Vehicle driving guide line generation method and device, equipment and storage medium

Abstract

The invention relates to a vehicle driving guide line generation method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring a road complexity evaluation value of a vehicle on a preset distance road on a current driving path; the road complexity evaluation value is used for representing the complexity of the road; under the condition that the road complexity evaluation value is greater than or equal to a preset evaluation value threshold value, obtaining lane line points of a preset distance road matched with the current driving path from a preset lightweight map, and obtaining a first guide line matched with the current driving path according to the lane line points; and acquiring a second guide line generated by the road sensing system and used for guiding the vehicle to run on the current driving path, and correcting the second guide line by using the first guide line to obtain a vehicle driving guide line of the current driving path on the preset distance road. The accuracy of generating the vehicle driving guide line on the complex road by the vehicle can be improved, and then the driving risk is reduced.

Description

Technical Field

[0001] The present application relates to the field of intelligent driving technology, and in particular to a method, device, computer equipment, computer-readable storage medium and computer program product for generating a vehicle driving guide line. Background Art

[0002] Highway navigation assistance (NOA, Navigate Of Assist) is a landmark feature of high-end intelligent driving. It can independently complete functions such as overtaking, avoidance, and on- and off-ramps. Early NOA was mainly based on the coordinated use of high-precision maps and high-precision positioning to achieve assisted driving. However, due to the slow update cycle of high-precision maps and the high resource consumption brought by high-precision maps and high-precision positioning, the solution using high-precision maps has gradually been replaced by lightweight map solutions.

[0003] However, when faced with complex road scenarios during assisted driving, lightweight map solutions are unable to provide accurate and stable vehicle driving guidance lines in a timely manner. Summary of the invention

[0004] Based on this, it is necessary to provide a vehicle driving guide line generation method, device, computer equipment, computer readable storage medium and computer program product that can accurately and stably provide vehicle driving guide lines in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for generating a vehicle driving guide line, comprising:

[0006] Obtaining a road complexity evaluation value of a road at a preset distance on a current driving path of the vehicle; the road complexity evaluation value is used to characterize the complexity of the road;

[0007] When the road complexity evaluation value is greater than or equal to a preset evaluation value threshold, a lane line point of a preset distance road matching the current driving path is obtained from a preset lightweight map, and a first guide line matching the current driving path is obtained according to the lane line point;

[0008] A second guide line generated by a road perception system for guiding a vehicle to travel on a current driving path is obtained, and the second guide line is corrected using the first guide line to obtain a vehicle travel guide line on a road of a preset distance for the current driving path.

[0009] In one embodiment, obtaining a road complexity evaluation value of a road at a preset distance on a current driving path of the vehicle includes:

[0010] Acquire current positioning information of the vehicle, determine the current positioning position in the preset navigation map based on the current positioning information, query the preset navigation map for road structure information of a road within a preset distance from the current positioning position, and acquire a lane line clarity evaluation result in real time from a road perception system;

[0011] A road complexity evaluation value of the road at the preset distance is obtained according to the road structure information and the lane line clarity evaluation result.

[0012] In one embodiment, the road structure information includes lane average radius, road slope, ramp lane interval and lane quantity change information; the lane line clarity evaluation result includes lane line visibility; the lane line visibility is obtained based on pixels identified as lane line areas in the road image acquired by the road perception system and a preset pixel brightness threshold;

[0013] The obtaining, according to the road structure information and the lane line clarity evaluation result, a road complexity evaluation value of the preset distance road includes:

[0014] Obtaining a curve evaluation value according to the average radius of the lane, the road slope, and the lane line visibility;

[0015] Obtaining a ramp evaluation value according to the ramp lane interval, a preset interval upper limit threshold, and a preset interval lower limit threshold;

[0016] Obtaining a lane quantity change evaluation value according to the lane quantity change information and a preset quantity change threshold;

[0017] The road complexity evaluation value of the preset distance road is obtained according to the curve evaluation value, the ramp evaluation value and the lane quantity change evaluation value.

[0018] In one embodiment, obtaining the curve evaluation value according to the average lane radius, the road slope and the lane line visibility includes:

[0019] Obtaining a lane radius evaluation value according to the lane average radius, a preset radius upper limit threshold, and a preset radius lower limit threshold;

[0020] Obtaining a road slope evaluation value according to the road slope, and a preset slope upper limit threshold and a preset slope lower limit threshold;

[0021] Obtaining a lane line visibility evaluation value according to the lane line visibility, a preset visibility upper threshold value, and a preset visibility lower threshold value;

[0022] A curve evaluation value is obtained according to the lane radius evaluation value, the road slope evaluation value and the lane line visibility evaluation value.

[0023] In one of the embodiments, the current driving path is obtained based on a preset navigation map;

[0024] The acquiring a lane line point matching the current driving path from a preset lightweight map, and obtaining a first guide line matching the current driving path according to the lane line point, includes:

[0025] Obtaining a lane line point set corresponding to the current driving path in the preset lightweight map, and obtaining a lane line point corresponding to each lane from the lane line point set, and obtaining a lane center point of each lane according to the lane line point of each lane;

[0026] Performing curve fitting using the lane center points of the lanes to obtain first guide lines of the lanes;

[0027] The first guide line of the lane corresponding to the current driving path is used as the first guide line matching the current driving path.

[0028] In one embodiment, the obtaining a lane line point set corresponding to the current driving path in the preset lightweight map includes:

[0029] Obtaining the path starting point coordinates, path ending point coordinates and target path point coordinates of the current driving path on a preset distance road from the preset navigation map; the target path point coordinates are the coordinates of any other path point in the current driving path except the path starting point and the path ending point;

[0030] Generating a candidate waypoint area in the preset lightweight map according to the path starting point coordinates and the path ending point coordinates;

[0031] According to the coordinates of the target path point, a target path point position with the shortest distance to the target path point is obtained from each path point position in the candidate path point area in the preset lightweight map; each path point position is obtained based on the lane line point of each lane;

[0032] According to the lane line points of each lane corresponding to each of the target path points, a lane line point set corresponding to the current driving path in the preset lightweight map is obtained.

[0033] In one embodiment, the step of obtaining a second guide line generated by a road perception system for guiding a vehicle to travel on a current driving path includes:

[0034] Obtain candidate guide lines generated by a road perception system corresponding to multiple lanes in the current driving direction, obtain the similarity between each of the candidate guide lines and the first guide line, and use the candidate guide line with the highest similarity as the second guide line for guiding the vehicle to travel on the current driving path.

[0035] In one embodiment, the method of using the first guide line to correct the second guide line to obtain a vehicle driving guide line of the current driving path on a road at a preset distance includes:

[0036] Acquire a plurality of interpolation points of the first guide line and the second guide line, and determine a plurality of geometric center points according to the plurality of interpolation points;

[0037] Fitting is performed based on the plurality of geometric center points to obtain a vehicle driving guide line of the current driving path on a road of a preset distance.

[0038] In one embodiment, the step of obtaining candidate guide lines corresponding to multiple lanes in the current driving direction generated by a road perception system includes:

[0039] Using the road perception system to collect multiple road images in the current driving direction;

[0040] Obtaining a multi-lane panoramic image from a bird's-eye view based on the plurality of road images;

[0041] The multi-lane panoramic image is input into a pre-trained lane centerline generation model to obtain lane centerlines of multiple lanes in the current driving direction, and the multiple lane centerlines are used as candidate guide lines for multiple lanes in the current driving direction.

[0042] In one embodiment, obtaining a multi-lane panoramic image from a bird's-eye view based on the plurality of road images includes:

[0043] Matching feature points of the plurality of road images, and splicing the plurality of road images based on the matched feature points to obtain a multi-lane panoramic image with a perspective view;

[0044] The multi-lane panoramic image with a perspective view is perspective-transformed to obtain a multi-lane panoramic image with a bird's-eye view.

[0045] In a second aspect, the present application also provides a vehicle driving guide line generating device, comprising:

[0046] A road complexity evaluation value acquisition module is used to acquire a road complexity evaluation value of a road at a preset distance on the current driving path of the vehicle; the road complexity evaluation value is used to characterize the complexity of the road;

[0047] A first guide line acquisition module, configured to acquire a lane line point of a road of a preset distance matching the current driving path from a preset lightweight map when the road complexity evaluation value is greater than or equal to a preset evaluation value threshold, and obtain a first guide line matching the current driving path according to the lane line point;

[0048] The vehicle driving guide line generation module is used to obtain the second guide line generated by the road perception system for guiding the vehicle to travel on the current driving path, and use the first guide line to correct the second guide line to obtain the vehicle driving guide line of the current driving path on the road at a preset distance.

[0049] In a third aspect, the present application further provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the method described in the first aspect when executing the computer program.

