Curve target identification method and system based on road information, and terminal equipment

Abstract

The invention discloses a curve target identification method and system based on road information and terminal equipment, and the method comprises the steps: carrying out the radius prediction through an initial curve radius according to the obtained current vehicle speed information, and obtaining a current curve radius; performing information fusion on the current curve radius and the current vehicle speed information to obtain a longitudinal distance threshold value; and based on a preset judgment condition and a longitudinal distance threshold value, curve target identification is carried out through the obtained vehicle driving information, and a vehicle following target in front of the vehicle is identified. By predicting the radius of the curve where the current vehicle is located, the vehicle following target of the side lane is accurately recognized, and the driving safety and the driving experience are improved.

Description

Technical Field

[0001] The present application belongs to the field of intelligent driving, and specifically relates to a method, system and terminal device for identifying curve targets based on road information. Background Art

[0002] Adaptive cruise control (ACC) is a driving assistance system that adaptively adjusts the vehicle's speed in multiple scenarios based on the speed and following distance set by the driver. The system builds a road environment based on perception modules, algorithms, data training and other information. When the target ahead is not identified, the system issues appropriate acceleration and deceleration based on the regulation and control algorithm to maintain the vehicle's speed at the speed set by the user. When the target is identified, that is, there is a vehicle ahead of the vehicle, the system can control the vehicle to follow the vehicle ahead and maintain a certain safety distance based on the set time interval level.

[0003] However, in some special scenarios, such as curved roads, due to the limitations of the monocular camera's perception performance, there is a certain probability that targets in the side lanes will be mis-screened, causing the intelligent driving control end to erroneously issue deceleration commands, causing the vehicle to decelerate unnecessarily, thus affecting the comfort and experience of the driver and passengers. Summary of the invention

[0004] This application proposes a curve target recognition method, system and terminal device based on road information, which accurately identifies the following vehicle targets in the side lane by predicting the curve radius of the current vehicle, thereby improving driving safety and driving experience.

[0005] A first aspect of the present application provides a method for identifying a curved target based on road information, the method comprising:

[0006] According to the current vehicle speed information obtained, the radius is predicted through the initial curve radius to obtain the current curve radius;

[0007] Fusing the current curve radius with the current vehicle speed information to obtain a longitudinal distance threshold;

[0008] Based on the preset judgment conditions and longitudinal distance threshold, the vehicle driving information is obtained to perform curve target recognition and identify the vehicle's front following target.

[0009] The above scheme first predicts the curve radius of the current vehicle based on the acquired speed information to obtain a more accurate current curve radius; then the current curve radius is integrated with the speed information to obtain the threshold value of the longitudinal distance between the vehicle and the vehicle in front relative to the current curve radius, providing data support for the subsequent determination of the following target that is closer to the vehicle; then according to the set judgment conditions, the vehicle driving information is used to identify the following target that is closest to the vehicle, has a safe lateral distance, and has no collision risk, so that the vehicle does not need to adjust the speed urgently during driving due to mistakenly identifying vehicles in other lanes as following targets, resulting in reduced passenger comfort. In addition, because the most suitable following target can be accurately identified in the curve scenario, driving safety can also be improved.

[0010] In a possible implementation method of the first aspect, according to the acquired current vehicle speed information, a radius prediction is performed using an initial curve radius to obtain a current curve radius, specifically:

[0011] According to a preset vehicle speed-radius increment table, the current vehicle speed information is interpolated to obtain a corresponding curve radius increment;

[0012] Outputting a curve reference value by conditionally judging the initial curve radius, the curve radius output in the previous cycle, and the curve radius increment;

[0013] The curve reference value and the curve radius output in the first threshold number of cycles before are calculated by sliding average to obtain the current curve radius of the current cycle.

[0014] The above scheme first interpolates the speed-radius increment table to find the curve radius increment corresponding to different speeds under standard conditions. This curve radius increment can prevent the curve radius collected by the sensor from fluctuating too much in a short period of time, thereby reducing subsequent calculation errors. Then, the initial curve radius collected currently and the curve radius predicted in the previous cycle are conditionally judged to output the curve reference value that can best reduce the calculation error. Then, the curve reference value and the curve radius predicted in the previous cycles are averaged to obtain the most accurate current curve radius, thereby reducing the risk of easily identifying vehicles in other lanes when identifying the following vehicle target.

[0015] In a possible implementation method of the first aspect, a curve reference value is output by conditionally judging an initial curve radius, a curve radius output in a previous cycle, and the curve radius increment, specifically:

[0016] If the initial curve radius is less than or equal to the sum of the curve radius output in the previous cycle and the curve radius increment, and the initial curve radius is greater than or equal to the difference between the curve radius output in the previous cycle and the curve radius increment, the curve radius output in the previous cycle is used as the curve reference value;

[0017] Otherwise, the initial curve radius is used as the curve reference value.

[0018] The above scheme uses the curve radius increment to determine whether the curve radius data collected by the sensor in a short period of time fluctuates too much. If the fluctuation is too large, the sensing data of the sensor is unreliable, because the lane width cannot change greatly in a short period of time, so it is more stable to use the curve radius output in the previous cycle as the curve reference value; if the fluctuation is not large, the initial curve radius currently collected can be used as the curve reference value.

