module 115 can also compute a rate of change of the justification value R . For example, the justifier module 115 can determine the rate of change by computing a derivative of the justification value R . The rate of change of the justification value R may be represented as .

[0111] The justification value R may represent an explicit quantification of error, e.g., an explicit quantification of a difference between map data, sensor data, and / or camera data. The adaptive justifier module 115 can combine positioning and perception data over a rolling time horizon with previous vehicle dynamic measurements. The rolling time horizon can include a predicted vehicle state based on received sensor data and / or camera data. For example, the computing device 205 can compute a rolling time horizon based on received sensor data and / or camera data. The computing device 205 can also construct a rolling time horizon that includes stored sensor data and / or camera data. In other words, the rolling time horizon can include a time window of stored vehicle dynamic data that can be used to construct a rolling time horizon prediction.

[0112] In various implementations, the computing device 205 can use the justification value R and its rate of change a suitable model to quantify error between the map data, sensor data, and / or camera data within a predetermined time period, i.e., steady state. In some cases, the computing device 205 generates a confidence interval using rolling and / or historical time domain data. For example, the computing device 205 can apply a suitable model error algorithm to the rationalized values based on the rolling and / or historical time domain data R and rate of change thereof .

[0113] rationalized values R may be used to modify operation of one or more vehicles 20. In some cases, when the rationalized values R exceed a predetermined rationalization threshold, the computing device 205 can de-weight camera data received from the camera 140. In some cases, when the rationalized values R exceed a predetermined rationalization threshold, the alert module 120 generates an alert, e.g., via the data input / output module 420. In some cases, the computing device 205 can cause the vehicle 20 to transition from a first vehicle operating state to a second vehicle operating state via the vehicle control module 125. For example, the computing device 205 can cause the vehicle 20 to transition from an autonomous state to a semi-autonomous or non-autonomous state, e.g., to reduce control authority in an uncertain environment represented by the rationalized values R The computing device 205 can also evaluate a localized lane likelihood based on the rationalized values for the identified lane lines and correct lane level localization. The computing device 205 can also adjust availability or control of autonomous or semi-autonomous features based on the evaluation of the rationalized error within the prediction time domain. The computing device 205 can also combine localization and perception data over a rolling time domain with historical vehicle dynamics measurements to isolate sources of potential discrepancies to enable robust lane following control.

[0114] In some cases, when the rationalized values R exceed a predetermined rationalization threshold, the computing device 205 can send a map database update trigger to indicate that an update to road segment map data is needed. This map database update trigger can be communicated to an entity that controls the map data provided to the map database 120. In some cases, the computing device 205 can determine that the vehicle 20 is traversing a construction zone. For example, the computing device 205 can compare the current rationalized values R to a historical time domain. If the deviation of the rationalized values R from the historical time domain is greater than a predetermined deviation amount, the computing device 205 can determine that the vehicle 20 is traversing a construction zone.

[0115] Figure 6An exemplary environment 600 is illustrated in which the computing device 205 can select lane 3 based on lane 3 having the lowest rationalization value R , i.e., relative to the lowest rationalization values of lanes 1, 2, and 4 R . It is to be understood that the rationalization value R may include additional inputs to the vehicle localization algorithm.

[0116] Figure 7 is a flowchart of an exemplary process 700 for computing the rationalization value R . The blocks of the process 700 can be performed by the computing device 205. The process 700 begins at block 705 where the computing device determines whether sensor data and / or camera data has been received. If the computing device 205 determines that neither sensor data nor camera data has been received, the process 700 returns to block 705. Otherwise, at block 710, the computing device 205 converts the vehicle dynamic state data to corresponding road geometry data.

[0117] At block 715, the computing device 205 computes the rationalization value R . At block 720, the computing device 205 computes the rate of change of the rationalization value R by computing the derivative of the rationalization value R . At block 725, the computing device 205 determines whether and , where comprises a predetermined rationalization threshold, and comprises a predetermined rationalization rate of change threshold.

[0118] If or , the process 700 returns to block 705. Otherwise, the computing device 205 can determine one or more vehicle actions at block 730 based on the rationalization value R . For example, the computing device 205 can disallow lane centering and / or lane changing when and . Additionally or alternatively, the computing device 205 can flag a segment of map data corresponding to the road segment for re-mapping. The computing device 205 can also apply a weight to the sign to indicate a potential severity of the rationalization value R. It is to be understood that the vehicle actions can include additional vehicle actions, such as those listed above. The process 700 then transfers to block 705.

[0119] While at least one exemplary embodiment has been presented in the foregoing detailed description of the application, it should be appreciated that a vast number of modifications can be made to the exemplary embodiments without departing from the scope of the present disclosure. Additionally, it should be appreciated that the exemplary embodiment or embodiments are only examples, and are not intended to limit the scope, applicability or configuration of the disclosure in any way. Rather, the foregoing detailed description will provide those skilled in the art with a convenient road map for implementing an exemplary embodiment of the disclosure, it being understood that various changes can be made in the function and arrangement of elements

[0120] This detailed description is merely intended to be exemplary and is not intended to limit the scope, applicability or configuration of the application. Furthermore, no theory of operation is intended by these presented detailed description. As used herein, the term "module" refers to any hardware, software, firmware, electronic control component, processing logic, and / or processor in isolation or any combination thereof, including but not limited to: application specific integrated circuit (ASIC), an electronic circuit, a processor (shared, dedicated or group) and memory that execute one or more software or firmware programs, a combinational logic circuit, and / or other suitable components that provide the described functionality.

