A lane line detection method, device, electronic equipment and storage medium

Abstract

The application relates to a lane line detection method, device, equipment and storage medium, the method comprising: acquiring a plurality of first lane line equations and a plurality of second lane line equations in the vehicle driving process, determining a fusion lane line curvature value and a fusion lane line curvature change rate value based on the plurality of first lane line equations and the plurality of second lane line equations, acquiring a real-time front-view lane line equation in the vehicle driving process, and determining a target lane line equation according to the real-time front-view lane line equation, the fusion lane line curvature value and the fusion lane line curvature change rate value; the application logically fuses the lane line equation determined by a high-precision map and the lane line equation determined by a front-view camera, takes the lane line equation recognized by the camera as the main part, takes the high-precision map lane line as the auxiliary part, optimizes the lane line recognized by the camera, can obtain a more accurate lane line equation, thereby providing a more reliable lane line for vehicle planning control, and making the driving experience of the user better.

Description

Technical Field

[0001] This invention relates to the field of automotive technology, and more specifically to a lane line detection method, device, electronic device, and storage medium. Background Technology

[0002] With the advancement of technology, automobiles have undergone revolutionary changes, and intelligent vehicle technology has been widely applied. Lane perception is a crucial component of intelligent driving. To ensure stable driving on the road, intelligent driving vehicles need to acquire accurate lane lines. In the development of intelligent driving perception technology, high-precision maps and high-precision positioning technology have been widely applied to intelligent driving vehicles to assist in accurate lane line recognition. However, existing intelligent driving vehicles over-rely on high-precision maps. When relying too heavily on high-precision map data, errors in the lane line data can cause abnormalities in vehicle planning and control. Summary of the Invention

[0003] In view of the above problems, embodiments of the present invention are proposed to provide a lane line detection method, apparatus, electronic device and storage medium that overcomes or at least partially solves the above problems.

[0004] To address the above problems, this invention discloses a lane line detection method, the method comprising:

[0005] Multiple first lane line equations are obtained during vehicle driving; the multiple first lane line equations are determined based on multiple sets of historical lane line data stored in a high-precision map.

[0006] Multiple second lane line equations are obtained during the vehicle's driving process; the multiple second lane line equations are determined based on multiple sets of historical lane line coordinate points collected by the forward-looking camera.

[0007] Based on the plurality of first lane line equations and the plurality of second lane line equations, determine the fused lane line curvature value and the fused lane line curvature change rate value;

[0008] The real-time forward-looking lane line equation is obtained during the vehicle's driving process; the real-time forward-looking lane line equation is determined based on the real-time lane line coordinate points collected by the forward-looking camera.

[0009] The target lane line equation is determined based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value.

[0010] Optionally, determining the target lane line equation based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value includes:

[0011] The offset distance of the rear axle center of the real-time forward-looking vehicle from the lane line and the real-time lane line yaw angle are determined from the real-time forward-looking lane line equation.

[0012] The target lane line equation is obtained by fitting the offset distance between the rear axle center of the real-time forward-looking vehicle and the lane line, the real-time lane line yaw angle, the fused lane line curvature value, and the fused lane line curvature change rate value.

[0013] Optionally, determining the fused lane line curvature value based on the plurality of first lane line equations and the plurality of second lane line equations includes:

[0014] Based on the plurality of first lane line equations, a plurality of first lane line curvature values ​​are obtained;

[0015] Based on the multiple second lane line equations, multiple second lane line curvature values ​​are obtained;

[0016] The standard deviation of lane line curvature is calculated based on the plurality of first lane line curvature values ​​and the plurality of second lane line curvature values.

[0017] The fused lane line curvature value is determined based on the standard deviation of the lane line curvature.

[0018] Optionally, determining the fused lane line curvature value based on the standard deviation of lane line curvature includes:

[0019] The real-time high-precision lane line equation is obtained during the vehicle's driving process. The real-time high-precision lane line equation is determined based on the real-time lane line data stored in the high-precision map.

[0020] Obtain the real-time forward lane curvature value of the real-time forward lane equation and the real-time high-precision lane curvature value of the real-time high-precision lane equation.

[0021] If the real-time forward-looking lane curvature value is greater than the real-time high-precision lane curvature value, then the negative number of the standard deviation of the lane curvature is determined as the lane curvature compensation amount.

[0022] Based on the lane line curvature compensation amount and the real-time forward-looking lane line curvature value, the fused lane line curvature value is determined.

[0023] Optionally, the method further includes:

[0024] If the real-time forward-looking lane curvature value is less than the real-time high-precision lane curvature value, then the standard deviation of the lane curvature is determined as the lane curvature compensation amount.

[0025] Optionally, determining the fused lane line curvature change rate value based on the plurality of first lane line equations and the plurality of second lane line equations includes:

[0026] Based on the multiple first lane line equations, multiple first lane line curvature change rate values ​​are obtained;

[0027] Based on the multiple second lane line equations, multiple second lane line curvature change rate values ​​are obtained;

[0028] The standard deviation of the lane curvature change rate is calculated based on the plurality of first lane line curvature change rate values ​​and the plurality of second lane line curvature change rate values.