[0019] In a possible implementation method of the first aspect, the current curve radius and the current vehicle speed information are fused to obtain a longitudinal distance threshold, specifically:

[0020] According to a preset vehicle speed-scaling factor table, the current vehicle speed information is interpolated to obtain a distance scaling factor;

[0021] According to a preset radius-longitudinal distance table, the current curve radius is interpolated to obtain a standard value of the longitudinal distance;

[0022] The distance scaling factor and the longitudinal distance standard value are calculated to obtain the longitudinal distance threshold.

[0023] In a possible implementation method of the first aspect, based on a preset judgment condition and a longitudinal distance threshold, a curved target is identified by using the acquired vehicle driving information to identify a vehicle following the vehicle in front of the vehicle, specifically:

[0024] Collecting the vehicle driving information at the current moment, and simulating the vehicle's self-driving trajectory and the target driving trajectory of the vehicle ahead according to the vehicle driving information and a preset trajectory simulation formula;

[0025] Based on the judgment conditions, several judgment criteria are set; wherein the judgment criteria include a confidence standard, a horizontal judgment standard and a no-cross risk standard;

[0026] determining whether the vehicle driving information, the self-driving trajectory, and the target driving trajectory meet the determination criteria, and when all the determination criteria are met, comparing the target longitudinal distance with the longitudinal distance threshold;

[0027] If the target longitudinal distance is greater than or equal to the longitudinal distance threshold, it is determined that the front vehicle is not a front following target;

[0028] If the target longitudinal distance is less than the longitudinal distance threshold, it is determined that the front vehicle is a front following target, and the vehicle speed is adjusted according to the target longitudinal distance and the longitudinal distance threshold.

[0029] In the above scheme, the lateral distance and longitudinal distance between the vehicle and the target vehicle are first collected to determine whether there is a collision risk between the vehicle and the target vehicle and whether the target vehicle is in the side lane according to the preset judgment conditions, so as to avoid identifying vehicles in other lanes; finally, when the judgment conditions are met, the target longitudinal distance is compared with the longitudinal distance threshold. If the target longitudinal distance exceeds the longitudinal distance threshold, it means that the vehicle is far away from the target vehicle. At this time, it is not suitable to use the target vehicle as a following target, because emergency acceleration to shorten the distance to the target vehicle will reduce driving comfort.

[0030] In a possible implementation method of the first aspect, determining whether the vehicle driving information, the self-driving trajectory, and the target driving trajectory meet the determination criteria is specifically as follows:

[0031] Acquiring a target longitudinal distance between the vehicle and a target vehicle from the vehicle travel information;

[0032] determining whether the confidence criterion is met according to the target longitudinal distance, the speed signal perception time and the longitudinal relative speed of the vehicle;

[0033] determining whether the lateral collision judgment standard is satisfied according to the predicted lateral collision time;

[0034] According to the self-driving trajectory and the target driving trajectory, it is determined whether the no-crossing risk standard is met.

[0035] The above scheme first determines whether the distance and relative speed between the vehicle and the vehicle in front will affect driving comfort based on the target longitudinal distance, etc.; then the predicted lateral collision time and the driving trajectory of the two vehicles are statistically analyzed to determine whether the vehicles are prone to collision due to being too close, thereby improving driving safety.

[0036] In a possible implementation method of the first aspect, judging whether the no-crossing risk criterion is met according to the self-driving trajectory and the target driving trajectory is specifically as follows:

[0037] According to a preset radius-lateral distance table, interpolation processing is performed on the current curve radius to obtain a lateral distance reference value;

[0038] Generate a lateral distance range according to the lateral distance reference value, and set the self-driving trajectory equal to the target driving trajectory for solving;

[0039] If within the lateral distance range, a solution exists where the self-driving trajectory is equal to the target driving trajectory, then the no-crossing risk criterion is satisfied;

[0040] Otherwise, the no cross-risk criterion is not met.

[0041] In a possible implementation method of the first aspect, the self-driving trajectory and the target driving trajectory are specifically:

[0042] The self-driving trajectory is specifically expressed as:

[0043] Y=C0+C1*X+C2*X 2 +C3*X 3 ;

[0044] Where Y is the self-driving trajectory, C0, C1, C2, C3 are the coefficients of the lane centerline, and X is the longitudinal coordinate of the vehicle;

[0045] The target driving trajectory is specifically expressed as:

[0046] W=A0+A1*X+A2*X 2 +A3*X 3 ;

[0047] Wherein, W is the target driving trajectory, A0, A1, A2, A3 are the coefficients of the target driving trajectory, and X is the longitudinal coordinate of the target vehicle.

[0048] A second aspect of the present application provides a curved target recognition system based on road information, the system comprising: a radius prediction module, a data fusion module and a vehicle following target recognition module;

[0049] The radius prediction module is used to predict the radius based on the initial curve radius according to the current vehicle speed information obtained, and obtain the current curve radius;

[0050] The data fusion module is used to fuse the current curve radius with the current vehicle speed information to obtain a longitudinal distance threshold;

[0051] The following vehicle target recognition module is used to perform curve target recognition based on preset judgment conditions and longitudinal distance thresholds through the acquired vehicle driving information, and identify the following vehicle target in front of the vehicle.

