Vehicle lateral control method, device, electronic device and storage medium

Abstract

The present specification provides a method, device, electronic device and storage medium for lateral control of a vehicle, the method comprising: identifying the lane line of the vehicle's driving direction, and at least when there are multiple lane lines in the recognition result, determining two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determining the first angle and second angle formed by the driving direction and the two target lane lines respectively. When the first angle and the second angle satisfy a relative size relationship, the vehicle is laterally controlled based on the two target lane lines. When the first angle and the second angle do not satisfy the relative size relationship, the vehicle is laterally controlled based on the target lane line corresponding to the relatively smaller angle between the first angle and the second angle.

Description

Technical Field

[0001] The present specification relates to the field of intelligent driving technology, and in particular to a vehicle lateral control method, device, electronic device and storage medium. Background Art

[0002] The lateral control of a vehicle is one of the key technologies in an intelligent driving system. It refers to controlling the lateral displacement of a vehicle so that it remains on a predetermined trajectory during driving. The goal of lateral control is to achieve stable driving, lane keeping or steering within the lane by adjusting the steering wheel angle and other means.

[0003] The lateral control function of the current intelligent driving system mainly relies on lane line information. It uses sensors such as cameras and lidar to identify the lane lines on the road the vehicle is traveling on, and performs laterally control of the vehicle based on the lane line information.

[0004] However, using the identified pseudo lane lines or cluttered lane lines to perform lateral control on the vehicle will affect the accuracy of lateral control. For example, if pseudo lane lines (such as roadside markings, shadows) are identified as the target lane lines of the target lane where the vehicle is located, performing lateral control based on the pseudo lane lines will cause the vehicle to make unnecessary turns, which may cause a traffic accident. Summary of the invention

[0005] In order to overcome the problems existing in the related art, this specification provides a lateral control method, device, electronic device and storage medium for a vehicle.

[0006] According to a first aspect of an embodiment of this specification, a lateral control method for a vehicle is provided, the method comprising:

[0007] Identify the lane line of the driving direction of the vehicle, and at least when there are multiple lane lines in the recognition result, determine two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determine a first angle and a second angle formed by the driving direction and the two target lane lines respectively;

[0008] When the first angle and the second angle satisfy a relative magnitude relationship, performing lateral control on the vehicle based on the two target lane lines;

[0009] When the first angle and the second angle do not satisfy the relative size relationship, the vehicle is laterally controlled based on a target lane line corresponding to a relatively smaller angle between the first angle and the second angle.

[0010] According to a second aspect of an embodiment of this specification, a lateral control device for a vehicle is provided, the device comprising:

[0011] An angle determination module, used to identify the lane line of the vehicle's driving direction, and at least when there are multiple lane lines in the recognition result, determine two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determine a first angle and a second angle formed by the driving direction and the two target lane lines respectively;

[0012] A first lateral control module, configured to perform lateral control on the vehicle based on the two target lane lines when the first angle and the second angle satisfy a relative magnitude relationship;

[0013] The second lateral control module is used to perform lateral control on the vehicle based on a target lane line corresponding to a relatively smaller angle between the first angle and the second angle when the first angle and the second angle do not satisfy a relative size relationship.

[0014] According to a second aspect of an embodiment of this specification, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the method described in the first aspect are implemented.

[0015] According to a third aspect of the embodiments of this specification, a computer-readable storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0016] The technical solutions provided by the embodiments of this specification may have the following beneficial effects:

[0017] In the embodiment of the present specification, the first angle and the second angle formed by the driving direction of the vehicle and the target lane line of the identified target lane are determined. When the first angle and the second angle satisfy the relative size relationship, it means that the two target lane lines are likely to be the real lane lines of the target lane and are suitable for lateral control of the vehicle, and the vehicle can be laterally controlled based on the two target lane lines. When the first angle and the second angle do not satisfy the relative size relationship, it means that one of the lane lines is likely to be a pseudo lane line or a cluttered lane line. At this time, the vehicle is laterally controlled from the target lane line corresponding to the relatively smaller angle between the first angle and the second angle, which can eliminate the interference of the pseudo lane line or the cluttered lane line on the lateral control algorithm, improve the accuracy of lateral control, and effectively avoid unnecessary steering of the vehicle.

[0018] It is to be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present specification. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the specification and, together with the description, serve to explain the principles of the specification.

[0020] Figure 1 It is a flow chart of a lateral control method of a vehicle shown in this specification according to an exemplary embodiment.

[0021] Figure 2 This is a schematic diagram of a scenario of lateral control of a vehicle according to an exemplary embodiment of the present specification.

[0022] Figure 3 Schematic diagram of a first angle and a second angle according to an exemplary embodiment of the present specification.

