Method and device for determining lane lines in high-precision maps

Abstract

A method for determining lane lines in a high-precision map includes: obtaining an environmental image and a three-dimensional point cloud of a vehicle driving environment (S1); determining a lane line area in the environmental image (S2); projecting points in the three-dimensional point cloud onto the environmental image to determine a target point located in the lane line area (S3); and fitting the target point to determine the lane line (S4). Thus, the environmental image and the three-dimensional point cloud can be combined to determine the target point located in the lane line area in the three-dimensional point cloud, and then the lane line can be determined by fitting the target point. The three-dimensional point cloud can be used as a high-precision map, and the process of determining the lane line does not require human intervention to a large extent, so it is conducive to semi-automatic or even fully automatic determination of lane lines in high-precision maps. When faced with repeated determination operations of a large number of lane lines, it can be completed at a high speed and efficiency, thereby improving the accuracy of determining the lane lines.

Description

Technical Field

[0001] The present disclosure relates to the field of map processing, and in particular to a method for determining lane lines in a high-precision map, a device for determining lane lines in a high-precision map, an electronic device, and an autonomous driving vehicle. Background Art

[0002] In the field of autonomous driving, the recognition of lane lines on the road is very important. In related technologies, there are two main ways to obtain lane lines: one is to detect lane lines in real time from the current environment image, and the other is to obtain pre-marked lane lines in high-precision maps to determine the location of lane lines in the environment.

[0003] Since the lane lines in high-precision maps need to be marked in advance, the existing method of marking lane lines in high-precision maps is mainly done manually. The manual marking operation in the map is relatively accurate for two-dimensional images, but three-dimensional images are generally generated based on lidar. The images generated by lidar generally have no color information, and the generated images are also affected by obstacles on the road surface, making it difficult for labelers to distinguish which positions on the road surface belong to lane lines, resulting in low accuracy in manual marking of lane lines in three-dimensional images.

[0004] High-precision maps are three-dimensional images, so it is difficult to achieve ideal accuracy when marking lane lines in high-precision maps. In addition, manually marking lane lines in high-precision maps requires a lot of repetitive operations, which is slow and inefficient. Summary of the invention

[0005] The present disclosure proposes a method for determining lane lines in a high-precision map, a device for determining lane lines in a high-precision map, an electronic device and an autonomous driving vehicle to solve the technical problems in related technologies of low accuracy and low efficiency in manually marking lane lines in high-precision maps.

[0006] According to a first aspect of an embodiment of the present disclosure, a method for determining lane lines in a high-precision map is proposed, comprising:

[0007] Acquire environmental images and three-dimensional point clouds of the vehicle's driving environment;

[0008] Determining a lane line area in the environment image;

[0009] Projecting points in the three-dimensional point cloud onto the environment image to determine target points located in the lane line area;

[0010] The target points are fitted to determine lane lines.

[0011] According to a second aspect of an embodiment of the present disclosure, a device for determining lane lines in a high-precision map is provided, wherein the device comprises one or more processors working alone or in collaboration, and the processors are used to execute:

[0012] Acquire environmental images and three-dimensional point clouds of the vehicle's driving environment;

[0013] Determining a lane line area in the environment image;

[0014] Projecting points in the three-dimensional point cloud onto the environment image to determine target points located in the lane line area;

[0015] The target points are fitted to determine lane lines.

[0016] According to a third aspect of an embodiment of the present disclosure, an electronic device is proposed, comprising a device for determining lane lines in a high-precision map as described in any of the above embodiments.

[0017] According to a fourth aspect of an embodiment of the present disclosure, an autonomous driving vehicle is proposed, comprising the electronic device described in the above embodiment.

[0018] According to the embodiments of the present disclosure, the target points in the lane line area in the three-dimensional point cloud can be determined by combining the environment image and the three-dimensional point cloud, and then the lane line can be determined by fitting the target points. The three-dimensional point cloud can be used as a high-precision map, and the above process of determining the lane line does not require human participation to a large extent, so it is conducive to semi-automatic or even fully automatic determination of lane lines in high-precision maps. When faced with repeated determination operations of a large number of lane lines, it can be completed at a high speed and efficiently, and the accuracy of determining the lane line can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present disclosure, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 It is a schematic flowchart of a method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0021] Figure 2 It is a schematic diagram showing a method of determining a lane line area in an environment image according to an embodiment of the present disclosure.

[0022] Figure 3 It is a schematic diagram showing a method of projecting points in the three-dimensional point cloud onto the environment image according to an embodiment of the present disclosure.

[0023] Figure 4 It is a schematic diagram showing a method of fitting target points to determine lane lines according to an embodiment of the present disclosure.

[0024] Figure 5 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0025] Figure 6 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0026] Figure 7 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0027] Figure 8 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0028] Fig. 9 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0029] Fig.10 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0030] Fig.11 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0031] Fig.12 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0032] Fig.13 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0033] Fig.14 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0034] Fig.15 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0035] Fig.16 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0036] Fig.17 It is a schematic flowchart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure.

[0037] Fig.18 It is a hardware structure diagram showing a device where a lane line determination device in a high-precision map is located according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0038] The following will be combined with the drawings in the embodiments of the present disclosure to clearly and completely describe the technical solutions in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present disclosure. In addition, the following embodiments and features in the embodiments can be combined with each other without conflict.

[0039] Figure 1 It is a schematic flow chart of a method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. The method for determining lane lines in a high-precision map described in the embodiment of the present disclosure can be applied to an image acquisition device, which can acquire environmental images and three-dimensional point clouds of a vehicle's driving environment, and can also be applied to other electronic devices that can analyze and process environmental images and three-dimensional point clouds, such as terminals, servers, and vehicle-mounted devices.

