Motor vehicle with turn signal based lane positioning

Abstract

A method increases the fidelity of a lane localization function on a motor vehicle by receiving input signals indicating a relative position of the vehicle with respect to a road. The input signals include GPS and geocoded map data, and an electronic turn signal indicating activation of a turn signal stalk. A sensor-specific lane probability distribution is calculated by the lane localization function using the input signals. The various distributions are fused by the localization function to generate a primary lane assignment. The primary lane assignment corresponds to the road lane with the highest probability among a set of possible lane assignments. In response to the primary lane assignment, an autonomous steering control action is performed on the motor vehicle using an advanced driver assistance system (ADAS). The motor vehicle has a controller that performs the method, for example, by executing instructions from a computer-readable medium.

Description

Technical Field

[0001] The present disclosure relates to methods and systems for increasing the fidelity of autonomous lane positioning functions on motor vehicles. Motor vehicles are typically equipped with lane positioning systems having logic and associated hardware. These systems work in conjunction to facilitate responsive, dynamic steering control actions. Modern lane positioning systems tend to rely heavily on digital geospatial map data and real-time GPS information when determining the current position of the motor vehicle. Autonomous steering control decisions can also be informed by data from evolving vision systems such as cameras, radar and / or lidar sensors, etc. The boundaries of paved roads are typically demarcated by detectable dashed or solid lane markings. On-board processing of various real-time lane positioning data enables an on-board controller to interpret the surrounding environment and make intelligent control decisions, such as when automatically adjusting the steering angle to maintain the motor vehicle's current lane position or to change lanes. Background Art

[0002] As is understood in the art, the lane localization function employed by advanced driver assistance systems (ADAS) in modern motor vehicles utilizes a suite of on-board sensors to collect different types of data that collectively describe the surrounding environment. The lane localization function also enables the driver of the motor vehicle to selectively offload certain driving tasks to an on-board controller. For example, in the case of automatic lane keeping / centering, the relevant data includes the GPS-informed / geocoded locations of the detected lane markings and other road surface boundaries described above. The controller continuously calculates the level of error between the motor vehicle's current position / heading and the trajectory of the detected lane markings and responds with autonomous steering control signals to generate a steering response appropriate to the situation. Lane change assist systems similarly use detected lane markings and other available sensor inputs to inform autonomous steering maneuvers when leaving the current lane, such as when passing another vehicle, while turning, or when merging onto a highway exit ramp. Summary of the Invention

[0003] The present disclosure relates to real-time operational control of a motor vehicle or other mobile platform using lane localization functions of the type generally described above. Automatic lane centering, lane change assist, and other advanced driver assistance systems (ADAS) perform control functions that rely on lane localization algorithms and associated hardware. However, such algorithms often rely heavily on lateral GPS accuracy when performing primary lane assignment, i.e., identifying the specific road lane in which the motor vehicle is currently traveling. GPS data has inherent errors, and therefore, incorrect primary lane assignments based solely or primarily on GPS data can result in suboptimal steering control performance.

[0004] Unlike traditional lane localization methods, the present method incorporates the on / off position of the turn signal stalk and its corresponding directional state ("turn signal state") as additional lane localization inputs. Correlation logic implemented using various electro-optical cameras, remote sensing devices such as radar and / or lidar systems, and digital mapping data is also used to mitigate instances of erroneous or inadvertent turn signal indications, thereby increasing the overall fidelity and robustness of the disclosed solution.

[0005] In a particular embodiment, a method for increasing the fidelity of a lane positioning function on a motor vehicle with a turn signal stalk includes receiving an input signal indicating the relative position of the motor vehicle with respect to a road, which occurs via an onboard controller. The received input signal includes GPS data and geocoded map data. In response to a set of enabling conditions, the method further includes receiving an electronic turn signal as an additional component of the input signal. The electronic turn signal indicates the current activation state of the turn signal stalk, i.e., an indication of an impending right or left turn. The lane positioning function then calculates a lane probability distribution for a plurality of specific sensors. This statistical calculation is performed using each input signal, including the electronic turn signal.

[0006] Thereafter, the method includes automatically fusing the various lane probability distributions, again using the lane localization function, to generate a primary lane assignment for the motor vehicle. The primary lane assignment corresponds to the road lane with the highest probability among a set of possible lane assignments. Through operation of the ADAS, the controller then executes an autonomous steering control action on the motor vehicle in response to the primary lane assignment, thereby changing the dynamic state of the motor vehicle.

