Vehicle driving control method and device and computer readable storage medium

Abstract

The invention discloses a vehicle driving control method and device and a computer readable storage medium, and relates to the field of auxiliary driving, and the method comprises the steps: verifying a visual lane line and a map reference line in front of a vehicle, and verifying the position information at a visual terminal point and a verification position point; judging whether the position relationship between the visual lane line and the map reference line is a preset position relationship or not; if yes, the position relation between the visual lane line and the map reference line is further verified according to a trend verification condition corresponding to the trend type between the visual lane line and the map reference line, and under the condition that the trend verification condition is met, the visual lane line is selected as a target passing route, and the vehicle is controlled to run. According to the invention, in two selectable routes of the visual lane line and the map reference line, the route which is more suitable for the current lane line condition and is relatively more accurate is selected, and the safety of vehicle driving is improved.

Description

Technical Field

[0001] This application relates to the field of assisted driving technology. More specifically, it particularly relates to a vehicle driving control method, device, and computer-readable storage medium. Background Art

[0002] Navigation-assisted driving is an intelligent driving technology that combines L2-level assisted driving functions and navigation information to achieve intelligent navigation-assisted driving of a vehicle from the starting point to the end point. Among them, the intelligent navigation-assisted driving function includes: intelligent entry and exit from ramps. That is, on highways or urban expressways, the vehicle can automatically enter and exit ramps and independently travel within the ramps. For example, when encountering a Y-shaped fork during driving, combined with navigation information or lane lines detected visually, it can intelligently select the correct fork to drive.

[0003] In the scenario of driving at a fork, the navigation-assisted driving system often follows the reference path provided by high-precision map navigation and ignores the lane lines detected visually. This may lead to situations such as the vehicle pressing on the line or deviating from the lane during driving. Moreover, the accuracy of the high-precision map is limited by the update cycle and the accuracy of high-precision positioning. Therefore, in the scenario of driving at a fork, the navigation-assisted driving cannot accurately provide the driving path, affecting driving safety. Summary of the Invention

[0004] In view of the above problems, this application provides a vehicle driving control method, device, and computer-readable storage medium to achieve the purpose of providing an accurate driving path in the scenario of driving at a fork. The specific solutions are as follows:

[0005] In the first aspect of this application, a vehicle driving control method is provided, including:

[0006] Obtain the visual lane lines and map reference lines in front of the vehicle;

[0007] Determine a verification position point between the visual end point of the visual lane line and the position of the vehicle;

[0008] Obtain position information for characterizing the positional relationship between the visual lane line and the map reference line. The position information at least includes: the horizontal distance between the visual end point and the map reference line, and the horizontal distance between the verification position point and the map reference line;

[0009] When the positional relationship between the visual lane line represented by the position information and the map reference line is a preset positional relationship, determine the trend types of the visual lane line and the map reference line, and determine the preset trend verification conditions corresponding to the trend types. The preset positional relationship is approaching parallel, and the trend types include: the same trend, opposite trends and approaching each other, and opposite trends and moving away from each other;

[0010] Judge whether the positional relationship between the visual lane line and the map reference line meets the trend verification conditions;

[0011] If the positional relationship between the visual lane line and the map reference line meets the trend verification conditions, use the visual lane line as the target driving route, and control the vehicle to drive according to the target communication route.

[0012] In one possible implementation, it further includes:

[0013] When the positional relationship between the visual lane line represented by the position information and the map reference line does not belong to the preset positional relationship, or when the positional relationship between the visual lane line and the map reference line does not meet the trend verification conditions, use the map reference line as the target driving route, and control the vehicle to drive according to the target driving route.

[0014] In one possible implementation, the determination of the verification position point between the visual end point of the visual lane line and the position of the vehicle includes:

[0015] Obtain at least one preset verification position point set in advance, and the verification distance corresponding to the preset verification position point. The verification distance is the distance between the preset verification position point and the position of the vehicle;

[0016] Use the maximum verification distance among the verification distances less than the visible distance of the lane line as the target verification distance. The visible distance of the lane line is the distance between the visual end point and the position of the vehicle;

[0017] Determine the preset verification position point corresponding to the target verification distance as the verification position point.

[0018] In one possible implementation, the obtaining of the position information used to represent the positional relationship between the visual lane line and the map reference line includes:

[0019] In the vehicle's own coordinate system, based on the visual end point of the visual lane line and the map reference line, determine the first lateral distance and the first slope difference between the visual end point and the map reference line as the end point position information;

[0020] In the vehicle's own coordinate system, based on the calibration position points of the visual lane line and the map reference line, determine the second lateral distance and the second slope difference between the calibration position points and the map reference line as calibration position information.

[0021] In a possible implementation, the process of determining that the positional relationship between the visual lane line represented by the position information and the map reference line is a preset positional relationship includes:

[0022] Determine two calibration distances adjacent to the visible distance of the lane line, and determine two preset calibration position points corresponding to the two calibration distances as reference calibration position points respectively;

[0023] Obtain the threshold group corresponding to each preset inspection position point set in advance, and the threshold group at least includes: a preset lateral distance threshold and a preset slope threshold;

[0024] Determine the intermediate calibration threshold group corresponding to the calibration position point, and the intermediate calibration threshold group includes: an intermediate lateral distance threshold and an intermediate slope threshold;

[0025] Based on the linear compensation principle, determine the end threshold group corresponding to the visible distance of the lane line based on the threshold groups corresponding to the two reference calibration position points, and the end threshold group includes: an end lateral distance threshold and an end slope threshold;

[0026] Judge whether the first lateral distance in the end position information is less than the end lateral distance threshold, and whether the first slope difference is less than the end slope threshold;

[0027] Judge whether the second lateral distance in the calibration position information is less than the intermediate lateral distance threshold, and whether the second slope difference is less than the intermediate slope threshold.

