A Vehicle Lane-Changing Decision Method, Device, Equipment and Medium

Abstract

The present application discloses a vehicle lane-changing decision method, device, equipment and medium, relating to the field of autonomous driving technology, including: determining a to-be-judged merging lane segment and a basic lane segment in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the subsequent lane segment of the to-be-judged merging lane segment; obtaining a curve model for the left and right boundary lines of the to-be-judged merging lane segment, and determining whether the to-be-judged merging lane segment is a merging lane segment based on the curve model; if the to-be-judged merging lane segment is a merging lane segment, determining the lane merging direction corresponding to the merging lane segment according to the included angles between the lane centerline vectors of the other preceding lane segments of the merging lane segment and the basic lane segment respectively and the basic lane segment; if there is no collision risk for the target vehicle in the merging lane segment, outputting a decision result for the target vehicle to change lanes in the merging lane segment based on the lane merging direction. The present application can make vehicle lane-changing decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of assisted driving, and particularly relates to a vehicle lane-changing decision method, device, equipment and medium. Background Art

[0002] Currently, for vehicle lane-changing decisions in lane merging scenarios, most rely on lane merging information sent by HDMap (High Definition Map, a high-precision map for autonomous driving), that is, specific lane merging information needs to be sent by HDMap. After the target vehicle receives the lane merging information sent by HDMap, it makes a decision judgment on vehicle lane-changing based on the lane merging information.

[0003] However, if HDMap does not send lane merging information or sends incorrect lane merging information, it will affect the decision judgment of vehicle lane-changing, resulting in the inability to make a lane-changing decision or making an incorrect decision in the merging area.

[0004] In summary, how to make a vehicle lane-changing decision is an urgent problem to be solved currently. Summary of the Invention

[0005] In view of this, the purpose of the present invention is to provide a vehicle lane-changing decision method, device, equipment and medium, which can make a vehicle lane-changing decision. The specific scheme is as follows:

[0006] In the first aspect, the present application discloses a vehicle lane-changing decision method, including:

[0007] Determine a to-be-judged merging lane segment and a basic lane segment in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the subsequent lane segment of the to-be-judged merging lane segment;

[0008] Obtain a boundary curve model based on the left and right two boundary lines of the to-be-judged merging lane segment, and judge whether the to-be-judged merging lane segment is a merging lane segment based on the curve model;

[0009] If the to-be-judged merging lane segment is the merging lane segment, determine the lane merging direction corresponding to the merging lane segment according to the target relationships between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors;

[0010] If there is no collision risk for the target vehicle in the merging lane segment, output a decision result for the target vehicle to change lanes in the merging lane segment based on the lane merging direction.

[0011] Wherein, determining the merging lane section to be determined in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes includes:

[0012] Acquire a target lane that is pre-divided into a plurality of segmented lane segments and surrounding same-direction lanes that are pre-divided into a plurality of segmented lane segments;

[0013] The segmented lane segment having multiple preceding lane segments in the target lane is used as a basic lane segment, and the lanes belonging to the target lane among the multiple preceding lane segments are used as merging lane segments to be determined.

[0014] The step of obtaining a boundary curve model based on the left and right boundary lines of the merging lane segment to be determined, and determining whether the merging lane segment to be determined is a merging lane segment based on the curve model includes:

[0015] Fitting the left and right boundary lines of the merging lane segment to be determined to obtain a left curve model corresponding to the left boundary line and a right curve model corresponding to the right boundary line;

[0016] Based on the left curve model and the right curve model, determining whether the left and right boundary lines intersect within a predetermined length and whether the left and right boundary lines gradually narrow;

[0017] If so, it is determined that the merging lane section to be determined is a merging lane section.

[0018] Wherein, the step of determining whether the left and right boundary lines intersect within a predetermined length includes:

[0019] The left curve model and the right curve model are combined to determine whether the left and right boundary lines intersect within the predetermined length within a predetermined range; the predetermined range is the range from the position of the target vehicle to a preset length position in front of the position of the target vehicle.

[0020] Wherein, judging whether the left and right boundary lines gradually narrow includes:

[0021] Based on the left curve model and the right curve model, a plurality of groups of corresponding boundary points are determined from the left and right boundary lines at intervals of the target distance;

[0022] The distance between each group of corresponding boundary points is calculated, and it is determined whether the left and right boundary lines gradually narrow according to the distance.

[0023] Wherein, before determining the lane merging direction corresponding to the merging lane segment according to the target relationships between the merging lane segment and other preceding lane segments of the basic lane segment respectively and the basic lane segment; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors, further including:

[0024] Based on the curve model, select the center point at any position where the distance between the left and right boundary lines of the merging lane segment is less than a first distance as the first center point, and use the center point at a second distance in front of the first center point as the second center point. The centerline pointing from the first center point to the second center point is used as the lane centerline vector of the merging lane segment; the first center point and the second center point are located in the merging lane segment;

[0025] Use the center point at a third distance from the initial center point of the basic lane segment as the third center point, and use the centerline pointing from the initial center point to the third center point as the lane centerline vector of the basic lane segment;

[0026] Determine the target center connection line between the first center point and the target vehicle. The target center connection line is translated along the lane segment segmentation line to the lane centerline in other preceding lane segments to determine the fourth center point corresponding to the first center point, and use the center point at the second distance in front of the fourth center point as the fifth center point. The centerline pointing from the fourth center point to the fifth center point is used as the lane centerline vector of other preceding lane segments.

