Vehicle lane line determination method and device, electronic equipment and storage medium
By generating lane lines through visual perception recognition and adaptive Kalman filtering algorithms, the problem of low accuracy in traditional lane line modeling under complex environments is solved, and high-quality lane line determination is achieved.
Patent Information
- Application Number
- CN202511557452.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-01-23
AI Technical Summary
Traditional lane line modeling struggles to cope with rapid changes in curved road conditions in complex environments, resulting in poor lane line quality and low accuracy.
Using two-dimensional lane line data based on visual perception and a preset spiral curve model, combined with an adaptive Kalman filter algorithm for state recursion, a fused target lane line in the vehicle coordinate system is generated.
It improves the quality and accuracy of lane markings in complex environments and enhances the ability to respond to rapid changes in curved road conditions.
Smart Images
Figure CN121375792A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of intelligent driving of vehicles, and in particular to a vehicle lane line determination method and device, an electronic device, and a storage medium. BACKGROUND
[0002] At present, with the development of society and the progress of science and technology, more and more new energy vehicles begin to support assisted driving, and lane line recognition and generation are the basis of assisted driving.
[0003] However, traditional lane line modeling is usually only applicable to describing simple curve shapes, but it performs poorly in accurately describing specific directions or changes in curvature, and in some cases does not conform to the smooth characteristics of actual lane lines, making it difficult to cope with rapid changes in curve road conditions in complex environments, thereby resulting in poor quality and low accuracy of the determined lane lines. SUMMARY
[0004] The present application provides a vehicle lane line determination method, device, electronic device, and storage medium, which solves the technical problem of being difficult to cope with rapid changes in curve road conditions in complex environments, thereby resulting in poor quality and low accuracy of the determined lane lines. The embodiments provided by the present application can accurately describe the curvature of the curve to achieve the greatest degree of response to rapid changes in curve road conditions in complex environments, thereby improving the quality and accuracy of the fused target lane line.
[0005] In a first aspect of the embodiments of the present application, a vehicle lane line determination method is provided, which comprises: obtaining target lane line perception data of a vehicle at a current time, wherein the target lane line perception data is two-dimensional lane line data obtained based on visual perception recognition; determining a lane line model corresponding to the target lane line based on the target lane line perception data and a preset spiral curve model, and a lane line state parameter corresponding to the lane line model at the current time; performing state recursion on the lane line state parameter at the current time based on a preset adaptive Kalman filtering algorithm to determine a fused lane line state parameter, wherein the preset adaptive Kalman filtering algorithm is a Kalman filtering algorithm determined based on an adaptive chi-square test model; generating a fused target lane line corresponding to the vehicle in a vehicle body coordinate system based on the fused lane line state parameter.
[0006] In a feasible implementation, before the target lane line perception data of the vehicle at the current time is obtained, the method further comprises: obtain pose information of the vehicle at the current time and initial lane line perception data of the vehicle at the current time; perform preprocessing on the initial lane line perception data to determine standard lane line perception data of the vehicle at the current time; perform coordinate system conversion of the standard lane line perception data from pixel coordinates to vehicle body coordinates based on the pose information to obtain target lane line perception data of the vehicle at the current time, wherein the target lane line perception data is three-dimensional lane line data.
[0007] In a feasible implementation, the preprocessing on the initial lane line perception data to determine the standard lane line perception data of the vehicle at the current time includes: based on a preset length threshold range, eliminate lane lines that do not satisfy the preset length threshold range from the initial lane line perception data to obtain first candidate lane line perception data of the vehicle at the current time; based on a preset lateral distance threshold, eliminate lane lines with a lateral distance less than or equal to the preset lateral distance threshold from the first candidate lane line perception data to obtain second candidate lane line perception data of the vehicle at the current time; perform down-sampling processing on the second candidate lane line perception data to determine the standard lane line perception data of the vehicle at the current time.
[0008] In a feasible implementation, the state recursion on the lane line state parameter at the current time based on the preset adaptive Kalman filtering algorithm to determine the fused lane line state parameter includes: based on the lane line state parameter at the current time and the driving speed of the vehicle, determine a predicted lane line state parameter of the vehicle at a next time node; based on noise data of the lane line state parameter at the current time and a Kalman filtering observation model in the preset adaptive Kalman filtering algorithm, determine an error covariance corresponding to the Kalman filtering observation model; based on the Kalman filtering observation model and a lane line observation matrix at the current time, determine a lane line observation vector at the current time; based on the error covariance and the lane line observation matrix at the current time, determine a Kalman gain of the vehicle at the next time node; based on the lane line observation vector at the current time and the predicted lane line state parameter at the next time node, determine an observation noise covariance of visual perception at the next time node; Adaptively adjust the filter parameter based on the observation noise covariance, the Kalman gain and an adaptive chi-square test model in the preset adaptive Kalman filter algorithm; Update the predicted lane line state parameter of the next time node based on the adjusted filter parameter, and determine the updated predicted lane line state parameter as the fused lane line state parameter.