[0029] The fused lane curvature change rate value is determined based on the standard deviation of the lane curvature change rate.

[0030] Optionally, determining the fused lane curvature change rate value based on the standard deviation of the lane curvature change rate includes:

[0031] The real-time high-precision lane line equation is obtained during the vehicle's driving process. The real-time high-precision lane line equation is determined based on the real-time lane line data stored in the high-precision map.

[0032] Obtain the real-time forward lane curvature change rate value of the real-time forward lane line equation and the real-time high-precision lane line curvature change rate value of the real-time high-precision lane line equation.

[0033] If the real-time forward-looking lane curvature change rate value is greater than the real-time high-precision lane curvature change rate value, then the negative number of the standard deviation of the lane curvature change rate is determined as the lane curvature change rate compensation amount.

[0034] Based on the lane line curvature change rate compensation amount and the real-time forward-looking lane line curvature change rate value, the fused lane line curvature change rate value is determined.

[0035] Optionally, the method further includes:

[0036] If the real-time forward-looking lane curvature change rate value is less than the real-time high-precision lane curvature change rate value, then the standard deviation of the lane curvature change rate is determined as the lane curvature change rate compensation amount.

[0037] Optionally, the method further includes:

[0038] If the number of second lane line equations obtained in the same time period is less than the number of first lane line equations, then the first lane line equation corresponding to the missing second lane line equation is determined as the reference lane line equation.

[0039] Obtain the high-precision lane line equations adjacent to the reference lane line equation;

[0040] Based on the reference lane line equation and the adjacent high-precision lane line equation, determine the reference lane line fusion curvature change rate value and the reference lane line fusion curvature value;

[0041] The missing second lane line equation is obtained by fitting the reference lane line equation, the reference lane line fusion curvature change rate value, and the reference lane line fusion curvature value.

[0042] The present invention also discloses a lane line detection device, the device comprising:

[0043] The first acquisition module is used to acquire multiple first lane line equations during vehicle driving; the multiple first lane line equations are determined based on multiple sets of historical lane line data stored in a high-precision map.

[0044] The second acquisition module is used to acquire multiple second lane line equations during the vehicle's driving process; the multiple second lane line equations are determined based on multiple sets of historical lane line coordinate points collected by the forward-looking camera.

[0045] The first determining module is used to determine the fused lane line curvature value and the fused lane line curvature change rate value based on the plurality of first lane line equations and the plurality of second lane line equations;

[0046] The third acquisition module is used to acquire the real-time forward lane line equation at the current moment during the vehicle's driving process; the real-time forward lane line equation is determined based on the real-time lane line coordinate points collected by the forward-looking camera at the current moment.

[0047] The first determining module is used to determine the target lane line equation based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value.

[0048] The present invention also discloses an electronic device, comprising: a processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the steps of the lane line detection method as described above.

[0049] The present invention also discloses a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the lane line detection method described above.

[0050] This invention logically fuses lane line equations determined by a high-precision map and those determined by a forward-looking camera. The lane line equations identified by the forward-looking camera are used as the primary equations, while those from the high-precision map are used as secondary equations. By optimizing the real-time lane lines identified by the forward-looking camera, more accurate lane line equations can be obtained. This provides more reliable lane lines for vehicle planning and control, and also improves the user's driving experience. Attached Figure Description

[0051] Figure 1 This is a flowchart of the steps of a lane line detection method provided in an embodiment of the present invention;

[0052] Figure 2 This is a flowchart of another lane line detection method provided in an embodiment of the present invention;

[0053] Figure 3 This is a flowchart of the steps of a lane line detection device provided in an embodiment of the present invention. Detailed Implementation

[0054] The embodiments of the present invention will be described below with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention and not for limiting the scope of protection of the present invention.

[0055] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. Therefore, the drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0056] In existing technologies, intelligent driving vehicles rely excessively on high-precision maps to identify lane lines during driving. However, high-precision maps may contain incorrect lane line data, leading to abnormal vehicle planning and control.

[0057] One of the core concepts of this invention is that by logically fusing the lane line equations determined by the high-precision map and the lane line equations determined by the forward-looking camera, with the lane line equations identified by the forward-looking camera as the main component and the lane line equations from the high-precision map as the auxiliary component, the real-time lane lines identified by the forward-looking camera are optimized to obtain more accurate lane line equations. This provides more reliable lane lines for vehicle planning and control, and also improves the user's driving experience.

[0058] Reference Figure 1 The present invention provides a flowchart of a lane line detection method according to an embodiment of the present invention, the method including the following steps:

[0059] Step 101: Obtain multiple first lane line equations during vehicle driving; the multiple first lane line equations are determined based on multiple sets of historical lane line data stored in the high-precision map.

