High-precision map generation method, device, equipment, medium and autonomous vehicle

Abstract

The disclosure provides a high-precision map generation method, a high-precision map generation device, an electronic device, a storage medium and an autonomous vehicle, relates to the technical field of computers, and in particular to the fields of high-precision maps, intelligent transportation, autonomous driving and autonomous parking, cloud services, Internet of Vehicles and intelligent cockpits. The specific implementation scheme is: determining a candidate camera image set according to position information of a target ground point, the candidate camera image set including at least one frame of candidate camera image, and each frame of candidate camera image being associated with the target ground point; obtaining a candidate projection pixel set according to the position information and a candidate camera pose corresponding to each frame of candidate camera image, at least one candidate projection pixel in the candidate projection pixel set being obtained by projecting the target ground point to at least one frame of candidate camera image; determining a target projection pixel corresponding to the target ground point from the candidate projection pixel set; and generating a projection map according to the target projection pixel, the projection map being used to generate a map.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the field of image processing, and in particular to high-definition map, intelligent transportation, autonomous driving and self-parking, cloud service, Internet of Vehicles and intelligent cockpit. Specifically, it relates to a high-definition map generation method, a high-definition map generation device, an electronic device, a storage medium and an autonomous vehicle. BACKGROUND

[0002] With the development of transportation, roads become more and more complex, and vehicle types also become more and more complex, so the quality requirements for maps are getting higher and higher. A map can include multiple layers. A layer can include a bottom layer (i.e., a projection map). The bottom layer of the map can be used to model the actual three-dimensional environment. The bottom layer of the map can be spliced with other layers to generate a map. For example, the map can be a high-definition map. The high-definition map, also known as high-precision map, is a map used by autonomous vehicles. The high-definition map has accurate vehicle position information and rich road element data information, which can help the vehicle to predict the complex information of the road surface, such as slope, curvature, heading, etc., and better avoid potential risks. SUMMARY

[0003] The present disclosure provides a map generation method, device, electronic device, medium and autonomous vehicle.

[0004] According to an aspect of the present disclosure, a map generation method is provided, comprising: determining a candidate camera image set according to position information of a target ground point, wherein the candidate camera image set includes at least one frame of candidate camera image, and each frame of the candidate camera image is associated with the target ground point; obtaining a candidate projection pixel set according to the position information and a candidate camera pose corresponding to each frame of the candidate camera image, wherein at least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to the at least one frame of candidate camera image; determining a target projection pixel corresponding to the target ground point from the candidate projection pixel set; and generating a projection map according to the target projection pixel, wherein the projection map is used to generate a map.

[0005] According to another aspect of the present disclosure, there is provided a map generation apparatus comprising: a first determining module configured to determine a candidate camera image set according to position information of a target ground point, wherein the candidate camera image set comprises at least one candidate camera image, and each candidate camera image is associated with the target ground point; a first obtaining module configured to obtain a candidate projected pixel set according to the position information and a candidate camera pose corresponding to each candidate camera image, wherein at least one candidate projected pixel in the candidate projected pixel set is obtained by projecting the target ground point to the at least one candidate camera image; a second determining module configured to determine a target projected pixel corresponding to the target ground point from the candidate projected pixel set; and a generating module configured to generate a projection map according to the target projected pixel, wherein the projection map is used to generate a map.

[0006] According to another aspect of the present disclosure, there is provided an electronic device comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method as described above.

[0007] According to another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to enable a computer to perform the method as described above.

[0008] According to another aspect of the present disclosure, there is provided a computer program product comprising a computer program, wherein the computer program, when executed by a processor, implements the method as described above.

[0009] It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0010] The accompanying drawings are used to better understand the present scheme, and do not limit the present disclosure. Among them:

[0011] Figure 1 An exemplary system architecture to which the map generation method and apparatus according to embodiments of the present disclosure can be applied is schematically shown;

[0012] Figure 2 A flowchart of a map generation method according to embodiments of the present disclosure is schematically shown;

[0013] Figure 3 An exemplary schematic diagram of a map generation process according to embodiments of the present disclosure is schematically shown.