[0009] In a possible implementation, the step of adaptively adjusting the filter parameter based on the observation noise covariance, the Kalman gain and an adaptive chi-square test model in the preset adaptive Kalman filter algorithm comprises: Calculate a chi-square value based on the observation noise covariance and the Kalman gain; Compare the chi-square value with a preset chi-square threshold; When the chi-square value is greater than or equal to the preset chi-square threshold, increase the error covariance based on a preset adjustment factor, and recalculate the Kalman gain to adaptively adjust the filter parameter.
[0010] In a possible implementation, the step of determining the predicted lane line state parameter of the vehicle at the next time node based on the lane line state parameter at the current time and the driving speed of the vehicle comprises: Determine a longitudinal displacement change and a heading angle change of the vehicle from the current time to the next time node based on the driving speed of the vehicle; Determine the predicted lane line state parameter of the vehicle at the next time node based on the longitudinal displacement change, the heading angle change and the lane line state parameter at the current time.
[0011] In a possible implementation, the step of generating the fused target lane line corresponding to the vehicle in the vehicle body coordinate system based on the fused lane line state parameter comprises: Reconstruct an updated lane line curve corresponding to the target lane line based on the fused lane line state parameter and the preset spiral curve model; Obtain at least one target sampling data point on the updated lane line curve, and determine a sequence composed of the target sampling data points as the fused target lane line corresponding to the vehicle in the vehicle body coordinate system.
[0012] In a second aspect, the embodiments of the present application provide a vehicle lane line determination device, which comprises: A second acquisition module is configured to acquire target lane line perception data of a vehicle at a current time, wherein the target lane line perception data is two-dimensional lane line data obtained based on visual perception. The second determining module is configured to determine, based on the target lane line perception data and a preset spiral curve model, a lane line model corresponding to the target lane line and a lane line state parameter corresponding to the lane line model at a current time; The third determining module is configured to perform state recursion on the lane line state parameter at the current time based on a preset adaptive Kalman filtering algorithm to determine a fused lane line state parameter, wherein the preset adaptive Kalman filtering algorithm is a Kalman filtering algorithm determined based on an adaptive chi-square test model. The generating module is configured to generate a fused target lane line corresponding to the vehicle in a vehicle body coordinate system based on the fused lane line state parameter.
[0013] In a third aspect, the embodiments of the present application provide an electronic device, which comprises a processor, a memory and a bus, the memory stores machine readable instructions executable by the processor, when the electronic device is running, the processor and the memory communicate through the bus, and the machine readable instructions are executed by the processor to perform the steps of the vehicle lane line determination method as described above.
[0014] In a fourth aspect, the embodiments of the present application provide a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to perform the steps of the vehicle lane line determination method as described above.
[0015] Compared with the prior art, the vehicle lane line determination method, device, electronic device and storage medium provided by the embodiments of the present application can accurately describe the curvature of a curve to achieve the maximum response to the rapid change of the curve road condition in a complex environment, and improve the quality and accuracy of the fused target lane line. BRIEF DESCRIPTION OF DRAWINGS
[0016] Figure 1 A flowchart of a vehicle lane line determination method provided by the embodiments of the present application is shown; Figure 2 A structural block diagram of a vehicle lane line determination device provided by the embodiments of the present application is shown; Figure 3 A structural schematic diagram of an electronic device provided by an embodiment of the application is shown.
[0017] Figure 2 and Figure 3 The correspondence between the reference signs in the drawings and the drawing names is as follows: 200, a vehicle lane line determination apparatus; 210, a first acquisition module; 220, a first determination module; 230, an obtaining module; 240, a second acquisition module; 250, a second determination module; 260, a third determination module; 270, a generation module; 300, an electronic device; 310, a processor; 320, a memory; 330, a bus. DETAILED DESCRIPTION
[0018] In order to better understand the technical solutions provided by the embodiments of the present specification, the technical solutions of the embodiments of the present specification will be described in detail below through the drawings and specific embodiments. It should be understood that the specific features in the embodiments of the present specification and the embodiments are detailed descriptions of the technical solutions of the embodiments of the present specification, and are not limitations of the technical solutions of the present specification. In the case of no conflict, the technical features in the embodiments of the present specification and the embodiments can be combined with each other.
[0019] In this document, relational terms such as first and second and the like can be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but can include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element preceded by "comprises... " does not, without more constraints, foreclose the existence of additional identical elements in the process, method, article, or apparatus that comprises the recited element. The term "two or more" includes two or more than two.
[0020] First, the application scenarios applicable to the present application are introduced. The embodiments provided by the present application are applicable to At present, the traditional lane line modeling is usually only applicable to describe simple curve shapes, but it performs poorly in accurately describing specific directions or curvature changes, and in some cases does not meet the actual lane line smoothness characteristics, and is difficult to cope with the rapid changes of curve road conditions in complex environments, thereby resulting in poor quality and low accuracy of the determined lane line.