[0060] In this embodiment of the invention, historical lane line data refers to lane lines traversed during a historical time period during vehicle travel. During vehicle travel, multiple first lane line equations can be determined based on multiple sets of historical lane line data from a high-precision map. In one example, if the current time is 10:00 AM, five sets of historical lane line data from 9:59 AM to 10:00 AM can be obtained. Based on these five sets of historical lane line data for this time period, five sets of first lane line equations can be determined. The five sets of first lane line equations are as follows:

[0061] y = A1 + B1*x + C1*x 2 +D1*x 3 , formula (1)

[0062] y = A² + B²x + C²x 2 +D2*x 3 , formula (2)

[0063] y = A³ + B³x + C³x 2 +D3*x 3 , formula (3)

[0064] y = A4 + B4*x + C4*x 2 +D4*x 3 , formula (4)

[0065] y = A5 + B5*x + C5*x 2 +D5*x 3 , formula (5)

[0066] Where A represents the offset distance between the rear axle center of the vehicle and the lane line, B represents the lane line yaw angle, C represents the lane line curvature value, D represents the lane line curvature change rate, y is the horizontal axis of the rear axle center coordinate, and x is the vertical axis of the rear axle center coordinate.

[0067] It should be noted that the number of sets of historical lane line data can be set according to the user's needs. The more sets there are, the more accurate the final lane line equation will be.

[0068] Step 102: Obtain multiple second lane line equations during vehicle driving; the multiple second lane line equations are determined by multiple sets of historical lane line coordinate points collected by the forward-looking camera.

[0069] In this embodiment of the invention, historical lane line coordinate points refer to the coordinate point data corresponding to lane lines traversed during a historical time period during the vehicle's driving process. The forward-looking camera can collect multiple sets of historical lane line coordinate points during the vehicle's driving process, and then identify the corresponding lane line equation based on each set of coordinate point data. In one example, if the current time is 10:00, the forward-looking camera can collect 5 sets of historical lane line coordinate points from 9:59 to 10:00, and then identify 5 sets of second lane line equations based on these 5 sets of historical lane line coordinate points. The 5 sets of second lane line equations are as follows:

[0070] y = A6 + B6*x + C6*x 2 +D6*x 3 , formula (6)

[0071] y = A7 + B7*x + C7*x 2 +D7*x 3 , formula (7)

[0072] y = A8 + B8*x + C8*x 2 +D8*x 3 , formula (8)

[0073] y = A9 + B9*x + C9*x 2 +D9*x 3 , formula (9)

[0074] y = A 10 +B 10 *x+C 10 *x 2 +D 10 *x 3 , formula (10)

[0075] It should be noted that the forward-facing camera uses a deep learning algorithm for recognition, but the specific type of deep learning algorithm is not specified here.

[0076] Step 103: Based on multiple first lane line equations and multiple second lane line equations, determine the fused lane line curvature value and the fused lane line curvature change rate value.

[0077] In this embodiment of the invention, a fused lane line curvature value C can be obtained by performing a fusion calculation based on multiple first lane line equations and multiple second lane line equations. 融 and the rate of change of lane curvature D 融 This makes the lane curvature values ​​and lane curvature change rate values ​​more reliable.

[0078] Step 104: Obtain the real-time forward-looking lane line equation during vehicle movement; the real-time forward-looking lane line equation is determined based on the real-time lane line coordinate points collected by the forward-looking camera.

[0079] In this embodiment of the invention, the real-time forward-looking lane line equation refers to the lane line equation determined in real time by the vehicle's forward-looking camera. In one example, if the current time is 10:00, the forward-looking camera can collect a set of real-time lane line coordinate points at 10:00, and then identify a set of real-time forward-looking lane line equations based on this set of real-time lane line coordinate points. The corresponding real-time forward-looking lane line equation is:

[0080] y = A 11 +B 11 *x+C 11 *x 2 +D 11 *x 3 , formula (11)

[0081] Step 105: Determine the target lane line equation based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value.

[0082] In this embodiment of the invention, the target lane line equation refers to the lane line equation obtained by fusing the first lane line equation and the second lane line equation according to the user's needs; it can be based on the obtained real-time forward-looking lane line equation (as shown in formula (11)), C 融 D 融 To obtain the target lane line equation, in one example, C in equation (11) 11 Replace with C 融 D 11 Replace with D 融 The resulting equation for the target lane line is:

[0083] y 目标 =A 11 +B 11 *x+C 融 *x 2 +D 融合 *x 3 , formula (12)

[0084] This invention logically fuses lane line equations determined by a high-precision map and those determined by a forward-looking camera. The lane line equations identified by the forward-looking camera are used as the primary equations, while those from the high-precision map are used as secondary equations. By optimizing the real-time lane lines identified by the forward-looking camera, more accurate lane line equations can be obtained. This provides more reliable lane lines for vehicle planning and control, and also improves the user's driving experience.

[0085] Reference Figure 2The diagram illustrates a flowchart of another lane line detection method provided by an embodiment of the present invention. The method may include the following steps:

[0086] Step 201: Obtain multiple first lane line equations during vehicle driving; the multiple first lane line equations are determined based on multiple sets of historical lane line data stored in the high-precision map.

[0087] Step 202: Obtain multiple second lane line equations during vehicle driving; the multiple second lane line equations are determined by multiple sets of historical lane line coordinate points collected by the forward-looking camera.