[0014] Figure 4 An example schematic diagram of a projection map is schematically shown according to an embodiment of the present disclosure;

[0015] Figure 5 A block diagram of a map generation apparatus is schematically shown according to an embodiment of the present disclosure; and

[0016] Figure 6 A block diagram of an electronic device adapted to implement a map generation method is schematically shown according to an embodiment of the present disclosure. DETAILED DESCRIPTION

[0017] Exemplary embodiments of the present disclosure are described herein with reference to the accompanying drawings, in which various details of embodiments of the present disclosure are set forth to assist in the understanding of the present disclosure. It will be apparent to those skilled in the art that various changes and modifications can be made thereto without departing from the scope and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted for clarity and conciseness.

[0018] Road surface elements are a kind of important basic elements in a map, which can be used for positioning vehicle pose and planning driving path. Therefore, in the case of making a map, road surface elements need to be extracted and labeled from a projection map. Therefore, the accuracy and information richness of the projection map will affect the generation effect of the map.

[0019] The projection map can be generated in the following manner. A laser radar can be configured on a collection vehicle, which can be used to collect relatively dense point cloud data. By traversing a target area, point cloud data of ground points on a road in the target area is collected. The point cloud data is projected on the top view to obtain a road reflectivity intensity map of the target area as a projection map. The ground points on the road in the target area can include target ground points.

[0020] The above-mentioned manner of generating a projection map is more susceptible to the quality of point cloud scanning. If the quality of the point cloud data is poor, it is more likely to cause the map to be unclear, i.e., it is more difficult to clearly express the texture information of the road.

[0021] To this end, the embodiment of the present disclosure proposes a map generation scheme. According to the position information of the target ground point, a candidate camera image set is determined. The candidate camera image set includes at least one frame of candidate camera image, and each frame of candidate camera image is associated with the target ground point. According to the position information and the candidate camera pose corresponding to each frame of candidate camera image, a candidate projection pixel set is obtained. At least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to at least one frame of candidate camera image respectively. A target projection pixel corresponding to the target ground point is determined from the candidate projection pixel set. A projection map is generated according to the target projection pixel. The projection map is used to generate a map.

[0022] The projection map is generated in the manner of the position information and the camera image, which improves the quality of the projection map, and thus can provide a clearer projection map for map labeling, which helps to improve the labeling accuracy and labeling efficiency. On this basis, the generation effect of the map is improved.

[0023] Figure 1 An exemplary system architecture to which the map generation method and device according to the embodiment of the present disclosure can be applied is schematically shown.

[0024] It should be noted that Figure 1 The shown is only an example of a system architecture to which the embodiment of the present disclosure can be applied, to help those skilled in the art understand the technical content of the present disclosure, but does not mean that the embodiment of the present disclosure cannot be used in other devices, systems, environments or scenarios.

[0025] As Figure 1 shown, the system architecture 100 according to the embodiment can include terminal devices 101, 102, 103, a network 104, a server 105 and a vehicle 106. The vehicle 106 can travel on the road. The network 104 is used to provide a communication link medium between the terminal devices 101, 102, 103 and the server 105, between the terminal devices 101, 102, 103 and the vehicle 106, and between the vehicle 106 and the server 105. The network 104 can include various connection types, such as wired and / or wireless communication links, etc.

[0026] The vehicle 106 can include a combustion engine powered vehicle, an electric vehicle or a hybrid electric vehicle, etc. For example, the vehicle 101 can be a vehicle configured with an automatic control system. The vehicle 101 can be an autonomous vehicle. The vehicle 101 can be installed with a collection device for collecting surrounding environment information. The collection device can include a laser radar 1060 and a camera 1061. The laser radar 1060 can include a laser scanner, at least one laser source and at least one detector.

[0027] The vehicle 106 can drive on the road so that the laser radar 1060 on the vehicle 106 can collect point cloud information associated with the target ground point. The camera 1061 can collect a camera image associated with the target ground point. The vehicle 106 can send the point cloud information and the camera image to the terminal device 101, 102, or 103. Or the vehicle 106 can send the point cloud information and the camera image to the server 105.

[0028] The user can use the terminal device 101, 102, or 103 to interact with the server 105 through the network 104 to receive or send messages, etc. Various communication client applications can be installed on the terminal device 101, 102, or 103, such as knowledge reading applications, web browser applications, search applications, instant messaging tools, email clients, and / or social platform software, etc. (only as examples).