[0021] Traditional lane line establishment can be divided into two categories, mainly including an optimization-based lane line establishment method and a filtering-based lane line establishment method, wherein the main disadvantage of the optimization-based lane line establishment method is high demand for computing power, that is, it cannot be used on a low-computing-power intelligent driving platform, and real-time performance is difficult to guarantee; while the filtering-based lane line establishment method has a smaller demand for computing power, that is, it can also be used for real-time fusion on a low-computing-power platform, but a traditional Kalman filter is difficult to cope with rapid changes in complex environments, and parameter adjustment needs manual intervention, lacking self-adaptability, and such a fixed-parameter filtering method has poor lane line quality and weak generalization in some scenes.
[0022] Based on this, the embodiments of the present application provide a vehicle lane line determination method, device, electronic equipment and storage medium, which solve the technical problem that the prior art is difficult to cope with rapid changes in complex curve road conditions, thereby resulting in poor lane line quality and low accuracy, and the embodiments of the present application can accurately describe the curvature of a curve to cope with rapid changes in complex curve road conditions to the greatest extent, thereby improving the quality and accuracy of the fused target lane line.
[0023] Figure 1 A flowchart of a vehicle lane line determination method provided by an embodiment of the present application is shown. As shown in Figure 1 The vehicle lane line determination method includes the following steps: S101, acquiring target lane line perception data of a vehicle at a current time, wherein the target lane line perception data is two-dimensional lane line data obtained based on visual perception.
[0024] In this step, the embodiments provided by the present application first acquire target lane line perception data of a vehicle in a driving process, that is, at a current time, by visual perception technology, and the acquired target lane line perception data is three-dimensional lane line data.
[0025] The visual perception technology is to perceive and recognize objects, structures and behaviors in the surrounding environment by using images or videos collected by a camera and through computer vision and deep learning algorithms.
[0026] Exemplarily, before acquiring the target lane line perception data of the vehicle at the current time, the method further includes: acquiring pose information of the vehicle at the current time and initial lane line perception data of the vehicle at the current time; preprocessing the initial lane line perception data to determine standard lane line perception data of the vehicle at the current time; performing coordinate system conversion of pixel coordinates to vehicle body coordinates on the standard lane line perception data based on the pose information to obtain the target lane line perception data of the vehicle at the current time, wherein the target lane line perception data is three-dimensional lane line data.
[0027] It can be understood that in the embodiments provided in the present application, first, initial lane line perception data of the vehicle at the current moment is acquired, and then preprocessing in the direction of including abnormal lane line elimination and lane line down-sampling is performed on the initial lane line perception data, and standard lane line perception data of the vehicle at the current moment is generated; meanwhile, the embodiments provided in the present application utilize the extrinsic parameters between the camera and the vehicle body and the vehicle pose information to perform coordinate system conversion of the generated standard lane line perception data from pixel coordinates to vehicle body coordinates, that is, conversion from a 2D pixel coordinate system to a vehicle body coordinate system, to generate target lane line perception data of the vehicle at the current moment.
[0028] In the above, the pose data in the embodiments provided in the present application can be specifically a 6-degree-of-freedom pose of the vehicle.
[0029] For example, preprocessing is performed on the initial lane line perception data to determine the standard lane line perception data of the vehicle at the current moment, including: Based on a preset length threshold range, lane lines that do not satisfy the preset length threshold range are eliminated from the initial lane line perception data to obtain first candidate lane line perception data of the vehicle at the current moment; based on a preset lateral distance threshold, lane lines with adjacent lateral distances less than or equal to the preset lateral distance threshold are eliminated from the first candidate lane line perception data to obtain second candidate lane line perception data of the vehicle at the current moment; down-sampling processing is performed on the second candidate lane line perception data to determine the standard lane line perception data of the vehicle at the current moment.
[0030] It should be noted that the embodiments provided in the present application will eliminate lane lines that do not satisfy the preset length threshold range from the initial lane line perception data based on the preset length threshold range, which is used to delete lane lines that are too short to achieve elimination of abnormal lane lines and obtain first candidate lane line perception data of the vehicle at the current moment; and the embodiments provided in the present application will continue to eliminate lane lines with adjacent lateral distances that are too close based on the obtained first candidate lane line perception data of the vehicle at the current moment, that is, lane lines with adjacent lateral distances less than or equal to the preset lateral distance threshold are eliminated from the first candidate lane line perception data to obtain second candidate lane line perception data of the vehicle at the current moment.
[0031] It can be understood that the embodiments provided in the present application will perform lane line down-sampling on the obtained second candidate lane line perception data, that is, the lane line shape points are thinned out, and then the standard lane line perception data of the vehicle at the current moment is determined.
[0032] In the embodiments provided in the present application, the preset length threshold range can be selected and used according to different application scenarios, and the preset transverse distance threshold can also be selected and used according to different application scenarios.
[0033] In S102, a lane line model corresponding to the target lane line and a lane line state parameter corresponding to the lane line model at the current time are determined based on the target lane line perception data and the preset spiral curve model.
[0034] In this step, after obtaining the two-dimensional target lane line perception data identified based on visual perception, the embodiments provided in the present application begin to establish the target lane line of the vehicle during driving based on the target lane line perception data. When establishing the target lane line, the lane line model corresponding to the target lane line and the lane line state parameter corresponding to the lane line model at the current time need to be determined.
