REDUCTION OF JERKY MOVEMENT DURING A TRANSITION BETWEEN A LANE GUIDANCE AND A LANE KEEPING SYSTEM

Abstract

Vehicle system of an autonomous steering control (100), comprising: a lane keeping system (105) configured to initiate a lane keeping system (105) in response to a failure of a lane guidance system (95), wherein the lane keeping system (105) is configured to generate multiple corrective steering commands, wherein one or more of the steering commands are at least partially defined by a minimized cost function, which is mitigated by a time-dependent steering angle limit that increases over time.

Description

AREA OF THE PRESENT INVENTION

[0001] The present invention relates to autonomous steering systems and specifically to the reduction of jerky movement during a transfer of steering control from a lane guidance system to a lane keeping system or from a lane keeping system to a lane guidance system. BACKGROUND

[0002] Examples of autonomous steering systems include a lane guidance system, which aims to keep the vehicle centered within a lane, and a lane keeping system, which aims to restore the vehicle to a lane-keeping position in the event of a failure of the lane guidance system.

[0003] A system of this kind, in which a cost function is applied to control the vehicle's return to its lane, is known, for example, from DE 10 2013 206 815 A1. DE 10 2012 104 766 A1 discloses a method for correcting the position of a vehicle in a lane of a roadway, in which a cost function is minimized. DE 10 2005 039 895 A1 describes a system for lane departure warning and / or lane keeping in a motor vehicle, which has a sensor unit oriented in the direction of travel of the motor vehicle for detecting the lanes and a sensor unit oriented laterally to the motor vehicle for scanning the median barrier.

[0004] The object of the invention is to provide a system and a method for autonomously steering a vehicle in response to a failure of a lane guidance system in order to reduce jerky movement resulting from a transition of steering control from a lane guidance system to a lane keeping system or vice versa.

[0005] The problem is solved by a system having the features of claim 1 and by a method having the features of claim 7. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The components and their designs, features, operation, and advantages of the steering control system can best be understood with reference to the following detailed description, which includes: Fig. 1 a schematic perspective view of a vehicle equipped with a steering control system according to an example; Fig. 2. A block diagram of a steering control system according to an example; Fig. 3 a schematic top view of a lane environment as seen by a sensor array of a steering control system, according to an example is; Fig. 4 a flowchart of a procedure for reducing jerky movement during a transfer of steering control from a lane guidance to a lane keeping system according to an example; Fig. 4A is a schematic view of a vehicle driving around a curve, showing several steering angle calculations based on vehicle and road curvature dynamics at given time intervals, according to an example; Fig. 4B a schematic view of a vehicle dynamics model of the vehicle from Fig. 4A according to an example is; Fig. 5 a graph showing several limiting functions, where a steering angle limit is shown as a function of time, according to an example; Fig. 5A is a data representation showing a steering angle command as a function of time, where no steering angle limit has been introduced, according to an example; Fig. 5B is a data representation showing a steering angle command as a function of time, with a steering angle limit of 1.0° introduced, according to an example; Fig. 5C is a data representation showing a steering angle command as a function of time, with a steering angle limit of 1.5° introduced, according to an example;

[0007] It should be noted that for clarity, certain elements in the figures may not be shown to scale and that reference symbols may be repeated in the figures to indicate corresponding or analogous elements. Detailed description

[0008] The following description elaborates on numerous details to provide a complete understanding of the system for reducing jerky motion, but can be performed without these specific details. It should also be noted that components and procedures widely known to those skilled in the art have been omitted for the sake of simplicity.

[0009] The terms “autonomous”, “semi-autonomous”, “automated” and “automatic” all refer to a motor vehicle steering control system designed to control a driving path on a road or lane with reduced driver input.

[0010] According to one example, the steering control system comprises a primary steering system, implemented as a lane guidance system and designed to keep a vehicle centered within a lane or intended driving zone, and a safety system, implemented as a lane keeping system and designed to restore the vehicle to the intended driving zone in response to the vehicle leaving the driving zone.