[0052] The third aspect of the present application provides a terminal device, which includes: a terminal device including a processor and a memory, the memory storing a computer program, and the processor implementing the steps of a curve target recognition method based on road information as described in any one of the embodiments of the present application when executing the computer program. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the implementation manner will be briefly introduced below. Obviously, the drawings described below are only some implementation manners of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0054] Figure 1 It is a specific flow chart of a method for identifying a curved target based on road information provided by a certain embodiment of the present application;

[0055] Figure 2 is a curve target misidentification diagram of a curve target recognition method based on road information provided by a certain embodiment of the present application;

[0056] Figure 3 A curve target recognition strategy of a curve target recognition method based on road information provided in a certain embodiment of the present application;

[0057] Figure 4 is a correct identification diagram of a curved target in a method for identifying a curved target based on road information provided by a certain embodiment of the present application;

[0058] Figure 5 is a structural diagram of a curved target recognition system based on road information provided by a certain embodiment of the present application;

[0059] Figure 6 A structural diagram of a terminal device is provided in accordance with a certain embodiment of the present application. DETAILED DESCRIPTION

[0060] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0061] It should be understood that the step numbers used in this article are only for the convenience of description and are not intended to limit the order in which the steps are executed.

[0062] First embodiment

[0063] Vehicles equipped with driving assistance cameras can realize ACC adaptive cruise control, maintain a steady speed and follow the vehicle in a stable manner, and adapt to the speed of the vehicle in front to realize stop-start-follow-go functions in all speed ranges and on all road types. However, when the vehicle is on a straight road and the target vehicles in front are all in a curved environment, or when the vehicle is in a curve and the target vehicle is on a straight road, it is difficult for the vehicle to directly confirm whether the target vehicle is the correct target to follow through pure visual judgment using the driving assistance camera. When the camera mistakenly selects a target and causes ACC to trigger a false deceleration, it will affect the driver's normal driving and reduce the driving experience.

[0064] like Figure 1 As shown, Figure 1 A specific flow chart of a method for identifying a curved target based on road information is provided for a certain embodiment of the present application. The method for identifying a curved target based on road information of this embodiment includes steps S1 to S3, which are described in detail as follows:

[0065] Step S1, based on the current vehicle speed information obtained, the radius is predicted through the initial curve radius to obtain the current curve radius.

[0066] Currently, vehicles equipped with ACC systems can accurately confirm the following target vehicle in front through auxiliary cameras. However, when the vehicle is on a straight road and the target vehicles in front are in a curved environment, or when the vehicle is in a curve and the target vehicle is on a straight road, the vehicle's selection of the vehicle in front mainly relies on the camera to identify the lane line, the vehicle's yaw angle driving trajectory prediction and the driving trajectory prediction of the target vehicle in front, and then judge the target based on the time distance to determine whether to use the target as a following target.

[0067] If a purely visual solution is adopted, that is, only a camera is used, when there is a blind spot on the curve or the light is too dark for the camera to recognize the lane line, the camera is likely to misselect the target, which in turn causes ACC to trigger a false deceleration, affecting the driver's normal driving. Figure 2 The figure shows a case of misidentification of a target on a curve. In the figure, the black one is the ego vehicle, and the red one is the misidentified target vehicle. The misidentified target vehicle is the closest vehicle to the ego vehicle in the virtual lane line. At this time, the misidentified target vehicle is in the curve and in the side lane. When the ACC system of the ego vehicle triggers the wrong deceleration to follow the misidentified target vehicle, the ego vehicle is likely to mistakenly enter the side lane and cause driving safety problems. Therefore, a more accurate vehicle following target recognition method is needed to accurately screen out suitable vehicle following targets in the curve driving scenario.

[0068] In the embodiment of the present application, the curve-related signals collected by the sensor are processed to limit the target recognition distance within a certain radius, thereby improving the misidentification phenomenon of target screening and finding a suitable vehicle-following target more accurately.

[0069] First, the vehicle speed signal is collected by the sensor to obtain the current speed information of the vehicle and interpolation is performed according to the preset speed-radius increment table to obtain the curve radius increment at different speeds. Among them, the table interpolation used in the embodiment of the present application is a linear interpolation method, which performs linear interpolation in each speed interval.

[0070] For example, the following Table 1 is a vehicle speed-radius increment table used in the embodiment of the present application, wherein v is the vehicle speed, dy is the curve radius increment, and when the current vehicle speed information is 26 m / s, it can be determined that the corresponding curve radius increment is 60 m.

[0071] Table 1 Speed-radius increment table

[0072] v 15m / s 20m / s 26m / s 35m / s 45m / s dy 12m 15m 60m 75m 90m

[0073] Specifically, the curve radius increment is mainly used to prevent the curve radius collected by the sensor at the front of the vehicle from fluctuating too much in a short period of time, thereby causing misidentification of the target vehicle. Because the lane radius will not change too much in a short period of time, and environmental factors such as light intensity and whether the lane line is clear will cause the sensor to collect data with large errors, the embodiment of the present application introduces a curve radius increment to adjust the length of the curve radius, so that the obtained curve radius value is stable, and the increase of errors such as value jumps or distortion caused by poor sensor recognition accuracy is prevented.

[0074] After the corresponding curve radius increment is obtained, a conditional judgment is performed based on the initial curve radius currently collected by the sensor, the curve radius output in the previous cycle, and the curve radius increment, and a curve reference value is output, specifically:

[0075] If the initial curve radius>=the curve radius output in the previous cycle-the curve radius increment, and the initial curve radius<=the curve radius output in the previous cycle+the curve radius increment, the curve radius output in the previous cycle is used as the curve reference value;

[0076] Otherwise, the initial curve radius is used as the curve reference value.