[0050] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the method described in the first aspect.

[0051] In a fifth aspect, the present application further provides a computer program product, comprising a computer program, which implements the steps of the method described in the first aspect when executed by a processor.

[0052] The above-mentioned vehicle driving guide line generation method, device, computer equipment, computer-readable storage medium and computer program product first obtain a road complexity evaluation value of a preset distance road on the current driving path of the vehicle. When the road complexity evaluation value is greater than or equal to the preset evaluation value threshold, the lane line points of the preset distance road matching the current driving path are obtained from the preset lightweight map, and based on the above-mentioned lane line points, a first guide line matching the current driving path is obtained; a second guide line generated by a road perception system for guiding the vehicle to travel on the current driving path is obtained, and the second guide line is corrected by using the first guide line to obtain the vehicle driving guide line on the preset distance road of the current driving path. In this embodiment, by obtaining a dynamic road complexity evaluation value, when it is determined to be a complex road, a first guide line matching the current driving path is obtained through static lightweight map data, and a second guide line for guiding the vehicle to travel on the current driving path is obtained through a dynamic road perception system. The first guide line obtained by combining the static map data is used to correct the second guide line obtained by the dynamic data to obtain a vehicle driving guide line. This can improve the accuracy of generating vehicle driving guide lines on complex roads, thereby reducing driving risks. In addition, by evaluating the road complexity to determine whether to generate the first guide line, when the road complexity evaluation value is less than or equal to the preset evaluation value threshold, the intelligent driving system does not need to generate the first guide line, which can save data processing and transmission resources and reduce the loss of operating resources of the intelligent driving system. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the drawings required for use in the embodiments of the present application or related technical descriptions will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.

[0054] Figure 1 A schematic diagram of a process of generating a vehicle driving guide line in one embodiment;

[0055] Figure 2 A schematic diagram of a process for obtaining a road complexity evaluation value of a road at a preset distance on a current driving path of a vehicle in one embodiment;

[0056] Figure 3 A structural block diagram of an intelligent driving system used in a method for generating a vehicle driving guide line in an embodiment;

[0057] Figure 4 A flowchart of a method for generating a vehicle driving guide line in another embodiment;

[0058] Figure 5 A schematic diagram of the distribution of vehicle cameras on a vehicle in one embodiment;

[0059] Figure 6 is a schematic diagram of the positional relationship between a lane line point set and a lane line center point in one embodiment;

[0060] Figure 7 A schematic diagram of the effect of lane line fitting in a panoramic image with a perspective view in one embodiment;

[0061] Figure 8 A schematic diagram of projecting a visual guide line and a map guide line onto a bird's-eye view in one embodiment;

[0062] Fig. 9 is a structural block diagram of a vehicle driving guide line generating device in one embodiment;

[0063] Fig.10 This is a diagram of the internal structure of a vehicle-side control device in one embodiment. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0065] In the prior art, the intelligent driving system based on the lightweight map architecture has insufficient performance in complex road scenarios. For example, lane changes are not smooth, the timing of lane changes is unreasonable, and the curves are not centered in scenarios such as on-ramps, off-ramps, and continuous curves.

[0066] After research and analysis, the applicant found that the reason for the above phenomenon is that the center line of the lane in the high-precision map is deleted in the lightweight map, and the center line generation method based on vehicle sensors, when faced with continuous ramps, continuous curves or other complex road scenes with high speeds and many road forks, is limited by the sensor's perception stability and the convergence speed of the perception system, resulting in the inability to provide vehicle driving guide lines in a timely and accurate manner.

[0067] Therefore, in order to improve the performance of the intelligent driving system under the lightweight map architecture in complex road scenes, so that it can achieve the complex road scene performance under high-precision maps, it is urgent to provide a vehicle driving guide line generation method that can improve the performance of the intelligent driving system under the lightweight map architecture.

[0068] In an exemplary embodiment, Figure 1As shown, a method for generating a vehicle driving guide line is provided, and the method is applied to an intelligent driving system, and the intelligent driving system is set in a vehicle-side control device, and the vehicle-side control device can be an intelligent driving domain controller, and the intelligent driving domain controller plays a central control and coordination role in automatic driving and assisted driving scenarios. The method for generating a vehicle driving guide line applied to an intelligent driving system provided in this embodiment includes the following steps S102 to S106, wherein:

[0069] Step S102, obtaining a road complexity evaluation value of a road at a preset distance on the current driving path of the vehicle; wherein the road complexity evaluation value is used to characterize the complexity of the road.

[0070] The current driving path may refer to a driving path at the current driving time in a global path based on the navigation map generated based on a destination selected by a user on the navigation map.

[0071] Among them, the preset distance road may refer to a road within a preset distance. For example, if the preset distance is 2 kilometers, then the preset distance road is a road within 2 kilometers. Therefore, the preset distance road of the vehicle on the current driving path can be based on the current driving path from the current position to the road within 2 kilometers in the path direction. Then, the road complexity evaluation value of the road within 2 kilometers can be obtained. It should be noted that the road complexity evaluation value is used to characterize the complexity of the road, and can also be used to evaluate the difficulty of the driving environment, so as to determine whether additional information is needed to further generate vehicle driving guide lines. The vehicle driving guide line is a specific path line used to guide the vehicle to travel on the road. Specifically, the road complexity evaluation value can be obtained based on the vehicle perception system (such as on-board sensors), and can also be obtained based on the road environment information in the preset map, or based on the traffic information in the cloud server.

[0072] Step S104, when the road complexity evaluation value is greater than or equal to the preset evaluation value threshold, a lane line point of a preset distance road matching the current driving path is obtained from a preset lightweight map, and a first guide line matching the current driving path is obtained based on the lane line point.

[0073] The preset evaluation value threshold may be a pre-set threshold used to determine whether the road complexity has reached a level requiring additional measures.

[0074] The preset lightweight map may refer to a lightweight high-precision map, which can be obtained by cropping the elements of the high-precision map over a large area, so as to reduce the overall data volume while retaining relatively important map elements. For example, the preset lightweight map may contain key information required for intelligent driving, such as lane geometry, traffic sign location, and other information.

[0075] The lane line points may refer to a series of points representing lane lines on a map, and these points may be used to construct the geometric shape of the lane.

[0076] Specifically, when the road complexity evaluation value exceeds or equals the preset evaluation value threshold, the lane of the preset distance road matching the current driving path is first determined from the preset lightweight map, and after the lane is determined, the lane line point of the determined lane is extracted, and then the corresponding lane center point is determined according to the lane line point, and then the lane center line of the lane is obtained according to the lane center point, and the lane center line is used as the first guide line matching the current driving path. Among them, the first guide line reflects the ideal driving path based on the static map data, and can provide a reference guide line when the road complexity is high.

[0077] Step S106, obtaining a second guide line generated by the road perception system for guiding the vehicle to travel on the current driving path, and using the first guide line to correct the second guide line to obtain a vehicle driving guide line on the current driving path on a road at a preset distance.

[0078] Among them, the road perception system can refer to a system that integrates multiple sensors and algorithms, which can be used to monitor and understand the environment around the vehicle in real time.

[0079] Specifically, in this embodiment, the road perception system may include a camera and a road modeling system. The camera can be used to obtain the current environment image, and then the road edges, lane lines and other elements of each lane in the current driving direction of the current driving path can be identified based on the current environment image. Then, based on the road edges, lane lines and other elements of each lane identified above, the road modeling system outputs the lane center lines of each lane in the current environment image, and uses the lane center line as the second guide line to guide the vehicle on the current driving path. Because the road complexity evaluation value is greater than or equal to the preset evaluation value threshold, it is necessary to use the first guide line obtained above to correct the second guide line to obtain the vehicle driving guide line on the current driving path at a preset distance on the road, and then the intelligent driving system can perform vehicle driving control according to the vehicle driving guide line.

[0080] In the above-mentioned vehicle driving guide line generation method, first, a road complexity evaluation value of a preset distance road on the current driving path of the vehicle is obtained. When the road complexity evaluation value is greater than or equal to a preset evaluation value threshold, lane line points of the preset distance road matching the current driving path are obtained from a preset lightweight map, and based on the above-mentioned lane line points, a first guide line matching the current driving path is obtained; a second guide line generated by a road perception system for guiding the vehicle to travel on the current driving path is obtained, and the second guide line is corrected using the first guide line to obtain a vehicle driving guide line on the preset distance road of the current driving path. In this embodiment, by obtaining a dynamic road complexity evaluation value, when it is determined to be a complex road, a first guide line matching the current driving path is obtained through static lightweight map data, and a second guide line for guiding the vehicle to travel on the current driving path is obtained through a dynamic road perception system. The first guide line obtained by combining the static map data is used to correct the second guide line obtained by the dynamic data to obtain a vehicle driving guide line. This can improve the accuracy of generating vehicle driving guide lines on complex roads, thereby reducing driving risks. In addition, by evaluating the complexity of the road to decide whether to generate the first guide line, when the road complexity evaluation value is less than the preset evaluation value threshold, the intelligent driving system does not need to generate the first guide line, which can save data processing and transmission resources and reduce the loss of operating resources of the intelligent driving system.