[0007] In some configurations, the motor vehicle may include a camera configured to collect real-time video image data of a road. In such embodiments, the input signal includes real-time video image data.

[0008] The method may also include determining, by the controller, a lane marking type using real-time video image data of the road. In this particular case, the enabling condition may include a predetermined "traversable" lane marking type on the side of the lane that matches the direction of the turn signal, such as a broken or dotted line demarcating a traversable boundary of a given lane according to prevailing traffic regulations.

[0009] In other embodiments, the motor vehicle may include a remote sensing system configured to collect radar data and / or lidar data of the road, the input signal including the radar data and / or lidar data.

[0010] The method may include automatically fusing radar data and / or lidar data with video image data, or fusing radar data with lidar data, using object fusion logic of the controller.

[0011] In an exemplary non-limiting embodiment of the present method, the lane localization function comprises a Markov localization function. In this case, calculating the plurality of lane probability distributions comprises using the Markov localization function.

[0012] The enabling conditions contemplated herein may also include a lane change value representing the time that has elapsed since the last detected lane change in the direction of the turn signal and / or the time that has elapsed since the turn signal has been set in a particular direction.

[0013] In one aspect of the present invention, the method includes determining, by the controller, a driver attention score. In this particular case, the enabling condition may include the driver attention score exceeding a calibrated threshold attention score.

[0014] The enabling conditions may also include an anticipatory value indicating the presence and / or estimated width of an upcoming lane of the road, such as an anticipatory value reported to the controller from an external application, map data, a crowd-sourced application, or the like.

[0015] Performing an autonomous steering control action may include performing one or more of a lane centering control maneuver, a driver-requested automatic lane change maneuver, and / or a controller-initiated automatic lane change maneuver.

[0016] This document also describes a motor vehicle having a set of (one or more) wheels connected to a vehicle body, a turn signal stalk, an ADAS configured to control the dynamic state of the motor vehicle based on a primary lane assignment, and a controller. The controller is further configured to execute instructions for increasing the fidelity of an autonomous lane positioning function on the motor vehicle using a turn signal generated by activation of the turn signal stalk. This is achieved by executing the method outlined above.

[0017] Also disclosed herein is a computer-readable medium having recorded thereon instructions for selectively increasing the fidelity of an autonomous lane positioning function on a motor vehicle having a turn signal lever. The instructions are selectively executed by a processor of the motor vehicle in response to an enabling condition, thereby causing the processor to perform the method in various disclosed embodiments thereof.

[0018] The above-mentioned features and advantages of the present disclosure and other features and attendant advantages will become apparent from the following detailed description of illustrative examples and modes for implementing the present disclosure when taken in conjunction with the accompanying drawings and the appended claims. In addition, the present disclosure explicitly includes combinations and sub-combinations of the elements and features presented above and below. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 is a schematic diagram of a representative motor vehicle configured with turn signal-based lane positioning logic in accordance with the present disclosure.

[0020] Figure 2is a plan view of an exemplary lane change maneuver, wherein when the lane change maneuver is initiated, Figure 1 Activate the turn signal of the motor vehicle.

[0021] Figure 3 is a flow chart describing an exemplary embodiment of a method according to the present disclosure.

[0022] Figure 4 Can be Figure 1 A schematic logic flow diagram of exemplary control logic for use with a motor vehicle.

[0023] Figure 5 Is the description when Figure 1 A representative series of probability distributions for an example application of turn signal usage when performing a lane positioning function on a motor vehicle is shown. DETAILED DESCRIPTION

[0024] The present disclosure is susceptible of embodiments in many different forms. Representative examples of the present disclosure are shown in the accompanying drawings and described in detail herein as non-limiting examples of the disclosed principles. For this reason, elements and limitations described in the Abstract, Introduction, Summary, and Detailed Description sections but not explicitly recited in the claims should not be incorporated, individually or collectively, into the claims by implication, inference, or otherwise.