[0121] Embodiments of the present disclosure can be described herein in terms of functional and / or logical block components and various processing steps. It should be appreciated that such block components can be realized by any number of hardware, software, and / or firmware components configured to perform the specified functions. For example, an embodiment of the present disclosure can employ various integrated circuit components, e.g., memory elements, digital signal processing elements, logic elements, look-up tables, or the like, which can carry out a variety of functions under the control of one or more microprocessors or other control devices. Furthermore, those skilled in the art will appreciate that embodiments of the present disclosure can be practiced with one or more systems including a single processor system or a multiprocessor system containing multiple processors, a distributed computing system, networked computing system, or the like.

Claims

1. A system including a computing device, the computing device including a processor and a memory, the memory including instructions that configure the processor to: Vehicle dynamic data is converted into corresponding sensor data representing road geometry; A rationalization value is calculated based on the difference between the sensor data and map data representing the road geometry; the rationalization value is a value representing the difference between the sensor data and the map data; and The decision to modify the vehicle status is based on the aforementioned rationalization value. The processor is also configured to calculate the rationalization value according to the following equation: in, Including curvature estimation error, Including tilt angle estimation error, This indicates the error in slope angle estimation. Including map-based curvature estimation, Including map-based tilt angle estimation, Including map-based slope angle estimation, Including vehicle curvature estimation, This includes vehicle tilt angle estimation, and Including vehicle slope angle estimation, The processor is also configured to, according to To calculate the rationalization value, wherein, R is the rationalized value. It is the transpose of E, and Q is adjusted based on lane centering error tracking.

2. The system according to claim 1, wherein, The processor is also configured to calculate the rate of change of the rationalization value.

3. The system according to claim 2, wherein, The processor is also configured to determine whether to modify the vehicle state based on the rationalization value and the rate of change of the rationalization value.

4. The system according to claim 1, wherein, The processor is also configured to label map data corresponding to road segments and correct the map database in the segments, wherein the vehicle dynamic data corresponds to the road segments.

5. The system according to claim 1, wherein, The processor is also configured to change the vehicle state from an autonomous state to at least one of a semi-autonomous state or a non-autonomous state based on the rationalization value.

6. The system according to claim 1, wherein, The processor is also configured to at least one of the following: generate an alarm based on the rationalization value or reduce control privileges in uncertain environments.

7. The system according to claim 1, wherein, The processor is also configured to assess local lane probabilities based on rationalization values ​​for the identified lane lines and to correct lane-level positioning.

8. The system according to claim 1, wherein, The processor is also configured to adjust at least one of the availability or control of autonomous or semi-autonomous features based on an evaluation of rationalized values ​​within the prediction time domain.

9. The system according to claim 1, wherein, The processor is also configured to combine positioning and perception data in the rolling time domain with historical vehicle dynamics measurements to isolate sources of potential discrepancies and achieve robust lane following control.

10. A vehicle including a system, the system including a controller, the controller including a processor and a memory, the memory including instructions that configure the processor to: Vehicle dynamic data is converted into corresponding sensor data representing road geometry; A rationalization value is calculated based on the difference between the sensor data and map data representing the road geometry; the rationalization value is a value representing the difference between the sensor data and the map data; and The decision to modify the vehicle status is based on the aforementioned rationalization value. The processor is also configured to calculate the rationalization value according to the following equation: in, Including curvature estimation error, Including tilt angle estimation error, This indicates the error in slope angle estimation. Including map-based curvature estimation, Including map-based tilt angle estimation, Including map-based slope angle estimation, Including vehicle curvature estimation, This includes vehicle tilt angle estimation, and Including vehicle slope angle estimation, The processor is also configured to, according to To calculate the rationalization value, wherein, R is the rationalized value. It is the transpose of E, and Q is adjusted based on lane centering error tracking.

11. The vehicle according to claim 10, wherein, The processor is also configured to calculate the rate of change of the rationalization value.

12. The vehicle according to claim 11, wherein, The processor is also configured to determine whether to modify the vehicle state based on the rationalization value and the rate of change of the rationalization value.

13. The vehicle according to claim 10, wherein, The processor is also configured to label map data corresponding to road segments, wherein the vehicle dynamic data corresponds to the road segments.

14. The vehicle according to claim 10, wherein, The processor is also configured to change the vehicle state from an autonomous state to at least one of a semi-autonomous state or a non-autonomous state based on the rationalization value.

15. The vehicle according to claim 10, wherein, The processor is also configured to generate an alarm based on the rationalization value.

16. The vehicle of claim 10, wherein local lane probability is assessed based on rationalized values ​​of the identified lane lines, and lane-level positioning is corrected.

Images