[0081] In an exemplary embodiment, Figure 2 As shown, step S102 includes steps S202 to S204, wherein:

[0082] Step S202, obtain the current positioning information of the vehicle, determine the current positioning position in the preset navigation map based on the current positioning information, query the road structure information of the road at a preset distance starting from the current positioning position from the preset navigation map, and obtain the lane line clarity evaluation result from the road perception system in real time.

[0083] Among them, the current positioning information may refer to the exact geographic location coordinates obtained from the positioning tool pre-configured on the vehicle. The current positioning position may refer to the current position of the vehicle in the preset navigation map determined according to the position coordinates in the current positioning information. The preset navigation map may be used for users to select the navigation destination and generate a global path based on the navigation map. It can also be used to display the position of the vehicle in the navigation map and display nearby traffic events. The road structure information may include information such as the geometric features of the lane (such as curve curvature, slope, etc.), the number of lanes, the type of road, and traffic signs. The lane line clarity assessment result may refer to the evaluation result obtained by the road perception system capturing the lane image and then using image processing technology to quantitatively evaluate the visibility and clarity of the lane line, which can be used to reflect the degree of recognizability of the lane line.

[0084] Step S204, obtaining a road complexity evaluation value of a road at a preset distance according to the road structure information and the lane line clarity evaluation result.

[0085] Specifically, this embodiment first obtains the current positioning information of the vehicle through a pre-configured positioning tool, and then determines the current positioning position with the same position coordinates in the preset navigation map based on the position coordinates in the current positioning information, and then queries the road result information of the road section of a preset distance based on the current positioning position from the preset navigation map, and obtains the lane line clarity evaluation result within the perception range from the road perception system in real time; finally, a preset processing method or algorithm is used to perform comprehensive calculations based on the road structure information and the lane line clarity evaluation result to obtain a road complexity evaluation value of the preset distance road.

[0086] In this embodiment, by comprehensively considering the road structure information and the lane line clarity, the accuracy of the road complexity assessment can be improved, thereby improving the accuracy of subsequent measures taken based on the road complexity evaluation value.

[0087] In an exemplary embodiment, the road structure information includes lane average radius, road slope, ramp lane interval, and lane quantity change information; the lane line clarity evaluation result includes lane line visibility; the lane line visibility is obtained based on pixels identified as lane line areas in the road image acquired by the road perception system and a preset pixel brightness threshold; the above step S204 may specifically include:

[0088] Step S2041, obtaining a curve evaluation value according to the average lane radius, road slope, and lane line visibility;

[0089] Step S2042, obtaining a ramp evaluation value according to the ramp lane interval, a preset interval upper limit threshold, and a preset interval lower limit threshold;

[0090] Step S2043, obtaining a lane quantity change evaluation value according to the lane quantity change information and a preset quantity change threshold;

[0091] Step S2044, obtaining a road complexity evaluation value of a preset distance road according to the curve evaluation value, the ramp evaluation value and the lane quantity change evaluation value.

[0092] Among them, road structure information may include lane average radius, road slope, ramp lane spacing and lane number change information, and lane line clarity assessment results include lane line visibility; lane line visibility is obtained based on pixels identified as lane line areas in road images acquired by the road perception system and a preset pixel brightness threshold. Specifically, after the road perception system acquires the road image, it identifies the image area corresponding to the lane lines in the road image, and then compares the pixels in the image area corresponding to the lane lines one by one, and takes pixels in the area whose brightness is greater than the preset pixel brightness threshold as visible pixels, and then takes the ratio of visible pixels to all pixels in the lane line image area as lane line visibility.

[0093] The road complexity described by the road complexity evaluation value in this embodiment is to evaluate the complexity from the perspective of whether the system can quickly predict the guide line based on vision. On the one hand, the future guide line (i.e., the second guide line generated by the road perception system to guide the vehicle on the current driving path) is predicted based on the image seen by the camera, which is limited by the distance and clarity seen by vision. Therefore, when driving on curves or ramps, weather, lighting, traffic volume, road curvature, road undulations, etc. should be considered as factors. On the other hand, when the road topology is complex, the system has a large amount of calculations, and it is impossible to solve the optimal solution in a short time, the visual guide line will be uneven and jumpy, so whether it is continuously merged, crossed, and bifurcated also needs to be considered as a factor. Specifically, the above steps S2041-step S2044 can be obtained by the following process:

[0094] 1) For curves, the main considerations are the decreased visibility of lane lines due to lighting and weather, and the limited visual distance and clarity of curves due to a small lane radius or a large slope. Therefore, factors that need to be considered in curve scenarios include lane line visibility, average lane radius, and road slope. In this embodiment, a curve evaluation value is obtained based on lane line visibility, average lane radius, and road slope. For example, the curve evaluation value can be obtained by a lane line visibility evaluation value T1, a lane average radius evaluation value T2, and a road slope evaluation value T3.

[0095] 2) The ramp mainly considers that the visual guide line does not have time to converge and stabilize due to the short-distance multiple lane separation scene. Therefore, for the ramp scene, if the lane separation scene occurs, the interval S (ramp lane separation interval) where the lane separation occurs is determined. In this embodiment, the ramp evaluation value is obtained according to the ramp lane separation interval, the preset interval upper limit threshold and the preset interval lower limit threshold, which can be obtained by the following formula:

[0096] T

[0097] Among them, T4 is the ramp evaluation value; S is the interval of lane division, when the lane division interval is less than or equal to 100m (meters), the score is 100 points; when the lane division interval is greater than 200m, the score is 0 points; when the interval is between 100 and 200, the score is -S+200; among them, the preset interval upper limit threshold can be 200m, and the preset interval lower limit threshold can be 100m. It should be noted that the preset interval upper limit threshold and the preset interval lower limit threshold can also be other values, which can be set or adjusted according to actual needs.

[0098] 3) The increase or decrease of lanes on expressways or partial roads will cause the visual guide line to jump. In this embodiment, the lane quantity change evaluation value is obtained based on the lane quantity change information and the preset quantity change threshold. The specific steps may be: the lane quantity change threshold is set to 1, that is, 100 points are directly accumulated when a lane is increased or reduced, and then the lane quantity change evaluation value T5 is obtained.

[0099] In summary, according to the curve evaluation value, the ramp evaluation value and the lane quantity change evaluation value, the road complexity evaluation value of the preset distance road is obtained, that is, the road complexity evaluation value T=T1+T2+T3+T4+T5.

[0100] In this embodiment, by considering lane line visibility, average lane radius and road slope, the system can better predict potential risks in curves, and the ramp evaluation value corresponding to the ramp lane interval can make driving responses in advance for complex traffic conditions. The lane number change evaluation value can timely adjust the vehicle driving guide line to prevent driving discontinuity caused by lane changes. By comprehensively considering multiple factors such as lane average radius, road slope, lane line visibility, ramp lane interval, lane number change information, the above embodiment provides a comprehensive road complexity scoring system. This multi-dimensional evaluation method can more accurately reflect the impact of road conditions on autonomous driving, thereby improving the adaptability and accuracy of the intelligent driving system on complex road sections.

[0101] In an exemplary embodiment, the above step S2041 may specifically include:

[0102] Step S21, obtaining a lane radius evaluation value according to the average lane radius, a preset radius upper limit threshold, and a preset radius lower limit threshold;

[0103] Step S22, obtaining a road slope evaluation value according to the road slope, and a preset upper slope threshold and a preset lower slope threshold;

[0104] Step S23, obtaining a lane line visibility evaluation value according to the lane line visibility, and a preset visibility upper threshold and a preset visibility lower threshold;

[0105] Step S24, obtaining a curve evaluation value according to the lane radius evaluation value, the road slope evaluation value and the lane line visibility evaluation value.

[0106] Among them, after the road perception system obtains the road image, it identifies the image area corresponding to the lane line in the road image, and then compares the pixels in the image area corresponding to the lane line one by one, and takes the pixels in the area whose brightness is greater than the preset pixel brightness threshold as visible pixels, and then takes the ratio of visible pixels to all pixels in the lane line image area as the lane line visibility.