[0029] The terminal device 101, 102, or 103 can be various electronic devices with a display screen and supporting web browsing, including but not limited to smartphones, tablet computers, laptop computers, desktop computers, etc.

[0030] The server 105 can be various types of servers providing various services. The server 106 can be a cloud server. The cloud server is a host product in the cloud computing service system, which solves the defects of large management difficulty and weak business scalability in traditional physical host and VPS (Virtual Private Server, VPS) services.

[0031] It should be noted that the map generation method provided by the embodiments of the present disclosure can generally be executed by the terminal device 101, 102, or 103. Correspondingly, the map generation apparatus provided by the embodiments of the present disclosure can also be arranged in the terminal device 101, 102, or 103.

[0032] For example, the terminal device 101, 102, or 103 determines a candidate camera image set according to the position information of the target ground point. The candidate camera image set includes at least one frame of candidate camera image, and each frame of candidate camera image is associated with the target ground point. According to the position information and the candidate camera pose corresponding to each frame of candidate camera image, a candidate projection pixel set is obtained. At least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to at least one frame of candidate camera image respectively. A target projection pixel corresponding to the target ground point is determined from the candidate projection pixel set. A projection map is generated according to the target projection pixel. The projection map is used to generate a map.

[0033] Alternatively, the map generation method provided by the embodiments of the present disclosure can also be generally executed by the server 105. Correspondingly, the map generation apparatus provided by the embodiments of the present disclosure can be generally arranged in the server 105. The map generation method provided by the embodiments of the present disclosure can also be executed by a server or a server cluster different from the server 105 and capable of communicating with the terminal device 101, 102, 103 and / or the server 105. Correspondingly, the map generation apparatus provided by the embodiments of the present disclosure can also be arranged in a server or a server cluster different from the server 105 and capable of communicating with the terminal device 101, 102, 103 and / or the server 105.

[0034] For example, the server 105 determines a candidate camera image set according to the position information of the target ground point. The candidate camera image set includes at least one frame of candidate camera image, and each frame of candidate camera image is associated with the target ground point. A candidate projection pixel set is obtained according to the position information and a candidate camera pose corresponding to each frame of candidate camera image. At least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to at least one frame of candidate camera image respectively. A target projection pixel corresponding to the target ground point is determined from the candidate projection pixel set. A projection map is generated according to the target projection pixel. The projection map is used to generate a map.

[0035] Alternatively, the map generation method provided by the embodiments of the present disclosure can also be generally executed by the vehicle 106. Correspondingly, the map generation apparatus provided by the embodiments of the present disclosure can be generally arranged in the vehicle 106. For example, the vehicle 106 is configured to execute the map generation method.

[0036] It should be understood that Figure 1 The number of terminal devices, networks, servers and vehicles in the system 100 is only illustrative. According to the needs of implementation, there can be any number of terminal devices, networks, servers and vehicles.

[0037] Figure 2 A flowchart of a map generation method according to the embodiments of the present disclosure is schematically shown.

[0038] As shown in FIG. 2, the method 200 includes operations S210-S240. Figure 2

[0039] At operation S210, a candidate camera image set is determined according to position information of a target ground point. The candidate camera image set includes at least one frame of candidate camera image, and each frame of candidate camera image is associated with the target ground point.

[0040] ​At operation S220, a candidate projection pixel set is obtained according to the position information and a candidate camera pose corresponding to each of the candidate camera images. At least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to at least one of the candidate camera images.

[0041] At operation S230, a target projection pixel corresponding to the target ground point is determined from the candidate projection pixel set.

[0042] At operation S240, a projection map is generated according to the target projection pixel. The projection map is used to generate a map.

[0043] According to an embodiment of the present disclosure, the target ground point can refer to a ground point on a road to be projected. The position information can refer to three-dimensional coordinate information. The camera image can be an image captured by a camera. The camera image has a camera pose corresponding to the camera image. That is, the camera image can be a camera image captured by the camera at the camera pose corresponding to the camera image. The camera pose can include a translation matrix and a rotation matrix. The candidate camera image can be a camera image associated with the target ground point. The candidate camera pose can refer to a camera pose corresponding to the candidate camera image. The projection pixel can be obtained by projecting the ground point to the camera image. The map can include a high-definition map. The high-definition map is a component in autonomous driving. The high-definition map provides technical support for scene perception and decision-making and other basic technologies for autonomous driving of an autonomous vehicle in combination with real-time positioning technology of the autonomous vehicle.