[0035] The lane line state parameter in the embodiments provided in the present application can specifically but not limitedly include an intercept, an angle, a curvature, and a curvature change rate.
[0036] It can be understood that the lane line model corresponding to the target lane line in the embodiments provided in the present application is determined based on the preset spiral curve model and the target lane line perception data, and the determined lane line model formula is specifically as follows: Formula 1 , wherein, is used to represent the intercept of the two-dimensional lane line. is used to represent the angle of the two-dimensional lane line. is used to represent the curvature of the two-dimensional lane line. is the curvature change rate of the two-dimensional lane line.
[0037] Here, in the embodiments provided in the present application, the lane line state parameter corresponding to the determined lane line model at the current time is specifically as follows: Formula 2 , wherein, is used to represent the intercept of the lane line at the time (i.e., the current time). is used to represent the angle of the lane line at the time (i.e., the current time). is used to represent the curvature of the lane line at the time (i.e., the current time). is used to represent the curvature change rate of the lane line at the time (i.e., the current time).
[0038] It should be noted that the type of the preset spiral curve model in the embodiments provided in the present application can be selected and used according to different application scenarios and use conditions. The preset spiral curve model in the embodiments provided in the present application can be specifically set to a cubic spiral curve equation.
[0039] It can be understood that the embodiments provided in the present application use a cubic spiral curve equation to replace a traditional cubic curve equation to model a lane line. Here, the lane line model corresponding to the target lane line generated by the cubic spiral curve equation modeling is more consistent with the vehicle lane line on an actual road. The cubic spiral curve equation is usually used in the technical field of device road generation such as road planning, path generation, and flight path design. It can accurately describe the initial direction and curvature change of a curve, and can generate a smoothly transitioned path according to a starting position and direction, especially in places where smooth turning is required. An ordinary cubic polynomial may not be able to provide such continuity and accuracy, but the lane line generated by the cubic spiral curve equation is more consistent with the actual scenario.
[0040] In S103, a preset adaptive Kalman filtering algorithm is used to perform state recursion on the lane line state parameter at the current time to determine the fused lane line state parameter. The preset adaptive Kalman filtering algorithm is a Kalman filtering algorithm determined based on an adaptive chi-square test model.
[0041] In this step, after the lane line state parameter of the vehicle at the current time is determined in the embodiments provided in the present application, the predicted lane line state parameter of the vehicle at the next time node is predicted. Then, the above lane line is fused and judged based on the predicted lane line state parameter at the next time node and the preset adaptive Kalman filtering algorithm, so as to generate a more accurate target lane line in the subsequent process and determine the fused lane line state parameter.
[0042] It can be understood that in the embodiments provided in the present application, the time difference between the current time and the next time node is preset, and the specific time difference size can be selected and used according to different application scenarios and use conditions.
[0043] For example, the preset adaptive Kalman filtering algorithm is used to perform state recursion on the lane line state parameter at the current time to determine the fused lane line state parameter, including: Based on the lane line state parameter at the current moment and the driving speed of the vehicle, a predicted lane line state parameter of the vehicle at a next time node is determined; based on the noise data of the lane line state parameter at the current moment and a Kalman filter observation model in a preset adaptive Kalman filter algorithm, an error covariance corresponding to the Kalman filter observation model is determined; based on the Kalman filter observation model and a lane line observation matrix at the current moment, a lane line observation vector at the current moment is determined; based on the error covariance and the lane line observation matrix at the current moment, a Kalman gain of the vehicle at the next time node is determined; based on the lane line observation vector at the current moment and the predicted lane line state parameter at the next time node, an observation noise covariance recognized by visual sense at the next time node is determined; based on the observation noise covariance, the Kalman gain and an adaptive chi-square test model in the preset adaptive Kalman filter algorithm, a filter parameter is adaptively adjusted; based on the adjusted filter parameter, the predicted lane line state parameter at the next time node is updated, and the updated predicted lane line state parameter is determined as a fused lane line state parameter.
[0044] It should be noted that the formula for determining the predicted lane line state parameter of the vehicle at the next time node is specifically:
[0045]
[0046]
[0047] Formula 3 Wherein, , is used to represent the driving speed of the vehicle; is used to represent the moving distance of the vehicle between the time and the time .
[0048] It can be understood that the state prediction algorithm in the preset adaptive Kalman filter algorithm in the embodiments provided in the present application can be specifically set as: Formula 4 Here, according to the state prediction algorithm in the above formula 4, it can be known that the state transition matrix corresponding to the Kalman filter observation model is specifically: Formula 5 Wherein, the control matrix corresponding to the Kalman filter observation model in the embodiments provided in the present application is specifically: Formula 6 The input matrix is specifically: Formula 7 According to the above formula 7, the predicted lane line state parameter of the vehicle at the next time node is determined based on the lane line state parameter at the current time and the driving speed of the vehicle, including: Based on the driving speed of the vehicle, the longitudinal displacement change and the heading angle change of the vehicle from the current time to the next time node are determined; based on the longitudinal displacement change, the heading angle change and the lane line state parameter at the current time, the predicted lane line state parameter of the vehicle at the next time node is determined.