[0088] Step 203: Based on multiple first lane line equations and multiple second lane line equations, determine the fused lane line curvature value and the fused lane line curvature change rate value.

[0089] In one embodiment of the present invention, determining the fused lane line curvature value based on multiple first lane line equations and multiple second lane line equations may include:

[0090] Based on multiple first lane line equations, multiple first lane line curvature values ​​are obtained; based on multiple second lane line equations, multiple second lane line curvature values ​​are obtained; based on the multiple first lane line curvature values ​​and multiple second lane line curvature values, the standard deviation of lane line curvature is calculated; based on the standard deviation of lane line curvature, the fused lane line curvature value is determined.

[0091] Specifically, the multiple first lane line equations are given by formulas (1), (2), (3), (4), and (5), resulting in five sets of first lane line curvature values: C1, C2, C3, C4, and C5. Similarly, the multiple second lane line equations are given by formulas (6), (7), (8), (9), and (10), resulting in five sets of second lane line curvature values: C6, C7, C8, C9, and C1. 10 Therefore, it can be determined based on C1, C2, C3, C4, C5, C6, C7, C8, C9, C 10 The standard deviation of lane curvature σ1 is calculated, and then the fused lane curvature value C can be determined based on σ1. 融 The formula for calculating σ1 is:

[0092]

[0093] Where σ1 is the standard deviation of lane curvature, n represents the index of the lane curvature value, xi is the lane curvature value with index i, and μ is the average of multiple lane curvature values, i.e., μ = (C1 + C2 + C3 + C4 + C5 + C6 + C7 + C8 + C9 + C 10 ) / 10.

[0094] In one embodiment of the present invention, determining the fused lane line curvature value based on the standard deviation of lane line curvature may include:

[0095] The system acquires real-time high-precision lane line equations during vehicle operation, which are determined based on real-time lane line data stored in a high-precision map. It also acquires the real-time forward-looking lane line curvature value from the real-time forward-looking lane line equation and the real-time high-precision lane line curvature value from the real-time high-precision lane line equation. If the real-time forward-looking lane line curvature value is greater than the real-time high-precision lane line curvature value, the negative of the standard deviation of the lane line curvature is determined as the lane line curvature compensation amount. Based on the lane line curvature compensation amount and the real-time forward-looking lane line curvature value, the system determines the fused lane line curvature value.

[0096] In this embodiment of the invention, the real-time high-precision lane line equation refers to the lane line equation determined based on the vehicle's real-time lane line data. In one example, if the current time is 10:00, the real-time high-precision lane line equation can be determined based on the lane line data for 10:00 stored in the high-precision map. The real-time high-precision lane line equation is as follows:

[0097] y = A 12 +B 12 *x+C 12 *x 2 +D 12 *x 3 , formula (14)

[0098] The real-time high-precision lane curvature value C can be obtained according to formula (14). 12 The real-time forward lane curvature value C is obtained according to formula (11). 11 If C 11 >C 12 Then, the negative of the standard deviation of lane curvature (i.e., -σ1) can be determined as the lane curvature compensation amount a1, and then based on a 1＝- σ1 and C 11 Calculate the curvature value C of the merged lane line. 融 The specific formula is as follows:

[0099]

[0100] In one embodiment of the present invention, the method may further include:

[0101] If the real-time forward-looking lane curvature value is less than the real-time high-precision lane curvature value, then the standard deviation of lane curvature is determined as the lane curvature compensation amount.

[0102] In this embodiment of the invention, if C 11 <C 12Then a1 = σ1, at this time

[0103] In one embodiment of the present invention, determining the fused lane line curvature change rate value based on multiple first lane line equations and multiple second lane line equations may include:

[0104] Based on multiple first lane line equations, multiple first lane line curvature change rate values ​​are obtained; based on multiple second lane line equations, multiple second lane line curvature change rate values ​​are obtained; based on the multiple first lane line curvature change rate values ​​and the multiple second lane line curvature change rate values, the standard deviation of lane line curvature change rate is calculated; based on the standard deviation of lane line curvature change rate, the fused lane line curvature change rate value is determined.

[0105] In this embodiment of the invention, the first lane line equations and the second lane line equations obtained above will continue to be introduced. The multiple first lane line equations are respectively formula (1), formula (2), formula (3), formula (4), and formula (5), and the five sets of first lane line curvature change rate values ​​obtained are D1, D2, D3, D4, and D5. The multiple second lane line equations are respectively formula (6), formula (7), formula (8), formula (9), and formula (10), and the five sets of second lane line curvature change rate values ​​obtained are D6, D7, D8, D9, and D10. 10 Therefore, based on D1, D2, D3, D4, D5, D6, D7, D8, D9, D 10 The standard deviation σ2 of the lane curvature change rate is calculated, and then the fused lane curvature change rate value D can be determined based on σ2. 融 The formula for calculating σ² is:

[0106]

[0107] Where σ² is the standard deviation of the lane curvature change rate, n represents the index of the lane curvature change rate value, xi is the lane curvature change rate value with index i, and μ is the average of multiple lane curvature change rate values, i.e., μ = (D1 + D2 + D3 + D4 + D5 + D6 + D7 + D8 + D9 + D 10 ) / 10.