[0027] Wherein, determining the lane merging direction corresponding to the merging lane segment according to the target relationships between the merging lane segment and other preceding lane segments of the basic lane segment respectively and the basic lane segment; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors, including:

[0028] If the included angle between the lane centerline vectors of the merging lane segment and the basic lane segment is greater than the included angle between the lane centerline vectors of other preceding lane segments of the basic lane segment, and the vehicle centerline vector of the merging lane segment is on the right side of the vehicle centerline vector of other preceding lane segments, then determine that the lane merging direction corresponding to the merging lane segment is left merging;

[0029] If the included angle between the lane centerline vectors of the confluence lane segment and the basic lane segment is less than the included angle between the lane centerline vectors of other predecessor lane segments of the basic lane segment and the basic lane segment, and the vehicle centerline of the confluence lane segment is on the left side of the vehicle centerline of the other predecessor lane segments, then it is determined that the lane confluence direction corresponding to the confluence lane segment is right confluence.

[0030] In a second aspect, the present application discloses a vehicle lane-changing decision-making device, including:

[0031] A lane determination module, configured to determine a to-be-determined confluence lane segment and a basic lane segment in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the successor lane segment of the to-be-determined confluence lane segment;

[0032] A lane judgment module, configured to obtain a boundary curve model based on the left and right two boundary lines of the to-be-determined confluence lane segment, and judge whether the to-be-determined confluence lane segment is a confluence lane segment based on the curve model;

[0033] A direction determination module, configured to, if the to-be-determined confluence lane segment is the confluence lane segment, determine the lane confluence direction corresponding to the confluence lane segment according to the target relationships between the confluence lane segment and other predecessor lane segments of the basic lane segment and the basic lane segment respectively; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors;

[0034] A vehicle lane-changing decision-making module, configured to, if there is no collision risk for the target vehicle in the confluence lane segment, output a decision result for the target vehicle to change lanes in the confluence lane segment based on the lane confluence direction.

[0035] In a third aspect, the present application discloses an electronic device, including:

[0036] A memory, configured to store a computer program;

[0037] A processor, configured to execute the computer program to implement the vehicle lane-changing decision-making method disclosed above.

[0038] In a fourth aspect, the present application discloses a computer-readable storage medium, configured to store a computer program; wherein, when the computer program is executed by a processor, the vehicle lane-changing decision-making method disclosed above is implemented.

[0039] It can be seen that the to-be-determined merging lane segment and the basic lane segment in the target lane are determined based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the subsequent lane segment of the to-be-determined merging lane segment in the target lane; a boundary curve model is obtained based on the left and right boundary lines of the to-be-determined merging lane segment, and it is determined whether the to-be-determined merging lane segment is a merging lane segment based on the curve model; if the to-be-determined merging lane segment is the merging lane segment, the lane merging direction corresponding to the merging lane segment is determined according to the target relationships between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors; it is determined whether there is a collision risk for the target vehicle in the merging lane segment, and if not, the decision result of the target vehicle changing lanes in the merging lane segment based on the lane merging direction is output. Thus, it can be seen that the present application can directly determine whether there is a merging lane segment and determine the lane merging direction of the merging lane segment. At this time, the lane merging information sent by the HDMap is not necessary, so the influence of the HDMap on the vehicle lane-changing decision is reduced, and the situation of being unable to make a vehicle lane-changing decision or making a wrong decision is avoided; in addition, when the lane merging information sent by the HDMap is available, the present application can further verify the lane merging information sent by the HDMap to avoid incorrect lane merging information; in addition, the present application determines whether the to-be-determined merging lane segment is a merging lane segment by curve fitting the left and right boundary lines of the to-be-determined merging lane segment, and determines the lane merging direction corresponding to the merging lane segment by using the included angle between the lane centerline vectors, so as to make a vehicle lane-changing decision subsequently and improve the accuracy of the vehicle lane-changing decision. Description of the Drawings

[0040] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0041] Figure 1 It is a flowchart of a vehicle lane-changing decision method disclosed in the present application;

[0042] Figure 2 It is a schematic diagram of a lane merging scenario disclosed in the present application;

[0043] Figure 3 It is another schematic diagram of a lane merging scenario disclosed in the present application;

[0044] Figure 4A schematic diagram of lane segmentation disclosed in this application;

[0045] Figure 5 A flowchart of a specific vehicle lane-changing decision-making method disclosed in this application;

[0046] Figure 6 A schematic diagram showing the lane centerline vector disclosed in this application;

[0047] Figure 7 A schematic structural diagram of a vehicle lane-changing decision-making device disclosed in this application;

[0048] Figure 8 A structural diagram of an electronic device disclosed in this application. Detailed implementation manners

[0049] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0050] Currently, for vehicle lane-changing decisions in lane confluence scenarios, most rely on the lane confluence information sent by the HDMap, that is, the HDMap needs to send specific lane confluence information. After the target vehicle receives the lane confluence information sent by the HDMap, it makes a decision judgment on vehicle lane-changing based on the lane confluence information.

[0051] However, if the HDMap does not send lane confluence information or sends incorrect lane confluence information, it will affect the decision judgment of vehicle lane-changing, resulting in the inability to make a lane-changing decision or making a wrong decision in the confluence area.

[0052] Therefore, the embodiments of this application propose a vehicle lane-changing decision-making scheme that can make vehicle lane-changing decisions.

[0053] The embodiments of this application disclose a vehicle lane-changing decision-making method. Refer to Figure 1 As shown, this method includes:

[0054] Step S11: Determine the to-be-judged confluence lane segment and the basic lane segment in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the subsequent lane segment of the to-be-judged confluence lane segment.