[0044] According to an embodiment of the present disclosure, at least one candidate camera image satisfying the camera image associated with the target ground point can be determined from a plurality of camera images according to the position information of the target ground point. A candidate camera image set is obtained according to the at least one candidate camera image.

[0045] According to an embodiment of the present disclosure, after obtaining the candidate camera image set, the target ground point can be projected to each of the candidate camera images respectively by using the position information of the target ground point and the camera pose of each of the candidate camera images, to obtain a candidate projection pixel corresponding to each of the candidate camera images. Thereby, a candidate projection pixel set obtained according to at least one candidate projection pixel can be obtained.

[0046] According to an embodiment of the present disclosure, after obtaining the candidate projection pixel set, a target projection pixel satisfying a predetermined condition can be determined from the candidate projection pixels included in the candidate projection pixel set. The projection map can be generated according to the target projection pixel. For example, the projection map can be generated according to the target projection pixel and other projection pixels. The other projection pixels can be target projection pixels corresponding to other ground points obtained by the ground points using the map generation method described in the embodiments of the present disclosure. The projection map can be used to generate the map. That is, the road pavement elements of the road can be extracted according to the projection map. The road pavement elements are labeled to generate the high-definition map. The road pavement elements can include at least one of the following: lane lines and road pavers.

[0047] According to an embodiment of the present disclosure, the candidate camera image set associated with the target ground point is determined according to the position information of the target ground point. The candidate projection pixel set is obtained according to the position information and the candidate camera poses corresponding to each of the candidate camera images. At least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to at least one candidate camera image. The target projection pixel corresponding to the target ground point is determined from the candidate projection pixel set. The projection map is generated according to the target projection pixel. The projection map is used to generate the map. The projection map is generated according to the position information and the camera images, which improves the quality of the projection map, and in turn provides a clearer projection map for labeling the map, which helps to improve the labeling accuracy and efficiency. On this basis, the generation effect of the map is improved.

[0048] In addition, not only the projection in the top view is performed, and therefore, if there are multiple layers of roads, the projection map of each layer of road can be identified separately.

[0049] According to an embodiment of the present disclosure, operation S210 can include the following operations.

[0050] The position information of the target ground point is determined. The candidate camera pose set is determined from the plurality of camera poses according to the position information of the target point. The camera image corresponding to each of the at least one candidate camera pose included in the candidate camera pose set is determined as the candidate camera image, and the candidate camera image set is obtained.

[0051] According to an embodiment of the present disclosure, the position information of the target ground point can be determined according to the two-dimensional ground network map and the three-dimensional ground network model. Then, the candidate camera pose is determined from the plurality of camera images according to the position information of the target ground point. The camera image corresponding to the candidate camera pose is determined as the candidate camera image.

[0052] According to an embodiment of the present disclosure, the above map generation method can further include the following operations.

[0053] According to the positioning track and the pose conversion relationship, the camera poses of the plurality of camera images are determined, and the plurality of camera poses are obtained. The positioning track includes positioning information of a plurality of track points.

[0054] According to an embodiment of the present disclosure, the track points can refer to track points of a collection vehicle. The positioning information can be obtained by processing point cloud information collected by a radar configured on the collection vehicle. The collection vehicle can also be configured with a camera. The camera can be used to collect camera images. The configuration positions of the camera and the radar can be set according to actual business needs, which are not limited herein. For example, the camera can be configured at the front of the collection vehicle. The radar can be configured at the top of the collection vehicle, so as to be able to collect camera images and point cloud information within a 360° range around the collection vehicle.

[0055] According to an embodiment of the present disclosure, the pose conversion relationship can refer to a pose conversion relationship between the radar and the camera configured on the collection vehicle. The pose conversion relationship can be determined according to relative position information of the radar and relative position information of the camera. That is, the relative position information of the radar and the relative position information of the camera can be determined. The pose conversion relationship between the camera and the radar is determined according to the relative position information of the radar and the relative position information of the camera.