[0025] For the purposes of this specification, unless otherwise stated, the use of the singular includes the plural and vice versa, the terms "and" and "or" shall be conjunctions and disjunctions, "any" and "all" shall mean "any and all", and the words "include", "comprising", "containing", "having" and the like shall mean "including but not limited to". In addition, approximate words such as "approximately", "almost", "substantially", "generally", "approximately" and the like may be used herein in the sense of "at, close to or nearly at", or "within 0-5%", or "within an acceptable manufacturing tolerance", or any logical combination thereof.

[0026] Referring to the drawings, wherein like reference numerals refer to like features throughout the several views, Figure 1 , a motor vehicle 10 is shown having a controller (C) 50 programmed to execute the method 100. The method 100 (an embodiment of which is described in Figure 3 Execution of computer-readable instructions (shown in and described in detail below) enables the controller 50 to selectively increase the fidelity of the on-board lane positioning function 51 by optionally incorporating the driver-activated turn signal on / off and direction state ("turn signal state") into its lane probability calculation. Increased fidelity, in the present teachings, refers to improved accuracy of lane positioning prediction relative to the actual position of the motor vehicle relative to methods that do not incorporate turn signal information as described herein.

[0027] To simplify the illustration, selected components of the motor vehicle 10 are shown and described, while other components are omitted. Figure 1 In the representative embodiment shown, a motor vehicle 10 includes wheels 12 positioned relative to a vehicle body 14, with the wheels 12 in rolling contact with a road surface 16, with at least one wheel 12 being powered / driven. The actual number of wheels 12 may vary depending on the particular configuration of the motor vehicle 10. That is, as few as one driven wheel 12 is possible, such as in the case of a motorcycle, scooter, or electric bicycle / e-bike, while two or more driven wheels 12 are possible in other configurations, such as in gasoline-powered and / or electric passenger or commercial vehicles, crossovers, sport utility vehicles, trucks, and the like.

[0028] Within the scope of the present disclosure, motor vehicle 10 is equipped with a plurality of lane positioning input sensors and / or devices, hereinafter referred to, for simplicity, as lane positioning suite 18. Generally speaking, the components of lane positioning suite 18 provide input signals (arrow 30) to controller 50 indicating the relative position of motor vehicle 10 with respect to / on a road having road surface 16. Consequently, controller 50 relies on the capabilities of lane positioning suite 18 in real time when performing autonomous or semi-autonomous steering functions, such as, but not necessarily limited to, lane keeping assist with lane departure warning, lane change assist, lane centering, and the like.

[0029] The configuration of lane location suite 18 will vary depending on the specific equipment configuration of motor vehicle 10. However, typically, lane location suite 18 will include or have access to at least a geocoded map database 22 and a GPS receiver 24 that receives GPS signals 25 from a constellation of orbiting GPS satellites (not shown), as is well known in the art. Thus, input signals (arrow 30) typically include geocoded map data and GPS signals 25 from the corresponding geocoded map database 22 and GPS receiver 24, with the map data provided by these sources being displayed to the driver of motor vehicle 10 via a touchscreen (not shown) or other suitable display disposed in the center stack or other convenient location within motor vehicle 10, or on a similar touchscreen of a smartphone or other portable electronic device.

[0030] In addition, the lane positioning kit 18 may include a camera 20 and one or more remote sensing transceivers 26, such as radar sensors and / or lidar sensors. With respect to the camera 20, such a device may be securely attached to the vehicle body 14 at a suitable forward-facing location on the vehicle body 14, such as behind a rearview mirror 17 attached to the windshield 19, attached to an instrument panel (not shown), or at another suitable application location that provides good visibility of the roadway in front of the motor vehicle 10. The camera 20 is configured to collect real-time video image data of the roadway, and the input signal (arrow 30) includes the real-time video image data. The remote sensing transceiver 26 is, in turn, configured to transmit electromagnetic energy at a sensor-specific wavelength to a target as an interrogation signal, and to receive a reflected portion of the electromagnetic waveform from the target as a response signal. In Figure 1 In the figure, for simplicity of explanation, such inquiry and response signals are abbreviated as WW.