[0108] In one embodiment of the present invention, determining the fused lane curvature change rate value based on the standard deviation of the lane curvature change rate may include:

[0109] The system acquires real-time high-precision lane line equations during vehicle operation, which are determined based on real-time lane line data stored in a high-precision map. It also acquires the real-time forward-looking lane line curvature change rate value of the real-time forward-looking lane line equation and the real-time high-precision lane line curvature change rate value of the real-time high-precision lane line equation. If the real-time forward-looking lane line curvature change rate value is greater than the real-time high-precision lane line curvature change rate value, the negative of the standard deviation of the lane line curvature change rate is determined as the lane line curvature change rate compensation amount. Based on the lane line curvature change rate compensation amount and the real-time forward-looking lane line curvature change rate value, the system determines the fused lane line curvature change rate value.

[0110] Specifically, the real-time high-precision lane curvature change rate value D can be obtained according to formula (14). 12 The real-time forward lane curvature value D is obtained according to formula (11). 11 If D 11 >D 12 Therefore, the negative of the standard deviation of lane curvature (i.e., -σ2) can be determined as the lane curvature change rate compensation amount a2, and then based on a 2＝- σ2 and D 11 Calculate the curvature value D of the merged lane line. 融 , specifically calculate D 融 The formula is:

[0111]

[0112] In one embodiment of the present invention, the method may further include:

[0113] If the real-time forward-looking lane curvature change rate is less than the real-time high-precision lane curvature change rate, then the standard deviation of the lane curvature change rate is determined as the lane curvature change rate compensation amount.

[0114] In this embodiment of the invention, if D 11 <D 12 Then a2 = σ2, at this time

[0115] Step 204: Obtain the real-time forward-looking lane line equation during vehicle movement; the real-time forward-looking lane line equation is determined based on the real-time lane line coordinate points collected by the forward-looking camera.

[0116] Step 205: Determine the offset distance of the rear axle center of the real-time forward-looking vehicle from the lane line and the real-time lane line yaw angle from the real-time forward-looking lane line equation.

[0117] In this embodiment of the invention, the offset distance A between the center of the rear axle of the forward-looking vehicle and the lane line can be obtained from formula (11). 11 Real-time lane line yaw angle B 11.

[0118] Step 206: The target lane line equation is obtained by fitting the real-time offset distance of the rear axle center of the forward-looking vehicle from the lane line, the real-time lane line yaw angle, the fused lane line curvature value, and the fused lane line curvature change rate value.

[0119] In this embodiment of the invention, the offset distance A between the center of the vehicle's rear axle and the lane line can be used as a reference. 11 Real-time lane line yaw angle B 11 C 融 D 融 The equation of the target lane line is obtained, as shown in formula (12).

[0120] In one embodiment of the present invention, the method may further include:

[0121] If the number of second lane line equations acquired in the same time period is less than the number of first lane line equations, then the first lane line equation corresponding to the missing second lane line equation is determined as the reference lane line equation; the high-precision lane line equations adjacent to the reference lane line equation are acquired; based on the reference lane line equation and the adjacent high-precision lane line equations, the reference lane line fusion curvature change rate value and the reference lane line fusion curvature value are determined; the missing second lane line equation is fitted according to the reference lane line equation, the reference lane line fusion curvature change rate value, and the reference lane line fusion curvature value.

[0122] In this embodiment of the invention, the number of first lane line equations and the number of second lane line equations within the same time period can be obtained simultaneously. If the number of first lane line equations and the number of second lane line equations are the same, it indicates that the second lane line equations are not missing. If the number of first lane line equations is greater than the number of second lane line equations, it indicates that the second lane line equations are missing. In one example, the time period is from 9:59 to 10:00. If the number of first lane line equations is 5 and the number of second lane line equations is 4, it indicates that the number of second lane line equations is missing. The five sets of first lane line equations are set as the lane line equations corresponding to formulas (1), (2), (3), (4), and (5), respectively. The order of these five sets of first lane line equations is arranged according to time. The missing second lane line equation is the second set of second lane line equations. Then, the first lane line equation corresponding to formula (2) can be determined as the reference lane line equation.

[0123] Furthermore, the high-precision lane line equations adjacent to the equation of formula (2) can be obtained, which can be formula (1) or formula (3). Here, formula (1) is used as an example. Based on formula (1) and formula (2), two sets of lane line curvature values ​​C1 and C2 and two sets of lane line curvature change rate values ​​D1 and D2 can be obtained. Then, the standard deviation of lane line curvature σ3 of the adjacent lane line equations can be calculated according to formula (18).

[0124]

[0125] Where σ3 is the standard deviation of the curvature of adjacent lane lines, n represents the index of the lane line curvature value, xi is the lane line curvature value with index i, and μ is the average value of multiple lane line curvature values, i.e. μ=(C1+C2) / 2.