[0077] The above process is to make a conditional judgment between the initial curve radius currently collected and the curve radius predicted in the previous cycle, and output a relatively stable curve reference value that can best reduce the calculation error, thereby avoiding the use of data with excessive fluctuations for subsequent calculations and increasing data errors.

[0078] Then the obtained curve reference value is filtered, and the curve radius and the curve reference value output in the first n cycles are taken for sliding average calculation, and the average value is taken as the current curve radius in the current cycle. The specific calculation formula is:

[0079] Rd_out=(Rd_fill+Rd_out(0)+Rd_out(1)+...+Rd_out(n)) / n;

[0080] Where Rd_out is the current curve radius, Rd_fill is the curve reference value, and n is the previous n cycles.

[0081] Optionally, in some embodiments, n is taken as 9, that is, the curve radius and curve reference value output in the first nine cycles are taken for sliding average calculation.

[0082] Step S2, fusing the current curve radius with the current vehicle speed information to obtain a longitudinal distance threshold.

[0083] In the embodiment of the present application, the distance scaling factor is first determined by looking up and interpolating the preset vehicle speed-scaling factor table according to the current vehicle speed information. The distance scaling factor is used to scale the longitudinal distance threshold according to the vehicle speed to ensure driving safety.

[0084] Table 2 below is a vehicle speed-scaling factor table used in the embodiment of the present application:

[0085] Table 2 Vehicle speed-scaling factor table

[0086] v 5.7m / s 8.3m / s 16.7m / s 22.2m / s 33.3m / s α 0.8 0.85 0.88 0.9 0.98

[0087] In the table, v is the vehicle speed and α is the distance scaling factor; when the vehicle speed is 16.7 m / s, the corresponding distance scaling factor is 0.88, that is, the longitudinal distance threshold is scaled at a ratio of 0.88.

[0088] Then, the obtained current curve radius is linearly interpolated through a preset radius-longitudinal distance table to obtain a standard value of the longitudinal distance. The radius-longitudinal distance table used in the embodiment of the present application is shown in Table 3 below:

[0089] Table 3 Radius-longitudinal distance table

[0090] Rd_out 100m 400m 800m 1500m 1800m 2000m d_long 50m 60m 75m 90m 108m 115m

[0091] In the table, Rd_out is the current curve radius, and d_long is the basic identification distance.

[0092] Finally, the basic recognition distance is multiplied by the distance scaling factor to obtain the corresponding longitudinal distance threshold. The longitudinal distance threshold is the conditional threshold set for the vehicle to determine the front target as a following target in the longitudinal distance when the radius of the curve is less than 1800m. If the front target vehicle exceeds the longitudinal distance threshold, it means that the distance from the vehicle is too far and it is not determined as a following target.

[0093] Step S3, based on the preset judgment condition and the longitudinal distance threshold, the curved target is identified through the acquired vehicle driving information to identify the vehicle following the vehicle in front.

[0094] In an embodiment of the present application, three judgment criteria are set according to preset judgment conditions, namely, a confidence standard, a lateral judgment standard and a no-crossing risk standard, which are used to judge whether the longitudinal distance between the vehicle and the target vehicle ahead is too close or too far, whether a lateral collision may occur between the vehicle and the target vehicle ahead, and whether there is a crossing risk between the two vehicles.

[0095] The vehicle driving information at the current moment is collected, including the target longitudinal distance, target lateral distance, longitudinal relative speed, etc. between the vehicle and the target vehicle in front.

[0096] Determine whether the vehicle's driving state meets the confidence standard: When the vehicle's sensor continuously senses the target vehicle in front for more than 0.15 seconds, the distance and relative speed between the vehicle and the target vehicle in front are determined. When the target longitudinal distance is less than 120 meters or the longitudinal relative speed is less than -5 meters / second, the vehicle's driving state meets the confidence standard.

[0097] Determine whether the vehicle's driving state meets the lateral judgment standard: In a curve scenario, if the predicted lateral collision time between the vehicle and the target vehicle in front is greater than 0.5s, it is considered that the lateral risk with the target vehicle is small and the lateral judgment standard is met. The lateral collision time is predicted based on the target lateral distance.

[0098] Determine whether the driving state of the vehicle meets the no-crossing risk standard: According to the vehicle driving information and the preset trajectory simulation formula, simulate the vehicle's self-driving trajectory and the target driving trajectory of the vehicle in front. Then, based on the current curve radius, interpolate the preset radius-lateral distance table to obtain a lateral distance reference value, as shown in Table 4 below:

[0099] Table 4 Radius-lateral distance table

[0100] Rd_out 100m 400m 800m 1500m 1800m 2000m X_rel 30m 35m 40m 50m 55m 60m

[0101] In the table, X_rel is the reference value of the lateral distance.

[0102] A corresponding lateral distance range is generated according to the lateral distance reference value, and then the self-driving trajectory is set equal to the target driving trajectory for solving. If there is a solution within the lateral distance range where the self-driving trajectory is equal to the target driving trajectory, then the no-crossing risk criterion is satisfied, indicating that there is no trajectory crossing risk between the self-vehicle and the target vehicle ahead.

[0103] Exemplarily, the lateral distance range of the embodiment of the present application is 0 to X_rel.