[0056] According to an embodiment of the present disclosure, the radar can collect point cloud information according to a first time interval, and obtain positioning information of a plurality of track points. The camera can collect camera images according to a second time interval, and obtain a plurality of frames of camera images. Each track point has a first timestamp corresponding to the track point. Each frame of camera image has a second timestamp corresponding to the camera image. The first time interval and the second time interval can be the same or different. If the first time interval is less than the second time interval, the number of track points is greater than the number of camera images in the same time period. In this case, if there is no camera image corresponding to the track point, an interpolation method can be used to obtain the camera image corresponding to the track point.

[0057] According to an embodiment of the present disclosure, after obtaining the positioning information of each track point, the camera pose of the camera image corresponding to the track point can be determined according to the pose conversion relationship and the positioning information of each track point.

[0058] According to an embodiment of the present disclosure, determining the position information of the target ground point can include the following operations.

[0059] According to the two-dimensional ground grid map, the two-dimensional coordinate information of the target ground point is determined. The height information of the target ground point from the target ground is determined by using the three-dimensional ground grid model. The position information of the target ground point is obtained according to the two-dimensional coordinate information and the height information.

[0060] According to an embodiment of the present disclosure, the two-dimensional ground grid map can be obtained by discretizing the target region on the map in a certain resolution. The three-dimensional ground grid model can be constructed from the point cloud information corresponding to the target region. The three-dimensional ground grid model can include a plurality of connected triangular facets. Each triangular facet can be attached to the ground. Each triangular facet can represent the height information from the ground at that location.

[0061] According to an embodiment of the present disclosure, for a target ground point, the height information of the target ground point can be determined from the three-dimensional ground grid model according to the index information of the target ground point. The position information of the target ground point can be obtained according to the two-dimensional coordinate information of the target ground point and the height information of the target ground point. If there are multiple layers of roads, multiple height information corresponding to the two-dimensional coordinate information can be obtained. Thus, multiple position information can be obtained. Each position information can be processed by using the map generation method according to the present disclosure.

[0062] According to an embodiment of the present disclosure, determining the candidate camera pose set from the plurality of camera poses according to the position information of the target point can include the following operations.

[0063] According to the predetermined position offset range and the position information, a target position range is determined. The camera poses in the plurality of camera poses that match the target position range are determined as the candidate camera poses, and a candidate camera pose set is obtained.

[0064] According to an embodiment of the present disclosure, the predetermined position offset range can be used as a basis for determining the candidate camera pose. The predetermined position offset range can be configured according to actual business needs, which is not limited herein. Each position information has a camera pose corresponding to the position information. The target position range can be determined according to the predetermined position offset range and the position information. The target position range can include multiple position information. The camera pose corresponding to each position information in the multiple position information included in the target position range can be determined as the candidate camera pose. Thus, the candidate camera pose set can be obtained.

[0065] According to an embodiment of the present disclosure, operation S220 can include the following operations.

[0066] According to the position information, the camera intrinsic parameters, and the candidate camera pose corresponding to each frame of candidate camera image included in the candidate camera image set, a candidate projection pixel set is obtained by using a projection equation.

[0067] According to an embodiment of the present disclosure, the projection equation can be determined according to the camera intrinsic parameters, the position information, and the candidate camera pose.

[0068] According to an embodiment of the present disclosure, for each candidate camera image included in the candidate camera set, the position information, the camera intrinsic parameter and the candidate camera pose corresponding to the candidate camera image are input into a projection equation to obtain a candidate projection pixel corresponding to the target ground point. Thus, a candidate projection pixel set can be obtained.

[0069] According to an embodiment of the present disclosure, the candidate projection pixel can be determined according to the following formulas (1) and (2).

[0070] I = D (K[R, T]P) (1)

[0071] I = [u, v] (2)

[0072] According to an embodiment of the present disclosure, D (K[R, T]P) represents the projection equation. I represents the candidate projection pixel. u and v represent the pixel coordinate information of the candidate projection pixel. K represents the camera intrinsic parameter. [R, T] represents the candidate camera pose corresponding to the candidate camera image. P represents the position information. R represents the candidate conversion matrix. T represents the candidate translation matrix. D represents the distortion correction.