[0126] Then, the reference lane line blending curvature value C is calculated according to formula (19). 参考融合 Formula (19) is:

[0127]

[0128] According to formula (20), the standard deviation σ4 of the rate of change of lane curvature of adjacent lane line equations can be calculated:

[0129]

[0130] Where σ4 is the standard deviation of the rate of change of curvature of adjacent lane lines, n represents the sequence number of the rate of change of curvature of lane lines, xi is the rate of change of curvature of lane line with index i, and μ is the average value of multiple rate of change of curvature of lane lines, i.e. μ=(D1+D2) / 2.

[0131] Then, the reference lane line fusion curvature change rate value D is calculated according to formula (21). 参考融合 Formula (21) is:

[0132]

[0133] Furthermore, A1 and B1, C can be obtained from the reference lane line equation formula (1). 参考融合 D 参考融合 The equation for the missing second lane line obtained by fitting is:

[0134] y 缺失 ==A1+B1*x+C 参考融合 *x 2 +D 参考融合 *x 3 , formula (22)

[0135] This invention logically fuses lane line equations determined by a high-precision map and those determined by a forward-looking camera. The lane line equations identified by the forward-looking camera are used as the primary equations, while those from the high-precision map are used as secondary equations. By optimizing the real-time lane lines identified by the forward-looking camera, more accurate lane line equations can be obtained. This provides more reliable lane lines for vehicle planning and control, and also improves the user's driving experience.

[0136] Reference Figure 3 The diagram illustrates a structural block diagram of a lane line detection device according to an embodiment of the present invention. The device may include:

[0137] The first acquisition module 301 is used to acquire multiple first lane line equations during vehicle driving; the multiple first lane line equations are determined based on multiple sets of historical lane line data stored in a high-precision map.

[0138] The second acquisition module 302 is used to acquire multiple second lane line equations during the vehicle's driving process; the multiple second lane line equations are determined based on multiple sets of historical lane line coordinate points collected by the forward-looking camera.

[0139] The first determining module 303 is used to determine the fused lane line curvature value and the fused lane line curvature change rate value based on the plurality of first lane line equations and the plurality of second lane line equations;

[0140] The third acquisition module 304 is used to acquire the real-time forward lane line equation at the current moment during the vehicle's driving process; the real-time forward lane line equation is determined based on the real-time lane line coordinate points collected by the forward-looking camera at the current moment.

[0141] The second determining module 305 is used to determine the target lane line equation based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value.

[0142] This invention discloses a lane line detection device. By logically fusing the lane line equations determined by a high-precision map and the lane line equations determined by a forward-looking camera, the device prioritizes the lane line equations identified by the forward-looking camera and uses the lane lines from the high-precision map as a supplement. This optimizes the real-time lane lines identified by the forward-looking camera, resulting in more accurate lane line equations. This provides more reliable lane lines for vehicle planning and control, and also improves the user's driving experience.

[0143] In one embodiment of the present invention, the second determining module 305 may include:

[0144] The first determining submodule is used to determine the offset distance of the rear axle center of the real-time forward-looking vehicle from the lane line and the real-time lane line yaw angle from the real-time forward-looking lane line equation.

[0145] The first fitting submodule is used to fit the target lane line equation using the offset distance of the rear axle center of the real-time forward-looking vehicle from the lane line, the real-time lane line yaw angle, the fused lane line curvature value, and the fused lane line curvature change rate value.

[0146] In one embodiment of the present invention, the first determining module 303 may include:

[0147] The first acquisition submodule is used to acquire multiple first lane line curvature values ​​based on the multiple first lane line equations;

[0148] The second acquisition submodule is used to acquire multiple second lane line curvature values ​​based on the multiple second lane line equations;

[0149] The first calculation submodule is used to calculate the standard deviation of lane line curvature based on the plurality of first lane line curvature values ​​and the plurality of second lane line curvature values.

[0150] The second determining submodule is used to determine the fused lane line curvature value based on the standard deviation of the lane line curvature.

[0151] In one embodiment of the present invention, the second determining submodule may include:

[0152] The first acquisition unit is used to acquire the real-time high-precision lane line equation during the vehicle's driving process. The real-time high-precision lane line equation is determined based on the real-time lane line data stored in the high-precision map.

[0153] The second acquisition unit is used to acquire the real-time forward lane curvature value of the real-time forward lane equation and the real-time high-precision lane curvature value of the real-time high-precision lane equation.

[0154] The first determining unit is used to determine the negative number of the standard deviation of the lane line curvature as the lane line curvature compensation amount if the real-time forward-looking lane line curvature value is greater than the real-time high-precision lane line curvature value.

[0155] The second determining unit is used to determine the fused lane line curvature value based on the lane line curvature compensation amount and the real-time forward-looking lane line curvature value.

[0156] In one embodiment of the present invention, the apparatus may further include:

[0157] The third determining unit is used to determine the standard deviation of lane line curvature as the lane line curvature compensation amount if the real-time forward-looking lane line curvature value is less than the real-time high-precision lane line curvature value.