[0023] Figure 4 It is a flowchart of determining whether a lateral control condition is satisfied according to an exemplary embodiment of the present specification.

[0024] Figure 5 It is a flowchart of another lateral control method of a vehicle according to an exemplary embodiment of the present specification.

[0025] Figure 6 It is a structural schematic diagram of an electronic device according to an exemplary embodiment of the present specification.

[0026] Figure 7 It is a block diagram of a lateral control device for a vehicle according to an exemplary embodiment of the present specification. DETAILED DESCRIPTION

[0027] The lateral control of a vehicle is one of the key technologies in an intelligent driving system. It refers to controlling the lateral displacement of a vehicle so that it remains on a predetermined trajectory during driving. The goal of lateral control is to achieve stable driving, lane keeping or steering within the lane by adjusting the steering wheel angle and other means.

[0028] The lateral control function of the current intelligent driving system mainly relies on lane line information. It uses sensors such as cameras and lidar to identify the lane lines on the road the vehicle is traveling on, and performs laterally control of the vehicle based on the lane line information.

[0029] However, using the identified pseudo lane lines or cluttered lane lines to perform lateral control on the vehicle will affect the accuracy of lateral control. For example, if pseudo lane lines (such as roadside markings, shadows) are identified as the target lane lines of the target lane where the vehicle is located, then performing lateral control based on the pseudo lane lines will cause the vehicle to make unnecessary turns. For example, cluttered lane lines can be lane lines that are not smooth, curved and do not conform to the normal driving trajectory, blurred, or have multiple lane lines overlapping, which are not suitable for lateral control.

[0030] In view of the above technical problems, this specification proposes a lateral control method for a vehicle, which performs lateral control on the vehicle by selecting a suitable lane line from two identified target lane lines to improve the accuracy of lateral control.

[0031] Next, the embodiments of this specification are described in detail:

[0032] like Figure 1 As shown, Figure 1 This is a flow chart of a lateral control method of a vehicle according to an exemplary embodiment of the present specification, comprising the following steps:

[0033] Step 101: Identify the lane lines of the vehicle's traveling direction. At least when there are multiple lane lines in the identification result, determine two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determine a first angle and a second angle formed by the traveling direction and the two target lane lines, respectively.

[0034] Step 102: When the first angle and the second angle satisfy a relative magnitude relationship, the vehicle is laterally controlled based on the two target lane lines.

[0035] Step 103: When the first angle and the second angle do not satisfy the relative size relationship, the vehicle is laterally controlled based on a target lane line corresponding to a relatively smaller angle between the first angle and the second angle.

[0036] like Figure 2 As shown, Figure 2This is a schematic diagram of a scenario for lateral control of a vehicle according to an exemplary embodiment of the present specification. Lane line 210A and lane line 210B constitute a target lane 250, and vehicle 230 is currently traveling on target lane 250. Lane line 210B and lane line 210C constitute an adjacent lane 260 adjacent to the target lane. Vehicle 230 can collect lane line information in front of the vehicle through sensor technologies such as laser radar and camera. In addition to lane line 210A and lane line 210B of the target lane 250 where the vehicle is located, the lane line can also be lane line 210C of an adjacent lane adjacent to the target lane 250. Assume that there is a pseudo lane line 220 within the range of the target lane 250. The pseudo lane line 220 may be a crack on the road, a repair mark, a shadow of a roadside obstacle or a tree, a wheel mark, and other false lane lines. Due to various reasons, such as the sensor being disturbed (such as dust, blur) resulting in unclear collected road images, or the recognition accuracy of the lane line recognition model being not high enough, the vehicle mistakenly recognizes the pseudo lane line 220 as the lane line of the lane. For example, in the recognition result 240, in addition to recognizing the recognized lane line 241A and the recognized lane line 241B corresponding to the real lane line 210A and the real lane line 210B of the target lane 250, and the recognized lane line 243 corresponding to the real lane line 210C of the adjacent lane 260, the vehicle 230 also mistakenly recognizes the pseudo lane line 220 formed by tree shadows or wheel tracks as the lane line 242.