[0107] Specifically, the curve mainly considers the reduced visibility of the lane line caused by lighting and weather, and the limited visual visibility and clarity of the curve caused by a small lane radius or a large slope. Therefore, the above steps S21 to S24 can be obtained by the following process:

[0108] 1) For lane line visibility in curved scenes, the current light intensity, snow and water on the road, and the wear of the lane line itself may cause the lane line visibility to decrease; specifically, the curve evaluation value is obtained based on the average lane radius, road slope, and lane line visibility. The evaluation value corresponding to the lane line visibility can be obtained by the following formula:

[0109] T

[0110] Wherein, T1 is the lane line visibility evaluation value, V represents the lane line visibility, when the lane line visibility is less than or equal to 40%, the score is 100; when the lane line visibility is greater than or equal to 90%, the score is 0; when the lane line visibility is between 40% and 90%, the score is -200V+180. Wherein, the preset upper limit threshold of visibility can be 90%, and the preset lower limit threshold of visibility can be 40%. It should be noted that the preset upper limit threshold of visibility and the preset lower limit threshold of visibility can also be other values, which can be set or adjusted according to actual needs.

[0111] 2) The lane radius evaluation value in the curve scene can be obtained based on the average lane radius, the preset radius upper limit threshold and the preset radius lower limit threshold, which can be obtained by the following formula:

[0112] T

[0113] Wherein, T2 is the lane radius evaluation value, R is the lane average radius, when the lane average radius is less than or equal to 30m, the integral is 100 points, when the lane average radius is greater than or equal to 80m, the integral is 0 points, and when the lane average radius is between 30 and 80m, the integral is -2R+160. Wherein, the preset radius upper limit threshold may be 80m, and the preset radius lower limit threshold may be 30m. It should be noted that the preset radius upper limit threshold and the preset radius lower limit threshold may also be other values, which may be set or adjusted according to actual needs.

[0114] 3) The road slope evaluation value in the curve scene can be obtained according to the road slope, and the preset upper slope threshold and the preset lower slope threshold. Specifically, it can be obtained by the following formula:

[0115] T

[0116] Wherein, T3 is the road slope evaluation value, G is the road slope; when the road slope is less than or equal to 3°, the points are 0; when the road slope is greater than or equal to 8°, the points are 100; when the road slope is between 3° and 8°, the points are 20G-60. Wherein, the preset upper limit threshold of the slope can be 8°, and the preset lower limit threshold of the slope can be 3°. It should be noted that the preset upper limit threshold of the slope and the preset lower limit threshold of the slope can also be other values, which can be set or adjusted according to actual needs.

[0117] In summary, the curve evaluation value is obtained according to the lane radius evaluation value, the road slope evaluation value and the lane line visibility evaluation value. Specifically, the curve evaluation value can be obtained by the sum of the lane line visibility evaluation value T1, the lane radius evaluation value T2 and the road slope evaluation value T3, that is, T1+T2+T3.

[0118] In this embodiment, a curve evaluation value is derived for the curve through the average lane radius, road slope and lane line visibility. This can fully consider the effects of reduced lane line visibility caused by lighting and weather, and limited visual visibility and clarity of the curve due to a small lane radius or excessive slope, thereby improving the accuracy and reliability of the curve evaluation value.

[0119] In an exemplary embodiment, the current driving path is obtained based on a preset navigation map; in the step S104, lane line points matching the current driving path are obtained from the preset lightweight map, and a first guide line matching the current driving path is obtained based on the lane line points. This content may specifically include: obtaining a lane line point set corresponding to the current driving path in the preset lightweight map, and obtaining lane line points corresponding to each lane from the lane line point set, and obtaining lane center points of each lane based on the lane line points of each lane; performing curve fitting using the lane center points of each lane to obtain a first guide line for each lane; and using the first guide line of the lane corresponding to the current driving path as the first guide line matching the current driving path.

[0120] The current driving path may refer to a future driving path obtained based on the current vehicle position in a global path in a navigation map, and the global path in the navigation map may be a global path generated based on a destination selected by a user in the navigation map. Lane line points may refer to a series of points representing lane lines on a map, and these points may be used to construct the geometry of a lane.

[0121] Specifically, the current driving path is obtained based on a preset navigation map, and then a lane line point set corresponding to the current driving path is obtained in the preset lightweight map according to the current driving path, wherein these waypoints may include lane line points of multiple lanes, and then the lane line points corresponding to each lane are obtained from the lane line point set; for example, the lane line point set includes lane line points of two lanes, namely A, B, and C waypoints, A and B waypoints surround the first lane, and B and C waypoints surround the second lane, and then the lane center point of the first lane can be obtained according to A and B waypoints, and the lane center point of the second lane can be obtained according to B and C waypoints; wherein A waypoint includes lane line points along the driving path There are multiple waypoints A1, A2, A3, etc. in the driving path direction, and B waypoints also include multiple waypoints B1, B2, B3, etc. along the driving path direction. Therefore, the lane center point of the first lane can be obtained through the A1 and B1 waypoints corresponding to the first lane, and multiple lane center points of the first lane can be obtained according to the remaining waypoints A2, B2, A3, B3, etc., and curve fitting is performed using multiple lane center points to obtain the first guide line of the first lane; similarly, the first guide lines of the remaining lanes can be obtained, and then the specific lanes included in the path are determined according to the current driving path, and the first guide line of the lane corresponding to the current driving path is used as the first guide line matching the current driving path.

[0122] In this embodiment, a lane line point set is obtained from a preset lightweight map through the current driving path, and then the lane center point is obtained based on the lane line point of each lane. Curve fitting is then performed using the lane center point to obtain the first guide line of each lane. The precise lane information in the preset lightweight map can improve the accuracy of the first guide line generation.

[0123] In an exemplary embodiment, the above-mentioned acquisition of a lane line point set corresponding to the current driving path in a preset lightweight map may specifically include: acquiring the path starting point coordinates, path ending point coordinates and target path point coordinates of the current driving path at a preset distance from a preset navigation map; the target path point coordinates are the coordinates of any other path point in the current driving path except the path starting point and the path ending point; generating a candidate path point area in the preset lightweight map based on the path starting point coordinates and the path ending point coordinates; acquiring the target path point with the shortest distance to the target path point from each path point in the candidate path point area in the preset lightweight map based on the target path point coordinates; each path point is obtained based on the lane line points of each lane; and obtaining a lane line point set corresponding to the current driving path in the preset lightweight map based on the lane line points of each lane corresponding to each target path point.

[0124] Among them, the preset navigation map can be used for users to select the navigation destination and generate a global path based on the navigation map. It can also be used to display the position of the vehicle in the navigation map, as well as display nearby traffic events, etc. The path starting point coordinates can be the geographical coordinates of the starting point of the current driving path within the preset distance road range, and the path ending point coordinates can be the geographical coordinates of the ending point of the current driving path within the preset distance road range. The candidate waypoint area can refer to an area containing relevant lane line points delineated in the lightweight map based on the path starting point coordinates and the path ending point coordinates. The target path point position can refer to the actual path point found in the candidate waypoint area with the smallest distance from the given target path point coordinates, and the target path point position can correspond to the lane line point of each lane.

[0125] Exemplarily, the path starting point coordinates (A, B), path ending point coordinates (C, D) and target path point coordinates (X, Y) of the current driving path at a preset distance are obtained from the preset navigation map, and then according to the path starting point coordinates (A, B) and the path ending point coordinates (C, D), a rectangular or polygonal area can be roughly circled on the preset lightweight map. This area is the candidate waypoint area, and the area should cover all possible road sections from the starting point to the end point; then in the candidate waypoint area, according to the target path point coordinates (X, Y), the actual path point closest to it is found, which can be obtained by calculating the Euclidean distance or Manhattan distance between each path point and the target path point To implement, assume that a nearest path point (M, N) is found and taken as the target path point, and the lane line points of each lane where the target path point (M, N) is located are obtained. For example, the lane line points corresponding to the target path point (M, N) in the first lane include lane line point a and lane line point b, and the lane line points corresponding to the second lane include lane line point b, lane line point c, and the third lane, etc., and then traverse all target path points in the current driving path, obtain the lane line points of each lane corresponding to all target path points corresponding to the current driving path, and form a lane line point set, that is, the lane line point set corresponding to the current driving path in the preset lightweight map is obtained.

[0126] In this embodiment, the preset navigation map is matched with the preset lightweight map through coordinates to obtain a lane line point set, which can accurately determine the lane line points related to the current driving path, improve the path accuracy of matching paths in different maps, and ensure the accuracy of the guide lines generated according to the lane line point set.

[0127] In an exemplary embodiment, the step 106 of acquiring the second guide line generated by the road perception system for guiding the vehicle on the current driving path may specifically include: acquiring candidate guide lines generated by the road perception system corresponding to multiple lanes in the current driving direction, acquiring the similarity between each candidate guide line and the first guide line, and using the candidate guide line with the highest similarity as the second guide line for guiding the vehicle on the current driving path.