[0028] In a possible implementation, the determining the trend calibration condition corresponding to the trend type according to the trends of the visual lane line and the map reference line includes:

[0029] Identify the curvatures corresponding to the visual lane line and the map reference line of the visual end point to obtain the visual lane line curvature and the map reference line curvature;

[0030] Based on the visual lane line curvature and the map reference line curvature, determine the trend type of the visual lane line relative to the map reference line;

[0031] Obtain the trend calibration condition corresponding to the trend type set in advance.

[0032] In a possible implementation, when the trend types are the same, the trend verification condition includes: in the preset range of the visual end point in the vehicle's own coordinate system, there is no intersection between the visual lane line and the map reference line.

[0033] In a possible implementation, when the trend types are opposite and approaching each other, the trend verification condition includes: within the interval between the vehicle's location and the visual end point, there is no intersection between the visual lane line and the map reference line, and the first lateral distance is greater than a preset threshold, where the first lateral distance is the lateral distance between the visual end point and the map reference line.

[0034] In a possible implementation, when the trend types are opposite and moving away from each other, the trend verification condition includes: the difference between the first lateral distance and the third lateral distance is not greater than a preset threshold, and the separation degree between the visual end point and the map reference line meets a preset condition.

[0035] The second aspect of the present application provides a vehicle driving control device, including:

[0036] A line acquisition unit, configured to acquire the visual lane line and the map reference line in front of the vehicle;

[0037] A verification point determination unit, configured to determine a verification position point located between the visual end point of the visual lane line and the vehicle's location;

[0038] A position information acquisition unit, configured to acquire position information for characterizing the position relationship between the visual lane line and the map reference line, where the position information at least includes: the horizontal distance between the visual end point and the map reference line, and the horizontal distance between the verification position point and the map reference line;

[0039] A trend condition confirmation unit, configured to determine the trend types of the visual lane line and the map reference line and determine the pre-set trend verification conditions corresponding to the trend types when the position relationship between the visual lane line and the map reference line characterized by the position information is a preset position relationship, where the preset position relationship is approaching parallel, and the trend types include: the same trend, opposite trends and approaching each other, opposite trends and moving away from each other;

[0040] A trend judgment unit, configured to judge whether the position relationship between the visual lane line and the map reference line meets the trend verification condition;

[0041] A vehicle control unit, which is configured to use the visual lane line as the target driving route if the judgment result of the trend judgment unit is yes, and control the vehicle to drive according to the target communication route.

[0042] The third aspect of the present application provides a computer-readable storage medium carrying one or more computer programs. When the one or more computer programs are executed by a processor, the device where the computer-readable storage medium is located can implement the vehicle driving control method in the first aspect or any implementation manner of the first aspect.

[0043] As can be seen from the above technical solutions, for the vehicle driving control method provided in the embodiments of the present application, first, based on the position information of the visual end point and the intermediate verification position point of the visual lane line, the positional relationship between the visual lane line and the map reference line is judged to ensure that the visual lane line and the map reference line tend to be parallel, and to avoid incorrect directions of the lane lines selected at fork intersections. Further, when the visual lane line and the map reference line tend to be parallel, the trend types of the visual lane line and the map reference line are identified, and the trend of the visual lane line is verified according to the trend verification conditions corresponding to their trend types, so as to ensure the safety and correctness of the trend of the visual lane line, and screen out visual lane lines with problems such as lane line disorder and intersection.

[0044] Based on this, when the visual lane line simultaneously meets the above preset positional relationship and trend verification conditions, the visual lane line is selected as the target driving route and the vehicle is controlled to drive, which can not only ensure the correctness of the driving direction selected by the vehicle at the fork intersection, but also ensure that the vehicle drives according to the lane line visually perceived, provide a more accurate driving route for the vehicle, and avoid situations such as the vehicle crossing the line or deviating from the lane.

[0045] In summary, the present application verifies the positional relationship between the visual lane line and the map reference line, and selects a more suitable and relatively more accurate route from the two optional routes of the visual lane line and the map reference line to improve the safety of vehicle driving. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only the embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained according to the provided drawings.

[0047] Figure 1 It is a flowchart showing a method for implementing vehicle driving control provided by an embodiment of the present application;

[0048] Figure 2 A schematic diagram of the visual lane line at the fork provided by the embodiment of the present application;

[0049] Figure 3 An example diagram of the trend type provided by the embodiment of the present application;

[0050] Figure 4 An example diagram of the preset verification position point provided by the embodiment of the present application;

[0051] Figure 5 A schematic structural diagram of a vehicle driving control device provided by the embodiment of the present application. Specific embodiments

[0052] Next, the technical solutions in the embodiments of the present application will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present application.

[0053] The main function of the ramp set on the highway or urban expressway is to connect roads at different heights, and there are often Y-shaped forks for lane separation. During the navigation assisted driving process, if only relying on the lane lines perceived visually for tracking, the mutated lane lines will cause the navigation assisted driving system to snake, or even drive into the wrong lane. Moreover, the clarity of the lane lines perceived visually is unstable. Therefore, the navigation assisted driving system often ignores the lane lines perceived visually and only relies on the high-precision map and high-precision positioning to obtain the driving path in front of the vehicle and complete the fork passing task based on this driving path.

[0054] However, the embodiments of the present application consider that the accuracy of the high-precision map is limited by factors such as the update frequency of the high-precision map and the accuracy of the high-precision positioning, and completely ignoring the lane lines actually perceived visually. In case of emergencies, such as lane maintenance, sudden accidents, etc., the map may not provide sufficient instructions, resulting in a decline in the vehicle's actual perception and emergency handling capabilities, and being unable to quickly and correctly decide the correct driving path, thus affecting vehicle driving safety.