[0104] Among them, the self-driving trajectory is specifically expressed as:

[0105] Y=C0+C1*X+C2*X 2 +C3*X 3 ;

[0106] Where, Y is the self-driving trajectory, C0, C1, C2, and C3 are the coefficients of the lane centerline, which are generally provided by the road information fusion module of the vehicle; X is the longitudinal coordinate of the vehicle;

[0107] The target driving trajectory is specifically expressed as:

[0108] W=A0+A1*X+A2*X 2 +A3*X 3 ;

[0109] Wherein, W is the target driving trajectory, A0, A1, A2, A3 are coefficients of the target driving trajectory, which are generally provided by the target trajectory processing module of the vehicle; X is the longitudinal coordinate of the target vehicle.

[0110] When the above three criteria are all met, it means that the longitudinal distance and lateral distance between the target vehicle in front and this vehicle are safe and reasonable, and there is no risk of intersection in the driving trajectories between the two vehicles, there is no risk of collision and it is not a vehicle in the side lane. At this time, the recognition and judgment mechanism can be activated to determine whether the target vehicle in front can be used as a following target through the target longitudinal distance and the longitudinal distance threshold.

[0111] Specifically, if the target longitudinal distance is greater than or equal to the longitudinal distance threshold, it is determined that the vehicle in front is not the front following target; if the target longitudinal distance is less than the longitudinal distance threshold, it is determined that the vehicle in front is the front following target, and the driving speed of the vehicle is adjusted according to the target longitudinal distance and the longitudinal distance threshold.

[0112] Figure 3 A curve target recognition strategy of an embodiment of the present application is provided, and the entire strategy is divided into three modules: signal filtering, information fusion, and target data. Signal filtering is to filter the vehicle speed information and curve radius collected by the vehicle sensor, select reasonable data for data fusion, and finally output the identified following vehicle target in combination with the simulated driving trajectory. Compared with the prior art, it not only relies on the visual solution provided by the camera, but also strictly controls the predicted curve radius range through the collected data to avoid the identified following vehicle target actually being in other lanes.

[0113] Under the curve target recognition strategy provided by the embodiment of the present application, the vehicle will Figure 4As shown in the demonstration, vehicles in the side lanes will not be identified as following targets.

[0114] Implementing the embodiments of the present application has the following beneficial effects:

[0115] The embodiment of the present application first predicts the curve radius of the current vehicle based on the acquired vehicle speed information to obtain a relatively accurate current curve radius; then the current curve radius is integrated with the vehicle speed information to obtain the threshold value of the longitudinal distance between the vehicle and the vehicle in front relative to the current curve radius, providing data support for the subsequent determination of the following vehicle target that is closer to the vehicle; then according to the set judgment conditions, the following vehicle target that is closest to the vehicle, has a safe lateral distance, and has no collision risk is identified through the acquired vehicle driving information, so that the vehicle does not need to adjust the speed urgently during driving due to mistakenly identifying vehicles in other lanes as following vehicle targets, resulting in reduced passenger comfort. Moreover, because the most suitable following vehicle target can be accurately identified in the curve scenario, driving safety can also be improved.

[0116] Second embodiment

[0117] Furthermore, in order to execute the curved target recognition system based on road information corresponding to the above method embodiment to achieve corresponding functions and technical effects, Figure 5 A structural diagram of a curved target recognition system based on road information is provided. For ease of description, only the parts related to this embodiment are shown. The curved target recognition system based on road information provided by the embodiment of the present application includes:

[0118] The radius prediction module 201 is used to perform radius prediction based on the initial curve radius according to the acquired current vehicle speed information to obtain the current curve radius.

[0119] In an embodiment of the present application, the current vehicle speed information is interpolated according to a preset vehicle speed-radius increment table to obtain a corresponding curve radius increment; a curve reference value is output by conditionally judging the initial curve radius, the curve radius output in the previous cycle, and the curve radius increment; a sliding average calculation is performed on the curve reference value and the curve radius output in the first threshold number of cycles to obtain the current curve radius of the current cycle.

[0120] The data fusion module 202 is used to fuse the current curve radius with the current vehicle speed information to obtain a longitudinal distance threshold.

[0121] In the embodiment of the present application, the distance scaling factor is first determined by looking up and interpolating the preset vehicle speed-scaling factor table according to the current vehicle speed information. The distance scaling factor is used to scale the longitudinal distance threshold according to the vehicle speed to ensure driving safety.

[0122] The following table 5 is a vehicle speed-scaling factor table used in the embodiment of the present application:

[0123] Table 5 Vehicle speed-scaling factor table

[0124]

[0125]

[0126] In the table, v is the vehicle speed and α is the distance scaling factor; when the vehicle speed is 16.7 m / s, the corresponding distance scaling factor is 0.88, that is, the longitudinal distance threshold is scaled at a ratio of 0.88.

[0127] Then, the obtained current curve radius is linearly interpolated through a preset radius-longitudinal distance table to obtain a longitudinal distance standard value. The radius-longitudinal distance table used in the embodiment of the present application is shown in Table 6 below:

[0128] Table 6 Radius-longitudinal distance table

[0129] Rd_out 100m 400m 800m 1500m 1800m 2000m d_long 50m 60m 75m 90m 108m 115m

[0130] In the table, Rd_out is the current curve radius, and d_long is the basic identification distance.

[0131] Finally, the basic recognition distance is multiplied by the distance scaling factor to obtain the corresponding longitudinal distance threshold. The longitudinal distance threshold is the conditional threshold set for the vehicle to determine the front target as a following target in the longitudinal distance when the radius of the curve is less than 1800m. If the front target vehicle exceeds the longitudinal distance threshold, it means that the distance from the vehicle is too far and it is not determined as a following target.