[0128] The candidate guide line may refer to a possible driving path generated by a road perception system based on multiple lanes in the current driving direction. The similarity may be used to measure the degree of matching between two guide lines, which may be determined by geometric distance, curve fitting error or other distance correlation analysis algorithms.

[0129] Specifically, a possible driving path generated by the road perception system based on multiple lanes in the current driving direction, that is, candidate guide lines of multiple lanes, is obtained, and then the similarity between each candidate guide line and the first guide line is obtained through a preset algorithm, and the candidate guide line with the highest similarity is used as the second guide line for guiding the vehicle to travel on the current driving path.

[0130] In this embodiment, by obtaining the similarity between the candidate guide line and the first guide line, the candidate guide line closest to the first guide line can be selected, thereby improving the reliability of correcting the second guide line using the first guide line in subsequent steps.

[0131] In an exemplary embodiment, the above step 106 uses the first guide line to correct the second guide line to obtain the vehicle driving guide line of the current driving path on the road at a preset distance. This content may specifically include: obtaining multiple interpolation points of the first guide line and the second guide line, and determining multiple geometric center points based on the multiple interpolation points; fitting based on the multiple geometric center points to obtain the vehicle driving guide line of the current driving path on the road at a preset distance.

[0132] The interpolation point may refer to a series of coordinate points selected from the first guide line and the second guide line, which are used to perform interpolation calculation between the two guide lines. The geometric center point may refer to a point obtained by geometrically averaging the interpolation points of the first guide line and the second guide line at the same position.

[0133] Specifically, multiple interpolation points of the first guide line and the second guide line are obtained, and then the geometric center points of the interpolation points at corresponding positions are determined based on the multiple interpolation points, and then all the geometric center points are fitted to obtain the vehicle driving guide line of the current driving path on the road at a preset distance.

[0134] In this embodiment, by combining the information of the first guide line and the second guide line, the generated vehicle driving guide line takes into account both static map data and real-time perception data (including data reflecting actual road conditions and traffic conditions), thereby improving the accuracy of the vehicle driving guide line.

[0135] In an exemplary embodiment, the above step of obtaining candidate guide lines corresponding to multiple lanes in the current driving direction generated by the road perception system may include: using the road perception system to collect multiple road images in the current driving direction; obtaining a multi-lane panoramic image from a bird's-eye view based on the multiple road images; inputting the multi-lane panoramic image into a pre-trained lane centerline generation model to obtain lane centerlines of multiple lanes in the current driving direction, and using the multiple lane centerlines as candidate guide lines for the multiple lanes in the current driving direction.

[0136] The road perception system may collect multiple road images in the current driving direction through multiple on-board cameras; the multi-lane panoramic image may refer to converting multiple road images into a panoramic image viewed from above through image stitching and transformation technology. The pre-trained lane centerline generation model may be a pre-trained machine learning or deep learning model for extracting lane centerlines from an input multi-lane panoramic image.

[0137] Specifically, a road perception system is used to collect multiple road images in the current driving direction, and then image stitching technology and image transformation technology are used to convert the multiple road images into a multi-lane panoramic image from a bird's-eye view. The multi-lane panoramic image is then input into a pre-trained lane centerline generation model to obtain lane centerlines of multiple lanes in the multi-lane panoramic image, that is, lane centerlines of multiple lanes in the current driving direction are obtained, and the multiple lane centerlines are used as candidate guide lines for multiple lanes in the current driving direction.

[0138] In this embodiment, by acquiring a multi-lane panoramic image from a bird's-eye view, more comprehensive road information can be acquired, thereby improving the accuracy of extracting lane center lines, that is, improving the accuracy of candidate guide lines.

[0139] In an exemplary embodiment, in the above embodiment, a multi-lane panoramic image with a bird's-eye view is obtained based on multiple road images. This content may specifically include: matching feature points of multiple road images, stitching the multiple road images based on the matched feature points to obtain a multi-lane panoramic image with a perspective view; performing perspective transformation on the multi-lane panoramic image with a perspective view to obtain a multi-lane panoramic image with a bird's-eye view.

[0140] Feature point matching can refer to a computer vision technology that is used to find points of correspondence between different images. These points are usually areas with unique properties in the image, such as corners, edges, etc. Image stitching can refer to the process of merging two or more partially overlapping images into a wide field of view or panoramic image. Perspective transformation can refer to a two-dimensional mapping that can simulate how objects in three-dimensional space are projected onto a two-dimensional plane. The perspective transformation can change the viewpoint of the image so that the image can be viewed from different angles.

[0141] Specifically, feature points of multiple road images are matched, matching feature points are found between adjacent road images, and a suitable geometric transformation matrix is ​​calculated using the found matching feature points. The geometric transformation matrix is ​​applied to splice the multiple road images in a correct order and position to form a multi-lane panoramic image with a perspective view. The multi-lane panoramic image with a perspective view is then converted into a bird's-eye view with a perspective transformation matrix to obtain a multi-lane panoramic image with a bird's-eye view.

[0142] In this embodiment, by performing feature point matching, image stitching and perspective transformation on the road image, a series of local road images can be effectively synthesized into a global bird's-eye view panoramic image, which can improve the comprehensiveness of the road information.

[0143] Figure 3 This is a structural block diagram of the intelligent driving system used in the vehicle driving guide line generation method in the embodiment of the present application. The intelligent driving system may include a camera (on-board camera), a navigation map SD map (preset navigation map), a lightweight high-precision map HD lite (lightweight map), an intelligent cockpit domain controller CDC (cockpit CDC), an intelligent driving domain controller ADC (intelligent driving ADC), and a PBOX (Positioning Box, a high-precision combined positioning system). Please refer to Figure 3 Among them, the car camera is mainly responsible for driving perception, identifying the current road edge, lane line and other elements for subsequent perception mapping; the navigation map SD map runs in the cockpit domain controller CDC, which is mainly used by users to select the destination of high-speed navigation assistance NOA, generate a global path based on the navigation map, and display the vehicle's position in the navigation map and nearby traffic events, etc.; PBOX is mainly used to locate the precise position of the vehicle and feed back the position information to the map module to update the vehicle's position; the lightweight high-precision map HD lite runs in the intelligent driving domain controller ADC, which mainly stores the road network in a fixed area and other road element information attached to it; the navigation map SD map sends the navigation destination and the global path based on the navigation map to the lightweight high-precision map HD lite. After receiving the data, the lightweight high-precision map HD lite performs path matching based on its own data to obtain a global path based on the lightweight high-precision map HD lite; as the vehicle position is updated, the navigation map and the lightweight high-precision map output the path and other traffic element information of the next 2 kilometers (preset distance road) near the vehicle; the intelligent driving ADC outputs the perception guide line (the second guide line) based on the road modeling system of the perception result, based on HD The map guide line (first guide line) output by lite data processing is fused to output the final guide line (vehicle driving guide line), which will be used in the subsequent path tracking module for vehicle control.

[0144] Figure 4 This is a flowchart of a method for generating a vehicle driving guide line in an embodiment of the present application. Figure 1-Figure 3 Reference Figure 4 The present application also provides an optional implementation method including steps S1 to S9, specifically:

[0145] Step S1, initialization: After the user gets on the car and sets the destination based on the navigation map, the system will perform global path planning and provide multiple optional paths for the user to choose from. When the user selects a route and turns on the NOA function in an area where the high-speed NOA conditions are met, the system will generate visual guide lines and perform road complexity scoring at the same time.

[0146] Step S2, road complexity scoring; when the system enters the NOA function, the road complexity scoring module will receive information from the navigation map and high-precision positioning, and query the road information of the next 2 kilometers based on the positioning on the navigation map to obtain the average curvature, slope, merging and splitting of the future road. In addition, the lane line clarity evaluation result of the perception module will be accepted for complexity scoring. The "complex road" described in this embodiment is to evaluate the complexity from the perspective of whether the system can quickly predict the guide line based on vision. On the one hand, the future guide line is predicted based on the image seen by the camera, which is limited by the distance and clarity seen by vision. Therefore, when driving on a curve or ramp, weather, lighting, traffic flow, road curvature, road undulations, etc. should be considered as factors. On the other hand, when the road topology is complex, the system has a large amount of calculations and cannot solve the optimal solution in a short time, the visual guide line will be not smooth, jumping, etc., so whether it is continuously imported and exported, crossed, and bifurcated also needs to be considered as a factor. In order to make the complex road judgment programmatic, an integral strategy is designed to integrate multiple factors for judgment, and the integral is used to reflect the complexity of the current road for the rapid visual generation of guide lines. The specific strategies are as follows:

[0147]

[0148] As described in the table above, this embodiment divides complex road types into three categories, namely, curves, ramps, and others. Among them, the curves mainly consider the reduced visibility of lane lines caused by lighting and weather, and the limited visual visibility and clarity of curves caused by too small lane radius or too large slope. The ramps mainly consider the lack of time for the visual guide lines to converge and stabilize due to the short-distance multiple lane separation scenes. In addition, the visual guide line jumps caused by the increase or decrease of lanes on expressways or some roads are also considered.