[0055] In this embodiment, the vehicle lane-changing decision-making method is mainly applicable to the lane confluence scenario. Refer to Figure 2As shown, it is a schematic diagram of a lane confluence scenario, including a two-in-one scenario of the right lane on the left side of the picture and a three-in-one scenario of the right lane on the right side of the picture. In this scenario, the end of the right lane becomes narrower and narrower and gradually disappears. Visually, the right lane merges into the left lane; see Figure 3 As shown, it is a schematic diagram of another lane confluence scenario, including a two-in-one scenario of the left lane on the left side of the picture and a three-in-one scenario of the left lane on the right side of the picture. In this scenario, the end of the left lane becomes narrower and narrower and gradually disappears. Visually, the left lane merges into the right lane; Figure 2 and Figure 3 In

[0056] It should be noted that the vehicle lane change decision is completed by ADAS. For scenarios where the shape of the lane line changes within a short distance, the forward-looking perception system such as a camera generally cannot accurately detect it. It is necessary to rely on the HDMap to give the lane confluence information for the ADAS system to make corresponding decisions to complete the lane confluence action; when the HDMap does not send clear lane confluence information, the lane confluence scenario can be judged through the lane connection relationship and lane boundary line information; among them, ADAS (Advanced Driving Assistance System, advanced driving assistance system) can integrate multiple sensors, controllers and actuators to realize real-time monitoring and control of the vehicle; HDMap; is a high-precision map for autonomous driving. As prior data for autonomous driving, it expresses map elements such as roads, lanes, roadside traffic signs and ground markings.

[0057] In this embodiment, segmented lane segments obtained by segmenting the target lane and surrounding same-direction lanes in advance can be obtained. The specific lane segmentation situation is sent by the HDMap. Specifically, determining the to-be-judged confluence lane segment in the target lane based on the target lane where the target vehicle is located and surrounding same-direction lanes includes: obtaining the target lane pre-segmented into a plurality of segmented lane segments and surrounding same-direction lanes pre-segmented into a plurality of segmented lane segments; taking the segmented lane segment in the target lane that has multiple predecessor lane segments as the basic lane segment, and taking the lanes belonging to the target lane among the multiple predecessor lane segments as the to-be-judged confluence lane segment.

[0058] In a specific embodiment, first, lane-related information (including lane segmentation situation) needs to be obtained from the HDMap, see Figure 4 As shown, it is a schematic diagram of a lane segmentation. The HDMap combined with RTK (Real Time Kinematic, high-precision positioning technology based on the global satellite navigation system) positioning can match the vehicle itself with the lane, and deduce which lane the vehicle itself is in, and obtain the vehicle lane information and surrounding lane information from the HDMap, Figure 4The map information in the middle contains lanes in the form of sections in the horizontal direction. The lanes in the section are emitted from right to left in sequence. There is a successor relationship between the lanes in the vertical direction. Obtain the following information: the lane ID number (different numbers can be set for different lanes before segmentation), the lane where the vehicle is located is lane1, the connection relationship of the lanes, such as the successor lane section of lane1 is lane2, the predecessor lane section of lane2 is lane1, the left lane of lane1 is lane4, the right lane of lane4 is lane1, the center line and the points of the left and right boundary lines of each lane in the vehicle body coordinate system. It should be noted that the map information provided by different high-precision map suppliers may vary. If the points of the lane center line cannot be obtained, they can also be calculated from the left and right boundary lines. Further, based on the obtained information, the confluence lane section to be judged and the basic lane section are determined.

[0059] Specifically, start searching for the successor lane section from the lane1 where the vehicle is located until there is no successor lane section, and thus form the path1 where the vehicle is located. Figure 4 In the scenario, path1 contains lane1, lane2, lane3, lane7; traverse all lanes in path1, and judge whether there is a lane in path1 that has two predecessor lane sections. If not, continue to traverse. If so, determine this lane as the basic lane section, and use the lanes belonging to path1 among the multiple predecessor lane sections as the confluence lane section to be judged. lane3 is the confluence lane section to be judged.

[0060] Step S12: Obtain the boundary curve model based on the left and right boundary lines of the confluence lane section to be judged, and judge whether the confluence lane section to be judged is a confluence lane section based on the curve model.

[0061] Step S13: If the confluence lane section to be judged is the confluence lane section, determine the lane confluence direction corresponding to the confluence lane section according to the target relationships between the confluence lane section and the other predecessor lane sections of the basic lane section and the basic lane section respectively; the target relationships include the included angle relationship and the position relationship between the lane center line vectors.

[0062] In this embodiment, if the included angle between the lane centerline vectors of the confluence lane segment and the basic lane segment is greater than the included angle between the lane centerline vectors of other preceding lane segments of the basic lane segment and the basic lane segment, and the vehicle centerline vector of the confluence lane segment is on the right side of the vehicle centerline vector of the other preceding lane segments, it is determined that the lane confluence direction corresponding to the confluence lane segment is left confluence; if the included angle between the lane centerline vectors of the confluence lane segment and the basic lane segment is less than the included angle between the lane centerline vectors of other preceding lane segments of the basic lane segment and the basic lane segment, and the vehicle centerline of the confluence lane segment is on the left side of the vehicle centerline of the other preceding lane segments, it is determined that the lane confluence direction corresponding to the confluence lane segment is right confluence.

[0063] Step S14: If there is no collision risk for the target vehicle in the confluence lane segment, output the decision result of the target vehicle changing lanes in the confluence lane segment based on the lane confluence direction.

[0064] In this embodiment, it is necessary to determine whether there is a collision risk for the target vehicle in the confluence lane segment. Before making the determination, first, the position at the third distance in front of the confluence end position of the confluence lane segment is used as the vehicle confluence preparation position, and the position at the fourth distance in front of the confluence end position of the confluence lane segment is used as the vehicle confluence dangerous position; the fourth distance is less than the third distance; the specific determination process is as follows: When the target vehicle is at the fifth distance in front of the vehicle confluence preparation position, determine whether there is a collision risk for the target vehicle in the confluence lane segment. If not, output the decision result of the target vehicle changing lanes in the confluence lane segment based on the lane confluence direction; correspondingly, it also includes: When the target vehicle reaches the sixth distance in front of the vehicle confluence dangerous position, remind the driver to take over the target vehicle; if the target vehicle reaches the vehicle confluence dangerous position and the driver still has not driven the vehicle and the driver still has not taken over the target vehicle, then end the assisted driving. It should be noted that specific distances are not specifically set here and can be set according to actual situations.