[0049] wherein, is used to represent the longitudinal displacement change of the vehicle from the current time to the next time node . is used to represent the predicted lane line state parameter of the vehicle from the current time to the next time node .
[0050] Here, the Kalman filter observation model in the preset adaptive Kalman filter algorithm in the embodiments provided in the application can be specifically set as: Formula 8 wherein, = 4; is used to represent the observation matrix at the current time , is used to represent the observation quantity at the current time . is used to represent the number of sampling points on the lane line.
[0051] Here, Formula 9 Formula 10 In the embodiments provided in the application, the formula for determining the observation noise covariance recognized by the visual sense at the next time node can be specifically: Formula 11 wherein, Q is used to represent the observation noise covariance recognized by the visual sense at the next time node; the four quantities on the diagonal line of the above matrix are the prior noise of the intercept, the angle, the curvature and the curvature change rate of the lane line.
[0052] Here, the formula for determining the error covariance corresponding to the Kalman filter observation model is specifically: Formula 12 Exemplarily, based on the error covariance and the lane line observation matrix at the current time, the formula for determining the lane line observation matrix in the Kalman gain of the vehicle at the next time node is specifically: Equation 13 Here, the observation noise covariance matrix is specifically: Equation 14 Wherein, the nine quantities on the diagonal are the prior observation noise of the nine sampling points of the lane line.
[0053] Therefore, according to the above Equation 13 and Equation 14, the Kalman gain of the vehicle at the next time node can be determined, and the formula for determining the Kalman gain at the next time node is specifically: Equation 15 For example, based on the observation noise covariance, the Kalman gain, and the adaptive chi-square test model in the preset adaptive Kalman filtering algorithm, the filtering parameters are adaptively adjusted, including: Based on the observation noise covariance and the Kalman gain, the chi-square value is calculated; the chi-square value is compared with the preset chi-square threshold; when the chi-square value is greater than or equal to the preset chi-square threshold, the error covariance is increased based on the preset adjustment factor, and the Kalman gain is recalculated to adaptively adjust the filtering parameters.
[0054] It should be noted that the formula of the observation noise covariance in the embodiments provided by the present application can be specifically: Equation 16 It can be understood that the formula for determining the chi-square value in the embodiments provided by the present application is specifically: Equation 17 And in the embodiments provided by the present application, the formula for fusing the joint covariance of the above lane line state parameters is specifically: Equation 18 Here, in the adaptive adjustment of the filtering parameters based on the observation noise covariance, the Kalman gain, and the adaptive chi-square test model in the preset adaptive Kalman filtering algorithm, the formula for adaptively adjusting the filtering parameters and updating the joint covariance is specifically: Equation 19 Wherein, is used to represent the preset chi-square threshold, and the preset chi-square threshold in the embodiments provided by the present application can be selected and used according to different application scenarios and use conditions; is used to represent a fixed adjustment parameter.
[0055] The formula for recalculating the Kalman gain, i.e., updating the Kalman filtering gain, provided by the embodiments of the present application is as follows: Equation 20 Here, the updated state quantity formula obtained according to Equation 20 is: Equation 21 The formula of the updated error covariance obtained according to Equation 20 is specifically: Equation 22 In the above, the chi-square test is used to detect outliers and evaluate the quality of the filtering result, especially in systems with noise and uncertainty. Specifically, the chi-square test can help determine whether the residuals of the Kalman filter meet the Gaussian distribution assumption, to identify abnormal phenomena or mutations in the filtering process, and to ensure the stability and accuracy of the filter; and the embodiments provided in the present application dynamically adjust the Kalman filter parameters through the chi-square test to achieve the purpose of adaptive filtering.
[0056] In S104, a fused target lane line corresponding to the vehicle in the vehicle coordinate system is generated based on the fused lane line state parameter.
[0057] In this step, after determining the fused lane line state parameter, the embodiments provided in the present application will update and generate a more accurate fused target lane line based on the fused lane line state parameter, while reducing the hardware computing power requirement, so that the generation method of the target lane line can also run smoothly on a low-computing-power platform.
[0058] Illustratively, generating a fused target lane line corresponding to the vehicle in the vehicle coordinate system based on the fused lane line state parameter comprises: reconstructing an updated lane line curve corresponding to the target lane line based on the fused lane line state parameter and a preset spiral curve model; obtaining at least one target sampling data point on the updated lane line curve, and determining a sequence composed of the target sampling data points as the fused target lane line corresponding to the vehicle in the vehicle coordinate system.
[0059] Compared with the prior art, the method for determining a vehicle lane line provided in the embodiments of the present application can accurately describe the curvature of a curve, so as to achieve the maximum response to the rapid change of the curve road condition in a complex environment, and improve the quality and accuracy of the fused target lane line.