[0149] For lane line visibility in curved scenes, the current light intensity, snow and water on the road, and the wear of the lane line itself may cause the lane line visibility to decrease. The lane line visibility is characterized based on the ratio of visible lane line pixels in the image to the total lane line pixels, and the road complexity integral corresponding to the lane line visibility is calculated using the following formula:

[0150] T

[0151] Wherein, T1 is the integral corresponding to the lane line visibility, V represents the lane line visibility, when the lane line visibility is less than or equal to 40%, the integral is 100 points; when the lane line visibility is greater than or equal to 90%, the integral is 0 points; when the lane line visibility is between 40% and 90%, the integral is -200V+180. Wherein, the preset upper limit threshold of visibility can be 90%, and the preset lower limit threshold of visibility can be 40%. It should be noted that the preset upper limit threshold of visibility and the preset lower limit threshold of visibility can also be other values, which can be set or adjusted according to actual needs.

[0152] The average radius integral evaluation strategy in the curve scene is as follows:

[0153] T

[0154] Among them, T2 is the integral corresponding to the average radius of the lane, R is the average radius of the lane, when the average radius of the lane is less than or equal to 30m, the integral is 100 points, when the average radius of the lane is greater than or equal to 80m, the integral is 0 points, and when the average radius of the lane is between 30 and 80m, the integral is -2R+160. Among them, the preset radius upper limit threshold can be 80m, and the preset radius lower limit threshold can be 30m. It should be noted that the preset radius upper limit threshold and the preset radius lower limit threshold can also be other values, which can be set or adjusted according to actual needs.

[0155] The road slope integral evaluation strategy in the curve scene is shown in the following formula:

[0156] T

[0157] Among them, T3 is the integral corresponding to the road slope, G is the road slope, when the slope is less than or equal to 3°, the integral is 0 points, when the slope is greater than or equal to 8°, the integral is 100 points, and when the radius is between 3° and 8°, the integral is 20G-60.

[0158] For the ramp scenario, if a road split occurs, the interval S where the lane split occurs is determined using the following formula:

[0159] T

[0160] Among them, T4 is the integral corresponding to the ramp lane interval, S is the interval at which the lanes appear, when the lane interval is less than or equal to 100m (meters), the integral is 100 points; when the lane interval is greater than 200m, the integral is 0 points; when the interval is between 100 and 200, the integral is -S+200; wherein, the preset interval upper limit threshold may be 200m, and the preset interval lower limit threshold may be 100m. It should be noted that the preset interval upper limit threshold and the preset interval lower limit threshold may also be other values, which can be set or adjusted according to actual needs.

[0161] For lane additions or lane reductions on expressways or parts of roads, 100 points will be awarded directly.

[0162] To sum up, the integral T=T1+T2+T3+T4+T5+T6.

[0163] Step S3, start map guide line determination; the above step S2 determines the complexity of the road in multiple dimensions, and finally obtains the integral T. If T>=100, start map guide line generation, if T<100, only use visual guide lines.

[0164] Step S4: Image stitching: When the system turns on NOA, the visual guide line will be used to control the vehicle by default, and the generation of the visual guide line requires image acquisition first. The specific camera layout is as follows Figure 5 As shown in Figure 2, the intelligent driving system uses a total of 7 cameras, of which 4 cameras are used for subsequent BEV perspective image stitching and road modeling. Figure 5 Cameras numbered 2, 4, 5, and 6.

[0165] The main steps of the image acquisition process are: the optical signal is converted into an electrical signal through the CMOS sensor; the electrical signal is digitized after being processed by the ADC; the digital signal is converted into a high-speed serial data stream through a serializer; it is transmitted to the domain controller over long distances through technologies such as GMSL; in the domain controller, the deserializer restores the serial data stream to parallel data.

[0166] The main steps of the image stitching process are: denoising, white balancing, and color correction of the original images captured by the camera; performing distortion correction on images from different cameras based on the camera's internal parameters; identifying and extracting key feature points such as corners or edges in the image through algorithms; finding matching feature point pairs of adjacent camera images; based on the feature point pairs, inversely calculating the geometric transformation model between images to achieve precise alignment of images; after alignment, stitching the images from multiple cameras to form a continuous panoramic image; and converting the panoramic image from a perspective perspective to a BEV (bird's eye view) perspective by calculating the perspective transformation matrix.

[0167] Step S5, visual guide line (second guide line) generation; based on the spliced ​​transformed BEV view, use the end-to-end centerline generation model to output the lane line. The input of this model is the spliced ​​BEV image, and the output is a 5th-order polynomial lane center guide line based on the ego-vehicle coordinate system. The general steps of using the end-to-end visual guide line generation model are: based on the driving data collection of human drivers; manual or automatic lane centerline annotation based on the spliced ​​BEV view; end-to-end model training based on the annotated image; use the end-to-end visual guide line generation model to generate visual guide lines for the spliced ​​images collected during driving.

[0168] Step S6, map matching and waypoint extraction: To achieve low computing power consumption, the evaluation of road complexity is based on the SD map. When the complexity of the path in step 3 is determined to require waking up the map guide line, the system will match the SD map with the HD lite map and extract the waypoints corresponding to the next 2 kilometers of the path on the HD lite map. The map matching and waypoint extraction steps are as follows: take out the latitude and longitude coordinates of the starting point SS (latSS, LonSS) and the end point SE (LatSE, LonSE) of the path points in the next 2 kilometers on the SD map; use the latitude and longitude of the two points SS and SE to draw a rectangle on the HD lite map as the optional point range (candidate waypoint area) for HD lite map matching; take out all road waypoints Si (LatSi, LonSi) (target path points) in the next 2 kilometers on the SD map, i is the waypoint number; calculate the distance between Si and all road waypoints Hj (LatHj, LonHj) in the rectangle on the HD lite map, and the distance is expressed as: Di=(LatSi-LatHj)*(LatSi-LatHj)+(LonSi- LonHj)*(LonSi-LonHj); define Hj with the smallest distance Di to Si as the corresponding point of Si on the SD map on the HDlite map; perform the above operations on all Si in turn to obtain the set H of all waypoints on the HD lite map corresponding to the path on the SD map (lane line point set).

[0169] Step S7, map guide line (first guide line) generation: In step S6, a set of waypoints H in the HD lite map of the road determined to be complex has been obtained, and HD lite element processing is performed based on these waypoints to generate a map-based guide line. The specific steps are as follows:

[0170] Lane line point set extraction: Lane lines are described in the form of points in the light map. The point attributes of lane lines contain predecessor and successor information. Straight lines can be used to connect the predecessor and successor points of each point to obtain the basic road lane envelope. Select the elements Hk of the waypoint set H corresponding to the path in the HD lite map in sequence, {Hk|Hk∈H (k is the subscript)}, and then extract the lane line point set Lk (k is the subscript) belonging to the Hk waypoint.

[0171] Lane centerline point set calculation: For the lane line points Lkj with the same index belonging to the same waypoint Hk, {Lkj|Lkj∈Lk (j is the subscript)}, the corresponding lane center point can be obtained by taking the average of the x and y coordinates of the two lane line points of a lane, such as Figure 6 As shown, Figure 6The edge lane line point sets LK on both sides are lane line point sets, and the lane line center points LC0, LC1, etc. are calculated lane center line point sets. There is no explicit lane line, so the dashed line in the figure represents the lane line based on point fitting. The relevant point with index 1 is used as an example to describe the relationship. S1 is a road point on the path on the SD map, H1 is the corresponding road point mapped to the HD lite map, L11 and L12 are two lane line points of a lane corresponding to the H1 road point (there are other Lkj for multiple lanes), with L11 coordinates (L11x, L11y), and L12 coordinates (L12x, L12y), then the corresponding lane line center point LC1 (x, y) has the following relationship: the horizontal coordinate x of the center line point is (L11x+L12x) / 2; the vertical coordinate y of the center line point is (L11y+L12y) / 2.

[0172] Lane line fitting: Based on the calculated center line point set and lane line set, a fifth-order polynomial is used to fit the lane line points. Every 6 points are grouped together and sequentially substituted into the formula y=ax^5+bx^4+cx^3+dx^2+e^x+f, where a to f are unknown coefficients. The lane line and center line equations within the current horizontal coordinate x range can be obtained. The final effect of lane line fitting in the panoramic image with perspective view is shown below. Figure 7 As shown in the figure, 701 and 702 are lane lines at the boundary, 703 is the fitted non-boundary lane line, 704 is the fitted lane center line, and 705 is the lane center line where the current vehicle is traveling. The fitted lane center line is used as the map guide line (first guide line) and sent to the subsequent guide line fusion module.