[0011] In some examples, the steering control system is designed to allow a driver to regain control of the vehicle from either the lane guidance or lane keeping system, either permanently or temporarily. In some examples, once the lane keeping system has successfully restored the vehicle to the correct driving zone, steering control is automatically returned to the lane guidance system.

[0012] A lane guidance system failure can be caused, among other things, by an erroneous deactivation, a miscalculation of a correction steering angle, and a misinterpretation of a driving position.

[0013] In certain examples, the lane guidance or lane keeping system sends corrective steering angle commands to steering systems such as an electric power steering (EPS) system, an active front steering (AFS) system, or other steering systems.

[0014] Now, referring to the characters... Fig. 1 a schematic representation of a vehicle 10 equipped with a system 100 of autonomous steering control comprising both a lane guidance and lane keeping system, sensor arrangements 40 and vehicle dynamics sensors 30.

[0015] The road sensor array 40 is designed to acquire data that can be used or processed by the steering control system 100 to identify a vehicle position. The sensor array 40 is implemented as any one or a combination of sensors, such as, but not limited to, video cameras, light detection and ranging sensors (LIDAR sensors), laser detection and ranging sensors (LADAR sensors), radar, remote optical sensors, or other sensors that provide data relating to a vehicle's location in relation to road features such as lane markings, shoulders, median barriers, the edge of the road, and other objects or features. Without diminishing the scope of protection, this explanation describes the sensor array 40 implemented as a camera array.

[0016] In some examples, the vehicle includes 10 vehicle dynamics sensors, 30 such as steering angle sensors, steering torque sensors, wheel speed sensors, inertial measurement units designed to capture vehicle dynamics parameters, such as steering angle and torque, lateral and longitudinal speed, changes in lateral and longitudinal speed, yaw rate and wheel rotation.

[0017] Fig. Figure 2 is a block diagram of a system 100 of an autonomous steering control system according to an example, comprising one or more processors 110, a vehicle dynamics arrangement 30, a road sensor arrangement 40, a non-transient long-term memory 130, a memory 120 into which a lane guidance module 95, a lane keeping module 105, and a database 170 containing vehicle dynamics information, digital road data, or other relevant information are loaded. The steering control system 100 also includes a long-term memory 170, a user input device 140, and a user output device 150. The input device 140 can be implemented as a touchscreen, keyboard, microphone, pointer device, or other devices providing such functionality.The user output device 150 can be implemented as any device or combination of devices such as a display screen, loudspeaker or headphones, or other devices that provide such functionality.

[0018] The long-term storage system 100 can also include a receiver 180 of a global positioning system (GPS).

[0019] The memory 120 can be implemented as random access memory (RAM) or as other memory types that provide such functionality.

[0020] The non-transient long-term storage 130 can be implemented as a hard disk drive or other suitable non-volatile storage.

[0021] For the purposes of this explanation, the lane guidance functionality provided by the lane guidance module 95 makes the steering control system 100 a lane guidance system, and the lane keeping functionality provided by the lane keeping module 105 makes the steering control system 100 analogously a lane keeping system, so that, according to an example, the lane steering system 100 includes both types of steering control.

[0022] Fig. Figure 3 is a schematic top view of the vehicle 10 traveling in a defined lane environment, and the respective fields of view of a camera arrangement according to an example. As shown, the road 300 is delimited into a lane 305 by lane markings 310, which also define an outer boundary of a safety zone 320, which is further bounded by an inner boundary 325. The inner boundaries 325 and the lane centerline 464 are virtually introduced by the steering control system according to an example. The center lane zone 330 between the safety zones 320 typically spans between 65% and 95% of the lane width in a given example.

[0023] It should be noted that the position of vehicle 10 in relation to road features can also be determined in combination with GPS data and vehicle motion sensor data, as in other examples.