[0031] In addition to the camera 20, geocoded map database 22, GPS receiver 24 and remote sensing transceiver 26, Figure 1 The method 100 and the controller 50 also selectively rely on the turn signal state of a turn signal lever 28 which is an additional component of the lane positioning kit 18. As is understood in the art, such a turn signal lever 28 is located near the steering wheel 29 of the motor vehicle 10. When the driver (DD) decides to change lanes or turn, the driver (DD) indicates the direction of the upcoming maneuver by moving the turn signal lever 28 up or down. In the corresponding turn signal circuit (which is omitted for simplicity of illustration), such activation of the turn signal lever 28 enables current to energize the turn signal lamps 31 provided at each corner of the motor vehicle 10. The flashing action of the turn signal lamps 31, in Figure 1 denoted as TS in FIG, for a typical left turn, visually alerting other drivers or pedestrians near the motor vehicle 10 of the impending turn. Thus, to improve the fidelity of the lane positioning function 51, the turn signal state is selectively used herein as an additional available input to the controller 50, subject to certain entry conditions.

[0032] To perform method 100, Figure 1The controller 50 shown schematically in FIG1 is equipped with a dedicated amount of volatile and non-volatile memory (M) and one or more processors (P), such as microprocessors or central processing units, and other related hardware and software, such as digital clocks or timers, input / output circuits, buffer circuits, application-specific integrated circuits (ASICs), systems on a chip (SOCs), electronic circuits, and other necessary hardware required to provide the programmed functions. The method 100 can be implemented in response to an input signal (arrow 30) by a computer-executable program of instructions, such as a program module, commonly referred to as a software application or application program or a variation thereof, executed by the controller 50. Thereafter, the controller 50 can output the output signal (arrow CC) to the controller 50. O ) as part of method 100, for example, transmitted to a set of automatic driver assistance systems (ADAS) equipment 70.

[0033] In a non-limiting example, software may include routines, programs, objects, components, and data structures that perform specific tasks or implement specific data types. Software may form an interface that allows a computer to react based on an input source. Software may also collaborate with other code segments to initiate various tasks in response to data received in conjunction with a data source. Software may be stored on various memories (M), such as, but not limited to, CD-ROMs, magnetic disks, solid-state memories, and the like. Similarly, method 100 or portions thereof may be executed by devices other than controller 50 and / or embodied in firmware or dedicated hardware in a usable manner, such as when implemented by an ASIC, a programmable logic device, a field programmable logic device, discrete logic, and the like.

[0034] Still refer to Figure 1 As part of the present solution, the controller 50 is equipped to perform control actions downstream of the lane location probability determination disclosed below. To this end, by way of example and not limitation, the ADAS device 70 may include a lane keeping assist (LKA) system 52 and a lane departure warning (LDW) system 54, which may operate separately or in conjunction with each other in different situations. When the controller 50 performs dynamic control actions, in response to the method 100, electronic control and feedback signals (arrows 30-1 and arrows 30-2) may be exchanged between the controller 50, the LKA system 52, and the LDW system 54. Thus, the input signal (arrow 30) may include corresponding electronic control and feedback signals (arrows 30-1 and 30-2). The motor vehicle 10 may also be equipped with these or more (N) additional ADAS components, namely ADAS N 58 , with corresponding electronic control and feedback signals (arrow 30 -N) to enable the controller 50 to perform autonomous steering control actions, including lane centering control maneuvers, driver-requested automatic lane change maneuvers, and / or system / controller 50-initiated automatic lane change maneuvers.

[0035] Now refer to Figure 2 , describes the process of a motor vehicle 10 performing a typical lane change maneuver while traveling on a multi-lane driving surface 40. In this case, the multi-lane driving surface 40 is represented as a section of a highway or expressway having four parallel lanes, namely L1, L2, L3 and L4, which are separated from each other by inner lane markings 44, which in this case are dashed lines and are therefore traversable according to prevailing traffic laws. The outer boundary of the driving surface is demarcated by outer lane markings / boundary lines 144. Such boundary lines 144 are typically solid to delineate the beginning of the shoulder at the edge of the multi-lane driving surface 40, but may also be other non-traversable lines, such as a centerline, or on a slope or hill, or on an extended section of the multi-lane driving surface with limited forward visibility and / or an obstructed view.

[0036] In the illustrated scenario, the driver of motor vehicle 10 traveling in lane L2 may decide to merge into lane L1, for example in preparation for an upcoming ramp or when overtaking another vehicle. Figure 2 Indicated by arrow AA. Figure 1 The controller 50 can assist in this effort as part of an autonomous or semi-autonomous steering control maneuver. However, in order to do so, the controller 50 requires accurate real-time data indicating the current position of the motor vehicle 10 relative to the multi-lane driving surface 40 and its lanes L1, L2, L3, and L4. This determination is made automatically by the controller 50 based on statistical probability analysis, as is generally understood in the art and described in further detail below. According to the method 100, and in order to increase Figure 1 The fidelity of the onboard lane positioning function 51, the controller 50 selectively Figure 1 The activation state of the turn signal lever 28 is incorporated into its lane positioning calculation.