[0132] The following vehicle target recognition module 203 is used to perform curve target recognition based on preset judgment conditions and longitudinal distance thresholds through the acquired vehicle driving information to identify the following vehicle target in front of the vehicle.

[0133] In the embodiment of the present application, the vehicle driving information at the current moment is collected, and the self-driving trajectory of the vehicle and the target driving trajectory of the vehicle in front are simulated according to the vehicle driving information and a preset trajectory simulation formula;

[0134] Based on the judgment conditions, several judgment criteria are set; wherein the judgment criteria include a confidence standard, a horizontal judgment standard and a no-cross risk standard;

[0135] determining whether the vehicle driving information, the self-driving trajectory, and the target driving trajectory meet the determination criteria, and when all the determination criteria are met, comparing the target longitudinal distance with the longitudinal distance threshold;

[0136] If the target longitudinal distance is greater than or equal to the longitudinal distance threshold, it is determined that the front vehicle is not a front following target;

[0137] If the target longitudinal distance is less than the longitudinal distance threshold, it is determined that the front vehicle is a front following target, and the vehicle speed is adjusted according to the target longitudinal distance and the longitudinal distance threshold.

[0138] In some embodiments, the radius prediction module 201 further includes:

[0139] Currently, vehicles equipped with ACC systems can accurately confirm the following target vehicle in front through auxiliary cameras. However, when the vehicle is on a straight road and the target vehicles in front are in a curved environment, or when the vehicle is in a curve and the target vehicle is on a straight road, the vehicle's selection of the vehicle in front mainly relies on the camera to identify the lane line, the vehicle's yaw angle driving trajectory prediction and the driving trajectory prediction of the target vehicle in front, and then judge the target based on the time distance to determine whether to use the target as a following target.

[0140] If a purely visual solution is adopted, that is, only a camera is used, when there is a blind spot on the curve or the light is too dark for the camera to recognize the lane line, the camera is likely to misselect the target, which in turn causes ACC to trigger a false deceleration, affecting the driver's normal driving.

[0141] In the embodiment of the present application, the curve-related signals collected by the sensor are processed to limit the target recognition distance within a certain radius, thereby improving the misidentification phenomenon of target screening and finding a suitable vehicle-following target more accurately.

[0142] First, the vehicle speed signal is collected by the sensor to obtain the current speed information of the vehicle and interpolation is performed according to the preset speed-radius increment table to obtain the curve radius increment at different speeds. Among them, the table interpolation used in the embodiment of the present application is a linear interpolation method, which performs linear interpolation in each speed interval.

[0143] For example, the following Table 7 is a vehicle speed-radius increment table used in the embodiment of the present application, wherein v is the vehicle speed, dy is the curve radius increment, and when the current vehicle speed information is 26 m / s, it can be determined that the corresponding curve radius increment is 60 m.

[0144] Table 7 Speed-radius increment table

[0145] v 15m / s 20m / s 26m / s 35m / s 45m / s dy 12m 15m 60m 75m 90m

[0146] Specifically, the curve radius increment is mainly used to prevent the curve radius collected by the sensor at the front of the vehicle from fluctuating too much in a short period of time, thereby causing misidentification of the target vehicle. Because the lane radius will not change too much in a short period of time, and environmental factors such as light intensity and whether the lane line is clear will cause the sensor to collect data with large errors, the embodiment of the present application introduces a curve radius increment to adjust the length of the curve radius, so that the obtained curve radius value is stable, and the increase of errors such as value jumps or distortion caused by poor sensor recognition accuracy is prevented.

[0147] After the corresponding curve radius increment is obtained, a conditional judgment is performed based on the initial curve radius currently collected by the sensor, the curve radius output in the previous cycle, and the curve radius increment, and a curve reference value is output, specifically:

[0148] If the initial curve radius>=the curve radius output in the previous cycle-the curve radius increment, and the initial curve radius<=the curve radius output in the previous cycle+the curve radius increment, the curve radius output in the previous cycle is used as the curve reference value;

[0149] Otherwise, the initial curve radius is used as the curve reference value.

[0150] The above process is to make a conditional judgment between the initial curve radius currently collected and the curve radius predicted in the previous cycle, and output a relatively stable curve reference value that can best reduce the calculation error, thereby avoiding the use of data with excessive fluctuations for subsequent calculations and increasing data errors.

[0151] Then the obtained curve reference value is filtered, and the curve radius and the curve reference value output in the first n cycles are taken for sliding average calculation, and the average value is taken as the current curve radius in the current cycle. The specific calculation formula is:

[0152] Rd_out=(Rd_fill+Rd_out(0)+Rd_out(1)+...+Rd_out(n)) / n;

[0153] Where Rd_out is the current curve radius, Rd_fill is the curve reference value, and n is the previous n cycles.

[0154] In some embodiments, the following vehicle target recognition module 203 further includes:

[0155] Three judgment criteria are set according to the preset judgment conditions, namely, confidence standard, lateral judgment standard and no-crossing risk standard, which are used to judge whether the longitudinal distance between the vehicle and the target vehicle in front is too close or too far, whether a lateral collision may occur between the vehicle and the target vehicle in front, and whether there is a crossing risk between the two vehicles.

[0156] The vehicle driving information at the current moment is collected, including the target longitudinal distance, target lateral distance, longitudinal relative speed, etc. between the vehicle and the target vehicle in front.