[0037] At least when there are multiple lane lines in the recognition result, two target lane lines of the target lane where the vehicle is located can be determined from the multiple lane lines. In the case where there is a lane line in the recognition result, the vehicle can be controlled laterally based on the lane line. Specifically, the lane line closest to the vehicle side can be selected from the lane lines on each side of the vehicle as the target lane line. For example, it is assumed that the lane lines on the left side of the vehicle include lane line 241A, the lane lines on the right side of the vehicle include lane line 241B and lane line 243, and lane line 242 is located in the middle of the vehicle body and is not located on either side of the vehicle. Lane line 241A can be used as the target lane line on the left side of target lane 250, and since lane line 241B is closest to the right side of the vehicle, lane line 241B can be used as the target lane line on the right side of target lane 250. In the above scenario, lane line 241A and lane line 241B can be accurately identified as the target lane lines of target lane 250. For another example, assuming that the lane lines on the left side of the vehicle include lane line 241A and pseudo lane line 242, and the lane lines on the right side of the vehicle include lane line 241B and lane line 243, since pseudo lane line 242 is closest to the left side of the vehicle, pseudo lane line 242 may be mistakenly identified as the target lane line on the left side of target lane 250. Similarly, assuming that the lane lines on the right side of the vehicle include pseudo lane line 242 and lane line 241B, pseudo lane line 242 may be mistakenly identified as the target lane line on the right side of target lane 250.

[0038] Of course, the above is only an exemplary description, and those skilled in the art can use recognition means well known in the art to determine the two target lane lines of the target lane where the vehicle is located from the recognition results, and this specification does not impose any restrictions on the technical means for identifying the target lane lines. Similarly, from the recognition results, a pseudo lane line or a lane line of an adjacent lane is mistakenly identified as the target lane line of the target lane where the vehicle is located, or the identified target lane line is the real lane line of the target lane, but it is a cluttered lane line and is not suitable for lateral control of the vehicle. For the above-mentioned situations, this solution can eliminate the target lane line that is not suitable for lateral control of the vehicle by comparing the relative size relationship between the first angle and the second angle of the two target lane lines. For details, please refer to the following embodiments:

[0039] After determining the two target lane lines, the first angle and the second angle formed by the vehicle's driving direction and the two target lane lines can be determined. Specifically, Figure 3 As shown, Figure 3It is a schematic diagram of the first angle and the second angle shown in this specification according to an exemplary embodiment. Arrow 330 indicates the driving direction of vehicle 230, and lane line 241A and lane line 241B are two target lane lines respectively. The reference point on the corresponding target lane line can be determined at a distance of n*v in front of the vehicle, and the tangent on the reference point is obtained to calculate the angle between the tangent and the driving direction of the vehicle. Wherein, n is a constant, and v is the current speed of the vehicle. For example, at the position n*v in front of the vehicle, the reference point of the corresponding lane line 241A is determined to be 340A, the reference point of the lane line 241B is determined to be 340B, and the first angle 310A formed by the target lane line 241A and the driving direction of the vehicle is determined, and the second angle 310B formed by the target lane line 241B and the driving direction of the vehicle is determined. Of course, the point with the largest curvature change on lane line 241A and lane line 241B can also be selected as a reference point, and a tangent is made based on the reference point to determine the first angle and the second angle formed by the driving direction of the vehicle and the two target lane lines respectively. The above is merely an exemplary description, and this specification does not limit the method of determining the first angle and the second angle formed by the vehicle's traveling direction and the two target lane lines respectively.

[0040] In one embodiment, when detecting the first angle and the second angle formed by the driving direction and two target lane lines, since the sensor (such as a camera, a radar) may be affected by factors such as jitter, light changes, or blurred road markings, resulting in high-frequency noise being mixed in the detected angle signal, the first angle and the second angle can be low-pass filtered to smooth the changes in the first angle and the second angle.

[0041] After determining the first angle and the second angle, the vehicle can be laterally controlled based on the two target lane lines when the first angle and the second angle meet the relative size relationship. When the first angle and the second angle do not meet the relative size relationship, the vehicle is laterally controlled based on the target lane line corresponding to the relatively smaller angle of the first angle and the second angle. Wherein, the first angle and the second angle meeting the relative size relationship may be that the absolute value of the difference between the first angle and the second angle is not greater than the deviation threshold, or the ratio of the relatively larger angle of the two angles to the relatively smaller angle of the two angles is not greater than the ratio threshold. Of course, it can also be other relative size relationships that are used to measure the proximity of the first angle and the second angle. The closer the first angle value and the second angle value are, the higher the directional consistency of the two target lane lines is, and the more likely it is to correspond to the two real lane lines of the target lane where the vehicle is located. Of course, even if the two target lane lines correspond to the two real lane lines of the target lane where the vehicle is located, if the directional consistency of the two target lane lines does not meet the requirements, that is, the first angle and the second angle do not meet the relative size relationship, the vehicle cannot be laterally controlled based on the two target lane lines. Therefore, by judging the proximity between the first angle and the second angle, it is possible to determine not only whether there is a false lane line in the two target lane lines, but also whether the two target lane lines are suitable for being used together to perform lateral control on the vehicle. Figure 2 In the example, it is assumed that the two identified target lane lines are lane line 241A and pseudo lane line 242, respectively, and the lane line 241A and pseudo lane line 242 respectively form a first angle and a second angle with the driving direction of the vehicle 230. The deviation degree of the first angle and the second angle is highly likely to be higher than the deviation threshold. At this time, it can also be determined based on the angle that the lane line with a smaller angle is highly likely to be the real lane line of the target lane. Therefore, the vehicle can be prevented from performing lateral control based on the pseudo lane line. In addition, the adjacent lane line can also be excluded by judging the relative size relationship. For example, the vehicle is located at the center line of the target lane. The lane line of the target lane and the lane line of the adjacent lane respectively form a first angle and a second angle with the driving direction of the vehicle. The first angle is highly likely to be much smaller than the second angle. At this time, it can also be determined based on the angle that the lane line with a smaller angle is highly likely to be the lane line of the target lane. The vehicle can be controlled laterally based on the lane line, thereby avoiding the interference of the adjacent lane line on the lateral control of the vehicle. In addition, even if the two target lane lines are the real lane lines of the target lane where the vehicle is located, if the first angle and the second angle do not satisfy the relative size relationship, selecting the target lane line with a relatively small angle to perform lateral control of the vehicle can also avoid severe shaking of the vehicle during lateral control.