[0037] refer to Figure 3 , the exemplary embodiment of method 100 is described in terms of programmed steps or algorithmic logic blocks (referred to as "blocks" for simplicity). Method 100 begins at block B102 ("REC 30"), Figure 1 The controller 50 receives input signals (arrow 30), including electronic control and feedback signals ( Figure 1 30 -N) and thus indicate the direction of the motor vehicle 10 relative to the road (e.g. Figure 2 In one embodiment of the method 100, block B102 includes receiving GPS data and geocoded map data from the GPS receiver 24 and the map database 22, respectively, as part of an input signal (arrow 30). Upon receiving the input signal (arrow 30), the method 100 proceeds to block B104.

[0038] In block B104 ("TS-ENBLCond?"), Figure 1 The controller 50 can determine whether certain turn signal based fidelity enhancement enabling conditions are met. Block B104 is used to ensure Figure 1 The turn signal state of the turn signal stalk 28 is used situationally, ie, when appropriate and possible to provide information, but not when consideration of the turn signal information would reduce the accuracy of the resident lane probability determination of the lane location function 51 .

[0039] As part of the method 100, Figure 1 The controller 50 can receive and evaluate an electronic signal indicating the activation state of the turn signal lever 28. For example, such a signal can be a current or voltage signal representing the left or right turn state of the turn signal lever 28, or a measured or detected switch position indicating such state. By way of example and not limitation, an activation condition can include the controller 50 determining that no lane change has been recently detected in the current turn direction of the turn signal lever 28, or that no turn signal has been recently set in the direction opposite to the current turn signal direction. The controller 50 can also determine whether the turn signal lever 28 has been set in a particular direction for a calibratable amount of time, i.e., an amount of time greater than a minimum value but less than a maximum value, such as by comparing the time elapsed in a given left or right turn state to the calibratable amount of time.

[0040] In an alternative embodiment of the motor vehicle 10, wherein interior cameras and / or other hardware and associated software assess and assign a numerical score to the driver's attention level, e.g. Figure 1 The camera 20 on the windshield 19 is juxtaposed with a gaze tracking camera mounted to the dashboard or other interior location of the motor vehicle 10, or a sensor that tracks erratic steering or braking inputs that indicate decreased attention, so that the controller 50 knows the driver's current attention level as a reporting score. The enabling conditions may also include a threshold minimum driver attention score.

[0041] Other exemplary enabling conditions that may be used as part of block B104 include a particular detected lane marking type. More specifically, the controller 50 may, for example, use resident image processing software to evaluate whether a line detected on the side of lane L1, L2, L3, or L4 that matches the turn signal direction of the turn signal stalk 28 is a dashed line or another type of line that can be crossed, and / or whether a line marking located above the side of the lane that matches the turn signal direction is valid, i.e., not corresponding to a lane marking. Figure 2 The edge of the road at the boundary line 144. Information such as whether the adjacent lane in the direction of the turn signal has an expected lane width within a calibrated tolerance may also be used, e.g. Figure 2 Lane L1. You can also use the previous data, such as Figure 2On one side of the multi-lane driving surface 40, there is no information about adding a new lane within a predetermined distance in the direction matching the turn signal direction. When this turn signal fidelity enhancement enabling condition is not met, method 100 proceeds to block B105, and alternatively proceeds to block B106.

[0042] In block B105 (“DISBL CC TS ”), the controller 50 temporarily disables the use of the turn signal state by preventing the use of the turn signal state in the lane positioning function 51, wherein Figure 3 The "CC TS Lane positioning based control is simplified based on turn signal status information. Method 100 proceeds to block B106 to disable turn signal based fidelity enhancement.