[0073] According to an embodiment of the present disclosure, operation S230 can include the following operations.

[0074] The target camera image corresponding to the target ground point is determined from the candidate camera image set. The candidate projection pixel corresponding to the target camera image is determined as the target projection pixel corresponding to the target ground point.

[0075] According to an embodiment of the present disclosure, the target camera image corresponding to the target ground point can be determined from the candidate camera image set based on a selection condition. The candidate projection pixel corresponding to the target camera image is determined as the target projection pixel corresponding to the target ground point.

[0076] According to an embodiment of the present disclosure, the selection condition can include at least one of the following: the target camera image corresponding to the target ground point is determined according to a target camera image corresponding to a first adjacent ground point. The first adjacent ground point can be determined according to the target ground point. The target camera image corresponding to the target ground point is determined from a target candidate camera image set. The target candidate camera image set is a candidate camera image set corresponding to the same single-circle track. The target camera image corresponding to the target ground point is a candidate projection pixel corresponding to a static object.

[0077] According to an embodiment of the present disclosure, determining the target camera image corresponding to the target ground point from the candidate camera image set can include the following operations.

[0078] A first adjacent ground point corresponding to the target ground point is determined. The target camera image corresponding to the target ground point is determined from the candidate camera image set according to a target camera image corresponding to the first adjacent ground point.

[0079] According to an embodiment of the present disclosure, the target camera image corresponding to the target ground point and the first neighboring ground point has a correlation relationship, and therefore, the target camera image corresponding to the target ground point can be determined according to the target camera image of the first neighboring ground point. The first neighboring ground point can refer to a ground point having a distance within a first predetermined distance range from the target ground point. The first predetermined distance range can be configured according to actual business requirements, which is not limited herein. For example, the target camera image corresponding to the first neighboring ground point can be determined as the target camera image corresponding to the target ground point.

[0080] According to an embodiment of the present disclosure, determining the target camera image corresponding to the target ground point from the candidate camera image set can include the following operations.

[0081] In a case where it is determined that the target ground point is a ground point on a one-way road, a target candidate camera image set is determined from the candidate camera image set. The target candidate camera image set is a candidate camera image set corresponding to the same single-loop track. The target camera image corresponding to the target ground point is determined from the target candidate camera image set.

[0082] According to an embodiment of the present disclosure, the one-way road can select a camera image set corresponding to the same single-loop track. The single-loop track can refer to a track belonging to the same loop. That is, a track formed by starting from a starting point of the road and returning to the same starting point again.

[0083] According to an embodiment of the present disclosure, it can be determined whether the target ground point is a ground point on a one-way road. If it is determined that the target ground point is a ground point on a one-way road, candidate camera images corresponding to the same single-loop track can be determined from the candidate camera image set, to obtain a target candidate camera image set. For example, candidate camera images corresponding to the same single-loop track in the candidate camera image set can be determined according to time stamps corresponding to the candidate camera images. After the target candidate camera set is determined, the target camera image corresponding to the target ground point can be determined from the target candidate camera set.

[0084] According to an embodiment of the present disclosure, determining the target camera image corresponding to the target ground point from the target candidate camera image set can include the following operations.

[0085] A second neighboring ground point corresponding to the target ground point is determined from a set of other ground points corresponding to the target candidate camera image set. The target camera image corresponding to the target ground point is determined from the target candidate camera image set according to the target camera image corresponding to the second neighboring ground point.

[0086] According to an embodiment of the present disclosure, each target candidate camera image has at least one ground point corresponding to the target candidate camera image. The ground point corresponding to the target candidate camera image can be referred to as another ground point. Thus, a set of another ground points corresponding to the target candidate camera image can be determined.

[0087] According to an embodiment of the present disclosure, the target camera image corresponding to the target ground point and the second neighboring ground point has an association relationship, and thus the target camera image corresponding to the target ground point can be determined according to the target camera image of the second neighboring ground point. The second neighboring ground point can refer to a ground point having a distance from the target ground point within a second predetermined distance range. The second predetermined distance range can be configured according to actual business requirements, which is not limited herein. For example, the target camera image corresponding to the second neighboring ground point can be determined as the target camera image corresponding to the target ground point.