[0042] In one embodiment, in addition to executing the processing logic of subsequent steps when there are multiple lane lines in the recognition result, the processing logic of subsequent steps may also be executed when the lateral control condition of the vehicle is met and there are multiple lane lines in the recognition result. The lateral control condition may refer to a prerequisite for the vehicle to perform lateral control according to the currently recognized lane lines. If the currently recognized lane lines meet the lateral control condition, these lane lines may be used to perform lateral control of the vehicle, and if the currently recognized lane lines do not meet the lateral control condition, lateral control is not performed according to the current lane lines. By setting the lateral control condition, false lane lines that are not suitable for lateral control can be preliminarily screened out according to the lateral control condition, thereby reducing the risk of vehicle control failure or unwarranted deviation.

[0043] Next, if Figure 4 As shown, Figure 4 This specification is based on an exemplary embodiment of the flow chart showing whether the lateral control condition is met. The specific judgment steps are as follows:

[0044] Step 401: Identify the lane line of the vehicle's driving direction. This step is the same as the step of "identify the lane line of the vehicle's driving direction" in step 101, and will not be repeated here.

[0045] Step 402: Determine a first number of lane lines to be compared from the recognition results.

[0046] Step 403: Determine a second number of lane lines at the driving position of the vehicle marked in the map.

[0047] The lane lines to be compared may be lane lines within the target lane range where the vehicle is located. Step 402 may specifically be determining a first number of lane lines within the target lane range where the vehicle is located from the recognition result. Accordingly, step 403 may specifically be a second number of lane lines marked on the target lane in the map. For example, Figure 1 As shown, the vehicle 230 is traveling on the target lane 250, and the lane lines within the target lane 250 may be all lane lines identified within the target lane. Figure 2As shown, lane line 241A is the recognition result of the real lane line 210A on the target lane 250, and lane line 241B is the recognition result of the real lane line 210B on the target lane 250. Then the lane lines within the target lane range can be all lane lines within the target lane 250 range formed by lane lines 241A and 241B in the recognition result 240, including lane line 241A, lane line 241B and pseudo lane line 242. The number of lane lines marked on the target lane 250 in the map can refer to the second number of real lane lines of the target lane 250. For example, if the target lane 250 contains two lane lines, the second number of lane lines marked on the map for the target lane 250 is 2.

[0048] The lane lines to be compared may be all the identified lane lines in the recognition result. Step 402 may specifically determine a first number of all the lane lines in the recognition result. Accordingly, step 403 may specifically determine a second number of lane lines within the recognition range of the driving position of the vehicle marked in the map. For example, Figure 1 As shown, the recognition range of the vehicle 230 covers the target lane 250 and the adjacent lane 260 where the vehicle is located, and the recognition result 240 includes the lane lines 241A, 241B, 243 and pseudo lane lines 242 of the target lane 250 and the adjacent lane 260. The lane lines within the recognition range of the vehicle's driving position that can be marked in the map are the second number of real lane lines of the target lane 250 and the adjacent lane 260. It should be noted that the second number of lane lines at the driving position of the vehicle marked in the map is related to the recognition range of the vehicle. If the vehicle can recognize the lane lines of all lanes ahead, the lane lines at the vehicle's position marked in the map are the number of lane lines of all lanes at the current position; if the vehicle recognizes the lane lines of the current lane or a part of the number of lanes, the lane lines at the vehicle's position marked in the map are the lane lines of the current lane or a part of the number of lanes.

[0049] Step 404: Determine whether the first quantity is equal to the second quantity. If the first quantity is equal to the second quantity, continue to step 405A; if the first quantity is not equal to the second quantity, continue to step 405B.