[0043] Block B106 (“CALC Bel(L T =1)”) including the use of Figure 1 The lane positioning function 51 shown calculates a multi-lane probability distribution using control and feedback signals (arrows 30-1, 30-2 ... 30-N). When block B106 is reached directly from block B104, such signals would include an electronic turn signal triggered by actuation of the turn signal lever 28. Block B106 also includes automatically fusing the lane probability distributions, again via the lane positioning function 51, to generate a primary lane assignment for the motor vehicle 10. To this end, various fusion techniques can be used, including weighting the various channel probability distributions in a particular manner. As is understood in the art, the resulting primary lane assignment corresponds to the road lane with the highest probability among a set of possible lane assignments.

[0044] Regarding lane positioning and its associated statistical probability analysis, generally, the controller 50 performs real-time calculations on the probability that the motor vehicle 10 is in a particular lane at a given moment. Figure 2 , for example, for Figure 1 For each sensor or combination thereof in the lane localization suite 18, this calculation results in a lane probability for each of the representative lanes L1, L2, L3, and L4. Various localization algorithms can be used for this purpose, including but not limited to Markov localization, Monte Carlo localization, etc. In terms of application, available sensor updates can be applied to each sensor type in the same manner by multiplying the current probability distribution by the prior distribution / prior confidence and then normalizing to derive an updated probability distribution as the new confidence.

[0045] How to accurately calculate the given input probability distribution (P(s) in a given application T |l)) may differ depending on the sensor considered, i.e. Figure 1 Various sensors or components in the lane positioning kit 18. Figure 4 The logic 50L shows one possible implementation, as described below. For example, based on Figure 1 The distribution probabilities of the GPS input to the GPS receiver 24 shown may assign higher priority to detected lanes that are closer to the GPS prediction. Similarly, for lane marking input, lanes with the best / most matches to lane markings detected by the camera 20 will have higher probabilities, while lanes with fewer / poorer matches will have lower probabilities.

[0046] For the turn signal indications contemplated herein to increase the fidelity of lane localization function 51, lanes that meet the criteria of having an adjacent available lane in the direction of the turn signal, such as indicating a lane to the left of motor vehicle 10 when turn signal stalk 28 is used to signal a left turn, are given a higher probability than lanes that do not meet this criteria. The exact number of high or low probability results is adjustable to adjust the weighting of a given sensor or sensor input. Once a given sensor input distribution is calculated, it can be applied in the same manner as other available sensor inputs to produce a final confidence / probability distribution.

[0047] As Figure 3 As part of block B106 of FIG. 5 , when turn signal based fidelity enhancement is enabled, the controller 50 next determines the turn signal P(s) for each lane (l) based on the turn signal P(s) for each lane (l). T |l) Calculate lane probability, for example:

[0048] Confidence (L T =l)=α T P(s T |l)Confidence (L T-1 =l)

[0049]

[0050] Among them, S T is the direction of the turn signal at time T, L T is the set of available lanes at time T, e.g., Figure 2 Lanes L1, L2, L3 and L4.

[0051] Brief Reference Figure 5 A representative probability sequence 80 for lanes L1-L4 exists, with a probability distribution 82 for each lane L1, L2, L3, and L4, with a probability score ranging from 0 (very unlikely) to 1 (very likely). For sensor update calculations, such as for a turn signal, the controller 50 calculates:

[0052] Confidence (L T =l)=α T P(s T|l)Confidence (L T-1 =l)

[0053]

[0054] The probability distribution 82 from the sensor update is then applied to the previous probability distribution 84 at time T-1, namely the confidence (L T -1=1). Therefore, the controller 50 multiplies the probability distributions 82 and 84 together to calculate the updated probability distribution 86 for the current time T, namely the confidence (L T =1). Figure 5 The effect in a representative scenario is that at time T-1, lane L1 is returned as the most likely lane, with a normalized probability of approximately 0.5 in this illustrative example. However, one iteration of method 100 augmented with turn signal information results in an updated probability distribution 86, in which lane L2 is now identified as the most likely lane. As a result, the new primary lane assignment using the turn signal information will be lane L2, which controller 50 uses as the basis for executing multiple possible autonomous steering control actions.

[0055] Reference again Figure 3 , block B108 ("EXEC CA") in the illustrated exemplary embodiment of method 100 includes, in response to the primary lane assignment determined in block B106, Figure 1 The controller 50 performs autonomous steering control actions on the motor vehicle 10. Exemplary steering control actions taken as part of block B108 may include one or more of a lane centering control maneuver, a driver-requested automatic lane change maneuver, and / or a system-initiated automatic lane change maneuver, among other possible control actions.