[0024] As shown, the forward-facing camera of vehicle 10, for example, has a forward field of view 340 of approximately 80 meters and a relatively narrow viewing angle of approximately 40 degrees. In one example, the forward-facing camera is coupled with the lane guidance system, whereas the side- and rear-facing cameras are coupled with the lane keeping system. The side- and rear-facing fields of view 360 and 365, respectively, have a relatively short range of approximately 10 to 20 meters and a viewing angle of 130 to 180 degrees in certain examples. In one example, the steering control system processes position data every 10 milliseconds and takes a corrective action. It should be noted that in other examples, the processing timing is set to different predefined or changing time intervals.

[0025] In operation, the lane guidance system acts as the primary steering control and is designed to keep the vehicle centered in a center lane zone 330 while the lane keeping system is in standby mode until the vehicle drifts 10 out of the safety zone 320, at which point steering control is transferred to the lane keeping system until the vehicle is restored to the center lane zone 330, as mentioned above.

[0026] Fig. Figure 4 is a flowchart showing the steps involved in generating corrective steering angle commands while minimizing a jerky movement or jolt of a vehicle, which in relation to Fig. 4A and Fig. 4B is described.

[0027] Generally, the lane keeping system is activated in step 400 in response to a failure of the lane guidance system. In step 410, a corrective steering angle is derived from a cost function, designed to minimize the lateral and heading angle deviations of the predicted path of the vehicle from the lane center. Consequently, the corrective steering angle becomes large as the vehicle moves 10 degrees away from the lane center. Often, this corrective steering angle is so large that it causes a jerky movement.

[0028] This cost function steering angle can cause excessive lateral jerkiness and discomfort with the driving experience. Therefore, in step 420, a vehicle dynamics steering angle is calculated from the road curvature, "p", and vehicle motion measurements and is used to mitigate the cost function steering angle.

[0029] Step 440 calculates the maximum rate at which the correction steering angle is applied.

[0030] In step 460, for a specific example, the maximum angle change rate calculated in step 440 is modified according to design factors.

[0031] In step 480, a steering angle command is generated according to a schedule.

[0032] In step 490, a steering control signal is returned to the guidance system.

[0033] Specifically, as mentioned above, in step 400 the lane keeping system is activated in response to a failure of the lane guidance system. Such a failure is considered to have occurred, for example, when the steering control system 100 identifies that a front corner of the vehicle 10 crosses the inner boundary of lane marking 310 within 0.5 seconds.

[0034] In step 410, a correction steering angle “δ” is derived by minimizing a cost function: J=∫0tlk{[yϕ]err⋅Q(x)⋅[yϕ]+δ⋅R(t)⋅δ}dt Where: t lk the duration that the lane keeping system is in operation, e.g. 6 s; y err The error of a lateral offset is, for example, (y desired - y predicted ); ϕ err the heading angle error is, for example, (ϕ desired - ϕ predicted ); and Q(t) and R(t) are empirically derived weighting factors.

[0035] In Fig. Figure 4A shows an exemplary correction cost function path 461 with a relatively sharp and uncomfortable correction steering angle 462, which needs to be mitigated.

[0036] In step 420, a vehicle dynamics steering angle is calculated from road curvature and vehicle motion measurements according to the following vehicle dynamics equation: [y˙φ˙v˙yr˙]=[0vx10000100−Cf+CrmvxbCr−aCfmvx−vx 00bCr−aCfIvxa2Cf+b2CrIvx]⋅[yφvyr]+[00CfmaCfI]⋅δ+[0vx00]⋅ρ As in Fig. 4B shown: Is “y” the lateral deviation from the track center? Is “φ” the heading angle deviation? Is “v x “The vehicle's longitudinal speed.” Is “v y “the vehicle's lateral speed.” Is “r” the vehicle yaw rate? Is “δ” the steering angle? Are “a” and “b” distances to the front and rear axles respectively, measured from the center of the vehicle? Is "m" the vehicle mass? Is “I” the vehicle inertia? Are “C f “, “C r “Factors of front and rear corner stiffness. Is “ρ” the road curvature?

[0037] The values ​​for y, φ, φ̇, v̇ yṙ can be measured by the vehicle camera array 40 or vehicle dynamics sensors 30, obtained from road data stored in a database 175, or from GPS data. Their values ​​are inserted into the vehicle dynamics equation above and solved for “δ” at fixed time intervals, which, according to examples, can be between 10 and 100 milliseconds.