[0157] Determine whether the vehicle's driving state meets the confidence standard: When the vehicle's sensor continuously senses the target vehicle in front for more than 0.15 seconds, the distance and relative speed between the vehicle and the target vehicle in front are determined. When the target longitudinal distance is less than 120 meters or the longitudinal relative speed is less than -5 meters / second, the vehicle's driving state meets the confidence standard.

[0158] Determine whether the vehicle's driving state meets the lateral judgment standard: In a curve scenario, if the predicted lateral collision time between the vehicle and the target vehicle in front is greater than 0.5s, it is considered that the lateral risk with the target vehicle is small and the lateral judgment standard is met. The lateral collision time is predicted based on the target lateral distance.

[0159] Determine whether the driving state of the vehicle meets the no-crossing risk standard: According to the vehicle driving information and the preset trajectory simulation formula, simulate the vehicle's self-driving trajectory and the target driving trajectory of the vehicle in front. Then, based on the current curve radius, interpolate the preset radius-lateral distance table to obtain a lateral distance reference value, as shown in Table 8 below:

[0160] Table 8 Radius-lateral distance table

[0161] Rd_out 100m 400m 800m 1500m 1800m 2000m X_rel 30m 35m 40m 50m 55m 60m

[0162] In the table, X_rel is the reference value of the lateral distance.

[0163] A corresponding lateral distance range is generated according to the lateral distance reference value, and then the self-driving trajectory is set equal to the target driving trajectory for solving. If there is a solution within the lateral distance range where the self-driving trajectory is equal to the target driving trajectory, then the no-crossing risk criterion is satisfied, indicating that there is no trajectory crossing risk between the self-vehicle and the target vehicle ahead.

[0164] Exemplarily, the lateral distance range of the embodiment of the present application is 0 to X_rel.

[0165] Among them, the self-driving trajectory is specifically expressed as:

[0166] Y=C0+C1*X+C2*X 2 +C3*X 3 ;

[0167] Where, Y is the self-driving trajectory, C0, C1, C2, and C3 are the coefficients of the lane centerline, which are generally provided by the road information fusion module of the vehicle; X is the longitudinal coordinate of the vehicle;

[0168] The target driving trajectory is specifically expressed as:

[0169] W=A0+A1*X+A2*X 2 +A3*X 3 ;

[0170] Wherein, W is the target driving trajectory, A0, A1, A2, A3 are coefficients of the target driving trajectory, which are generally provided by the target trajectory processing module of the vehicle; X is the longitudinal coordinate of the target vehicle.

[0171] When the above three criteria are all met, it means that the longitudinal distance and lateral distance between the target vehicle in front and this vehicle are safe and reasonable, and there is no risk of intersection in the driving trajectories between the two vehicles, there is no risk of collision and it is not a vehicle in the side lane. At this time, the recognition and judgment mechanism can be activated to determine whether the target vehicle in front can be used as a following target through the target longitudinal distance and the longitudinal distance threshold.

[0172] Specifically, if the target longitudinal distance is greater than or equal to the longitudinal distance threshold, it is determined that the vehicle in front is not the front following target; if the target longitudinal distance is less than the longitudinal distance threshold, it is determined that the vehicle in front is the front following target, and the driving speed of the vehicle is adjusted according to the target longitudinal distance and the longitudinal distance threshold.

[0173] Implementing the embodiments of the present application has the following beneficial effects:

[0174] The embodiment of the present application first predicts the curve radius of the current vehicle based on the acquired vehicle speed information to obtain a relatively accurate current curve radius; then the current curve radius is integrated with the vehicle speed information to obtain the threshold value of the longitudinal distance between the vehicle and the vehicle in front relative to the current curve radius, providing data support for the subsequent determination of the following vehicle target that is closer to the vehicle; then according to the set judgment conditions, the following vehicle target that is closest to the vehicle, has a safe lateral distance, and has no collision risk is identified through the acquired vehicle driving information, so that the vehicle does not need to adjust the speed urgently during driving due to mistakenly identifying vehicles in other lanes as following vehicle targets, resulting in reduced passenger comfort. Moreover, because the most suitable following vehicle target can be accurately identified in the curve scenario, driving safety can also be improved.

[0175] Furthermore, Figure 6 This is a structural diagram of a terminal device provided in one embodiment of the present application. Figure 6 As shown, the terminal device 3 of this embodiment includes: at least one processor 30 (in Figure 6Only one is shown) and a memory 31 and a computer program 32 stored in the memory 31 and executable on the at least one processor. When the processor 30 executes the computer program 32, the steps of a curve target recognition method based on road information described in any one of the embodiments of the present application can be implemented.

[0176] The terminal device 3 may be a computing device such as a desktop computer, a cloud server, or a laptop computer. The computing device may include but is not limited to a processor 30 and a memory 31 . Figure 6 This is merely an example of the terminal device 3 and does not constitute a limitation on the terminal device 3 , which may include more or fewer components than those shown in the figure.

[0177] The specific embodiments described above further describe the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only a specific embodiment of the present application and is not intended to limit the scope of protection of the present application. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of protection of the present application.

Claims

1. A method for identifying curved targets based on road information, characterized in that: include: According to the current vehicle speed information obtained, the radius is predicted through the initial curve radius to obtain the current curve radius; Fusing the current curve radius with the current vehicle speed information to obtain a longitudinal distance threshold; Based on the preset judgment conditions and longitudinal distance threshold, the vehicle driving information is obtained to perform curve target recognition and identify the vehicle's front following target.