[0173] Step S8, guide line fusion; the visual guide line module does not know the future global path and planning of the vehicle, but recommends the guide lines of each lane that the vehicle may currently go to based on the lane it currently belongs to. Limited by the camera recognition distance and FOV (Field of View), the visual guide line is generally shorter than the map guide line, but it is generated based on the current real-time road conditions and more realistically reflects the current road and traffic conditions. The map guide line is based on the known global path planning solution and fitting, has a clear road choice, and considers a longer distance, so the choice is more reasonable. However, it uses non-real-time data from the cloud, and unreasonable planning may occur (such as temporary road damage or obstacles). In order to make full use of the advantages of the above two types of guide lines, in complex sections where map guide lines are enabled, the visual guide lines need to be fused with the map guide lines and output to the subsequent vehicle planning control module. The schematic diagram of projecting the visual guide lines and map guide lines onto the BEV view is shown in the figure. Figure 8 As shown, Figure 8The three guide lines 801, 802, and 803 are visual guide lines, and the guide line 804 is a map guide line. The arrows in the lanes in the figure only indicate the driving direction. The following steps are for guide line fusion:

[0174] Visual guide line selection: First, based on the map guide lines, suitable future road guide lines are selected from the visual guide lines. The specific method can use distance correlation analysis, that is, interpolate points for each visual and map guide line at the same horizontal x interval, and then solve the distance between the corresponding points of different visual guide lines and map guide lines. Finally, the distance between each visual guide line and the map guide line is summed up, and the visual guide line with the smallest total distance to the map guide line is selected as the representative to participate in the subsequent guide line fusion.

[0175] Guideline fusion: After selecting the appropriate visual guideline to be fused, based on the interpolation points in the previous step, the geometric center points of the map guideline and the visual guideline with the same index are calculated in turn. The specific method is the same as the lane center point solution in step 7. Finally, based on the solved vision and map center point set, a fifth-order polynomial fitting is performed to obtain the final guideline. The specific fitting method is the same as the lane line fitting in step 7.

[0176] Step S9, end; after the vehicle passes through the complex route section, that is, after the vehicle passes through the current map guide line range, the entire map guide line related program will exit, and the system will reuse the default visual guide line and wait for the next wake-up.

[0177] In this embodiment, the following beneficial effects can be achieved through the above technical solution:

[0178] Improve the accuracy of road complexity assessment: This embodiment provides a comprehensive road complexity scoring system by comprehensively considering multiple factors such as lane visibility, road curvature, slope, lane spacing, etc. This multi-dimensional evaluation method can more accurately reflect the impact of road conditions on the autonomous driving system, thereby improving the adaptability and accuracy of the autonomous driving system on complex roads.

[0179] Enhanced safety of automated driving systems: The system can activate map guidance lines when it identifies a highly complex road, which helps provide additional safety on complex roads where visual guidance lines may be inaccurate or unreliable. By combining visual guidance lines with map guidance lines, the system can provide drivers with more reliable navigation information and reduce the risk of accidents.

[0180] Optimize resource utilization and cost-effectiveness: By accurately assessing road complexity and intelligently choosing to use visual guide lines or fused guide lines, the system can use computing resources more efficiently. On sections of road that do not require high-precision map guide lines, the system only relies on visual guide lines, saving data processing and transmission resources and reducing system operation costs.

[0181] Enhanced system compatibility and scalability: The technical solution of this embodiment can be seamlessly integrated with the existing autonomous driving system without the need for large-scale modification of existing hardware. At the same time, the system's scoring and guide line generation strategies can be adjusted according to different autonomous driving levels and vehicle models, with good compatibility and scalability, which is conducive to promotion and application.

[0182] Real-time response: The system can quickly evaluate the complexity of the road during real-time driving and dynamically adjust the guide line generation strategy based on the evaluation results to ensure that the autonomous vehicle can generate the best guide line under various road conditions.

[0183] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0184] Based on the same inventive concept, the embodiment of the present application also provides a vehicle driving guide line generation device for implementing the vehicle driving guide line generation method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in one or more vehicle driving guide line generation device embodiments provided below can refer to the limitations of the vehicle driving guide line generation method above, and will not be repeated here.

[0185] In an exemplary embodiment, Fig. 9 As shown, a vehicle driving guide line generating device 900 is provided, comprising: a road complexity evaluation value obtaining module 901, a first guide line obtaining module 902 and a vehicle driving guide line generating module 903, wherein:

[0186] The road complexity evaluation value acquisition module 901 is used to acquire the road complexity evaluation value of the road at a preset distance on the current driving path of the vehicle; the road complexity evaluation value is used to characterize the complexity of the road;

[0187] A first guide line acquisition module 902 is used to acquire a lane line point of a preset distance road matching the current driving path from a preset lightweight map when the road complexity evaluation value is greater than or equal to a preset evaluation value threshold, and obtain a first guide line matching the current driving path according to the lane line point;

[0188] The vehicle driving guide line generation module 903 is used to obtain the second guide line generated by the road perception system for guiding the vehicle to travel on the current driving path, and use the first guide line to correct the second guide line to obtain the vehicle driving guide line on the current driving path on the road at a preset distance.

[0189] In an exemplary embodiment, the road complexity evaluation value acquisition module 901 is also used to obtain the current positioning information of the vehicle, determine the current positioning position in a preset navigation map based on the current positioning information, query the road structure information of a preset distance road based on the current positioning position from the preset navigation map, and obtain lane line clarity evaluation results in real time from the road perception system; based on the road structure information and the lane line clarity evaluation results, obtain the road complexity evaluation value of the preset distance road.

[0190] In an exemplary embodiment, the road structure information includes the average radius of the lane, the road slope, the ramp lane spacing and the lane number change information; the lane line clarity evaluation result includes the lane line visibility; the lane line visibility is obtained based on the pixels identified as the lane line area in the road image acquired by the road perception system and the preset pixel brightness threshold; the above-mentioned road complexity evaluation value acquisition module 901 is also used to obtain the curve evaluation value according to the average radius of the lane, the road slope and the lane line visibility; obtain the ramp evaluation value according to the ramp lane spacing, and the preset interval upper limit threshold and the preset interval lower limit threshold; obtain the lane number change evaluation value according to the lane number change information and the preset number change threshold; obtain the road complexity evaluation value of the preset distance road according to the curve evaluation value, the ramp evaluation value and the lane number change evaluation value.

[0191] In an exemplary embodiment, the road complexity evaluation value acquisition module 901 is also used to obtain a lane radius evaluation value based on the average radius of the lane, and a preset radius upper limit threshold and a preset radius lower limit threshold; to obtain a road slope evaluation value based on the road slope, and a preset slope upper limit threshold and a preset slope lower limit threshold; to obtain a lane line visibility evaluation value based on lane line visibility, and a preset visibility upper limit threshold and a preset visibility lower limit threshold; to obtain a curve evaluation value based on the lane radius evaluation value, the road slope evaluation value and the lane line visibility evaluation value.

[0192] In an exemplary embodiment, the current driving path is obtained based on a preset navigation map; the above-mentioned first guide line acquisition module 902 is also used to obtain a lane line point set corresponding to the current driving path in the preset lightweight map, and obtain lane line points corresponding to each lane from the lane line point set, and obtain the lane center point of each lane according to the lane line points of each lane; use the lane center point of each lane to perform curve fitting to obtain the first guide line of each lane; use the first guide line of the lane corresponding to the current driving path as the first guide line matching the current driving path.

[0193] In an exemplary embodiment, the above-mentioned first guide line acquisition module 902 is also used to obtain the path starting point coordinates, path ending point coordinates and target path point coordinates of the current driving path at a preset distance from the preset navigation map; the target path point coordinates are the coordinates of any other path point in the current driving path except the path starting point and the path ending point; based on the path starting point coordinates and the path ending point coordinates, a candidate path point area is generated in the preset lightweight map; based on the target path point coordinates, the target path point with the smallest distance to the target path point is obtained from each path point in the candidate path point area in the preset lightweight map; each path point is obtained based on the lane line points of each lane; based on the lane line points of each lane corresponding to each target path point, a lane line point set corresponding to the current driving path in the preset lightweight map is obtained.

[0194] In an exemplary embodiment, the above-mentioned vehicle driving guide line generation module 903 is also used to obtain candidate guide lines generated by the road perception system corresponding to multiple lanes in the current driving direction, obtain the similarity between each candidate guide line and the first guide line, and use the candidate guide line with the highest similarity as the second guide line for guiding the vehicle to travel on the current driving path.