[0055] To solve the above problems, the embodiments of the present application provide a vehicle driving control method, which can be applied to navigation assisted driving systems, intelligent driving systems, autonomous driving systems, etc. Referring to Figure 1 , a flowchart of implementing the vehicle driving control method provided by the embodiment of the present application is used to illustrate the implementation process of the vehicle driving control method. The process specifically includes:

[0056] Step S110: Obtain the visual lane lines and map reference lines in front of the vehicle.

[0057] The visual lane lines are the lane lines in front of the vehicle visually perceived. Among them, the visually perceived lane lines can be obtained by collecting the image information in front of the vehicle through a camera installed on the vehicle body and analyzing the image information. The map reference line is the driving route provided by the high-precision map for the vehicle at the current moment.

[0058] Specifically, reference can be made to Figure 2 , a schematic diagram of the visual lane lines at a fork in the road provided by an embodiment of the present application. Generally, the map reference line represents highway guardrails, green belts, etc., and is used to distinguish lanes in different driving directions; the visual reference line represents white solid lines, arrows, etc. marked on the road surface, such as the lane dividing line for separating traffic flows moving in the same direction and the guiding lane line for indicating the driving direction. Figure 2 It is a schematic diagram of the visual lane lines at a Y-shaped fork. In fact, there should be two visual lane lines respectively pointing to intersection A and intersection B for the visual lane lines. However, during the vehicle driving process, the navigation assistance system will determine the only driving direction according to the pre-set driving route. In the embodiment of the present application, the only driving direction of the vehicle is the direction where intersection A is located. Therefore, the vehicle collects the image information in front of the vehicle and processes the image information to obtain the visual lane lines.

[0059] Step S120: Determine the verification position points between the visual end points of the visual lane lines and the position of the vehicle.

[0060] It can be understood that the vehicle camera can only collect the image information within a certain distance range in front of the vehicle. For example, with reference to Figure 2 , taking the vehicle coordinate system as a reference, the vehicle collects the visual lane lines from 0 to m in the longitudinal direction. Among them, the visual lane line at the horizontal line of the ordinate m can be used as the visual end point.

[0061] In the embodiment of the present application, to verify the consistency between the visual lane lines and the map reference lines, the positional relationships between the visual lane lines and the map reference lines at different points are respectively verified. Optionally, with reference to the figure, at least one position point is selected between the position of the vehicle (x = 0) and the visual end point (x = m) as the position point for verification. For example: the midpoint at a longitudinal distance x = m / 2 relative to the vehicle is used as the verification position point, or the visual lane line is trisected, and a position point is randomly determined at each segment as the verification position point to obtain three verification position points.

[0062] It can be understood that the more the number of verification position points, the more accurate the visual lane line inspection result. In the embodiments of the present application, the number, position of the verification position points, and the method of selecting the verification position points are not particularly limited, and can be adaptively selected according to the road conditions during actual application. For example, in the case of relatively complex road conditions, the number of verification position points can be adaptively increased to improve the accuracy of verification and ensure driving safety; while in the case of fewer vehicles and good road conditions, the number of verification position points can be adaptively reduced to improve the verification efficiency.

[0063] Step S130, obtain position information for characterizing the positional relationship between the visual lane line and the map reference line.

[0064] Among them, the position information at least includes: the horizontal distance between the visual end point and the map reference line, and the horizontal distance between the verification position point and the map reference line.

[0065] On the premise of ensuring the driving safety of the vehicle, it is necessary to verify whether the visual lane line is parallel to the map reference line to avoid the visual lane line intersecting with the map reference line, resulting in traffic accidents when the vehicle travels along the visual lane line.

[0066] To verify whether the positional relationship between the visual lane line and the map reference line is tending to be parallel, it can be judged by position information such as the lateral distance between the two lines at different points or the angle between the two lines. In the embodiments of the present application, the lateral distance S between the visual lane line and the map reference line on the horizontal line where the longitudinal distance m is located is obtained. Similarly, the lateral distance between the visual lane line and the map reference line on the horizontal line where the verification position point is located is obtained. Specifically, at a certain longitudinal distance relative to the vehicle, the lateral distance S between the visual lane line and the map reference line dis is calculated as shown in formula (1).

[0067] (1)

[0068] Among them, S v is the lateral coordinate corresponding to the visual lane line at the longitudinal distance x, and S m is the lateral coordinate corresponding to the map reference line at the longitudinal distance x, and abs is the absolute value calculation.

[0069] Optionally, the calculation methods of the lateral coordinate S of the visual lane line and the lateral coordinate S v of the map reference line, and the lateral coordinate S m corresponding to the map reference line at the longitudinal distance x can refer to formula (2).

[0070] (2)

[0071] Among them, x is the longitudinal distance, and C is the polynomial parameter.

[0072] Optionally, the position information characterizing the positional relationship between the visual lane line and the map reference line may further include: the slope difference, included angle, curvature, etc. between the visual lane line and the map reference line. It can be understood that in order to accurately verify the visual lane line, multiple pieces of position information can be selected for verification. The above is only an example and does not represent the only information that can be used for verification.

[0073] Step S140: Determine whether the positional relationship between the visual lane line characterized by the position information and the map reference line is a preset positional relationship. If so, execute step S150; if not, execute step S180.

[0074] The preset positional relationship is approaching parallel.

[0075] Based on the position information obtained in step S130, verify whether the positional relationship between the visual lane line and the map reference line is approaching parallel. Specifically, compare whether the horizontal distance between the verification position point and the horizontal distance of the visual end point are equal, or determine whether the horizontal distance of the verification position point is less than threshold 1 and the horizontal distance of the visual end point is less than threshold 2. Threshold 1 and threshold 2 are relatively close values. If the above conditions are met, it can be preliminarily determined that the visual lane line and the map reference line are approaching parallel. Further, execute step S150 to verify the trend relationship between the visual lane line and the map reference line. If the positional relationship between the visual lane line characterized by the position information and the map reference line does not belong to the preset positional relationship, execute step S180, and use the map reference line as the target driving path to control the vehicle to drive.