[0060] The adaptive Kalman filtering algorithm is used to fuse the lane line, and the filtering parameters can be dynamically adjusted, so as to improve the fusion accuracy and real-time performance of the lane line, and better cope with the lane line fusion problem in a complex environment. This can more significantly improve the adaptability and real-time performance of the prior art.
[0061] Figure 2 A structural block diagram of a determination device for a vehicle lane line provided in the embodiments of the present application is shown in FIG. 1. Figure 2 As shown in FIG. 1, the determination device for a vehicle lane line 200 includes: A first obtaining module 210 is configured to obtain pose information of a vehicle at a current time and initial lane line perception data of the vehicle at the current time, wherein the target lane line perception data is two-dimensional lane line data obtained based on visual perception.
[0062] A first determining module 220 is configured to pre-process the initial lane line perception data to determine standard lane line perception data of the vehicle at the current time.
[0063] A obtaining module 230 is configured to perform coordinate system conversion from pixel coordinates to vehicle coordinates on the standard lane line perception data based on the pose information to obtain target lane line perception data of the vehicle at the current time.
[0064] A second obtaining module 240 is configured to obtain target lane line perception data of the vehicle at the current time, wherein the target lane line perception data is two-dimensional lane line data obtained based on visual perception.
[0065] A second determining module 250 is configured to determine, based on the target lane line perception data and a preset spiral curve model, a lane line model corresponding to the target lane line and lane line state parameters corresponding to the lane line model at the current time.
[0066] The third determination module 260 is configured to perform state recursion on the lane line state parameter at the current time based on a preset adaptive Kalman filtering algorithm to determine the fused lane line state parameter, wherein the preset adaptive Kalman filtering algorithm is a Kalman filtering algorithm determined based on an adaptive chi-square test model.
[0067] The generation module 270 is configured to generate the fused target lane line corresponding to the vehicle in the vehicle coordinate system based on the fused lane line state parameter.
[0068] For example, the first determination module 210 is specifically configured to: remove lane lines that do not meet the preset length threshold range from the initial lane line perception data based on the preset length threshold range to obtain first candidate lane line perception data of the vehicle at the current time; remove lane lines with a neighboring lateral distance less than or equal to a preset lateral distance threshold from the first candidate lane line perception data based on the preset lateral distance threshold to obtain second candidate lane line perception data of the vehicle at the current time; and perform down-sampling processing on the second candidate lane line perception data to determine standard lane line perception data of the vehicle at the current time, wherein the target lane line perception data is three-dimensional lane line data.
[0069] For example, the third determination module 260 is specifically configured to: determine a predicted lane line state parameter of the vehicle at a next time node based on the lane line state parameter at the current time and a driving speed of the vehicle.
[0070] determine an error covariance corresponding to a Kalman filtering observation model based on noise data of the lane line state parameter at the current time and the Kalman filtering observation model in the preset adaptive Kalman filtering algorithm.
[0071] determine a lane line observation vector at the current time based on the Kalman filtering observation model and a lane line observation matrix at the current time.
[0072] determine a Kalman gain of the vehicle at the next time node based on the error covariance and the lane line observation matrix at the current time.
[0073] determine an observation noise covariance of visual perception at the next time node based on the lane line observation vector at the current time and the predicted lane line state parameter at the next time node.
[0074] adaptively adjust a filtering parameter based on the observation noise covariance, the Kalman gain, and an adaptive chi-square test model in the preset adaptive Kalman filtering algorithm.
[0075] Based on the adjusted filtering parameter, the predicted lane line state parameter at the next time node is updated, and the updated predicted lane line state parameter is determined as the fused lane line state parameter.
[0076] For example, based on the observation noise covariance, the Kalman gain and the adaptive chi-square test model in the preset adaptive Kalman filtering algorithm, the filtering parameter is adaptively adjusted, including: Based on the observation noise covariance and the Kalman gain, the chi-square value is calculated.
[0077] The chi-square value is compared with the preset chi-square threshold.
[0078] When the chi-square value is greater than or equal to the preset chi-square threshold, the error covariance is increased based on the preset adjustment factor, and the Kalman gain is recalculated to adaptively adjust the filtering parameter.
[0079] For example, based on the lane line state parameter at the current time and the driving speed of the vehicle, the predicted lane line state parameter of the vehicle at the next time node is determined, including: Based on the driving speed of the vehicle, the longitudinal displacement change and the heading angle change of the vehicle from the current time to the next time node are determined.
[0080] Based on the longitudinal displacement change, the heading angle change and the lane line state parameter at the current time, the predicted lane line state parameter of the vehicle at the next time node is determined.
[0081] For example, the generation module is specifically configured to: Based on the fused lane line state parameter and the preset spiral curve model, the updated lane line curve corresponding to the target lane line is reconstructed.
[0082] At least one target sampling data point on the updated lane line curve is obtained, and a sequence composed of the target sampling data points is determined as the fused target lane line corresponding to the vehicle in the vehicle coordinate system.