[0038] As it is in Fig. As shown in Figure 4A, the vehicle dynamics path 463, defined by the vehicle dynamics equation above, does not lead vehicle 10 along the centerline 464 to the center of lane 305; instead, it follows the road curvature defined by the lane markings 310 and based on the parameters mentioned above. Therefore, its correction steering angles are smaller than those generated by the cost function. Accordingly, the vehicle dynamics path 463 serves as a basis for mitigating relatively sharp correction angles of the cost function path 461, as illustrated by the following example.

[0039] Each vehicle dynamics steering angle value “δ” derived from the vehicle dynamics equation is modified by increasing it by approximately 50% as a safety margin and then compared to the corresponding cost function steering angle. If the cost function steering angle exceeds this modified vehicle dynamics steering angle, the modified vehicle dynamics steering angle is selected, according to a specific example, as the correction steering angle value that is used to issue a steering command.

[0040] For example, if the vehicle dynamics steering angle is calculated as 2°, the value is increased by 50%, resulting in 3°. If the steering angle value calculated from the cost function optimization is 3.5°, only a 3° steering angle is used. It should be noted that the vehicle dynamics steering angle is increased by different values ​​in other examples.

[0041] It should be noted that different steering angles and their rates can be converted into corresponding torques using known relationships between steering angle and steering torque.

[0042] In step 440, the steering control system calculates a maximum rate of angle change by setting the following derivative matrix equal to an empirically derived limit of a lateral jerky movement of 0.13 g / 200 ms and solving for δ. All vehicle parameters are known constant vehicle parameters, where V x , V y as can be obtained from the vehicle sensors 30, for example. V x is assumed to be a constant for a short period of time in the time derivative. [v˙yw˙]=[−Cf+CrmvxbCr−aCfmvx−vxbCr−aCfIvxa2Cf+b2CrIvx]⋅[vyw]+[CfmaCfI]⋅δ

[0043] In step 460, the steering control system 100 modifies the maximum rate of change according to design parameters in certain examples.

[0044] The modification of the maximum rate of change is implemented in a specific example as follows: Modified maximum angular rate of change (MMRAC) (° / s) = (δ point) (kspeed) (kcurvature) Where: t: elapsed time in seconds since activation of the lane keeping controller k speed Weighting factor for speed depending on speed ranges k curvature : Weighting factor for curvature

[0045] For example: will k speed to a value of "1.1" for a speed range, V x , set between 0 - 48 km / h or 0 - 30 mph and will k speed to a value of "1" for a speed range, V x , set between 48 - 105 km / h or 30 - 65 mph and will k speed to a value of "0.9" for a speed range, V x , which exceeds 105 km / h or 65 mph.

[0046] Thus, the rate of steering angle change decreases as the speed, V x , increased, as it is in Fig. 5 is shown.

[0047] Similarly, k curvature according to areas of road curvature, p.

[0048] For example: will k curvature set to a value of "1.1" for curvatures, p, that are less than 0.001 (1 / m), will k curvature set to a value of "1.0" for curvatures, p, that are greater than 0.001 and less than 0.0005 (1 / m), will k curvature set to a value of “0.9” for curvatures, ρ, that are greater than 0.0005 (1 / m).

[0049] The weighting factors k curvature and k curvatureare derived empirically and, in certain examples, are stored in database 170 of the steering control system 100. The weighting factors k curvature and k curvature are design parameters that are selected according to the vehicle design concept. For example, k curvature in the above ranges, set to 1.2, 1.0 and 0.9 for a sports car or 1.05, 10.0 and 0.95 for a luxury sedan.

[0050] In step 480, the steering control system generates a steering angle command that includes the correction steering angle described above, derived from the cost function and vehicle dynamics, and an application rate defined by the modified maximum angle change rate (MMRAC) in a time sequence according to the following example: Applied steering limit = (MMRAC)(t−1)+1° Where: "t", for example, is the time in seconds between 1.0 and 6.0 seconds after the lane keeping controller is activated. It should be noted that different time increments and durations can also be used.