[0076] Step S150: Determine the trend types of the visual lane line and the map reference line, and determine the pre-set trend verification conditions corresponding to the trend types.

[0077] Step S160: Determine whether the positional relationship between the visual lane line and the map reference line meets the trend verification conditions. If so, execute step S170; if not, execute step S180.

[0078] Step S170: Use the visual lane line as the target driving route, and control the vehicle to drive according to the target communication route.

[0079] Step S180: Use the map reference line as the target driving route, and control the vehicle to drive according to the target driving route.

[0080] It can be understood that the above steps only verify the positional relationship between several position points of the visual lane line and the map reference line, and cannot observe the overall trend of the visual lane line. There may be a situation where the lateral distance and included angle between the visual lane line and the map lane line at position point 1 and position point 2 meet the preset requirements, but the visual lane line between position point 1 and position point 2 intersects with the map reference line, or the directions of the visual lane line and the map reference line are opposite, etc. Based on this, in order to achieve a comprehensive verification of the visual lane line, the embodiments of the present application also need to verify the visual lane line according to the trend types of the visual lane line and the map reference line.

[0081] Referring to Figure 3 , the exemplary diagrams of the trend types provided by the embodiments of the present application. Among them, observing the positional relationship between the map reference line and the visual lane line in two major categories: the fork in the road to the left (1, 2, 3, 4) and the fork in the road to the right (5, 6, 7, 8), the trend types of the map reference line and the visual lane line can be summarized as the same trend (2, 3, 6, 7), opposite trends and approaching each other (1, 5), and opposite trends and moving away from each other (4, 8).

[0082] Figure 3 These are the extreme cases under the above three trend types provided by the embodiments of the present application, that is, the visual lane lines intersect or the direction deviation is too large, which affects the driving safety of the vehicle. If the actual verification result is Figure 3 any one of them, then the visual lane line cannot be used as the target driving route, and the map reference line is used as the target driving route to control the vehicle to drive.

[0083] Specifically, the characteristics such as the end direction, end curvature, and curve separation trend of the visual lane line and the map reference line can be used to classify the trends of the visual lane line and the map reference line, and obtain the pre-set trend verification conditions corresponding to this trend type, and judge whether the trend between the visual lane line and the map reference line is Figure 3 one of them.

[0084] Optionally, the trend verification conditions can be set according to the characteristics of each trend type. For example, for the trend type with the same trend, the decisive factor affecting whether the visual lane line verification is successful is: whether the visual lane line and the map reference line intersect. Based on this, the trend verification condition corresponding to this trend type can be set as: between the vehicle's position and the visual end point, there is no intersection between the visual lane line and the map reference line.

[0085] For the trend types with opposite trends and moving away from each other, the decisive factor affecting whether the visual lane line is successfully verified is: the degree of opposite direction between the visual lane line and the map reference line (hereinafter simply referred to as the two lines). Based on this, the trend verification condition can be set as: the parameter condition for verifying the degree of opposite direction between the two lines. For example, the difference in curvature between the two lines is less than the preset difference, the lateral distance between the two lines is less than the preset distance, etc., which proves that the trends are opposite but the degree of opposite direction is small, and the visual lane line can still be used as the target driving route.

[0086] For the trend types with opposite trends and moving closer to each other, the decisive factor affecting whether the visual lane line is successfully verified is: the degree of closeness between the visual lane line and the map reference line, whether they intersect, resulting in a wrong fork direction of the visual lane line. Based on this, the trend verification condition corresponding to this trend type can be set as: the parameter condition for verifying the degree of closeness between the two lines, such as: whether there is an intersection between the two lines, whether the difference in curvature between the two lines is less than the preset difference, the lateral distance between the verification position points of the two lines is less than the preset threshold, etc., to prove that the degree of closeness between the two lines is small, the influence on the direction of the visual lane line is small, and it can be used as the target driving route.

[0087] Based on the above trend verification conditions, verify the trend relationship between the visual lane line and the map reference line collected at the current moment. If the verification is successful, that is, the positional relationship between the two lines meets the trend verification conditions, then execute step S170. If the verification fails, that is, the positional relationship between the two lines does not meet the trend verification conditions, then execute step S180. The navigation assisted driving system controls the driving vehicle to travel based on the finally determined target driving route.

[0088] In summary, for the vehicle driving control method provided by the embodiments of the present application, first, based on the position information of the visual end point and the intermediate verification position point of the visual lane line, judge the positional relationship between the visual lane line and the map reference line to ensure that the visual lane line and the map reference line tend to be parallel, and avoid the wrong direction of the lane line selected at the fork. Further, in the case where the visual lane line and the map reference line tend to be parallel, identify the trend types of the visual lane line and the map reference line, and verify the trend of the visual lane line according to the trend verification conditions corresponding to their trend types to ensure the safety and correctness of the trend of the visual lane line, and screen out the visual lane lines with problems such as lane line disorder and intersection.

[0089] Based on this, when the visual lane line simultaneously meets the above preset positional relationship and trend verification conditions, select the visual lane line as the target driving route and control the vehicle to travel, which can not only ensure the correctness of the driving direction selected by the vehicle at the fork, but also ensure that the vehicle travels according to the lane line visually perceived, provide a more accurate driving route for the vehicle, and avoid situations such as the vehicle pressing the line and deviating from the lane.

[0090] By verifying the positional relationship between the visual lane line and the map reference line, this application selects a more suitable and relatively more accurate route between the visual lane line and the map reference line among the two optional routes, improving the safety of vehicle driving.