[0040] like Figure 1 As shown, the method for determining lane lines in the high-precision map may include the following steps:

[0041] In step S1, an environmental image and a three-dimensional point cloud of a vehicle driving environment are obtained;

[0042] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle's driving environment can be obtained, wherein the environmental image can be obtained by an image acquisition device such as a camera, and the three-dimensional point cloud can be obtained by a laser radar.

[0043] In step S2, a lane line area is determined in the environment image;

[0044] In one embodiment, a lane line area may be determined in the acquired environment image.

[0045] Among them, the lane line area can be determined in the environmental image according to a predetermined image recognition model. For example, the image recognition model can be obtained in advance through machine learning (for example, it can be a designed neural network). The image recognition model can determine the lane line area in the image based on the input image. Then, the acquired environmental image can be input into the image recognition model to determine the lane line area in the environmental image.

[0046] In other embodiments, the road surface area may be first determined in the environmental image, and then the lane line area may be determined in the road surface area, thereby eliminating the need to analyze all information in the environmental image, thereby reducing the amount of information used to determine the lane line area, and helping to reduce misjudgment.

[0047] Figure 2 FIG. 1 is a schematic diagram showing a method of determining a lane line area in an environment image according to an embodiment of the present disclosure. Figure 2 As shown, the lane line area is determined in the environment image, and a specific color, such as white, can be marked for the lane line area to distinguish the lane line area from other areas in the environment image. Different objects in other areas, such as static objects and dynamic objects, can also be marked with different colors for differentiation.

[0048] In step S3, the points in the three-dimensional point cloud are projected onto the environment image to determine the target points located in the lane line area;

[0049] In one embodiment, the internal parameters of the image acquisition device that captures the environmental image can be obtained, and then based on the internal parameters and the rotation relationship and displacement relationship from the world coordinate system to the coordinate system of the image acquisition device, the first coordinate of the point in the three-dimensional point cloud in the world coordinate system is converted into a second coordinate in the environmental image.

[0050] For example, the first coordinate of a point in a 3D point cloud in the world coordinate system is (x w ,y w ,z w ), the corresponding second coordinate of the point in the environment image is (μ, ν, 1), and the relationship between the first coordinate and the second coordinate is as follows:

[0051]

[0052] in, The image acquisition device contains five internal parameters, namely a x =fm x , a y =fm y , γ, μ0 and ν0, where f is the focal length of the image acquisition device, m x is the number of pixels per unit distance in the x direction (scale factors), m y is the number of pixels per unit distance in the y direction, γ is the skew parameters between the x and y directions, and μ0 and ν0 correspond to the principal point of the optical center.

[0053] z cis the scale factor of any homogeneous coordinates. Matrix R is the rotation matrix, which is used to represent the rotation relationship from the world coordinate system to the coordinate system of the image acquisition device. Matrix T is the translation matrix, which is used to represent the displacement relationship from the world coordinate system to the coordinate system of the image acquisition device. Matrix R and matrix t belong to the extrinsic matrix of the image acquisition device.

[0054] After converting the first coordinate of the point in the three-dimensional point cloud in the world coordinate system into the second coordinate in the environment image, the corresponding coordinate of the point in the three-dimensional point cloud in the environment image, i.e., the second coordinate, is determined. The lane line area has been determined. Then, based on the relationship between the second coordinate and the lane line area, the target point located in the lane line area can be determined, i.e., the point in the three-dimensional point cloud located in the lane line area.

[0055] In step S4, the target point is fitted to determine the lane line.

[0056] In one embodiment, after determining the target point located in the lane line area, the target point can be fitted to determine the lane line, for example, the target point can be fitted with a Bezier curve. Since the target point is located in the lane line area, the curve obtained by fitting the target point can be used as the lane line.

[0057] According to the embodiments of the present disclosure, the target points in the lane line area in the three-dimensional point cloud can be determined by combining the environment image and the three-dimensional point cloud, and then the lane line can be determined by fitting the target points. The three-dimensional point cloud can be used as a high-precision map, and the above process of determining the lane line does not require human participation to a large extent, so it is conducive to semi-automatic or even fully automatic determination of lane lines in high-precision maps. When faced with repeated determination operations of a large number of lane lines, it can be completed at a high speed and efficiently, and the accuracy of determining the lane line can be improved.

[0058] Figure 3 It is a schematic diagram showing a method of projecting points in the three-dimensional point cloud onto the environment image according to an embodiment of the present disclosure. Figure 4 It is a schematic diagram showing a method of fitting target points to determine lane lines according to an embodiment of the present disclosure.

[0059] like Figure 3 As shown, the points in the three-dimensional point cloud can be projected onto the environment image to determine the points in the lane line area as the target points. The target points can be fitted, for example, by using a Bezier curve. The three-dimensional bird's-eye view of the lane line obtained by fitting is shown in FIG. Figure 4 shown.

[0060] Taking fitting the target point through the second-order Bezier curve as an example, a starting point A, an end point C and a point B can be selected from multiple target points, and point B is located between the starting point A and the end point C.

[0061] Determine a control point D on line segment AB, and determine a control point E on line segment BC, wherein AD / AB=BE / BC=k, where k is a preset ratio;

[0062] Then determine point F on line segment DE, where DF / DE=k;

[0063] On the basis of ensuring AD / AB=BE / BC=k, change the positions of control points D and E, and determine the new point F after changing control points D and E when DF / DE=k, and determine the point F in this way.

[0064] Finally, starting from the starting point A and heading toward point C, all points F are connected in sequence. The final curve is the curve fitted by the second-order Bezier curve to points A, B, and C.