[0158] In one embodiment of the present invention, the first determining module 303 may include:

[0159] The third acquisition submodule is used to acquire multiple first lane line curvature change rate values ​​based on the multiple first lane line equations;

[0160] The fourth acquisition submodule is used to acquire multiple second lane line curvature change rate values ​​based on the multiple second lane line equations;

[0161] The second calculation submodule is used to calculate the standard deviation of the lane line curvature change rate based on the plurality of first lane line curvature change rate values ​​and the plurality of second lane line curvature change rate values.

[0162] The third determining submodule is used to determine the fused lane line curvature change rate value based on the standard deviation of the lane line curvature change rate.

[0163] In one embodiment of the present invention, the third determining submodule may include:

[0164] The third acquisition unit is used to acquire the real-time high-precision lane line equation during the vehicle's driving process. The real-time high-precision lane line equation is determined based on the real-time lane line data stored in the high-precision map.

[0165] The fourth acquisition unit is used to acquire the real-time forward lane line curvature change rate value of the real-time forward lane line equation and the real-time high-precision lane line curvature change rate value of the real-time high-precision lane line equation.

[0166] The fourth determining unit is used to determine the negative number of the standard deviation of the lane curvature change rate as the lane curvature change rate compensation amount if the real-time forward-looking lane curvature change rate value is greater than the real-time high-precision lane curvature change rate value.

[0167] The fifth determining unit is used to determine the fused lane line curvature change rate value based on the lane line curvature change rate compensation amount and the real-time forward-looking lane line curvature change rate value.

[0168] In one embodiment of the present invention, the apparatus may further include:

[0169] The sixth determining unit is used to determine the standard deviation of the lane curvature change rate as the lane curvature change rate compensation amount if the real-time forward-looking lane curvature change rate value is less than the real-time high-precision lane curvature change rate value.

[0170] In one embodiment of the present invention, the apparatus may further include:

[0171] The third determining module is used to determine the first lane line equation corresponding to the missing second lane line equation as the reference lane line equation if the number of the second lane line equations obtained in the same time period is less than the number of the first lane line equations.

[0172] The fourth acquisition module is used to acquire high-precision lane line equations adjacent to the reference lane line equation;

[0173] The fourth determining module is used to determine the reference lane line fusion curvature change rate value and the reference lane line fusion curvature value based on the reference lane line equation and the adjacent high-precision lane line equation.

[0174] The fitting module is used to fit the missing second lane line equation based on the reference lane line equation, the reference lane line fusion curvature change rate value, and the reference lane line fusion curvature value.

[0175] This invention discloses a lane line detection device. By logically fusing the lane line equations determined by a high-precision map and the lane line equations determined by a forward-looking camera, the device prioritizes the lane line equations identified by the forward-looking camera and uses the lane lines from the high-precision map as a supplement. This optimizes the real-time lane lines identified by the forward-looking camera, resulting in more accurate lane line equations. This provides more reliable lane lines for vehicle planning and control, and also improves the user's driving experience.

[0176] As the device embodiment is basically similar to the method embodiment, the description is relatively simple, and relevant parts can be found in the description of the method embodiment.

[0177] This invention also provides an electronic device, comprising:

[0178] It includes a processor, a memory, and a computer program stored in the memory and capable of running on the processor. When the computer program is executed by the processor, it implements the various processes of the above-described lane line detection method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0179] This invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the above-described lane line detection method embodiments and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0180] 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 between the various embodiments can be referred to each other.

[0181] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, apparatus, or computer program products. Therefore, embodiments of the present invention can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of the present invention can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0182] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0183] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing terminal device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0184] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal equipment, causing a series of operational steps to be performed on the computer or other programmable terminal equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable terminal equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0185] Although preferred embodiments of the present invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of the embodiments of the present invention.

[0186] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal device. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal device that includes said element.

[0187] The present invention provides a detailed description of a lane line detection method, apparatus, electronic device, and computer-readable storage medium. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A lane line detection method, characterized in that, The method includes: Multiple first lane line equations are obtained during vehicle driving; the multiple first lane line equations are determined based on multiple sets of historical lane line data stored in a high-precision map. Multiple second lane line equations are obtained during the vehicle's driving process; the multiple second lane line equations are determined based on multiple sets of historical lane line coordinate points collected by the forward-looking camera. Based on the plurality of first lane line equations and the plurality of second lane line equations, determine the fused lane line curvature value and the fused lane line curvature change rate value; The real-time forward-looking lane line equation is obtained during the vehicle's driving process; the real-time forward-looking lane line equation is determined based on the real-time lane line coordinate points collected by the forward-looking camera. The target lane line equation is determined based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value. The step of determining the target lane line equation based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value includes: The offset distance of the rear axle center of the real-time forward-looking vehicle from the lane line and the real-time lane line yaw angle are determined from the real-time forward-looking lane line equation. The target lane line equation is obtained by fitting the offset distance between the rear axle center of the real-time forward-looking vehicle and the lane line, the real-time lane line yaw angle, the fused lane line curvature value, and the fused lane line curvature change rate value.