[0091] Next, in combination with the following application embodiments, other possible implementations of the above vehicle driving control method will be described.

[0092] In a possible implementation, the process of selecting a verification position point between the visual end point of the visual lane line and the vehicle's location may include: obtaining at least one preset verification position point set in advance and the verification distance corresponding to the preset verification position point, where the verification distance is the distance between the preset verification position point and the vehicle's location; taking the maximum verification distance among the verification distances less than the visible distance of the lane line as the target verification distance, and the visible distance of the lane line is the distance between the visual end point and the vehicle's location; determining the preset verification position point corresponding to the target verification distance as the verification position point.

[0093] Refer to Figure 4 , the example diagram of the preset verification position point provided by the embodiment of this application, still taking the visual lane line and the map reference line of Figure 2 as an example, preset verification position points are set with verification distances 1, 2, and 3 respectively. Among them, the verification distances can be 30m, 50m, and 70m respectively. It can be understood that the setting of the verification distance can be set according to the accuracy requirements of the verification. As can be seen from Figure 4 , between the visual end point and the vehicle's location, there are preset verification position point 1 and preset verification position point 2. Both position points can be used as the verification position point, or either one can be selected as the verification position point.

[0094] In the embodiment of this application, the preset verification position point corresponding to the maximum verification distance less than the longitudinal distance at the visual end point is selected, that is, the preset verification position point closest to the visual end point is selected from the vehicle's location to the visual end point as the verification position point. Specifically, the maximum verification distance less than the longitudinal distance at the visual end point, that is, the target verification distance, can be determined with reference to formula (3), and the verification position point corresponding to the target verification distance can be determined accordingly. For example, the longitudinal distance corresponding to the visual end point is m, 30 < 50 < m, then 50m is the maximum verification distance less than the longitudinal distance at the visual end point, and preset verification position point 2 is used as the verification position point.

[0095] (3)

[0096] Among them, DIS is the set of verification distances, dis is the target verification distance, and VR is the visible distance of the lane line.

[0097] Further, position information representing the positional relationship between the visual lane line and the map reference line corresponding to the calibration position point and the visual end point is obtained respectively. Specifically, the implementation process may include: in the vehicle's ego - coordinate system, based on the visual end point of the visual lane line and the map reference line, determining the first lateral distance and the first slope difference between the visual end point and the map reference line as the end - point position information; in the vehicle's ego - coordinate system, based on the calibration position point of the visual lane line and the map reference line, determining the second lateral distance and the second slope difference between the calibration position point and the map reference line as the calibration position information.

[0098] The first lateral distance and the second lateral distance corresponding to the visual end point and the calibration position point of the visual lane line can be calculated with reference to the above formulas (1) and (2), which will not be elaborated in this example.

[0099] The first slope difference θ between the visual end point and the map reference line dis can be calculated with reference to formula (4), and formula (4) is as follows:

[0100] (4)

[0101] where θ v is the slope of the visual lane line at the longitudinal distance x, θ m is the slope of the map reference line at the longitudinal distance x, and abs is the absolute - value calculation.

[0102] Optionally, the calculation method of the slope θ v of the visual lane line at the longitudinal distance x and the slope θ m of the map reference line at the longitudinal distance x can be referred to formula (5).

[0103] (5)

[0104] where x is the longitudinal distance and C is the polynomial parameter.

[0105] It can be understood that the slope represents the inclination degree of the visual lane line or the map reference line, and can also indirectly represent the angle between the visual lane line or the map reference line and the horizontal line. Therefore, the slope difference between the two lines can characterize the relative inclination degree of the two lines and whether the condition that the two lines tend to be parallel is met.

[0106] The process of determining whether the positional relationship between the visual lane line and the map reference line is a preset positional relationship by using the above-mentioned end position information and calibration position information includes: determining two calibration distances adjacent to the visible distance of the lane line, and determining two preset calibration position points corresponding to the two calibration distances, respectively serving as reference calibration position points; obtaining a threshold group corresponding to each preset inspection position point preset in advance, and the threshold group at least includes: a preset lateral distance threshold and a preset slope threshold; determining an intermediate calibration threshold group corresponding to the calibration position point, and the intermediate calibration threshold group includes: an intermediate lateral distance threshold and an intermediate slope threshold; based on the linear compensation principle, determining an end threshold group corresponding to the visible distance of the lane line based on the threshold groups corresponding to the two reference calibration position points, and the end threshold group includes: an end lateral distance threshold and an end slope threshold; determining whether the first lateral distance in the end position information is less than the end lateral distance threshold, and whether the first slope difference is less than the end slope threshold; determining whether the second lateral distance in the calibration position information is less than the intermediate lateral distance threshold, and whether the second slope difference is less than the intermediate slope threshold.

[0107] According to the path characteristics of the Y-shaped intersection collected historically, the range values of the lateral distance and the slope between the visual lane line and the map reference line are respectively set at different longitudinal distances of the vehicle, and based on this range value, a lateral distance threshold and a slope threshold corresponding to the lateral distance and the slope are set respectively. For example, for the collected standard visual lane line and map reference line historically, at the calibration distance of 70m, the range of the slope difference is 0.1 - 0.05, and the range of the lateral distance is 3.75 - 0. Based on this, the slope threshold can be set to 0.1, and the lateral distance threshold can be set to 3.75, which are used as the threshold group of the preset calibration position point corresponding to the calibration distance of 70m.