[0065] In a specific embodiment, refer to Figure 4As shown in the figure, the merging lane section is lane3, and the merging end point is p1 (i.e., the intersection of the two boundary lines of lane3). Select a position p2 for vehicle merging preparation d1m before the merging end point p1 (set a threshold such as 200m), and take a position p3 for vehicle merging danger d2m before point p1 (set a threshold such as 50m). When merging from the left, when the host vehicle travels a certain distance before point p2 (a distance threshold can be set, such as a fixed distance or vehicle speed multiplied by a set time), judge whether the line types of the left lane line sent by the HDMap and the left lane line (right lane line when merging from the right) sent by perception are both dashed lines. Calculate whether there is a collision risk through the left obstacle information sent by perception (right obstacle information when merging from the right) (the TTC (Time to Collision) time can be calculated through the relative distance / relative speed between the obstacle and the host vehicle. When TTC < the time threshold, it is considered that there is a collision risk). If the line types of the left lane lines are both dashed lines and there is no collision risk, then make a decision result of changing lanes to the left; otherwise, continue to judge. When the host vehicle travels to a certain distance before point p3 and still does not meet the lane change condition, remind the driver to take over the vehicle. If the driver still does not take over the vehicle when the host vehicle reaches point p3, then exit the ADAS function.

[0066] In this embodiment, after receiving the HDMap information, if the decision-making module receives that a certain lane has a merging attribute sent by the HDMap, it can be verified according to the above vehicle lane change decision-making method. If it can judge the merging scenario and the direction is consistent with the HDMap, then it is considered that the merging information of the HDMap passes the verification, and then enter the decision-making logic of merging lane change. If the merging attribute sent by the HDMap is not received, then execute the above vehicle lane change decision-making judgment logic. If the merging scenario and the merging direction are judged, then enter the decision-making logic of merging lane change.

[0067] It can be seen that the to-be-determined merging lane segment and the basic lane segment in the target lane are determined based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the subsequent lane segment of the to-be-determined merging lane segment in the target lane; a boundary curve model is obtained based on the left and right boundary lines of the to-be-determined merging lane segment, and it is determined whether the to-be-determined merging lane segment is a merging lane segment based on the curve model; if the to-be-determined merging lane segment is a merging lane segment, the lane merging direction corresponding to the merging lane segment is determined according to the target relationships between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors; it is determined whether there is a collision risk for the target vehicle in the merging lane segment, and if not, the decision result of the target vehicle changing lanes in the merging lane segment based on the lane merging direction is output. Thus, it can be seen that this application can directly determine whether there is a merging lane segment and determine the lane merging direction of the merging lane segment. At this time, the lane merging information sent by the HDMap is not necessary, so the influence of the HDMap on the vehicle lane-changing decision is reduced, and the situation of being unable to make a vehicle lane-changing decision or making a wrong decision is avoided; in addition, when the lane merging information sent by the HDMap is available, this application can further verify the lane merging information sent by the HDMap to avoid incorrect lane merging information; in addition, this application uses the method of curve fitting for the left and right boundary lines of the to-be-determined merging lane segment to determine whether the to-be-determined merging lane segment is a merging lane segment, and uses the included angle of the lane centerline vectors to determine the lane merging direction corresponding to the merging lane segment, so as to make a vehicle lane-changing decision later and improve the accuracy of the vehicle lane-changing decision.

[0068] An embodiment of this application discloses a specific vehicle lane-changing decision method. Compared with the previous embodiment, this embodiment further describes and optimizes the technical solution. Refer to Figure 5 as shown, which specifically includes:

[0069] Step S21: Determine the to-be-determined merging lane segment and the basic lane segment in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the subsequent lane segment of the to-be-determined merging lane segment.

[0070] In this embodiment, segmented lane segments obtained by pre-segmenting the target lane and surrounding lanes in the same direction may be obtained. The specific lane segmentation is sent by HDMap. Specifically, determining the merging lane segment to be determined in the target lane based on the target lane where the target vehicle is located and the surrounding lanes in the same direction includes: obtaining the target lane pre-divided into a plurality of segmented lane segments and the surrounding lanes in the same direction pre-divided into a plurality of segmented lane segments; taking the segmented lane segment in the target lane having multiple predecessor lane segments as the basic lane segment, and taking the lanes belonging to the target lane among the multiple predecessor lane segments as the merging lane segment to be determined.

[0071] Step S22: Fitting the left and right boundary lines of the merging lane section to be determined to obtain a left curve model corresponding to the left boundary line and a right curve model corresponding to the right boundary line; based on the left curve model and the right curve model, determining whether the left and right boundary lines intersect within a predetermined length and whether the left and right boundary lines gradually narrow; if so, determining that the merging lane section to be determined is a merging lane section.

[0072] It should be pointed out that the fitting is curve fitting, specifically polynomial fitting; by simultaneously determining whether the left and right boundary lines intersect within a predetermined length and whether the left and right boundary lines gradually narrow, it can be ensured that lane merging occurs. If only the left and right boundary lines gradually narrow, it cannot be ruled out that the lane narrows but there is no merging.

[0073] In this embodiment, the determination of whether the left and right boundary lines intersect within a predetermined length includes: combining the left curve model and the right curve model to determine that the left and right boundary lines intersect within a predetermined range within the predetermined length; the predetermined range is the range from the position of the target vehicle to the position of the target vehicle in front of the position of the target vehicle by a preset length. It should be noted that the preset length can be 200, which is specifically set according to actual conditions.