[0050] Step 405A: Determine whether the first lateral control condition is met. If so, go to step 406, indicating that the lateral control condition of the vehicle is met. If not, the vehicle is not laterally controlled based on the currently identified lane line, and the lane line information in front of the vehicle continues to be collected.

[0051] Step 405B: Determine whether the second lateral control condition is met. If so, go to step 406, indicating that the lateral control condition of the vehicle is met. If not, the vehicle is not laterally controlled based on the currently identified lane line, and the lane line information in front of the vehicle continues to be collected.

[0052] The constraint degree of the first lateral control condition is lower than that of the second lateral control condition. The higher the constraint degree, the lower the error tolerance of the lateral control condition to the lane line.

[0053] Step 406: The lateral control condition of the vehicle is satisfied.

[0054] Before determining whether the lateral control condition is met, this embodiment first compares whether the number of lane lines in the recognition result is consistent with the number of lane lines marked in the map. If they are consistent, it indicates that there is a high probability that there are no false lane lines in the vehicle recognition result, so the first lateral control condition with a lower degree of constraint can be used to determine whether the currently recognized lane line is available. If they are inconsistent, it indicates that there is a high probability that there are false lane lines in the vehicle recognition result, so the second lateral control condition with a higher degree of constraint can be used to determine whether the currently recognized lane line is available. This solution can be applied to scenarios where the accuracy of vehicle lane line recognition is not high, such as mistakenly identifying various false lane lines as real lane lines. By determining whether the number of recognized lane lines is consistent with that of real lane lines, lateral control conditions with different degrees of constraint can be flexibly switched, so that false lane lines are not easy to meet the lateral control conditions, while real lane lines are easy to meet the lateral control conditions. The filtering requirements for false lane lines can be comprehensively enhanced, and the potential risks caused by false lane lines can be significantly reduced.

[0055] Next, this specification provides the specific conditions of the first lateral control condition and the second lateral control condition:

[0056] The length of at least one lane line in the target lane range where the vehicle is located, the quality score of at least one lane line in the target lane range, the angle between at least one lane line in the target lane range and the driving direction, and / or the width between any two lane lines in the target lane range can be determined from the recognition result. The target lane may refer to the lane where the vehicle is traveling, and the target lane range may refer to the range between two lane lines (including two lane lines) in the lane where the vehicle is located.

[0057] The length of the lane line can be estimated based on the collected lane line image information.

[0058] The dimensions of the lane line quality score include, but are not limited to, the integrity and clarity of the lane line, etc. It can be a score of the comprehensive quality of the lane line, or a quality score of a certain dimension, and this specification does not impose any restrictions on this.

[0059] The method for determining the angle between the lane line and the driving direction can be referred to the above embodiment and will not be repeated in this specification.

[0060] The width between any two lane lines within the target lane range can be obtained by randomly sampling several points on the two lane lines, calculating the distance between the sampling points at the corresponding parallel positions of the two lane lines, and averaging all the distances as the width between the two lane lines. Figure 2 As shown, the target lane range includes lane lines 241A and 241B of the target lane 250, and a pseudo lane line 242 within the target lane range. The width between any two lane lines includes the width between lane line 241A and pseudo lane line 242, the width between lane line 241A and pseudo lane line 242, and / or the width between lane line 241B and pseudo lane line 242.

[0061] The first lateral control condition includes at least one of the following:

[0062] The length is greater than the first length threshold, which means that the lengths of all lane lines within the target lane range are greater than the first length threshold, or the lengths of one or more lane lines randomly selected from all lane lines are greater than the first length threshold. If the length is less than the first length threshold, it means that the lane line is very likely to be a false lane line. For example, the length of a tree shadow is relatively short.

[0063] The quality score is greater than the first scoring threshold, which means that the quality scores of all lane lines within the target lane range are greater than the first scoring threshold, or the quality scores of one or more lane lines randomly selected from all lane lines are greater than the first scoring threshold. If the quality score is less than the first scoring threshold, data for lateral control cannot be accurately extracted from the lane line information.

[0064] The angle is not greater than the first angle threshold, which means that the angles corresponding to all lane lines within the target lane range are not greater than the first angle threshold, or the angles corresponding to one or more lane lines randomly selected from all lane lines are not greater than the first angle threshold. If the angle is greater than the angle threshold, there is a high probability that the lane line is a false lane line or an adjacent lane line.

[0065] The width is not less than the first lower width threshold and not greater than the first upper width threshold. It can be that the width between any two lane lines within the target lane range is not less than the first lower width threshold and not greater than the first upper width threshold, or the width between any two or more lane lines within the target lane range is not less than the first lower width threshold and not greater than the first upper width threshold. If the width is too small or too large, it is very likely not a lane in the real world.