[0108] Based on this, obtain the threshold group corresponding to each preset calibration position point / inspection distance preset in advance. First, the threshold group of the calibration position point can be determined therefrom as the intermediate calibration threshold group. Further, determine two calibration distances adjacent to the visible distance of the lane line, referring to Figure 4 , the calibration distances adjacent to the visual end are calibration distance 2 and calibration distance 3. Based on the threshold groups corresponding to calibration distance 2 and calibration distance 3, and combining the linear step principle, determine the end threshold group corresponding to the visible distance of the lane line. Specifically, the end lateral distance threshold in the end threshold group can be calculated with reference to formula (6) and the end slope threshold .

[0109] (6)

[0110] where dis1 and dis2 are respectively the two calibration distances adjacent to the visual end, and To verify the preset horizontal distance threshold and preset slope threshold corresponding to the distance dis1, and to verify the preset horizontal distance threshold and preset slope threshold corresponding to the distance dis2, where VR is the visible distance of the lane line.

[0111] Furthermore, based on the obtained verification position information, end position information, intermediate verification threshold group, and end threshold group, determine whether the positional relationship between the two lines is the preset positional relationship. Specifically, determine whether the second horizontal distance in the verification position information is less than the intermediate horizontal distance threshold, and whether the second slope difference is less than the intermediate slope threshold. At the same time, determine whether the first horizontal distance in the end position information is less than the end horizontal distance threshold, and whether the first slope difference is less than the end slope threshold. When the judgment results of the above judgment steps are all yes, determine that the positional relationship between the two lines is the preset positional relationship.

[0112] In a possible implementation, when the positional relationship between the two lines is the preset positional relationship, identify the curvatures corresponding to the visual lane line and the map reference line of the visual end point, and obtain the visual lane line curvature and the map reference line curvature; based on the visual lane line curvature and the map reference line curvature, determine the trend type of the visual lane line relative to the map reference line; obtain the preset trend verification conditions corresponding to the trend type.

[0113] In the embodiments of the present application, the curvatures of the visual lane line and the map reference line at the visual end point are used to respectively determine the trends of the two lines, thereby determining the corresponding trend types of the two lines, and obtaining the trend verification conditions corresponding to the trend types.

[0114] Similarly, according to the path characteristics of the Y-shaped intersection collected historically, determine Figure 3 the occurrence conditions corresponding to each trend type in, and thereby reversely determine the trend verification conditions. For example: Historically collected, the trends of the visual lane line and the map reference line are opposite and far away, the difference between the lateral distance of the vehicle's position and the lateral distance at the visual end point is more than 3m, and at the visual end point, the curvature difference between the two lines is greater than 0.004. Reversely, determine the trend verification conditions corresponding to the trend type of opposite and far away trends as: the difference between the lateral distance of the vehicle's position and the lateral distance at the visual end point is not greater than 3m, and at the visual end point, the curvature difference between the two lines is not greater than 0.004.

[0115] Based on the above method, set the trend verification conditions corresponding to different trend types respectively. The specific trend verification conditions may include:

[0116] When the trend type is the same trend, the trend verification conditions include: within the preset range of the visual end point in the vehicle's own coordinate system, there is no intersection between the visual lane line and the map reference line.

[0117] When the trend types are opposite and approaching each other, the trend verification conditions include: within the interval between the vehicle's location and the visual end point, there is no intersection between the visual lane line and the map reference line, and the first horizontal distance is greater than a preset threshold, where the first horizontal distance is the horizontal distance between the visual end point and the map reference line.

[0118] When the trend types are opposite and moving away from each other, the trend verification conditions include: the difference between the first horizontal distance and the third horizontal distance is not greater than a preset threshold, and the degree of separation between the visual end point and the map reference line meets a preset condition.

[0119] Optionally, in the embodiments of the present application, the degree of separation between the visual end point and the map reference line can be represented by the difference in the second derivatives of the two lines at the visual end point to indicate the classification degree.

[0120] In summary, the embodiments of the present application verify the visual lane line from two aspects, namely the positional relationship between the two lines and the trend between the two lines, so as to determine whether the visual lane line matches the map lane line and ensure the correct direction of the visual lane line.

[0121] Next, the vehicle driving control device provided by the embodiments of the present application will be described. The vehicle driving control device described below can be correspondingly referred to the vehicle driving control method described above.

[0122] First, in combination with Figure 5 , the vehicle driving control device will be introduced. As Figure 5 shown, the vehicle driving control device may include:

[0123] A line acquisition unit 100 for acquiring the visual lane line and the map reference line in front of the vehicle;

[0124] A verification point determination unit 200 for determining a verification position point located between the visual end point of the visual lane line and the vehicle's location;

[0125] A position information acquisition unit 300 for acquiring position information for characterizing the positional relationship between the visual lane line and the map reference line, where the position information at least includes: the horizontal distance between the visual end point and the map reference line, and the horizontal distance between the verification position point and the map reference line;

[0126] A trend condition confirmation unit 400 is configured to determine the trend types of the visual lane line and the map reference line and determine a pre-set trend verification condition corresponding to the trend type when the positional relationship between the visual lane line represented by the position information and the map reference line is a pre-set positional relationship, where the pre-set positional relationship is approaching parallel, and the trend types include: same trend, opposite trends and approaching each other, and opposite trends and moving away from each other.

[0127] A trend judgment unit 500 is configured to judge whether the positional relationship between the visual lane line and the map reference line meets the trend verification condition.

[0128] A vehicle control unit 600 is configured to, if the judgment result of the trend judgment unit is yes, use the visual lane line as the target driving route and control the vehicle to travel according to the target communication route.

[0129] In summary, the vehicle driving control device provided in the embodiment of the present application first judges the positional relationship between the visual lane line and the map reference line based on the position information of the visual end point of the visual lane line and the verification position point in the middle, ensuring that the visual lane line and the map reference line tend to be parallel and avoiding incorrect directions of the lane lines selected at the fork. Further, when the visual lane line and the map reference line tend to be parallel, identify the trend types of the visual lane line and the map reference line, and verify the trend of the visual lane line according to the trend verification condition corresponding to the trend type, so as to ensure the safety and correctness of the trend of the visual lane line and screen out the visual lane lines with problems such as lane line disorder and intersection.