[0074] It should be noted that when the predetermined range is the range from the position of the target vehicle to a position of a preset length in front of the position of the target vehicle, and a real number solution is determined, it means that the lanes intersect in front of the vehicle.

[0075] In this embodiment, judging whether the left and right boundary lines are gradually narrowing includes: determining a plurality of groups of corresponding boundary points from the left and right boundary lines based on the left curve model and the right curve model with the target distance as an interval; calculating the distance between each group of the corresponding boundary points, and judging whether the left and right boundary lines are gradually narrowing according to the distance.

[0076] In a specific embodiment, see Figure 4As shown, lane 7 has two preceding lane segments, lane 3 and lane 6, of which lane 3 exists in path 1. The points of the left and right boundary lines of lane 3 are obtained from HDMap, and a cubic curve model is fitted using the least squares method. The equations of the left and right boundary lines are as follows:

[0077] ;

[0078] ;

[0079] in, , , and is the curve coefficient of the left boundary line equation, , , and is the curve coefficient of the right boundary line equation, x represents the longitudinal displacement, and y represents the lateral displacement.

[0080] By combining the left and right boundary line equations to form a two-variable equation system, eliminating the variable y, we get a cubic equation about x. We solve this equation system by Newton's method or gradient descent method. If there is at least one solution in the real number domain and x is in the interval [0,200], the two curves are considered to intersect. It should be pointed out that the control domain x is within a certain range to avoid invalid judgment of curve intersection. If the two curves intersect at infinity, it is not the desired result.

[0081] The fitted equations of the left and right boundary lines are discretized into points at intervals of 1 m in the longitudinal direction. The Euclidean distance between the two points is calculated in sequence as the lane width. The lane widths of three consecutive points are taken for judgment. Assuming that the lane widths of a certain longitudinal distance i meters and the following two points are , , ,like > and > It is considered that the lane width begins to narrow; i can be set according to actual conditions.

[0082] If all the above conditions are met, lane 3 is considered to be a merging lane section.

[0083] Step S23: If the merging lane segment to be determined is the merging lane segment, the lane merging direction corresponding to the merging lane segment is determined according to the target relationship between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationship includes the angle relationship between lane centerline vectors and the position relationship between lane centerline vectors.

[0084] In this embodiment, after determining the confluence lane section, it is also necessary to determine the lane confluence direction of the confluence lane section in order to determine the vehicle driving direction. In most existing methods, the confluence direction is determined by judging whether there are left and right lanes, but this method may not be applicable to the three-in-one or other multi-confluence scenarios. Therefore, this application proposes a method for determining the lane confluence direction corresponding to the confluence lane section based on the included angle of the lane centerline vectors. Specifically, the lane confluence direction corresponding to the confluence lane section is determined according to the target relationships between the confluence lane section and other preceding lane sections of the basic lane section and the basic lane section respectively; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors, including: if the included angle between the lane centerline vectors of the confluence lane section and the basic lane section is greater than the included angle between the lane centerline vectors of other preceding lane sections of the basic lane section and the basic lane section, and the vehicle centerline vector of the confluence lane section is on the right side of the vehicle centerline vector of the other preceding lane sections, it is determined that the lane confluence direction corresponding to the confluence lane section is left confluence; if the included angle between the lane centerline vectors of the confluence lane section and the basic lane section is less than the included angle between the lane centerline vectors of other preceding lane sections of the basic lane section and the basic lane section, and the vehicle centerline of the confluence lane section is on the left side of the vehicle centerline of the other preceding lane sections, it is determined that the lane confluence direction corresponding to the confluence lane section is right confluence.

[0085] It should be noted that the center line of the confluence lane section is in a curved state. Therefore, two center points can be selected from the confluence lane section and connected to obtain the lane center line vector. Specifically, before determining the lane confluence direction corresponding to the confluence lane section according to the target relationships between the confluence lane section and other preceding lane sections of the basic lane section respectively and the basic lane section, the following steps are also included: Based on the curve model, select a center point at any position where the distance between the left and right boundary lines of the confluence lane section is less than a first distance as the first center point, and use the center point at a second distance in front of the first center point as the second center point. The center line pointing from the first center point to the second center point is used as the lane center line vector of the confluence lane section; the first center point and the second center point are located in the confluence lane section; use the center point at a third distance from the initial center point of the basic lane section as the third center point, and use the center line pointing from the initial center point to the third center point as the lane center line vector of the basic lane section; determine the target center connection line between the first center point and the target vehicle, translate the target center connection line along the lane section segmentation line direction to the lane center line in other preceding lane sections to determine the fourth center point corresponding to the first center point, and use the center point at the second distance in front of the fourth center point as the fifth center point. The center line pointing from the fourth center point to the fifth center point is used as the lane center line vector of other preceding lane sections.

[0086] It should be noted that when determining the lane center line, it can be obtained by selecting two center points and connecting them. However, if the lane is not a straight lane, the selection of the center point will affect the lane center line and even the included angle (for example, in the right three-in-one scenario, if the vehicle's own lane is the rightmost lane and the other preceding lane section is the second lane from the right, at this time the two lanes are not straight lanes. Therefore, center points at different positions will result in different directions of the lane center line vector. If the two center points of the rightmost lane are close to the starting position of the lane and the other preceding lane section is close to the ending position of the lane, it may cause the included angle between the rightmost lane and the basic lane section to be smaller than the included angle between the other preceding lane section and the basic lane section. At this time, the vehicle center line vector of the confluence lane section is on the right side of the vehicle center line vector of the other preceding lane section. Due to the incorrect judgment of the included angle caused by the selection of the center point, there will be no result). Therefore, two center points of the confluence lane section can be determined first, and then two center points of other preceding lane sections can be determined based on the two center points of the confluence lane section to determine the lane center line vector. Figure 2

[0087] Figure 6 It should be noted that, referring to Figure 6As shown, it is a schematic diagram of the display of the lane centerline vector. In the figure, the target center connection line between the first center point and the target vehicle is determined. The target center connection line is translated along the lane segment dividing line to the lane centerline in the other preceding lane segment to determine the fourth center point corresponding to the first center point, and the center point at the second distance in front of the fourth center point is used as the fifth center point. When the other preceding lane segment is a straight lane, the fourth center point and the fifth center point can also be determined in the following way. Specifically: the center point where the connection line between the first center point in the other preceding lane segment and the target lane line is perpendicular to the target lane line is used as the fourth center point, and the center point where the connection line between the second center point in the other preceding lane segment and the target lane line is perpendicular to the target lane line is used as the fifth center point.