[0065] It should be noted that, in addition to fitting by a second-order Bezier curve, fitting by a third-order or higher-order Bezier curve may also be performed, and the specific selection may be made according to needs.

[0066] Figure 5 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Figure 5 As shown, determining the lane line area in the environment image includes:

[0067] In step S201, the lane line area is determined in the environment image according to a predetermined image recognition model.

[0068] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be acquired first, and then a lane line area may be determined in the environmental image according to a predetermined image recognition model.

[0069] For example, an image recognition model (such as a designed neural network) can be obtained in advance through machine learning. The image recognition model can determine the lane line area in the image based on the input image. The acquired environmental image can then be input into the image recognition model to determine the lane line area in the environmental image.

[0070] Then, the points in the three-dimensional point cloud can be projected onto the environment image to determine the target points located in the lane line area, and finally the target points are fitted to determine the lane line.

[0071] Figure 6FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Figure 6 As shown, determining the lane line area in the environment image includes:

[0072] In step S202, a road surface area is determined in the environment image;

[0073] In step S203, a lane line area is determined in the road surface area.

[0074] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be acquired first, and then the road surface area may be determined in the acquired environmental image.

[0075] For example, a road surface recognition model can be obtained in advance through machine learning. The road surface recognition model can determine the road surface area in the image based on the input image. The acquired environmental image can be input into the road surface recognition model to determine the road surface area in the environmental image. The image of the determined road surface area can then be input into the image recognition model obtained in advance through machine learning. The image recognition model can determine the lane line area in the image based on the input image, thereby determining the lane line area in the road surface area based on the input image of the road surface area.

[0076] Then, the points in the three-dimensional point cloud can be projected onto the environment image to determine the target points located in the lane line area, and finally the target points are fitted to determine the lane line.

[0077] Figure 7 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Figure 7 As shown, before projecting the points in the three-dimensional point cloud onto the environment image, the method further includes:

[0078] In step S5, obstacle points belonging to obstacles are determined in the three-dimensional point cloud;

[0079] In step S6, the obstacle points are removed from the three-dimensional point cloud;

[0080] The projecting of the points in the three-dimensional point cloud onto the environment image comprises:

[0081] In step S301, points in the three-dimensional point cloud excluding the obstacle points are projected onto the environment image.

[0082] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be acquired first, and then the road surface area may be determined in the acquired environmental image.

[0083] For a 3D point cloud, the points therein may be projected onto an environment image, and before projecting the points in the 3D point cloud onto the environment image, the obstacle points in the 3D point cloud that are obstacles may be removed. For example, an obstacle recognition model may be obtained in advance through machine learning (for example, deep learning), and the obstacle recognition model may determine obstacles in the 3D point cloud based on the input 3D point cloud. Then, by inputting the 3D point cloud into the obstacle recognition model, obstacles in the 3D point cloud may be determined, for example, the obstacle region corresponding to the obstacle in the 3D point cloud may be determined, and the points in the obstacle region may be removed from the 3D point cloud as obstacle points, so that there are no obstacle points in the remaining points in the 3D point cloud.

[0084] Therefore, when performing subsequent projection operations, the points in the three-dimensional point cloud with obstacle points removed can be projected onto the environmental image to determine the target points located in the lane line area, and finally the target points are fitted to determine the lane lines. This helps to avoid the obstacle points belonging to obstacles being projected into the lane line area of ​​the environmental image, affecting the accuracy of determining the target points.

[0085] Figure 8 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Figure 8 As shown, determining the obstacle points belonging to obstacles in the three-dimensional point cloud includes:

[0086] In step S501, obstacle points belonging to obstacles are determined in the three-dimensional point cloud according to a predetermined deep learning model.

[0087] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be acquired first, and then the road surface area may be determined in the acquired environmental image.

[0088] For a three-dimensional point cloud, points therein may be projected onto an environment image, and before projecting the points in the three-dimensional point cloud onto the environment image, obstacle points in the three-dimensional point cloud that are obstacles may be removed.

[0089] A deep learning model can be obtained in advance through deep learning. The deep learning model can use a three-dimensional point cloud as input and output information of obstacle points belonging to obstacles. According to the information, the obstacle points belonging to obstacles in the three-dimensional point cloud can be determined. Obstacles include but are not limited to vehicles, pedestrians, traffic signs, etc. The determined obstacle points can be removed from the three-dimensional point cloud.

[0090] Therefore, when performing subsequent projection operations, the points in the three-dimensional point cloud with obstacle points removed can be projected onto the environmental image to determine the target points located in the lane line area, and finally the target points are fitted to determine the lane lines. This helps to avoid the obstacle points belonging to obstacles being projected into the lane line area of ​​the environmental image, affecting the accuracy of determining the target points.

[0091] Fig. 9 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig. 9 As shown, the three-dimensional point cloud is a three-dimensional point cloud of the environment at the target time. Before projecting the points in the three-dimensional point cloud onto the environment image, the method further includes:

[0092] In step S7, determining a three-dimensional point cloud of the vehicle driving environment at at least one other time before or after the target time, and a predicted point cloud at the target time;

[0093] In step S8, the predicted point cloud is stacked into the three-dimensional point cloud of the environment at the target time;

[0094] The projecting of the points in the three-dimensional point cloud onto the environment image comprises:

[0095] In step S302, the points in the stacked three-dimensional point cloud are projected onto the environment image.

[0096] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be acquired first, and then the road surface area may be determined in the acquired environmental image.

[0097] For a three-dimensional point cloud, the points therein can be projected onto an environmental image. Before projecting the points in the three-dimensional point cloud onto the environmental image, the three-dimensional point cloud operation collected from the vehicle's driving environment can be performed continuously (the number of executions and the time of execution can be set as needed), thereby collecting three-dimensional point clouds of the environment at multiple times.