2. The method for identifying curved targets based on road information according to claim 1, characterized in that: According to the obtained current vehicle speed information, the radius is predicted by using the initial curve radius to obtain the current curve radius, which is specifically: According to a preset vehicle speed-radius increment table, the current vehicle speed information is interpolated to obtain a corresponding curve radius increment; Outputting a curve reference value by conditionally judging the initial curve radius, the curve radius output in the previous cycle, and the curve radius increment; The curve reference value and the curve radius output in the first threshold number of cycles before are calculated by sliding average to obtain the current curve radius of the current cycle.

3. The method for identifying curved targets based on road information according to claim 2, characterized in that: The curve reference value is output by conditionally judging the initial curve radius, the curve radius output in the previous cycle, and the curve radius increment, specifically: If the initial curve radius is less than or equal to the sum of the curve radius output in the previous cycle and the curve radius increment, and the initial curve radius is greater than or equal to the difference between the curve radius output in the previous cycle and the curve radius increment, the curve radius output in the previous cycle is used as the curve reference value; Otherwise, the initial curve radius is used as the curve reference value.

4. The method for identifying curved targets based on road information according to claim 1, characterized in that: The current curve radius and the current vehicle speed information are fused to obtain the longitudinal distance threshold, which is specifically: According to a preset vehicle speed-scaling factor table, the current vehicle speed information is interpolated to obtain a distance scaling factor; According to a preset radius-longitudinal distance table, the current curve radius is interpolated to obtain a longitudinal distance standard value; The distance scaling factor and the longitudinal distance standard value are calculated to obtain the longitudinal distance threshold.

5. The method for identifying curved targets based on road information according to claim 1, characterized in that: Based on the preset judgment conditions and the longitudinal distance threshold, the curved target recognition is performed through the acquired vehicle driving information to recognize the vehicle following the vehicle in front, specifically: Collecting the vehicle driving information at the current moment, and simulating the vehicle's self-driving trajectory and the target driving trajectory of the vehicle ahead according to the vehicle driving information and a preset trajectory simulation formula; Based on the judgment conditions, several judgment criteria are set; wherein the judgment criteria include a confidence standard, a horizontal judgment standard and a no-cross risk standard; determining whether the vehicle driving information, the self-driving trajectory, and the target driving trajectory meet the determination criteria, and when all the determination criteria are met, comparing the target longitudinal distance with the longitudinal distance threshold; If the target longitudinal distance is greater than or equal to the longitudinal distance threshold, it is determined that the front vehicle is not a front following target; If the target longitudinal distance is less than the longitudinal distance threshold, it is determined that the front vehicle is a front following target, and the vehicle speed is adjusted according to the target longitudinal distance and the longitudinal distance threshold.

6. The method for identifying curved targets based on road information according to claim 5, characterized in that: The determining whether the vehicle driving information, the self-driving trajectory and the target driving trajectory meet the determination criteria is specifically: Acquiring a target longitudinal distance between the vehicle and a target vehicle from the vehicle travel information; determining whether the confidence criterion is met according to the target longitudinal distance, the speed signal perception time and the vehicle longitudinal relative speed; determining whether the lateral collision judgment standard is satisfied according to the predicted lateral collision time; According to the self-driving trajectory and the target driving trajectory, it is determined whether the no-crossing risk standard is met.

7. The method for identifying curved targets based on road information according to claim 6, characterized in that: The determining whether the no-crossing risk standard is met according to the self-driving trajectory and the target driving trajectory is specifically as follows: According to a preset radius-lateral distance table, interpolation processing is performed on the current curve radius to obtain a lateral distance reference value; Generate a lateral distance range according to the lateral distance reference value, and set the self-driving trajectory equal to the target driving trajectory for solving; If within the lateral distance range, a solution exists where the self-driving trajectory is equal to the target driving trajectory, then the no-crossing risk criterion is satisfied; Otherwise, the no cross-risk criterion is not met.

8. The method for identifying curved targets based on road information according to claim 5, characterized in that: The self-driving trajectory and the target driving trajectory are specifically: The self-driving trajectory is specifically expressed as: Y=C0+C1*X+C2*X 2 +C3*X 3 ; Where Y is the self-driving trajectory, C0, C1, C2, C3 are the coefficients of the lane centerline, and X is the longitudinal coordinate of the vehicle; The target driving trajectory is specifically expressed as: W=A0+A1*X+A2*X 2 +A3*X 3 ; Wherein, W is the target driving trajectory, A0, A1, A2, A3 are the coefficients of the target driving trajectory, and X is the longitudinal coordinate of the target vehicle.

9. A curve target recognition system based on road information, characterized in that: include: Radius prediction module, data fusion module and vehicle following target recognition module; The radius prediction module is used to predict the radius based on the initial curve radius according to the current vehicle speed information obtained, and obtain the current curve radius; The data fusion module is used to fuse the current curve radius with the current vehicle speed information to obtain a longitudinal distance threshold; The following vehicle target recognition module is used to perform curve target recognition based on preset judgment conditions and longitudinal distance thresholds through the acquired vehicle driving information, and identify the following vehicle target in front of the vehicle.

10. A terminal device, characterized in that: The method comprises a processor and a memory, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a curve target recognition method based on road information as described in any one of claims 1 to 8 are implemented.

Images