[0130] Based on this, when the visual lane line simultaneously meets the above pre-set positional relationship and trend verification condition, select the visual lane line as the target driving route and control the vehicle to travel, which can not only ensure the correctness of the driving direction selected by the vehicle at the fork, but also ensure that the vehicle travels according to the lane line visually perceived, provide a more accurate driving route for the vehicle, and avoid situations such as the vehicle crossing the line or deviating from the lane.

[0131] In summary, the present application verifies the positional relationship between the visual lane line and the map reference line, and selects a more suitable and relatively more accurate route in the two optional routes of the visual lane line and the map reference line, improving the safety of vehicle driving.

[0132] In a possible implementation, the vehicle driving control device may further include:

[0133] When the positional relationship between the visual lane line represented by the position information and the map reference line does not belong to the preset positional relationship, or when the positional relationship between the visual lane line and the map reference line does not meet the trend verification condition, the map reference line is used as the target driving route, and the vehicle is controlled to drive according to the target driving route.

[0134] In a possible implementation, the verification point determination unit 200 includes:

[0135] A preset position point acquisition subunit, configured to acquire at least one preset verification position point set in advance, and a verification distance corresponding to the preset verification position point, where the verification distance is the distance between the preset verification position point and the position of the vehicle;

[0136] A target verification distance determination subunit, configured to use the maximum verification distance among the verification distances less than the visible distance of the lane line as the target verification distance, where the visible distance of the lane line is the distance between the visual end point and the position of the vehicle;

[0137] A verification position point determination subunit, configured to determine the preset verification position point corresponding to the target verification distance as the verification position point.

[0138] In a possible implementation, the position information acquisition unit 300 includes:

[0139] An end point position information determination subunit, configured to determine, in the vehicle's ego - coordinate system, a first lateral distance and a first slope difference between the visual end point of the visual lane line and the map reference line as the end point position information;

[0140] A verification position information determination subunit, configured to determine, in the vehicle's ego - coordinate system, a second lateral distance and a second slope difference between the verification position point of the visual lane line and the map reference line as the verification position information.

[0141] In a possible implementation, the functional unit for determining that the positional relationship between the visual lane line represented by the position information and the map reference line is a preset positional relationship includes:

[0142] A reference verification position point determination subunit, configured to determine two verification distances adjacent to the visible distance of the lane line, and determine two preset verification position points corresponding to the two verification distances as reference verification position points respectively;

[0143] A threshold group acquisition subunit, configured to acquire a threshold group corresponding to each of the preset inspection position points set in advance, where the threshold group at least includes: a preset horizontal distance threshold and a preset slope threshold;

[0144] An intermediate verification threshold group determination subunit, configured to determine an intermediate verification threshold group corresponding to the verification position point, where the intermediate verification threshold group includes: an intermediate horizontal distance threshold and an intermediate slope threshold;

[0145] An end point threshold group determination subunit, configured to determine an end point threshold group corresponding to the visible distance of the lane line based on the linear compensation principle and based on the threshold groups corresponding to two reference verification position points, where the end point threshold group includes: an end point horizontal distance threshold and an end point slope threshold;

[0146] A first judgment subunit, configured to judge whether the first horizontal distance in the end point position information is less than the end point horizontal distance threshold and whether the first slope difference is less than the end point slope threshold;

[0147] A second judgment subunit, configured to judge whether the second horizontal distance in the verification position information is less than the intermediate horizontal distance threshold and whether the second slope difference is less than the intermediate slope threshold.

[0148] In a possible implementation, the trend condition confirmation unit 400 includes:

[0149] A curvature identification subunit, configured to identify the curvatures corresponding to the visual lane line and the map reference line of the visual end point, and obtain a visual lane line curvature and a map reference line curvature;

[0150] A trend type determination subunit, configured to determine a trend type of the visual lane line relative to the map reference line based on the visual lane line curvature and the map reference line curvature;

[0151] A trend condition acquisition subunit, configured to acquire a pre-set trend verification condition corresponding to the trend type.

[0152] In a possible implementation, when the trend type is the same trend, the trend verification condition includes: in the vehicle's own vehicle coordinate system, within a preset range of the visual end point, there is no intersection between the visual lane line and the map reference line.

[0153] In a possible implementation, when the trend types are opposite and approaching each other, the trend verification condition includes: within the interval between the position of the vehicle and the visual end point, there is no intersection between the visual lane line and the map reference line, and the first lateral distance is greater than a preset threshold, where the first lateral distance is the lateral distance between the visual end point and the map reference line.

[0154] In a possible implementation, when the trend types are opposite and moving away from each other, the trend verification condition includes: the difference between the first lateral distance and the third lateral distance is not greater than a preset threshold, and the degree of separation between the visual end point and the map reference line meets a preset condition.

[0155] A third aspect of the present application provides a computer-readable storage medium carrying one or more computer programs, which, when executed by a processor, can enable the device where the computer-readable storage medium is located to implement the vehicle driving control method according to the first aspect or any implementation manner of the first aspect above.

[0156] Finally, it should also be noted that in this article, 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 also includes 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 element.

[0157] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts among the various embodiments can be referred to each other.