[0083] Compared with the prior art, the vehicle lane line determination device 200 provided in the embodiments of the present application can accurately describe the curvature of a curve, so as to realize the maximum response to the rapid change of the curve road condition in a complex environment, and improve the quality and accuracy of the fused target lane line.
[0084] The adaptive Kalman filtering algorithm is used to fuse the lane line, and the filtering parameters can be dynamically adjusted, so as to improve the fusion accuracy and real-time performance of the lane line, and better cope with the lane line fusion problem in a complex environment. This can more significantly improve the adaptability and real-time performance of the prior art.
[0085] Figure 3 The structure of the electronic device provided in the embodiments of the present application is shown in a structural schematic diagram. Figure 3 As shown in the figure, the electronic device 300 includes a processor 310, a memory 320 and a bus 330.
[0086] The memory 320 stores machine readable instructions executable by the processor 310. When the electronic device 300 is running, the processor 310 and the memory 320 communicate through the bus 330. When the machine readable instructions are executed by the processor 310, the steps of the vehicle lane line determination method in the method embodiment shown in the above Figure 1 The specific implementation can be referred to the method embodiment, and will not be described here.
[0087] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program. When the computer program is run by a processor, the steps of the vehicle lane line determination method in the method embodiment shown in the above Figure 1 The specific implementation can be referred to the method embodiment, and will not be described here.
[0088] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the system, device and unit described above can refer to the corresponding process in the foregoing method embodiments, and will not be described here.
[0089] It should be noted that in the above embodiments, the description of each embodiment has its own focus, and the parts not described in detail in a certain embodiment can be referred to the related description of other embodiments.
[0090] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-readable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) containing computer readable program code.
[0091] The present application is described with reference to flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and the combination of flows and / or blocks in the flowcharts 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, a special-purpose computer, an embedded computer, or other programmable data processing devices to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The means for implementing the functions specified in one or more flows and / or blocks.
[0092] These computer program instructions can also be stored in a computer readable memory capable of directing the computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer readable memory produce a product including instruction means, which implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks Figure 1 The means for implementing the functions specified in one or more flows and / or blocks.
[0093] These computer program instructions can also be loaded into a computer or other programmable data processing device, so that a series of operation steps are performed on the computer or other programmable device to produce a computer implemented process, so that the instructions executed on the computer or other programmable device provide a means for implementing the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in one or more flows and / or blocks The means for implementing the functions specified in one or more flows and / or blocks.
[0094] The embodiments of the present application also provide a computer program product, which includes computer software instructions, when the computer software instructions run on a processing device, so that the processing device executes the flow of the vehicle lane line determination method.
[0095] The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on the computer, the processes or functions according to the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium, for example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center through wired (for example, coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (for example, infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that the computer can store or be integrated into a data storage device such as a server, data center, etc. containing one or more available media sets. The available media can be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, a solid state disk (SSD)), etc.
[0096] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, device and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be repeated here.
[0097] In several embodiments provided in the present application, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the apparatus embodiments described above are only schematic. The division of the units is only a logical function division. Actual implementation can have another division manner. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the coupling or direct coupling or communication connection between the units shown or discussed can be indirect coupling or communication connection through some interfaces, devices or units, and can be electrical, mechanical or other forms.
[0098] The units described as separate components can or can not be physically separated, and the components shown as units can or can not be physical units, that is, they can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiments.
[0099] In addition, each of the function units in each of the embodiments of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit. The integrated unit can be realized in the form of hardware or in the form of a software function unit.
[0100] When the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or in other words, the part that contributes to the prior art or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods of the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various media that can store program codes.
[0101] The above embodiments are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.
[0102] Although the preferred embodiments of the present specification have been described, those skilled in the art can make additional changes and modifications to the embodiments once they know the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present specification.
[0103] Obviously, those skilled in the art can make various modifications and variations to the present specification without departing from the spirit and scope of the present specification. Thus, if these modifications and variations of the present specification fall within the scope of the claims of the present specification and their equivalent technologies, the present specification also intends to include these modifications and variations.
Claims
1. A method of determining a vehicle lane line, characterized by, The method for determining the vehicle lane line comprises the following steps: acquiring target lane line perception data of the vehicle at a current time point, wherein the target lane line perception data is two-dimensional lane line data obtained based on visual perception; determining a lane line model corresponding to the target lane line based on the target lane line perception data and a preset spiral curve model, and determining a lane line state parameter corresponding to the lane line model at the current time point; based on a preset adaptive Kalman filtering algorithm, recursively updating the lane line state parameter at the current time point to determine a fused lane line state parameter, wherein the preset adaptive Kalman filtering algorithm is a Kalman filtering algorithm determined based on an adaptive chi-square test model; generating a fused target lane line corresponding to the vehicle in a vehicle coordinate system based on the fused lane line state parameter.
2. The method of determining a lane line of a vehicle according to claim 1, wherein, Before the step of acquiring the target lane line perception data of the vehicle at the current time point, the method further comprises the following steps: acquiring pose information of the vehicle at the current time point and initial lane line perception data of the vehicle at the current time point; preprocessing the initial lane line perception data to determine standard lane line perception data of the vehicle at the current time point; based on the pose information, performing coordinate system conversion from pixel coordinates to vehicle coordinates on the standard lane line perception data to obtain the target lane line perception data of the vehicle at the current time point, wherein the target lane line perception data is three-dimensional lane line data.