[0098] Project the points in the three-dimensional point cloud onto the environment image. If only the three-dimensional point cloud of the environment at the target moment is projected onto the environment image, due to the small number of lines of the laser radar used to collect the three-dimensional point cloud, the point cloud density in the collected three-dimensional point cloud is low, and the number of target points located in the lane line area projected into the environment image is small, which is not conducive to obtaining accurate fitting results.

[0099] According to this embodiment, three-dimensional point clouds at multiple moments can be collected. Specifically, in addition to collecting the three-dimensional point cloud at the target moment, the three-dimensional point cloud of the vehicle driving environment at at least one other moment before or after the target moment can also be collected. For the three-dimensional point clouds collected at other moments, the predicted point cloud at the target moment can be determined. For example, the three-dimensional point cloud of the vehicle driving environment at other moments can be predicted based on the posture difference and position difference between the vehicle at other moments and the target moment (the prediction method can be selected as needed, for example, it can be implemented by Kalman filtering, or it can be implemented based on the prediction model obtained by pre-machine learning) to determine the three-dimensional point cloud of the vehicle driving environment at other moments, and the predicted point cloud at the target moment.

[0100] The predicted point cloud is then stacked into the three-dimensional point cloud of the environment at the target moment to increase the point cloud density. The points in the three-dimensional point cloud are then projected onto the environmental image. Specifically, the points in the stacked three-dimensional point cloud can be projected onto the environmental image, and then the target points located in the lane line area are determined. Since the stacked three-dimensional point cloud contains more points, the number of target points located in the lane line area projected into the environmental image can be increased, and then the target points can be fitted to determine the lane lines. Fitting can be performed on a larger number of target points, which is conducive to obtaining accurate fitting results.

[0101] Fig.10 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.10 As shown, the three-dimensional point cloud of at least one other time before or after the target time is determined, and the predicted point cloud at the target time includes:

[0102] In step S701, determining a posture difference between the vehicle at the other time and the target time, and a position difference between the vehicle at the other time and the target time;

[0103] In step S702, the three-dimensional point cloud of the vehicle driving environment at the other time is determined according to the posture difference and the position difference, and the predicted point cloud at the target time.

[0104] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be acquired first, and then the road surface area may be determined in the acquired environmental image.

[0105] For a three-dimensional point cloud, the points therein can be projected onto an environmental image, and before projecting the points in the three-dimensional point cloud onto the environmental image, the three-dimensional point cloud operation for collecting the vehicle's driving environment can be performed continuously (the number of executions and the time of execution can be set as needed), thereby collecting three-dimensional point clouds of the environment at multiple times. Specifically, in addition to collecting the three-dimensional point cloud at the target time, the three-dimensional point cloud of the vehicle's driving environment at at least one other time before or after the target time can also be collected. For the three-dimensional point clouds collected at other times, the predicted point cloud at the target time can be determined.

[0106] In one embodiment, the differences in the posture and position of the vehicle at other times and the target time may cause the three-dimensional point cloud of the vehicle's environment at other times to be different from that at the target time. Then, the three-dimensional point cloud of the vehicle's driving environment at other times can be determined based on the posture differences and position differences, and the predicted point cloud at the target time.

[0107] For example, if time t is the current time, time t+1 is the next time after the current time, the position coordinates of the vehicle at time t are (x1, y1), the position coordinates of the vehicle at time t+1 are (x1+x0, y1+y0), and the posture (such as driving direction) has not changed, then for the three-dimensional point cloud at time t+1, to determine its predicted point cloud at time t, all points in the three-dimensional point cloud at time t+1 can be subtracted from x0 in the x direction and from y0 in the y direction.

[0108] It should be noted that the method for determining the predicted point cloud is not limited to the above method, but can be selected as needed. For example, the target point cloud at other times can be predicted based on Kalman filtering to determine the predicted point cloud, or the target point cloud at other times can be predicted based on the prediction model obtained by pre-machine learning to determine the predicted point cloud.

[0109] The predicted point cloud is then stacked into the three-dimensional point cloud of the environment at the target moment to increase the point cloud density. The points in the three-dimensional point cloud are then projected onto the environmental image. Specifically, the points in the stacked three-dimensional point cloud can be projected onto the environmental image, and then the target points located in the lane line area are determined. Since the stacked three-dimensional point cloud contains more points, the number of target points located in the lane line area projected into the environmental image can be increased, and then the target points can be fitted to determine the lane lines. Fitting can be performed on a larger number of target points, which is conducive to obtaining accurate fitting results.

[0110] Fig.11 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.11 As shown, projecting the points in the three-dimensional point cloud onto the environment image includes:

[0111] In step S303, based on the internal parameters of the image acquisition device that acquires the environmental image, the rotation relationship and displacement relationship from the world coordinate system to the coordinate system of the image acquisition device, the first coordinate of the point in the three-dimensional point cloud in the world coordinate system is converted into the second coordinate in the environmental image.

[0112] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be acquired first, and then a road surface area may be determined in the acquired environmental image, and then points in the three-dimensional point cloud may be projected onto the environmental image.

[0113] For example, the internal parameters of the image acquisition device that captures the environmental image can be obtained, and then based on the internal parameters and the rotation and displacement relationship from the world coordinate system to the coordinate system of the image acquisition device, the first coordinate of the point in the three-dimensional point cloud in the world coordinate system is converted into the second coordinate in the environmental image.