[0158] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. A vehicle driving control method, characterized in that, Including: Obtain a visual lane line and a map reference line in front of the vehicle; Determine a calibration position point between the visual end point of the visual lane line and the position of the vehicle; Obtain position information for characterizing the positional relationship between the visual lane line and the map reference line, where the position information at least includes: the horizontal distance between the visual end point and the map reference line, and the horizontal distance between the calibration position point and the map reference line; When the positional relationship between the visual lane line and the map reference line characterized by the position information is a preset positional relationship, determine the trend types of the visual lane line and the map reference line, and determine a preset trend calibration condition corresponding to the trend type. The preset positional relationship is approaching parallel, and the trend types include: the same trend, opposite trends and approaching each other, opposite trends and moving away from each other; Judge whether the positional relationship between the visual lane line and the map reference line meets the trend calibration condition; If the positional relationship between the visual lane line and the map reference line meets the trend calibration condition, use the visual lane line as the target driving route, and control the vehicle to drive according to the target communication route.

2. The vehicle driving control method according to claim 1, characterized in that, Also including: When the positional relationship between the visual lane line and the map reference line characterized by the position information does not belong to the preset positional relationship, or when the positional relationship between the visual lane line and the map reference line does not meet the trend calibration condition, use the map reference line as the target driving route, and control the vehicle to drive according to the target driving route.

3. The vehicle driving control method according to claim 1, characterized in that The determining of the calibration position point between the visual end point of the visual lane line and the position of the vehicle includes: Obtain at least one preset calibration position point set in advance, and the calibration distance corresponding to the preset calibration position point. The calibration distance is the distance between the preset calibration position point and the position of the vehicle; Use the maximum calibration distance among the calibration distances less than the visible distance of the lane line as the target calibration distance. The visible distance of the lane line is the distance between the visual end point and the position of the vehicle; Determine the preset calibration position point corresponding to the target calibration distance as the calibration position point.

4. The vehicle driving control method according to claim 3, wherein The obtaining of the position information for characterizing the positional relationship between the visual lane line and the map reference line includes: In the vehicle's own vehicle coordinate system, based on the visual end point of the visual lane line and the map reference line, determine the first lateral distance and the first slope difference between the visual end point and the map reference line as the end point position information; In the vehicle's own vehicle coordinate system, based on the calibration position point of the visual lane line and the map reference line, determine the second lateral distance and the second slope difference between the calibration position point and the map reference line as the calibration position information.

5. The vehicle driving control method according to claim 4, wherein The process of judging that the positional relationship between the visual lane line and the map reference line characterized by the position information is a preset positional relationship includes: Determine two calibration distances adjacent to the visible distance of the lane line, and determine two preset calibration position points corresponding to the two calibration distances, respectively serving as reference calibration position points; Obtain the threshold group corresponding to each of the preset inspection position points set in advance, where the threshold group at least includes: a preset lateral distance threshold and a preset slope threshold; Determine the intermediate calibration threshold group corresponding to the calibration position point, where the intermediate calibration threshold group includes: an intermediate lateral distance threshold and an intermediate slope threshold; Based on the linear compensation principle, determine the end threshold group corresponding to the visible distance of the lane line based on the threshold groups corresponding to the two reference calibration position points, where the end threshold group includes: an end lateral distance threshold and an end slope threshold; Judge whether the first lateral distance in the end position information is less than the end lateral distance threshold, and whether the first slope difference is less than the end slope threshold; Judge whether the second lateral distance in the calibration position information is less than the intermediate lateral distance threshold, and whether the second slope difference is less than the intermediate slope threshold.

6. The vehicle driving control method according to claim 1, wherein, The determining the trend calibration condition corresponding to the trend type according to the trends of the visual lane line and the map reference line includes: Identify the curvatures corresponding to the visual lane line and the map reference line of the visual end point, and obtain the visual lane line curvature and the map reference line curvature; Based on the visual lane line curvature and the map reference line curvature, determine the trend type of the visual lane line relative to the map reference line; Obtain the trend calibration condition corresponding to the trend type set in advance.

7. The vehicle driving control method according to claim 6, wherein, In the case where the trend type is the same trend, the trend calibration condition includes: within the preset range of the visual end point in the vehicle's own coordinate system, there is no intersection between the visual lane line and the map reference line.

8. The vehicle driving control method according to claim 6, wherein In the case where the trend type is opposite and approaching each other, the trend calibration condition includes: within the interval between the vehicle's position and the visual end point, there is no intersection between the visual lane line and the map reference line, and the first lateral distance is greater than a preset threshold, where the first lateral distance is the lateral distance between the visual end point and the map reference line; In the case where the trend type is opposite and moving away from each other, the trend calibration condition includes: the difference between the first lateral distance and the third lateral distance is not greater than a preset threshold, and the separation degree between the visual end point and the map reference line satisfies a preset condition.

9. A vehicle driving control device, characterized in that, Includes: A line acquisition unit for acquiring the visual lane line and the map reference line in front of the vehicle; A calibration point determination unit for determining a calibration position point between the visual end point of the visual lane line and the vehicle's position; A position information acquisition unit for acquiring position information used to characterize the positional relationship between the visual lane line and the map reference line, where the position information at least includes: the horizontal distance between the visual end point and the map reference line, and the horizontal distance between the verification position point and the map reference line; A trend condition confirmation unit for determining the trend types of the visual lane line and the map reference line and determining the pre-set trend verification conditions corresponding to the trend types when the positional relationship between the visual lane line and the map reference line characterized by the position information is a preset positional relationship, where the preset positional relationship is approaching parallel, and the trend types include: the same trend, opposite trends and approaching each other, opposite trends and moving away from each other; A trend judgment unit for judging whether the positional relationship between the visual lane line and the map reference line meets the trend verification conditions; A vehicle control unit for, if the judgment result of the trend judgment unit is yes, using the visual lane line as the target driving route and controlling the vehicle to travel according to the target communication route.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium carries one or more computer programs, and when the one or more computer programs are executed by a processor, the device where the computer-readable storage medium is located can implement the vehicle driving control method as described in any one of claims 1 to 8.

Images