2. The method according to claim 1, characterized in that, The step of determining the fused lane curvature value based on the plurality of first lane line equations and the plurality of second lane line equations includes: Based on the plurality of first lane line equations, a plurality of first lane line curvature values ​​are obtained; Based on the multiple second lane line equations, multiple second lane line curvature values ​​are obtained; The standard deviation of lane line curvature is calculated based on the plurality of first lane line curvature values ​​and the plurality of second lane line curvature values. The fused lane line curvature value is determined based on the standard deviation of the lane line curvature.

3. The method according to claim 2, characterized in that, The step of determining the fused lane line curvature value based on the standard deviation of lane line curvature includes: The real-time high-precision lane line equation is obtained during the vehicle's driving process. The real-time high-precision lane line equation is determined based on the real-time lane line data stored in the high-precision map. Obtain the real-time forward lane curvature value of the real-time forward lane equation and the real-time high-precision lane curvature value of the real-time high-precision lane equation. If the real-time forward-looking lane curvature value is greater than the real-time high-precision lane curvature value, then the negative number of the standard deviation of the lane curvature is determined as the lane curvature compensation amount. Based on the lane line curvature compensation amount and the real-time forward-looking lane line curvature value, the fused lane line curvature value is determined.

4. The method according to claim 3, characterized in that, The method further includes: If the real-time forward-looking lane curvature value is less than the real-time high-precision lane curvature value, then the standard deviation of the lane curvature is determined as the lane curvature compensation amount.

5. The method according to claim 1, characterized in that, The step of determining the fused lane curvature change rate value based on the plurality of first lane line equations and the plurality of second lane line equations includes: Based on the multiple first lane line equations, multiple first lane line curvature change rate values ​​are obtained; Based on the multiple second lane line equations, multiple second lane line curvature change rate values ​​are obtained; The standard deviation of the lane curvature change rate is calculated based on the plurality of first lane line curvature change rate values ​​and the plurality of second lane line curvature change rate values. The fused lane curvature change rate value is determined based on the standard deviation of the lane curvature change rate.

6. The method according to claim 5, characterized in that, The step of determining the fused lane curvature change rate value based on the standard deviation of the lane curvature change rate includes: The real-time high-precision lane line equation is obtained during the vehicle's driving process. The real-time high-precision lane line equation is determined based on the real-time lane line data stored in the high-precision map. Obtain the real-time forward lane curvature change rate value of the real-time forward lane line equation and the real-time high-precision lane line curvature change rate value of the real-time high-precision lane line equation. If the real-time forward-looking lane curvature change rate value is greater than the real-time high-precision lane curvature change rate value, then the negative number of the standard deviation of the lane curvature change rate is determined as the lane curvature change rate compensation amount. Based on the lane line curvature change rate compensation amount and the real-time forward-looking lane line curvature change rate value, the fused lane line curvature change rate value is determined.

7. The method according to claim 6, characterized in that, The method further includes: If the real-time forward-looking lane curvature change rate value is less than the real-time high-precision lane curvature change rate value, then the standard deviation of the lane curvature change rate is determined as the lane curvature change rate compensation amount.

8. The method according to claim 1, characterized in that, The method further includes: If the number of second lane line equations obtained in the same time period is less than the number of first lane line equations, then the first lane line equation corresponding to the missing second lane line equation is determined as the reference lane line equation. Obtain the high-precision lane line equations adjacent to the reference lane line equation; Based on the reference lane line equation and the adjacent high-precision lane line equation, determine the reference lane line fusion curvature change rate value and the reference lane line fusion curvature value; The missing second lane line equation is obtained by fitting the reference lane line equation, the reference lane line fusion curvature change rate value, and the reference lane line fusion curvature value.

9. A lane line detection device, characterized in that, The device includes: The first acquisition module is used to acquire multiple first lane line equations during vehicle driving; the multiple first lane line equations are determined based on multiple sets of historical lane line data stored in a high-precision map. The second acquisition module is used to acquire multiple second lane line equations during the vehicle's driving process; the multiple second lane line equations are determined based on multiple sets of historical lane line coordinate points collected by the forward-looking camera. The first determining module is used to determine the fused lane line curvature value and the fused lane line curvature change rate value based on the plurality of first lane line equations and the plurality of second lane line equations; The third acquisition module is used to acquire the real-time forward lane line equation at the current moment during the vehicle's driving process; the real-time forward lane line equation is determined based on the real-time lane line coordinate points collected by the forward-looking camera at the current moment. The second determining module is used to determine the target lane line equation based on the real-time forward-looking lane line equation, the fused lane line curvature value, and the fused lane line curvature change rate value. The second determining module includes: The first determining submodule is used to determine the offset distance of the rear axle center of the real-time forward-looking vehicle from the lane line and the real-time lane line yaw angle from the real-time forward-looking lane line equation. The first fitting submodule is used to fit the target lane line equation using the offset distance of the rear axle center of the real-time forward-looking vehicle from the lane line, the real-time lane line yaw angle, the fused lane line curvature value, and the fused lane line curvature change rate value.

10. An electronic device, characterized in that, include: A processor, a memory, and a computer program stored in the memory and capable of running on the processor, wherein the computer program, when executed by the processor, implements the steps of the lane line detection method as described in any one of claims 1-8.

11. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the steps of the lane line detection method as described in any one of claims 1-8.

Images