[0114] After converting the first coordinate of the point in the 3D point cloud in the world coordinate system to the second coordinate in the environment image, the corresponding coordinate of the point in the 3D point cloud in the environment image, i.e., the second coordinate, is determined. The lane line area has been determined, so according to the relationship between the second coordinate and the lane line area, the target point located in the lane line area can be determined, i.e., the point in the 3D point cloud located in the lane line area. Finally, the target point can be fitted to determine the lane line.

[0115] Fig.12 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.12 As shown, the method also includes:

[0116] In step S9, marking the target point;

[0117] In step S10, the target point with annotation is displayed.

[0118] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle's driving environment can be obtained first, and then the road surface area can be determined in the acquired environmental image, and then the points in the three-dimensional point cloud can be projected onto the environmental image to determine the target points located in the lane line area, and then the target points can be fitted to determine the lane line.

[0119] As for the target points, they can be marked, for example, they can be marked with specific colors, such as marking the target points in the three-dimensional point cloud as white, and other non-target points can be marked as black. The target points can also be marked with specific identifiers, such as giving target points in the three-dimensional point cloud a different identifier from non-target points.

[0120] Accordingly, the target points with annotations can be displayed so that when the user views the three-dimensional point cloud, the user can distinguish the target points and non-target points in the three-dimensional point cloud according to the annotations.

[0121] Fig.13 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.13 As shown, the method also includes:

[0122] In step S11, the lane line is corrected according to the received correction instruction;

[0123] In step S12, the corrected lane line is projected into the environment image to determine whether the projection of the corrected lane line in the environment image matches the lane line area;

[0124] In step S13, response information is generated according to the matching result between the projection of the corrected lane line in the environment image and the lane line area.

[0125] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle's driving environment can be obtained first, and then the road surface area can be determined in the acquired environmental image, and then the points in the three-dimensional point cloud can be projected onto the environmental image to determine the target points located in the lane line area, and then the target points can be fitted to determine the lane line.

[0126] After the lane line is determined, a manually input correction instruction may be received to correct the lane line. However, the result of the manual correction may also have errors. Therefore, the corrected lane line may be projected into the environment image to determine whether the projection of the corrected lane line in the environment image matches the lane line area. Then, response information is generated based on the matching result between the projection of the corrected lane line in the environment image and the lane line area.

[0127] If the projection of the corrected lane line in the environmental image does not match the lane line area, for example, the projection of the corrected lane line in the environmental image exceeds a preset ratio and falls outside the lane line area, then the generated response information can be used to prompt the user that the correction result is unreasonable so that the user can make corrections again; if the projection of the corrected lane line in the environmental image matches the lane line area, for example, the projection of the corrected lane line in the environmental image is less than a preset ratio and falls outside the lane line area, then the generated response information can be used to prompt the user that the correction result is reasonable.

[0128] It should be noted that in addition to manual correction after fitting the target points to obtain the lane lines, this embodiment may also require manual correction in the process of determining the lane line area and projecting the points in the three-dimensional point cloud onto the environmental image. For example, in the process of determining the lane lines, manually input instructions may be received to correct, supplement, or delete the lane line area in the environmental image. For example, in the process of projecting the points in the three-dimensional point cloud onto the environmental image, the projected target may be adjusted.

[0129] Fig.14 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.14 As shown, before fitting the target point to determine the lane line, the method further includes:

[0130] In step S14, among the non-target points in the three-dimensional point cloud, candidate points whose distance to the target point is less than a preset distance are determined;

[0131] In step S15, an extension point whose similarity with preset attribute information of the target point is greater than a preset similarity is determined among the candidate points;

[0132] In step S16, the extension point and the target point are used as new target points;

[0133] Wherein, fitting the target point to determine the lane line includes:

[0134] In step S401, the new target point is fitted to determine the lane line.

[0135] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle's driving environment can be obtained first, and then the road surface area can be determined in the acquired environmental image, and then the points in the three-dimensional point cloud can be projected onto the environmental image to determine the target points located in the lane line area, and then the target points can be fitted to determine the lane line.

[0136] However, there will be more or less some deviations when projecting points in the three-dimensional point cloud into the environment image. For example, it may be caused by inaccurate external parameters of the image acquisition device that acquires the environment image. In this case, some target points located in the lane line area in the three-dimensional point cloud will not be projected into the lane line area of ​​the environment image. This may cause the fitting result to be inaccurate, that is, there is a difference between the lane line determined by fitting and the actual lane line in the three-dimensional point cloud.

[0137] However, since the deviation is generally not very large, the distance between these target points not projected to the lane line area and the target points projected to the area is generally close. Therefore, before fitting the target points to determine the lane line, candidate points whose distance to the target point is less than the preset distance can be determined from the non-target points in the 3D point cloud. These candidate points may be the target points not projected to the lane line area.

[0138] For the candidate points, extension points whose similarity with the preset attribute information of the target point is greater than the preset similarity can be determined. For example, a floodfill algorithm can be used to determine the extension points, wherein the preset attributes can be set as needed, such as reflection brightness (intensity). Since these extension points are very close to the preset attribute information of the target point, they are very likely to be target points that are not projected to the lane line area, so the new target points are fitted to determine the lane line. Specifically, the extension points and the original target points can be used as new target points for lane line fitting.

[0139] Based on this, the problem of inaccurate fitting results caused by deviation in projecting points in the three-dimensional point cloud into the environment image can be alleviated.

[0140] Fig.15 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.15 As shown, fitting the target point to determine the lane line includes:

[0141] In step S402, the target point is fitted by a curve model to determine the lane line.

[0142] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be first acquired, and then the road surface area may be determined in the acquired environmental image, and then the points in the three-dimensional point cloud may be projected onto the environmental image to determine the target points located in the lane line area, and then the target points may be fitted to determine the lane line. The fitted target points may be fitted by selecting a curve model as required, for example, a Bezier curve may be selected to fit the target points to determine the lane line.