[0066] The second lateral control condition includes at least one of the following:

[0067] The length is greater than the second length threshold, which means that the lengths of all lane lines within the target lane range are greater than the second length threshold, or the lengths of one or more lane lines randomly selected from all lane lines are greater than the second length threshold. And the second length threshold is greater than the first length threshold.

[0068] The quality score is greater than the second scoring threshold, which means that the quality scores of all lane lines within the target lane range are greater than the second scoring threshold, or the quality scores of one or more lane lines randomly selected from all lane lines are greater than the second scoring threshold. And the second scoring threshold is greater than the first scoring threshold.

[0069] The angle is not greater than the second angle threshold, which may be that the angles corresponding to all lane lines within the target lane range are not greater than the second angle threshold, or that the angles corresponding to one or more lane lines randomly selected from all lane lines are not greater than the second angle threshold. And the second angle threshold is less than the first angle threshold.

[0070] The width is not less than the second lower width threshold and not greater than the second upper width threshold, which may be that the width between any two lane lines within the target lane range is not less than the second lower width threshold and not greater than the second upper width threshold, or that the width between any two or more lane lines within the target lane range is not less than the second lower width threshold and not greater than the second upper width threshold. The second lower width threshold is greater than the first lower width threshold and / or the second upper width threshold is greater than the first upper width threshold.

[0071] If any of the sub-judgment conditions of the first lateral control condition or the second lateral control condition does not meet the requirement, the first or second lateral control condition may not be met. For example, assuming that in the first or second lateral control condition, the length of the lane line does not meet the requirement, while the other sub-judgment conditions meet the requirement, the first or second lateral control condition may not be met.

[0072] In addition, in addition to adjusting the constraint degree of the first lateral control condition and the second lateral control condition, the second lateral control condition may have more component sub-judgment conditions than the first lateral control condition to increase the difficulty of satisfying the lane line. For example, the first lateral control condition may include a length sub-judgment condition and a quality score sub-judgment condition, while the second lateral control condition may include a length sub-judgment condition, a quality score sub-judgment condition, an angle sub-judgment condition, and a width sub-judgment condition.

[0073] In one embodiment, if Figure 5 As shown, this specification provides another lateral control method for a vehicle, and the specific steps are as follows:

[0074] First stage:

[0075] Step 401: Identify the lane line of the vehicle's driving direction.

[0076] Step 402: Determine a first number of lane lines to be compared from the recognition results.

[0077] Step 403: Determine a second number of lane lines at the driving position of the vehicle marked in the map.

[0078] Step 404: Determine whether the first quantity is equal to the second quantity. If so, go to step 405A; if not, go to step 405B.

[0079] Step 405A: Determine whether the first lateral control condition is satisfied. If so, go to step 406; if not, end.

[0080] Step 405B: Determine whether the second lateral control condition is satisfied, if so, go to step 406, if not, end.

[0081] Step 406: The lateral control condition is satisfied.

[0082] Second stage:

[0083] Step 407: Determine whether there are multiple lane lines in the recognition result. If so, go to step 409; if not, go to step 408.

[0084] Step 408: Perform lateral control on the vehicle based on the identified lane line.

[0085] Step 409: Determine two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determine a first angle and a second angle formed by the driving direction and the two target lane lines respectively.

[0086] Step 410: Determine whether the first angle and the second angle satisfy a relative size relationship. If so, go to step 411; if not, go to step 412.

[0087] Step 411: Perform lateral control on the vehicle based on the two target lane lines.

[0088] Step 412: Perform lateral control on the vehicle based on a target lane line corresponding to a relatively smaller angle between the first angle and the second angle.

[0089] In this embodiment, in the first link, by comparing whether the number of identified lane lines in the recognition result is consistent with the number of real lane lines, it can be preliminarily determined whether there are false lane lines in the identified lane lines. And in the case where there are false lane lines, the second lateral control condition with a higher degree of constraint is used for judgment, so as to avoid the false lane lines from satisfying the second lateral control condition. At the same time, in the case where there are no false lane lines, the first lateral control condition with a lower degree of constraint is used to avoid the real lane lines from not satisfying the first lateral control condition. Therefore, by flexibly switching the first lateral control condition and the second lateral control condition, it is not easy for the false lane lines to satisfy the lateral control conditions, while the real lane lines are easy to satisfy the lateral control conditions, which can comprehensively enhance the filtering requirements for the false lane lines while avoiding the mistaken identification of the real lane lines as false lane lines. In addition, in the second link, when the lateral control condition is met, the present scheme further compares the relative size relationship between the first angle and the second angle formed by the two target lane lines and the driving direction of the vehicle, and can select the lane line suitable for lateral control of the vehicle. Through comprehensive judgment of the first link and the second link, a lane line suitable for lateral control of the vehicle can be selected to avoid unnecessary lateral control of the vehicle based on false lane lines or inappropriate lane lines (such as cluttered lane lines).