[0195] In an exemplary embodiment, the above-mentioned vehicle driving guide line generation module 903 is also used to obtain multiple interpolation points of the first guide line and the second guide line, determine multiple geometric center points based on the multiple interpolation points; and perform fitting based on the multiple geometric center points to obtain the vehicle driving guide line of the current driving path on the road at a preset distance.

[0196] In an exemplary embodiment, the vehicle driving guide line generation module 903 is also used to collect multiple road images in the current driving direction using a road perception system; obtain a multi-lane panoramic image from a bird's-eye view based on the multiple road images; input the multi-lane panoramic image into a pre-trained lane centerline generation model to obtain lane centerlines of multiple lanes in the current driving direction, and use the multiple lane centerlines as candidate guide lines for the multiple lanes in the current driving direction.

[0197] In an exemplary embodiment, the vehicle driving guide line generation module 903 is also used to match feature points of multiple road images, and to splice the multiple road images based on the matched feature points to obtain a multi-lane panoramic image with a perspective view; and to perform perspective transformation on the multi-lane panoramic image with a perspective view to obtain a multi-lane panoramic image with a bird's-eye view.

[0198] Each module in the above-mentioned vehicle driving guide line generating device can be implemented in whole or in part by software, hardware and their combination. Each of the above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the corresponding operations of each of the above modules.

[0199] In an exemplary embodiment, a vehicle-side control device is also provided, and its internal structure diagram can be as follows: Fig.10 As shown, it includes a memory and a processor, the memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0200] Those skilled in the art will understand that Fig.10 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the vehicle-side control device to which the scheme of the present application is applied. The specific vehicle-side control device may include more or fewer components than those shown in the figure, or combine certain components, or have a different component arrangement.

[0201] In one embodiment, a computer device is further provided, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiments when executing the computer program.

[0202] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0203] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0204] It should be noted that the user information (including but not limited to user device information, user personal information, user navigation information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0205] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, an artificial intelligence (AI) processor, etc., but are not limited to this.

[0206] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0207] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for generating a vehicle driving guide line, characterized in that: The method comprises: Obtaining a road complexity evaluation value of a road at a preset distance on a current driving path of the vehicle; the road complexity evaluation value is used to characterize the complexity of the road; When the road complexity evaluation value is greater than or equal to a preset evaluation value threshold, a lane line point of a preset distance road matching the current driving path is obtained from a preset lightweight map, and a first guide line matching the current driving path is obtained according to the lane line point; A second guide line generated by a road perception system for guiding a vehicle to travel on a current driving path is obtained, and the second guide line is corrected using the first guide line to obtain a vehicle travel guide line on a road of a preset distance for the current driving path.

2. The method according to claim 1, characterized in that The step of obtaining a road complexity evaluation value of a road at a preset distance on the current driving path of the vehicle includes: Acquire current positioning information of the vehicle, determine the current positioning position in the preset navigation map based on the current positioning information, query the preset navigation map for road structure information of a road within a preset distance from the current positioning position, and acquire a lane line clarity evaluation result in real time from a road perception system; A road complexity evaluation value of the road at the preset distance is obtained according to the road structure information and the lane line clarity evaluation result.

3. The method according to claim 2, characterized in that The road structure information includes lane average radius, road slope, ramp lane interval and lane quantity change information; the lane line clarity evaluation result includes lane line visibility; the lane line visibility is obtained based on pixels identified as lane line areas in the road image acquired by the road perception system and a preset pixel brightness threshold; The obtaining, according to the road structure information and the lane line clarity evaluation result, a road complexity evaluation value of the preset distance road includes: Obtaining a curve evaluation value according to the average radius of the lane, the road slope, and the lane line visibility; Obtaining a ramp evaluation value according to the ramp lane interval, a preset interval upper limit threshold, and a preset interval lower limit threshold; Obtaining a lane quantity change evaluation value according to the lane quantity change information and a preset quantity change threshold; The road complexity evaluation value of the preset distance road is obtained according to the curve evaluation value, the ramp evaluation value and the lane quantity change evaluation value.

4. The method according to claim 3, characterized in that: The obtaining of the curve evaluation value according to the lane average radius, the road slope and the lane line visibility includes: Obtaining a lane radius evaluation value according to the lane average radius, a preset radius upper limit threshold, and a preset radius lower limit threshold; Obtaining a road slope evaluation value according to the road slope, and a preset slope upper limit threshold and a preset slope lower limit threshold; Obtaining a lane line visibility evaluation value according to the lane line visibility, a preset visibility upper threshold value, and a preset visibility lower threshold value; A curve evaluation value is obtained according to the lane radius evaluation value, the road slope evaluation value and the lane line visibility evaluation value.

5. The method according to claim 1, characterized in that The current driving path is obtained based on a preset navigation map; The acquiring a lane line point matching the current driving path from a preset lightweight map, and obtaining a first guide line matching the current driving path according to the lane line point, includes: Obtaining a lane line point set corresponding to the current driving path in the preset lightweight map, and obtaining a lane line point corresponding to each lane from the lane line point set, and obtaining a lane center point of each lane according to the lane line point of each lane; Performing curve fitting using the lane center points of the lanes to obtain first guide lines of the lanes; The first guide line of the lane corresponding to the current driving path is used as the first guide line matching the current driving path.

6. The method according to claim 5, characterized in that The obtaining of a lane line point set corresponding to the current driving path in the preset lightweight map includes: Obtaining the path starting point coordinates, path ending point coordinates and target path point coordinates of the current driving path on a preset distance road from the preset navigation map; the target path point coordinates are the coordinates of any other path point in the current driving path except the path starting point and the path ending point; Generating a candidate waypoint area in the preset lightweight map according to the path starting point coordinates and the path ending point coordinates; According to the coordinates of the target path point, a target path point position with the shortest distance to the target path point is obtained from each path point position in the candidate path point area in the preset lightweight map; each path point position is obtained based on the lane line point of each lane; According to the lane line points of each lane corresponding to each of the target path points, a lane line point set corresponding to the current driving path in the preset lightweight map is obtained.

7. The method according to claim 1, characterized in that The step of obtaining a second guide line generated by a road perception system and used to guide the vehicle to travel on the current driving path includes: Obtain candidate guide lines generated by a road perception system corresponding to multiple lanes in the current driving direction, obtain the similarity between each of the candidate guide lines and the first guide line, and use the candidate guide line with the highest similarity as the second guide line for guiding the vehicle to travel on the current driving path.

8. The method according to claim 7, characterized in that The method of using the first guide line to correct the second guide line to obtain a vehicle driving guide line of the current driving path on a road of a preset distance includes: Acquire a plurality of interpolation points of the first guide line and the second guide line, and determine a plurality of geometric center points according to the plurality of interpolation points; Fitting is performed based on the plurality of geometric center points to obtain a vehicle driving guide line of the current driving path on a road of a preset distance.

9. The method according to claim 7, characterized in that: The step of acquiring candidate guide lines corresponding to a plurality of lanes in a current driving direction generated by a road perception system includes: Using the road perception system to collect multiple road images in the current driving direction; Obtaining a multi-lane panoramic image from a bird's-eye view based on the plurality of road images; The multi-lane panoramic image is input into a pre-trained lane centerline generation model to obtain lane centerlines of multiple lanes in the current driving direction, and the multiple lane centerlines are used as candidate guide lines for multiple lanes in the current driving direction.

10. The method according to claim 9, characterized in that The step of obtaining a multi-lane panoramic image from a bird's-eye view based on the plurality of road images comprises: Matching feature points of the plurality of road images, and splicing the plurality of road images based on the matched feature points to obtain a multi-lane panoramic image with a perspective view; The multi-lane panoramic image with a perspective view is perspective-transformed to obtain a multi-lane panoramic image with a bird's-eye view.

11. A vehicle driving guide line generating device, characterized in that: The device comprises: A road complexity evaluation value acquisition module is used to acquire a road complexity evaluation value of a road at a preset distance on the current driving path of the vehicle; the road complexity evaluation value is used to characterize the complexity of the road; A first guide line acquisition module, configured to acquire a lane line point of a road of a preset distance matching the current driving path from a preset lightweight map when the road complexity evaluation value is greater than or equal to a preset evaluation value threshold, and obtain a first guide line matching the current driving path according to the lane line point; The vehicle driving guide line generation module is used to obtain the second guide line generated by the road perception system for guiding the vehicle to travel on the current driving path, and use the first guide line to correct the second guide line to obtain the vehicle driving guide line of the current driving path on the road at a preset distance.

12. A vehicle-side control device, comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 10 are implemented.

Images