[0143] Fig.16 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.16 As shown, fitting the target point by a Bezier curve to determine the lane line includes:

[0144] In step S4021, the target point is fitted by multiple third-order Bezier curves to determine the lane line.

[0145] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle driving environment may be first obtained, and then a road surface area may be determined in the obtained environmental image, and then points in the three-dimensional point cloud may be projected onto the environmental image to determine a target point located in the lane line area, and then the target point may be fitted to determine the lane line. The fitted target point may be fitted by a plurality of third-order Bezier curves.

[0146] The equation for a third-order Bezier curve is as follows:

[0147] P(t)=A(1-t) 3 +B·3(1-t) 2 ·t+C·3(1-t)·t 2 +D·t 3 ;

[0148] A, B, C and D are the coordinates of the target points used as control points. The selection of control points has been described in the previous text and will not be repeated here.

[0149] Among them, a specific fitting method can be to determine the two farthest points among the target points as the starting point and the end point for fitting, and then for the fitted curve, determine whether there is a target point whose distance from the curve is greater than a preset distance. If such a target point exists, it means that the fitting effect does not meet the requirements, then a perpendicular line is drawn from the target point to the curve, and then the curve is divided into two parts from the intersection of the perpendicular line and the curve, and the target point is continued to be fitted for each part of the curve. If for the further fitted curve, there is still a target point whose distance to the curve is greater than the preset distance, then a perpendicular line is continued to be drawn from the target point to the curve, and then the curve is further divided from the intersection of the perpendicular line and the curve, and then each divided part of the curve is continued to be fitted to the target point until, for the fitted curve, the distance from all target points to the curve is less than or equal to the preset distance.

[0150] Fig.17 FIG. 1 is a schematic flow chart of another method for determining lane lines in a high-precision map according to an embodiment of the present disclosure. Fig.17 As shown, the method also includes:

[0151] In step S17, a control instruction is generated based on the lane line, wherein the control instruction is used to control the driving of the vehicle.

[0152] In one embodiment, an environmental image and a three-dimensional point cloud of the vehicle's driving environment can be obtained first, and then the road surface area can be determined in the acquired environmental image, and then the points in the three-dimensional point cloud can be projected onto the environmental image to determine the target points located in the lane line area, and then the target points can be fitted to determine the lane line.

[0153] The method for determining lane lines in the high-precision map can be applied to the field of autonomous driving technology. For example, when applied to autonomous driving vehicles, control instructions can be generated based on the fitted lane lines to control the vehicle's driving. For example, control instructions can be generated to ensure that the vehicle, during autonomous driving, stays between two lane lines so as to avoid interfering with or colliding with vehicles in other lanes, thereby ensuring traffic safety during autonomous driving.

[0154] The embodiment of the device for determining lane lines in the high-precision map disclosed in the present invention can be applied to vehicle-mounted devices. The device embodiment can be implemented by software, or by hardware, or by a combination of software and hardware. Taking software implementation as an example, as a device in a logical sense, it is formed by the processor of the device in which it is located reading the corresponding computer program instructions in the non-volatile memory into the memory and running them. From the hardware level, if Fig.18 As shown, it is a hardware structure diagram of the device where the lane line determination device in the high-precision map of the present disclosure is located. Fig.18 In addition to the processor, network interface, memory and non-volatile memory shown, the device in the embodiment where the apparatus is located may generally include other hardware, such as a forwarding chip responsible for processing messages, etc.; from a hardware structure perspective, the device may also be a distributed device, which may include multiple interface cards to expand message processing at the hardware level.

[0155] The embodiments of the present disclosure also provide a device for determining lane lines in a high-precision map. The device for determining lane lines in a high-precision map described in the embodiments of the present disclosure can be applied to image acquisition devices that can acquire environmental images and three-dimensional point clouds of a vehicle's driving environment, and can also be applied to other electronic devices that can analyze and process environmental images and three-dimensional point clouds, such as terminals, servers, and vehicle-mounted devices.

[0156] In one embodiment, the lane line determination device includes one or more processors working alone or in collaboration, and the processors are used to execute:

[0157] Acquire environmental images and three-dimensional point clouds of the vehicle's driving environment;

[0158] Determining a lane line area in the environment image;

[0159] Projecting points in the three-dimensional point cloud onto the environment image to determine target points located in the lane line area;

[0160] The target points are fitted to determine lane lines.

[0161] In one embodiment, the processor is configured to perform:

[0162] The lane line area is determined in the environment image according to a predetermined image recognition model.

[0163] In one embodiment, the processor is configured to perform:

[0164] determining a road surface area in the environment image;

[0165] A lane line area is determined in the road surface area.

[0166] In one embodiment, the processor is further configured to execute:

[0167] Before projecting the points in the three-dimensional point cloud onto the environment image, determining obstacle points belonging to obstacles in the three-dimensional point cloud;

[0168] Eliminating the obstacle points in the three-dimensional point cloud;

[0169] The projecting of the points in the three-dimensional point cloud onto the environment image comprises:

[0170] Projecting points in the three-dimensional point cloud excluding the obstacle points onto the environment image.

[0171] In one embodiment, the processor is configured to perform:

[0172] Obstacle points belonging to obstacles are determined in the three-dimensional point cloud according to a predetermined deep learning model.