[0090] Corresponding to the embodiments of the aforementioned method, this specification also provides embodiments of a device and a terminal to which it is applied.

[0091] like Figure 6 As shown, Figure 6 It is a structural diagram of an electronic device shown in this specification according to an exemplary embodiment. At the hardware level, the electronic device 600 includes a processor 602, an internal bus 604, a network interface 606, a memory 608 and a non-volatile memory 610, and of course may also include hardware required for other services. One or more embodiments of this specification can be implemented based on software, such as the processor 602 reading the corresponding computer program from the non-volatile memory 610 into the memory 608 and then running it. Of course, in addition to the software implementation, one or more embodiments of this specification do not exclude other implementations, such as logic devices or a combination of software and hardware, etc., that is to say, the execution subject of the following processing flow is not limited to each logic module, but can also be hardware or logic devices.

[0092] like Figure 7 As shown, Figure 7 This is a block diagram of a vehicle lateral control device according to an exemplary embodiment of the present specification. The device can be applied to Figure 6 In the electronic device 600 shown in the figure, the technical solution of this specification is implemented. The device includes:

[0093] Angle determination module 702, used to identify the lane line of the vehicle's driving direction, and at least when there are multiple lane lines in the recognition result, determine two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determine a first angle and a second angle formed by the driving direction and the two target lane lines respectively;

[0094] A first lateral control module 704, configured to perform lateral control on the vehicle based on the two target lane lines when the first angle and the second angle satisfy a relative magnitude relationship;

[0095] The second lateral control module 706 is used to perform lateral control on the vehicle based on a target lane line corresponding to a relatively smaller angle between the first angle and the second angle when the first angle and the second angle do not satisfy a relative size relationship.

[0096] Optionally, the at least case where there are multiple lane lines in the recognition result includes: a lateral control condition of the vehicle is met and there are multiple lane lines in the recognition result.

[0097] Optionally, the device also includes a lane line number determination module 708, which is used to determine a first number of lane lines to be compared from the recognition results, and a second number of lane lines at the driving position of the vehicle marked in the map; the angle determination module 702 is specifically used to determine whether the lateral control condition of the vehicle is met, including: if the first number is equal to the second number, then determine whether the first lateral control condition is met; if the first number is not equal to the second number, then determine whether the second lateral control condition is met; wherein the constraint degree of the first lateral control condition is lower than the constraint degree of the second lateral control condition.

[0098] Optionally, the lane line number determination module 708 is specifically used to determine a first number of lane lines within the target lane range where the vehicle is located from the recognition result, and a second number of lane lines marked for the target lane in the map; or, to determine a first number of all lane lines in the recognition result, and a second number of lane lines within the recognition range of the vehicle's driving position marked in the map.

[0099] Optionally, the device also includes a lane line indicator determination module 710, which is used to determine the length of the lane line within the target lane range of the vehicle, the quality score of the lane line, the angle between the lane line and the driving direction, and / or the width between any two lane lines; the first lateral control condition includes at least one of the following: the length is greater than a first length threshold; the quality score is greater than a first score threshold; the angle is not greater than a first angle threshold; the width is not less than a first lower width threshold and not greater than a first upper width threshold; the second lateral control condition includes at least one of the following: the length is greater than a second length threshold, and the second length threshold is greater than the first length threshold; the quality score is greater than a second score threshold, and the second score threshold is greater than the first score threshold; the angle is not greater than a second angle threshold, and the second angle threshold is less than the first angle threshold; the width is not less than a second lower width threshold and not greater than a second upper width threshold, and the second lower width threshold is greater than the first lower width threshold and / or the second upper width threshold is greater than the first upper width threshold.

[0100] Optionally, the angle determination module 702 is specifically configured to select a lane line closest to the vehicle from the lane lines on each side of the vehicle as the target lane line.

[0101] Optionally, the first angle and the second angle satisfy a relative size relationship, including: an absolute value of a difference between the first angle and the second angle is not greater than a deviation threshold.

[0102] The implementation process of the functions and effects of each module in the above-mentioned device is specifically described in the implementation process of the corresponding steps in the above-mentioned method, which will not be repeated here.

[0103] For the device embodiment, since it basically corresponds to the method embodiment, the relevant parts can refer to the partial description of the method embodiment. The device embodiment described above is only schematic, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, that is, they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this specification. Ordinary technicians in this field can understand and implement it without paying creative work.