[0056] Use as Figure 4 The representative control logic 50L schematically depicted in FIG can facilitate execution of the method 100 described above. The various components or sensors of the lane positioning kit 18 receive information from the camera 20, the geocoded map database 22 ("HD Map"), the GPS receiver 24 ("GPS"), the remote sensing transceiver 26 ("Radar / LiDAR"), and the turn signal lever 28 ("Turn Signal"). Figure 1 The above input signal 30 is represented here as CC 30 The respective output signal from each constituent sensor of lane localization suite 18 is transmitted to lane localization function 51 (“LLA”).

[0057] In a possible signal configuration, for example, the camera 20 may output camera data CC indicating the type of lane marking detected. 20B , camera data CC indicating the presence, size, shape, and extent of detected objects near the motor vehicle 1020A and indicating the detected dashed or solid lane markings (e.g. Figure 2 The camera data CC of the horizontal position of lines 44 and 144) 20C . Geocoded map database 22 output lane data CC 22A , such as the GPS position and width of the detected lane, lane marking data CC indicating the type of lane marking detected 22B and lane layout data CC indicating the detected lane layout 22C The GPS receiver 24 also outputs GPS position data CC 24 as the current GPS location of the motor vehicle 10. The remaining data may include remote sensing data CC from the remote sensing transceiver 26. 26 , which indicates the position and range of objects detected by radar and / or lidar, while the turn signal data CC from the turn signal lever 28 28 Indicates the direction of an upcoming right turn, left turn, or lane change.

[0058] Lane localization function 51, i.e., the coded or programmed implementation of the present method 100, is then used to generate a multi-lane probability distribution. Specifically, logic blocks 60, 62, 64, 66, and 68 may be used to independently generate corresponding sensor-specific probability distributions, which are then processed together via fusion block 69 ("fusion") to generate the aforementioned primary lane assignment. The ADAS device 70 is then informed of the primary lane assignment, and the controller 50 then uses the ADAS device 70 to determine the lane location based on the primary lane assignment. Figure 1 The corresponding control action is performed on the motor vehicle 10.

[0059] The logic block 60 may receive the fused object data CC from the object fusion block 27 27 and receives lane layout data CC from the geocoded map database 22 22C , then outputs a lane probability distribution indicating the relative lane of the fused object relative to the map-based lane layout (arrow P 60 ). With respect to the data fusion at block 27, it can generally be implemented in a manner similar to the implementation of the fusion block 69 described below, with various possibilities within the scope of the present disclosure including fusing video image data, radar data, and / or lidar data, i.e., any or all available sensor data depending on sensor availability and application requirements. Similarly, logic block 62 can use camera 20 and geocoded map database 22 to determine a lane probability distribution (arrow P) indicating the lane marking type. 62 Likewise, logic block 64 may generate a lane probability distribution (arrow P) indicative of the GPS position of motor vehicle 10. 64), while logic block 66 generates a lane probability distribution based on the detected lane changes notified only by camera 20 (arrow P 66 ). To illustrate the turn signal information, logic block 68 is based only on Figure 1 The state of the turn signal lever 28 generates a lane probability distribution (arrow P 68 ).

[0060] Sensor updates are applied in the same way as described above, i.e. by multiplying the probability distribution by the previous belief and then normalizing to produce the new belief. The input probability distribution is calculated differently based on the specific sensor used, e.g. Figure 4 P in 60 ,P 62 ,P 64 ,P 66 ,P 68 For example, using the turn signal stalk 28 as an input, a lane with a standard adjacent available lane in the direction of the turn signal may be given higher priority in logic block 68. In fusion block 69, the relative weight of each input may be assigned a calibrated value to produce a desired control result for the ADAS device 70.

[0061] Those skilled in the art will appreciate that by using the present method 100, the fidelity of the lane positioning function 51 can be improved by using a vehicle with a turn signal lever 28. Figure 1 By generating a lane probability distribution based on the turn signal informed by the state of the turn signal lever 28, Figure 1 The controller 50 is capable of generating a primary lane assignment for the motor vehicle 10 in a more precise manner. While the generation of the primary lane assignment by the controller 50 is a logical change of state with various benefits, including the ability to depict the relative position of the motor vehicle 10, potentially in real time, on a heads-up display (not shown) of the motor vehicle 10, dynamic control actions are also enabled. For example, the controller 50 can command an autonomous steering control action in response to the primary lane assignment based on a turn signal, or the controller 50 can perform various other dynamic control actions informed thereby. These and other benefits will be readily apparent to those skilled in the art in light of the foregoing disclosure.