[0173] In one embodiment, the three-dimensional point cloud is a three-dimensional point cloud of the environment at the target time, and the processor is further configured to execute:

[0174] Before projecting the points in the three-dimensional point cloud onto the environment image, determining a three-dimensional point cloud of the vehicle driving environment at at least one other time before or after the target time, and a predicted point cloud at the target time;

[0175] Stacking the predicted point cloud into the three-dimensional point cloud of the environment at the target time;

[0176] The projecting of the points in the three-dimensional point cloud onto the environment image comprises:

[0177] Projecting the points in the stacked three-dimensional point cloud onto the environment image.

[0178] In one embodiment, the determining of the three-dimensional point cloud at at least one other time before or after the target time comprises the processor executing:

[0179] Determine a difference in posture of the vehicle at the other time and the target time, and a difference in position of the vehicle at the other time and the target time;

[0180] The three-dimensional point cloud of the vehicle driving environment at the other time is determined according to the posture difference and the position difference, and the predicted point cloud at the target time.

[0181] In one embodiment, the processor is configured to perform:

[0182] According to the internal parameters of the image acquisition device that acquires the environmental image, the rotation relationship and displacement relationship between the world coordinate system and the coordinate system of the image acquisition device, the first coordinate of the point in the three-dimensional point cloud in the world coordinate system is converted into the second coordinate in the environmental image.

[0183] In one embodiment, the processor is further configured to execute:

[0184] marking the target point;

[0185] The target point is displayed with a label.

[0186] In one embodiment, the processor is further configured to execute:

[0187] Correcting the lane line according to the received correction instruction;

[0188] Projecting the corrected lane line into the environment image to determine whether the projection of the corrected lane line in the environment image matches the lane line area;

[0189] Response information is generated according to a matching result between the projection of the corrected lane line in the environment image and the lane line area.

[0190] In one embodiment, the processor is further configured to execute:

[0191] Before fitting the target point to determine the lane line, determining, among the non-target points in the three-dimensional point cloud, candidate points whose distance to the target point is less than a preset distance;

[0192] Determine, among the candidate points, an extension point whose similarity with the preset attribute information of the target point is greater than a preset similarity;

[0193] Taking the extension point and the target point as new target points;

[0194] Wherein, fitting the target point to determine the lane line includes:

[0195] The new target points are fitted to determine the lane lines.

[0196] In one embodiment, the processor is configured to perform:

[0197] The lane line is determined by fitting the target point with a curve model.

[0198] In one embodiment, the processor is configured to perform:

[0199] The target point is fitted by a plurality of third-order Bezier curves to determine the lane line.

[0200] In one embodiment, the processor is further configured to execute:

[0201] A control instruction is generated based on the lane line, wherein the control instruction is used to control vehicle driving.

[0202] An embodiment of the present disclosure further provides an electronic device, comprising a device for determining lane lines in a high-precision map as described in any of the above embodiments.

[0203] An embodiment of the present disclosure also provides an autonomous driving vehicle, comprising the electronic device described in the above embodiment.

[0204] The systems, devices, modules or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described separately by functions and are divided into various units. Of course, when implementing this application, the functions of each unit can be implemented in the same one or more software and / or hardware. It should be understood by those skilled in the art that the embodiments of the present disclosure can be provided as methods, systems, or computer program products. Therefore, the present disclosure can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present disclosure can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0205] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the system embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0206] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. The terms "include", "comprises" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the statement "comprises a ..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0207] The above is only an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application should be included in the scope of the claims of the present application.

Claims

1. A method for determining lane lines in a high-precision map, characterized in that: include: Acquire environmental images and three-dimensional point clouds of the vehicle's driving environment; The three-dimensional point cloud is a three-dimensional point cloud of the environment at the target moment; Determining a lane line area in the environment image; Projecting points in the three-dimensional point cloud onto the environment image to determine target points located in the lane line area; Fitting the target point to determine the lane line; Before projecting the points in the three-dimensional point cloud onto the environment image, the method further includes: Determine a three-dimensional point cloud of the vehicle driving environment at at least one other time before or after the target time, and a predicted point cloud at the target time; Stacking the predicted point cloud into the three-dimensional point cloud of the environment at the target time; The projecting of the points in the three-dimensional point cloud onto the environment image comprises: Projecting the points in the stacked three-dimensional point cloud onto the environment image.

2. The method according to claim 1, characterized in that Determining the lane line area in the environment image includes: The lane line area is determined in the environment image according to a predetermined image recognition model.

3. The method according to claim 1, characterized in that Determining the lane line area in the environment image includes: determining a road surface area in the environment image; A lane line area is determined in the road surface area.

4. The method according to claim 1, characterized in that Before projecting the points in the three-dimensional point cloud onto the environment image, the method further includes: Determining obstacle points belonging to obstacles in the three-dimensional point cloud; Eliminating the obstacle points in the three-dimensional point cloud; The projecting of the points in the three-dimensional point cloud onto the environment image comprises: Projecting points in the three-dimensional point cloud excluding the obstacle points onto the environment image.

5. The method according to claim 4, characterized in that Determining the obstacle points belonging to obstacles in the three-dimensional point cloud comprises: Obstacle points belonging to obstacles are determined in the three-dimensional point cloud according to a predetermined deep learning model.

6. The method according to claim 1, characterized in that The determining of the three-dimensional point cloud at at least one other time before or after the target time, the predicted point cloud at the target time comprises: Determine a difference in posture of the vehicle at the other time and the target time, and a difference in position of the vehicle at the other time and the target time; The three-dimensional point cloud of the vehicle driving environment at the other time is determined according to the posture difference and the position difference, and the predicted point cloud at the target time.

7. The method according to any one of claims 1 to 6, characterized in that The projecting of the points in the three-dimensional point cloud onto the environment image comprises: According to the internal parameters of the image acquisition device that acquires the environmental image, the rotation relationship and displacement relationship between the world coordinate system and the coordinate system of the image acquisition device, the first coordinate of the point in the three-dimensional point cloud in the world coordinate system is converted into the second coordinate in the environmental image.