[0088] It should be noted that the first distance, the second distance, and the third distance can be specifically set according to the situation; see Figure 6 As shown, the second distance and the predetermined distance can be 2m, L represents the first distance, represents the lane centerline vector of the confluence lane segment, represents the lane centerline vector of the other preceding lane segment, represents the lane centerline vector of the basic lane segment. This scenario is a two-in-one scenario, which requires a lane centerline vector of a confluence lane segment, a lane centerline vector of an other preceding lane segment, and a lane centerline vector of a basic lane segment. If it is Figure 6 a three-in-one scenario in

[0089] It should be noted that the calculation formula for the included angle of the lane centerline vector is as follows:

[0090] ;

[0091] ;

[0092] is the included angle of the lane centerline vector between the confluence lane segment and the basic lane segment; is the included angle of the lane centerline vector between the other preceding lane segment of the basic lane segment and the basic lane segment.

[0093] If is greater than , it is determined that the lane confluence direction corresponding to the confluence lane segment is left confluence; if is less than , it is determined that the lane confluence direction corresponding to the confluence lane segment is right confluence. Among them, Figure 6 is left confluence.

[0094] Step S24: If there is no collision risk for the target vehicle in the merging lane section, output the decision result of the target vehicle to change lanes in the merging lane section based on the lane merging direction.

[0095] It can be seen that this application determines the to-be-determined merging lane section and the basic lane section in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane section is the subsequent lane section of the to-be-determined merging lane section; fit the left and right boundary lines of the to-be-determined merging lane section to obtain the left curve model corresponding to the left boundary line and the right curve model corresponding to the right boundary line; based on the left curve model and the right curve model, determine whether the left and right boundary lines intersect within a predetermined length and whether the left and right boundary lines gradually narrow; if so, determine the to-be-determined merging lane section as a merging lane section; if the to-be-determined merging lane section is the merging lane section, determine the lane merging direction corresponding to the merging lane section according to the target relationships between the merging lane section and other preceding lane sections of the basic lane section and the basic lane section respectively; the target relationships include the included angle relationship between the lane centerline vectors and the position relationship between the lane centerline vectors; if there is no collision risk for the target vehicle in the merging lane section, output the decision result of the target vehicle to change lanes in the merging lane section based on the lane merging direction. Thus, this application discloses a specific curve fitting method to determine the merging lane section and a specific method to determine the merging direction by the included angle of the lane centerline vectors.

[0096] Correspondingly, the embodiment of this application also discloses a vehicle lane-changing decision device. Refer to Figure 7 as shown, the device includes:

[0097] A lane determination module 11, configured to determine the to-be-determined merging lane section and the basic lane section in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane section is the subsequent lane section of the to-be-determined merging lane section;

[0098] A lane judgment module 12, configured to obtain a boundary curve model based on the left and right boundary lines of the to-be-determined merging lane section, and determine whether the to-be-determined merging lane section is a merging lane section based on the curve model;

[0099] A direction determination module 13, configured to, if the to-be-determined merging lane section is the merging lane section, determine the lane merging direction corresponding to the merging lane section according to the target relationships between the merging lane section and other preceding lane sections of the basic lane section and the basic lane section respectively; the target relationships include the included angle relationship between the lane centerline vectors and the position relationship between the lane centerline vectors;

[0100] The vehicle lane-changing decision-making module 14 is configured to output a decision result for the target vehicle to change lanes in the merging lane segment based on the lane merging direction if there is no collision risk for the target vehicle in the merging lane segment.

[0101] Among them, the more specific working processes of the above-mentioned various modules can refer to the corresponding content disclosed in the foregoing embodiments, and will not be elaborated here.

[0102] It can be seen that the to-be-judged merging lane segment and the basic lane segment in the target lane are determined based on the target lane where the target vehicle is located and the surrounding same-direction lanes; the basic lane segment is the subsequent lane segment of the to-be-judged merging lane segment in the target lane; a boundary curve model is obtained based on the left and right boundary lines of the to-be-judged merging lane segment, and it is determined whether the to-be-judged merging lane segment is a merging lane segment based on the curve model; if the to-be-judged merging lane segment is the merging lane segment, the lane merging direction corresponding to the merging lane segment is determined according to the target relationships between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationships include the included angle relationship and the position relationship between the lane centerline vectors; it is judged whether there is a collision risk for the target vehicle in the merging lane segment, and if not, a decision result for the target vehicle to change lanes in the merging lane segment based on the lane merging direction is output. Thus, it can be seen that the present application can directly judge whether there is a merging lane segment and determine the lane merging direction of the merging lane segment. At this time, the lane merging information sent by the HDMap is not necessary, so the influence of the HDMap on the vehicle lane-changing decision-making is reduced, and the situation of being unable to make a vehicle lane-changing decision or making a wrong decision is avoided; in addition, when the lane merging information sent by the HDMap is received, the present application can further verify the lane merging information sent by the HDMap to avoid incorrect lane merging information; in addition, the present application uses the method of curve fitting for the left and right boundary lines of the to-be-judged merging lane segment to judge whether the to-be-judged merging lane segment is a merging lane segment, and uses the included angle of the lane centerline vectors to determine the lane merging direction corresponding to the merging lane segment, so as to perform vehicle lane-changing decision-making subsequently and improve the accuracy of vehicle lane-changing decision-making.