3. The method of determining a lane line of a vehicle according to claim 2, wherein, The step of preprocessing the initial lane line perception data to determine the standard lane line perception data of the vehicle at the current time point comprises the following steps: based on a preset length threshold range, eliminating lane lines not satisfying the preset length threshold range from the initial lane line perception data to obtain first candidate lane line perception data of the vehicle at the current time point; based on a preset lateral distance threshold, eliminating lane lines with a lateral distance less than or equal to the preset lateral distance threshold from the first candidate lane line perception data to obtain second candidate lane line perception data of the vehicle at the current time point; performing down-sampling processing on the second candidate lane line perception data to determine the standard lane line perception data of the vehicle at the current time point.
4. The method of claim 1, wherein The step of recursively updating the lane line state parameter at the current time point based on the preset adaptive Kalman filtering algorithm to determine the fused lane line state parameter comprises the following steps: based on the lane line state parameter at the current time point and a driving speed of the vehicle, determining a predicted lane line state parameter of the vehicle at a next time node; based on noise data of the lane line state parameter at the current time point and a Kalman filtering observation model in the preset adaptive Kalman filtering algorithm, determining an error covariance corresponding to the Kalman filtering observation model; based on the Kalman filtering observation model and a lane line observation matrix at the current time point, determining a lane line observation vector at the current time point; based on the error covariance and the lane line observation matrix at the current time point, determining a Kalman gain of the vehicle at the next time node; determine, based on the lane line observation vector at the current moment and the predicted lane line state parameter at the next time node, an observation noise covariance recognized by the visual perception at the next time node; adaptively adjust a filtering parameter based on the observation noise covariance, the Kalman gain, and an adaptive chi-square test model in the preset adaptive Kalman filtering algorithm; update the predicted lane line state parameter at the next time node based on the adjusted filtering parameter, and determine the updated predicted lane line state parameter as the fused lane line state parameter.
5. The method of determining a lane line of a vehicle according to claim 4, wherein, The adaptive adjustment of the filtering parameter based on the observation noise covariance, the Kalman gain, and the adaptive chi-square test model in the preset adaptive Kalman filtering algorithm comprises: calculating a chi-square value based on the observation noise covariance and the Kalman gain; comparing the chi-square value with a preset chi-square threshold value; when the chi-square value is greater than or equal to the preset chi-square threshold value, increasing the error covariance based on a preset adjustment factor, and recalculating the Kalman gain to adaptively adjust the filtering parameter.
6. The method of determining a lane line of a vehicle according to claim 4, wherein, The determination of the predicted lane line state parameter of the vehicle at the next time node based on the lane line state parameter at the current moment and the driving speed of the vehicle comprises: determining a longitudinal displacement change and a heading angle change of the vehicle from the current moment to the next time node based on the driving speed of the vehicle; determining the predicted lane line state parameter of the vehicle at the next time node based on the longitudinal displacement change, the heading angle change, and the lane line state parameter at the current moment.
7. The method of determining a lane line of a vehicle according to claim 1, wherein, The generation of the fused target lane line of the vehicle in the vehicle body coordinate system based on the fused lane line state parameter comprises: reconstructing an updated lane line curve corresponding to the target lane line based on the fused lane line state parameter and the preset spiral curve model; obtaining at least one target sampling data point on the updated lane line curve, and determining a sequence composed of the target sampling data points as the fused target lane line of the vehicle in the vehicle body coordinate system.
8. A vehicle lane line determination device characterized by comprising: The vehicle lane line determination device comprises: a second acquisition module configured to acquire target lane line perception data of a vehicle at a current moment, wherein the target lane line perception data is two-dimensional lane line data recognized based on visual perception; a second determination module configured to determine a lane line model corresponding to the target lane line based on the target lane line perception data and a preset spiral curve model, and determine a lane line state parameter corresponding to the lane line model at the current moment; a third determination module configured to perform state recursion on the lane line state parameter at the current moment based on a preset adaptive Kalman filtering algorithm to determine a fused lane line state parameter, wherein the preset adaptive Kalman filtering algorithm is a Kalman filtering algorithm determined based on an adaptive chi-square test model; a generation module configured to generate a fused target lane line of the vehicle in a vehicle body coordinate system based on the fused lane line state parameter.
9. An electronic device, comprising: A processor, a memory, and a bus, the memory storing machine readable instructions executable by the processor, the processor in communication with the memory via the bus when the electronic device is running, the machine readable instructions, when executed by the processor, performing steps of the method for determining a vehicle lane as claimed in any one of claims 1-7.
10. A computer-readable storage medium, characterized in that, A computer program stored on the computer readable storage medium, the computer program, when executed by the processor, performing steps of the method for determining a vehicle lane as claimed in any one of claims 1-7.