8. The method according to any one of claims 1 to 6, characterized in that The method further comprises: marking the target point; The target point is displayed with a label.

9. The method according to any one of claims 1 to 6, characterized in that The method further comprises: Correcting the lane line according to the received correction instruction; Projecting the corrected lane line into the environment image to determine whether the projection of the corrected lane line in the environment image matches the lane line area; Response information is generated according to a matching result between the projection of the corrected lane line in the environment image and the lane line area.

10. The method according to any one of claims 1 to 6, characterized in that Before fitting the target point to determine the lane line, the method further includes: Determine, among the non-target points in the three-dimensional point cloud, candidate points whose distance to the target point is less than a preset distance; Determine, among the candidate points, an extension point whose similarity with the preset attribute information of the target point is greater than a preset similarity; Taking the extension point and the target point as new target points; Wherein, fitting the target point to determine the lane line includes: The new target points are fitted to determine the lane lines.

11. The method according to any one of claims 1 to 6, characterized in that The fitting of the target point to determine the lane line comprises: The lane line is determined by fitting the target point with a curve model.

12. The method according to claim 11, characterized in that The determining the lane line by fitting the target point through a curve model comprises: The target point is fitted by a plurality of third-order Bezier curves to determine the lane line.

13. The method according to any one of claims 1 to 6, characterized in that The method further comprises: A control instruction is generated based on the lane line, wherein the control instruction is used to control vehicle driving.

14. A device for determining lane lines in a high-precision map, characterized in that: The determining device includes one or more processors working alone or in collaboration, and the processors are used to execute: Acquire an environmental image and a three-dimensional point cloud of the vehicle driving environment; the three-dimensional point cloud is a three-dimensional point cloud of the environment at the target moment; Determining a lane line area in the environment image; Projecting points in the three-dimensional point cloud onto the environment image to determine target points located in the lane line area; Fitting the target point to determine the lane line; Before projecting the points in the three-dimensional point cloud onto the environment image, the processor is further configured to execute: Determine a three-dimensional point cloud of the vehicle driving environment at at least one other time before or after the target time, and a predicted point cloud at the target time; Stacking the predicted point cloud into the three-dimensional point cloud of the environment at the target time; The projecting of the points in the three-dimensional point cloud onto the environment image comprises: Projecting the points in the stacked three-dimensional point cloud onto the environment image.

15. The device according to claim 14, characterized in that The processor is configured to execute: The lane line area is determined in the environment image according to a predetermined image recognition model.

16. The device according to claim 14, characterized in that The processor is configured to execute: determining a road surface area in the environment image; A lane line area is determined in the road surface area.

17. The device according to claim 14, characterized in that The processor is further configured to execute: Before projecting the points in the three-dimensional point cloud onto the environment image, determining obstacle points belonging to obstacles in the three-dimensional point cloud; Eliminating the obstacle points in the three-dimensional point cloud; The projecting of the points in the three-dimensional point cloud onto the environment image comprises: Projecting points in the three-dimensional point cloud excluding the obstacle points onto the environment image.

18. The device according to claim 17, characterized in that The processor is configured to execute: Obstacle points belonging to obstacles are determined in the three-dimensional point cloud according to a predetermined deep learning model.

19. The device according to claim 14, characterized in that The determining of the three-dimensional point cloud of at least one other time before or after the target time, the processor is configured to execute: Determine a difference in posture of the vehicle at the other time and the target time, and a difference in position of the vehicle at the other time and the target time; The three-dimensional point cloud of the vehicle driving environment at the other time is determined according to the posture difference and the position difference, and the predicted point cloud at the target time.

20. The device according to any one of claims 14 to 19, characterized in that The processor is configured to execute: According to the internal parameters of the image acquisition device that acquires the environmental image, the rotation relationship and displacement relationship between the world coordinate system and the coordinate system of the image acquisition device, the first coordinate of the point in the three-dimensional point cloud in the world coordinate system is converted into the second coordinate in the environmental image.

21. The device according to any one of claims 14 to 19, characterized in that The processor is further configured to execute: marking the target point; The target point is displayed with a label.

22. The device according to any one of claims 14 to 19, characterized in that The processor is further configured to execute: Correcting the lane line according to the received correction instruction; Projecting the corrected lane line into the environment image to determine whether the projection of the corrected lane line in the environment image matches the lane line area; Response information is generated according to a matching result between the projection of the corrected lane line in the environment image and the lane line area.

23. The device according to any one of claims 14 to 19, characterized in that The processor is further configured to execute: Before fitting the target point to determine the lane line, determining, among the non-target points in the three-dimensional point cloud, candidate points whose distance to the target point is less than a preset distance; Determine, among the candidate points, an extension point whose similarity with the preset attribute information of the target point is greater than a preset similarity; Taking the extension point and the target point as new target points; Wherein, fitting the target point to determine the lane line includes: The new target points are fitted to determine the lane lines.

24. The device according to any one of claims 14 to 19, characterized in that The processor is configured to execute: The lane line is determined by fitting the target point with a curve model.

25. The device according to claim 24, characterized in that The processor is configured to execute: The target point is fitted by a plurality of third-order Bezier curves to determine the lane line.

26. The device according to any one of claims 14 to 19, characterized in that The processor is further configured to execute: A control instruction is generated based on the lane line, wherein the control instruction is used to control vehicle driving.

27. An electronic device, characterized in that: A device for determining lane lines in a high-precision map comprising any one of claims 14 to 26.

28. An autonomous driving vehicle, characterized in that: An electronic device comprising the electronic device described in claim 27.

Images