[0103] Furthermore, the embodiment of the present application also provides an electronic device. Figure 8 It is a structural diagram of the electronic device 20 shown according to an exemplary embodiment, and the content in the figure should not be regarded as any limitation on the scope of use of the present application.

[0104] Figure 8Schematic diagram of the structure of an electronic device 20 provided by an embodiment of the present application. The electronic device 20 may specifically include: at least one processor 21, at least one memory 22, a display screen 23, an input / output interface 24, a communication interface 25, a power supply 26, and a communication bus 27. Among them, the memory 22 is used to store a computer program, and the computer program is loaded and executed by the processor 21 to implement the relevant steps in the vehicle lane-changing decision-making method disclosed in any of the foregoing embodiments. In addition, the electronic device 20 in this embodiment may specifically be an electronic computer.

[0105] In this embodiment, the power supply 26 is used to provide working voltage for each hardware device on the electronic device 20; the communication interface 25 can create a data transmission channel between the electronic device 20 and external devices, and the communication protocol it follows is any communication protocol applicable to the technical solution of the present application, and specific limitations are not imposed on it here; the input / output interface 24 is used to obtain external input data or output data to the outside, and its specific interface type can be selected according to specific application needs, and no specific limitations are made here.

[0106] In addition, as a carrier for resource storage, the memory 22 may be a read-only memory, a random access memory, a disk, or an optical disc, etc., and the resources stored thereon may include a computer program 221, and the storage method may be temporary storage or permanent storage. Among them, in addition to the computer program capable of implementing the vehicle lane-changing decision-making method executed by the electronic device 20 disclosed in any of the foregoing embodiments, the computer program 221 may further include a computer program capable of performing other specific tasks.

[0107] Furthermore, an embodiment of the present application also discloses a computer-readable storage medium for storing a computer program; wherein, when the computer program is executed by a processor, the vehicle lane-changing decision-making method disclosed above is implemented.

[0108] For the specific steps of this method, reference may be made to the corresponding content disclosed in the foregoing embodiments, and details are not repeated here.

[0109] The various embodiments in this application are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. For the same or similar parts between the various embodiments, reference may be made to each other. For the device disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and reference may be made to the description in the method part for the relevant parts.

[0110] Those skilled in the art may further realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, computer software, or a combination of the two. To clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described according to functions in the above description. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0111] The steps of the methods or algorithms described in combination with the embodiments disclosed herein can be directly implemented by hardware, software modules executed by a processor, or a combination of the two. The software modules can be placed in a random access memory (RAM), internal memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium well-known in the technical field.

[0112] Finally, it should also be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0113] The above has introduced in detail a vehicle lane-changing decision method, device, equipment, and storage medium provided by this application. Specific examples are used herein to elaborate on the principle and implementation manner of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to this application.

Claims

1. A vehicle lane change decision method, characterized in that: include: Determine a merging lane segment to be determined and a basic lane segment in the target lane based on the target lane where the target vehicle is located and surrounding same-direction lanes; The basic lane segment is a subsequent lane segment of the merging lane segment to be determined; Obtaining a boundary curve model based on the left and right boundary lines of the merging lane segment to be determined, and determining whether the merging lane segment to be determined is a merging lane segment based on the curve model; If the merging lane segment to be determined is the merging lane segment, the lane merging direction corresponding to the merging lane segment is determined according to the target relationship between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationship includes the angle relationship between lane centerline vectors and the position relationship between lane centerline vectors; If there is no collision risk for the target vehicle in the merging lane section, outputting a decision result of the target vehicle changing lanes in the merging lane section based on the lane merging direction; Wherein, determining the merging lane section to be determined in the target lane based on the target lane where the target vehicle is located and the surrounding same-direction lanes includes: Acquire a target lane that is pre-divided into a plurality of segmented lane segments and surrounding same-direction lanes that are pre-divided into a plurality of segmented lane segments; The segmented lane segment having multiple preceding lane segments in the target lane is used as a basic lane segment, and a lane belonging to the target lane among the multiple preceding lane segments is used as a merging lane segment to be determined; Wherein, judging whether the to-be-determined merging lane segment is a merging lane segment based on the curve model includes: Based on the curve model, determining whether the left and right boundary lines intersect within a predetermined length and whether the left and right boundary lines gradually narrow; If yes, determining that the merging lane section to be determined is a merging lane section; Wherein, the lane merging direction corresponding to the merging lane segment is determined according to the target relationship between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationship includes the angle relationship between lane centerline vectors and the position relationship between lane centerline vectors, including: If the lane centerline vector angle between the merging lane segment and the basic lane segment is greater than the lane centerline vector angle between other preceding lane segments of the basic lane segment and the basic lane segment, and the vehicle centerline vector of the merging lane segment is located to the right of the vehicle centerline vector of other preceding lane segments, then the lane merging direction corresponding to the merging lane segment is determined to be left merging; If the lane centerline vector angle between the merging lane segment and the basic lane segment is smaller than the lane centerline vector angle between other preceding lane segments of the basic lane segment and the basic lane segment, and the vehicle centerline of the merging lane segment is located to the left of the vehicle centerline of the other preceding lane segments, then the lane merging direction corresponding to the merging lane segment is determined to be right merging.