[0051] In step 490, the steering control is returned to the lane guidance system according to an example in response to a detection of travel in the center lane zone 330 by the steering control system.

[0052] Fig. Figure 5 shows, according to an example, various time-dependent steering limit functions “A”, “B”, and “C” of a steering angle limit (degrees) 500 as a function of time (seconds) 502 during a response period that begins one second after the lane keeping system assumes steering control, until the end of the response period when correct vehicle movement has been restored and the lane guidance system has resumed steering control. The six-second example is merely an illustration of a time period in which a vehicle's driving path is considered restored. As shown, the slope of each steering limit function represents a limit with respect to the steering angle change rate as described above.

[0053] As shown, steering limiting function “A” does not show any steering angle limits in contrast to 5C, with different steering angle responses shown according to the degree of steering.

[0054] According to an example, the steering limiting function “A” achieves a maximum steering limit of 2.5° at six seconds with a maximum angle change rate of 0.5, as represented by the slope.

[0055] According to an example, the steering limiting function “B” achieves a maximum steering limit of 1.5° at six seconds with a maximum angle change rate of 0.3, as represented by the slope.

[0056] According to an example, the steering limiting function “C” achieves a maximum steering limit of 1.0° at six seconds with a maximum angle change rate of 0.2, as represented by the slope.

[0057] Fig. 5A is a data representation of a steering angle command (degrees) 504 as a function of time (seconds) 506 and is shown within the first six seconds; the steering level command never exceeds 2.5°. Fig. 5B and Fig.According to examples, 5C are also data representations of the steering angle command (degrees) 504 as a function of time (seconds) 506 and show a maximum steering angle command within the first six seconds of 1.5° or 1.0°.

[0058] It should be noted that further examples may be combinations of different features that were explicitly described in the examples above.

Claims

[1] Vehicle system of an autonomous steering control (100), comprising: a lane keeping system (105) configured to initiate a lane keeping system (105) in response to a failure of a lane guidance system (95), wherein the lane keeping system (105) is configured to generate multiple corrective steering commands, wherein one or more of the steering commands are at least partially defined by a minimized cost function, which is mitigated by a time-dependent steering angle limit that increases over time. [2] System (100) according to claim 1, wherein the maximum rate of change of angle is defined at least partially by one or more vehicle dynamics parameters. [3] System (100) according to claim 2, wherein the maximum rate of change of angle is defined at least partially by the road curvature. [4] System (100) according to claim 1, wherein the lane keeping system (105) is further configured to transfer the steering control to the lane guidance system (95) in response to reaching a corrected driving path. [5] System (100) according to claim 1, wherein the lane keeping system (105) configured to initiate a lane keeping system (105) in response to a failure of a lane guidance system (95) based on data is implemented on a basis of data obtained from at least one sideways-facing sensor. [6] System (100) according to claim 1, wherein the lane keeping system (105) configured to initiate a lane keeping system (105) in response to a failure of a lane guidance system (95) is implemented on the basis of data obtained from at least one rearward-facing sensor. [7] Method for autonomously steering a vehicle, comprising: a lane keeping system (105) is used to generate multiple corrective lane keeping steering commands in response to a failure of a lane guidance system (95), wherein one or more of the steering commands are at least partially defined by a minimized cost function, which is mitigated by a time-dependent steering angle limit that increases over time. [8] Method according to claim 7, wherein one or more of the corrective steering commands is or are defined at least partially by a maximum angle change rate. [9] Method according to claim 7, wherein the lane keeping system (105) is configured to initiate a lane keeping system in response to a failure of a lane guidance system (95) and is implemented on the basis of data obtained from at least one sideways-facing sensor. [10] Method according to claim 7, wherein the lane keeping system (105) is configured to initiate a lane keeping system in response to a failure of a lane guidance system (95) and is implemented on the basis of data obtained from at least one rearward-facing sensor.

Images