[0062] The detailed description and accompanying drawings support and describe the present teachings, but the scope of the present teachings is limited only by the claims. Although some best modes and other embodiments for implementing the present teachings have been described in detail, there are various alternative designs and embodiments for practicing the present teachings as defined in the appended claims. In addition, the present disclosure expressly includes combinations and sub-combinations of the elements and features presented above and below.

Claims

1. A motor vehicle comprising: body; a set of wheels connected to the body of the vehicle; a turn signal lever configured to generate an electronic turn signal; an advanced driver assistance system (ADAS) configured to control the dynamic state of the motor vehicle based on the primary lane assignment; as well as A controller configured to execute instructions for performing a lane positioning function on a motor vehicle using an electronic turn signal, wherein the controller is configured to: receiving a set of input signals indicative of a relative position of a motor vehicle with respect to a road, the set of input signals comprising GPS data and geocoded map data; In response to an enabling condition, receiving an electronic turn signal as part of the set of input signals, the electronic turn signal indicating a current activation state of the turn signal stalk; Using the set of input signals, a multi-lane probability distribution is calculated through a lane localization function; automatically fusing lane probability distributions via a lane localization function to generate a primary lane assignment, wherein the primary lane assignment corresponds to a road lane with the highest probability among a set of possible lane assignments; performing an autonomous steering control action on the motor vehicle in response to a primary lane assignment using the ADAS; wherein the motor vehicle includes a camera configured to collect real-time video image data of a road, and wherein the set of input signals includes the real-time video image data; and The controller is configured to determine a lane marking type via the controller using real-time video image data of a road, wherein the enabling condition includes a predetermined traversable lane marking type on a side of a lane that matches a direction of the electronic turn signal.

2. The motor vehicle of claim 1 , further comprising a radar system and / or a lidar system configured to collect radar data and / or lidar data, respectively, of a road as part of the set of input signals, and wherein The controller is configured to automatically fuse radar data and / or lidar data together or with other sensor data using object fusion logic.

3. The motor vehicle according to claim 1, wherein: The enabling condition includes a lane change value representing the time that has passed since a lane change was detected in the direction of the electronic turn signal and / or the time that has passed since the electronic turn signal has been set in a direction.

4. The motor vehicle according to claim 1, wherein: The controller is configured to determine a driver attention score, wherein the enabling condition includes the driver attention score exceeding a calibrated threshold attention score.

5. The motor vehicle of claim 1, wherein: The enabling conditions include a look-ahead value indicative of the presence and / or estimated width of an upcoming lane of the road.

6. The motor vehicle of claim 1, wherein: The autonomous steering control action includes a lane centering control maneuver, a driver-requested automatic lane change maneuver, and / or a controller-initiated automatic lane change maneuver.

7. A computer readable medium having recorded thereon instructions for selectively increasing the fidelity of a lane positioning function on a motor vehicle having a turn signal lever, wherein: The instructions are selectively executable by a processor of a motor vehicle in response to a set of fidelity enhancement enabling conditions, thereby causing the processor to: receiving an electronic turn signal indicating a current activation state of a turn signal lever, wherein the electronic turn signal is part of a set of input signals; receiving GPS data and geocoded map data as part of the set of input signals and indicating a relative position of the motor vehicle with respect to a road; Calculating a plurality of lane probability distributions using the set of input signals through a lane localization function; automatically fusing the plurality of lane probability distributions via a lane localization function to generate a primary lane assignment for the motor vehicle, wherein the primary lane assignment corresponds to a road lane having a highest probability among a set of possible lane assignments; In response to the primary lane assignment, sending a control signal to an advanced driver assistance system (ADAS) of the motor vehicle to perform an autonomous steering control action of the motor vehicle, including one or more of a lane centering control maneuver, a driver-requested automatic lane change maneuver, and / or a system-initiated automatic lane change maneuver; and The enabling condition includes a predetermined traversable lane marking type detected on a side of a lane that matches an activation direction of the turn signal lever.

Images