[0104] The present specification also provides a computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of any of the aforementioned vehicle lateral control methods provided in the present application are implemented.

[0105] Specifically, computer-readable media suitable for storing computer program instructions and data include all forms of non-volatile memory, media and memory devices, including, for example, semiconductor memory devices (such as EPROM, EEPROM and flash memory devices), magnetic disks (such as internal hard disks or removable disks), magneto-optical disks, and CD ROM and DVD-ROM disks.

Claims

1. A lateral control method for a vehicle, characterized in that: The method comprises: Identify the lane lines of the driving direction of the vehicle, determine a first number of lane lines to be compared from the recognition result, and a second number of lane lines at the driving position of the vehicle marked in the map; at least when there are multiple lane lines in the recognition result, determine two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determine a first angle and a second angle formed by the driving direction and the two target lane lines respectively; at least when there are multiple lane lines in the recognition result, including: when the lateral control condition of the vehicle is satisfied and there are multiple lane lines in the recognition result; wherein the lateral control condition includes a first lateral control condition and a second lateral control condition, judging whether the lateral control condition of the vehicle is satisfied, including: if the first number is equal to the second number, judging whether the first lateral control condition is satisfied, and if the first number is not equal to the second number, judging whether the second lateral control condition is satisfied; wherein the constraint degree of the first lateral control condition is lower than the constraint degree of the second lateral control condition; When the first angle and the second angle satisfy a relative magnitude relationship, performing lateral control on the vehicle based on the two target lane lines; When the first angle and the second angle do not satisfy the relative size relationship, the vehicle is laterally controlled based on a target lane line corresponding to a relatively smaller angle between the first angle and the second angle.

2. The method according to claim 1, characterized in that: The determining, from the recognition result, a first number of lane lines to be compared, and a second number of lane lines at the driving position of the vehicle marked in the map, comprises: Determine a first number of lane lines within the target lane where the vehicle is located and a second number of lane lines marked on the target lane in the map from the recognition result; or, A first number of all lane lines in the recognition result and a second number of lane lines within a recognition range of a driving position of the vehicle marked in the map are determined.

3. The method according to claim 1, characterized in that The method further comprises: Determine the length of a lane line within the target lane where the vehicle is located, a quality score of the lane line, an angle between the lane line and the driving direction, and / or a width between any two lane lines; The first lateral control condition includes at least one of the following: The length is greater than a first length threshold; The quality score is greater than a first score threshold; The angle is not greater than a first angle threshold; The width is not less than a first lower width threshold and not greater than a first upper width threshold; The second lateral control condition includes at least one of the following: The length is greater than a second length threshold, and the second length threshold is greater than the first length threshold; The quality score is greater than a second score threshold, and the second score threshold is greater than the first score threshold; The angle is not greater than a second angle threshold, and the second angle threshold is less than the first angle threshold; The width is not less than a second lower width threshold and not greater than a second upper width threshold, the second lower width threshold is greater than the first lower width threshold and / or the second upper width threshold is greater than the first upper width threshold.

4. The method according to claim 1, characterized in that: The determining two target lane lines of the target lane where the vehicle is located from the plurality of lane lines comprises: The lane line closest to the vehicle is selected from the lane lines on each side of the vehicle as the target lane line.

5. The method according to claim 1, characterized in that The first angle and the second angle satisfy a relative magnitude relationship, including: An absolute value of a difference between the first angle and the second angle is not greater than a deviation threshold.

6. A lateral control device for a vehicle, characterized in that: The device comprises: An angle determination module is used to identify the lane line of the driving direction of the vehicle, determine a first number of lane lines to be compared from the recognition result, and a second number of lane lines at the driving position of the vehicle marked in the map; at least when there are multiple lane lines in the recognition result, determine two target lane lines of the target lane where the vehicle is located from the multiple lane lines, and determine a first angle and a second angle formed by the driving direction and the two target lane lines respectively; at least when there are multiple lane lines in the recognition result, including: when the lateral control condition of the vehicle is satisfied and there are multiple lane lines in the recognition result; wherein the lateral control condition includes a first lateral control condition and a second lateral control condition, judging whether the lateral control condition of the vehicle is satisfied, including: if the first number is equal to the second number, judging whether the first lateral control condition is satisfied, and if the first number is not equal to the second number, judging whether the second lateral control condition is satisfied; wherein the constraint degree of the first lateral control condition is lower than the constraint degree of the second lateral control condition; A first lateral control module, configured to perform lateral control on the vehicle based on the two target lane lines when the first angle and the second angle satisfy a relative magnitude relationship; The second lateral control module is used to perform lateral control on the vehicle based on a target lane line corresponding to a relatively smaller angle between the first angle and the second angle when the first angle and the second angle do not satisfy the relative size relationship.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images