2. The vehicle lane change decision method according to claim 1, characterized in that: The step of obtaining a boundary curve model based on the left and right boundary lines of the merging lane segment to be determined, and determining whether the merging lane segment to be determined is a merging lane segment based on the curve model includes: Fitting the left and right boundary lines of the merging lane segment to be determined to obtain a left curve model corresponding to the left boundary line and a right curve model corresponding to the right boundary line; Based on the left curve model and the right curve model, determining whether the left and right boundary lines intersect within a predetermined length and whether the left and right boundary lines gradually narrow; If so, it is determined that the merging lane section to be determined is a merging lane section.

3. The vehicle lane change decision method according to claim 2, characterized in that: The step of determining whether the left and right boundary lines intersect within a predetermined length includes: Based on the left curve model and the right curve model, it is determined that the left and right boundary lines intersect within the predetermined length within a predetermined range; the predetermined range is a range from a position of a target vehicle to a position of a preset length in front of the position of the target vehicle.

4. The vehicle lane change decision method according to claim 2, characterized in that: Determining whether the left and right boundary lines gradually narrow includes: Based on the left curve model and the right curve model, a plurality of groups of corresponding boundary points are determined from the left and right boundary lines at intervals of the target distance; The distance between each group of corresponding boundary points is calculated, and it is determined whether the left and right boundary lines gradually narrow according to the distance.

5. The vehicle lane change decision method according to claim 1, characterized in that: Before determining the lane merging direction corresponding to the merging lane segment according to the target relationship between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment, the method further includes: Based on the curve model, a center point of any position where the distance between the left and right boundary lines of the merging lane segment is less than a first distance is selected as a first center point, and a center point at a second distance in front of the first center point is selected as a second center point, and a center line pointing from the first center point to the second center point is selected as a lane center line vector of the merging lane segment; the first center point and the second center point are located in the merging lane segment; The center point of the basic lane segment at a third distance from the initial center point of the basic lane segment is used as the third center point, and the center line from the initial center point to the third center point is used as the lane center line vector of the basic lane segment; the initial center point is the first lane center point of the basic lane segment; Determine a target center line between the first center point and the target vehicle, translate the target center line along the lane segment dividing line direction to the lane center line in the other preceding lane segment to determine a fourth center point corresponding to the first center point, and use the center point at the second distance in front of the fourth center point as the fifth center point, and use the center line from the fourth center point to the fifth center point as the lane center line vector of the other preceding lane segment.

6. The vehicle lane change decision method according to any one of claims 1 to 5, characterized in that: If there is no collision risk for the target vehicle in the merging lane section, before outputting a decision result of the target vehicle changing lanes in the merging lane section based on the lane merging direction, the method further includes: When the target vehicle is located at the fifth distance in front of the vehicle merging preparation position, the lane line shape is obtained and the collision time between the obstacle and the target vehicle is calculated; wherein the vehicle merging preparation position is the third distance in front of the merging end position of the merging lane segment. Based on the lane line type and the collision time, it is determined whether the target vehicle has a collision risk in the merging lane section.

7. The vehicle lane change decision method according to claim 6, characterized in that: Also includes: If the target vehicle reaches the sixth distance ahead of the vehicle merging danger position, if the lane change condition is still not met, the driver is reminded to take over the target vehicle; wherein the vehicle merging danger position is the fourth distance ahead of the merging end position of the merging lane section; the fourth distance is less than the third distance; If the target vehicle reaches the dangerous position for merging vehicles and the driver has not yet taken over the target vehicle, the assisted driving is terminated.

8. A vehicle lane change decision device, characterized in that: include: A lane determination module, used to determine a merging lane segment to be determined and a basic lane segment in the target lane based on the target lane where the target vehicle is located and surrounding same-direction lanes; The basic lane segment is a subsequent lane segment of the merging lane segment to be determined; a lane determination module, configured to obtain a boundary curve model based on the left and right boundary lines of the merging lane segment to be determined, and determine whether the merging lane segment to be determined is a merging lane segment based on the curve model; a direction determination module, for determining, if the merging lane segment to be determined is the merging lane segment, a lane merging direction corresponding to the merging lane segment according to target relationships between the merging lane segment and other preceding lane segments of the basic lane segment and the basic lane segment respectively; the target relationship includes an angle relationship between lane centerline vectors and a position relationship between lane centerline vectors; a vehicle lane change decision module, configured to output a decision result of the target vehicle changing lanes in the merging lane section based on the lane merging direction if there is no collision risk for the target vehicle in the merging lane section; The lane determination module is specifically used to obtain a target lane pre-divided into a plurality of segmented lane segments and surrounding same-direction lanes pre-divided into a plurality of segmented lane segments; The segmented lane segment having multiple preceding lane segments in the target lane is used as a basic lane segment, and a lane belonging to the target lane among the multiple preceding lane segments is used as a merging lane segment to be determined; The lane determination module is specifically used to determine whether the left and right boundary lines intersect within a predetermined length and whether the left and right boundary lines gradually narrow based on the curve model; If yes, determining that the merging lane section to be determined is a merging lane section; The direction determination module is specifically configured to determine that the lane merging direction corresponding to the merging lane segment is left merging if the lane centerline vector angle between the merging lane segment and the basic lane segment is greater than the lane centerline vector angle between other preceding lane segments of the basic lane segment and the basic lane segment, and the vehicle centerline vector of the merging lane segment is located to the right of the vehicle centerline vector of other preceding lane segments; If the lane centerline vector angle between the merging lane segment and the basic lane segment is smaller than the lane centerline vector angle between other preceding lane segments of the basic lane segment and the basic lane segment, and the vehicle centerline of the merging lane segment is located to the left of the vehicle centerline of the other preceding lane segments, then the lane merging direction corresponding to the merging lane segment is determined to be right merging.

9. An electronic device, characterized in that: include: Memory, used to store computer programs; A processor, configured to execute the computer program to implement the vehicle lane change decision method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: Used to store a computer program; wherein, when the computer program is executed by a processor, the vehicle lane change decision method as described in any one of claims 1 to 7 is implemented.

Images