[0088] According to an embodiment of the present disclosure, the target projection pixel is a candidate projection pixel corresponding to a static object.

[0089] According to an embodiment of the present disclosure, the target projection pixel can be a candidate projection pixel corresponding to a static object, so as to improve the projection quality. The static object can refer to an object that is stationary on the road.

[0090] According to an embodiment of the present disclosure, operation S240 can include the following operations.

[0091] The color information of the target projection pixel is determined. According to the color information of the target projection pixel, the color information of the target ground point is determined. According to the color information of the target ground point, the projection map is generated.

[0092] According to an embodiment of the present disclosure, the color information can include RGB color information. R (Red), G (Green), and B (Blue). The color information of the target projection pixel can be assigned to the target ground point, so that the color information of the target ground point is consistent with the color information of the target projection pixel. The projection map can be generated according to the color information of the target projection pixel and the color information of another projection pixel. The another projection pixel can be obtained by using the map generation method according to an embodiment of the present disclosure.

[0093] The map generation method according to an embodiment of the present disclosure will be further described below with reference to Figures 3-4 , in combination with specific embodiments.

[0094] Figure 3 An example schematic diagram of a map generation process according to an embodiment of the present disclosure is schematically shown.

[0095] As Figure 3As shown, in the map generation process 300, according to the positioning trajectory 301 and the pose conversion relationship 302, the camera poses 303 of the plurality of camera images are determined respectively, and the plurality of camera poses 303 are obtained. The positioning trajectory includes the positioning information of each trajectory point.

[0096] According to the position information 304 and the predetermined position offset range 305, the target position range 306 is determined. The camera poses 303 in the plurality of camera poses 303 that match the target position range 306 are determined as candidate camera poses, and the candidate camera pose set 307 is obtained. The position information 304, the camera intrinsic parameter 308, and the candidate camera poses included in the candidate camera pose set 307 are input into the projection equation 309, and the candidate projection pixel set 310 is obtained. The target projection pixel 311 corresponding to the target ground point is determined from the candidate projection pixel set 310. The color information 312 of the target projection pixel is determined. According to the color information 313 of the target projection pixel 311, the color information 314 of the target ground point is determined. According to the color information 314 of the target ground point, the projection map 315 is generated. The projection map 315 is used to generate the map 316.

[0097] Figure 4 An example schematic diagram of a projection map according to an embodiment of the present disclosure is schematically shown.

[0098] As Figure 4 shown, the projection map 400 can be obtained by using the map generation method according to the embodiment of the present disclosure. The projection effect of the projection map 400 is relatively clear.

[0099] The above is only an example embodiment, but is not limited thereto, and other map generation methods known in the art can also be included, as long as the quality of the projection map can be improved.

[0100] Figure 5 A block diagram of a map generation apparatus according to an embodiment of the present disclosure is schematically shown.

[0101] As Figure 5 shown, the map generation apparatus 500 can include a first determination module 510, a first obtaining module 520, a second determination module 530, and a generation module 540.

[0102] The first determination module 510 is configured to determine a candidate camera image set according to position information of a target ground point. The candidate camera image set includes at least one frame of candidate camera image, and each frame of candidate camera image is associated with the target ground point.

[0103] The first obtaining module 520 is configured to obtain a candidate projection pixel set according to the position information and a candidate camera pose corresponding to each frame of candidate camera image. At least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to at least one frame of candidate camera image respectively.

[0104] The second determining module 530 is configured to determine a target projection pixel corresponding to the target ground point from the candidate projection pixel set.

[0105] The generating module 540 is configured to generate a projection map according to the target projection pixel. The projection map is used to generate a map.

[0106] According to an embodiment of the present disclosure, the first determining module 510 can include a first determining sub-module, a second determining sub-module, and a third determining sub-module.

[0107] The first determining sub-module is configured to determine position information of the target ground point.

[0108] The second determining sub-module is configured to determine a candidate camera pose set from a plurality of camera poses according to the position information of the target point.

[0109] The third determining sub-module is configured to determine a camera image corresponding to each of at least one candidate camera pose included in the candidate camera pose set as the candidate camera image, to obtain the candidate camera image set.

[0110] According to an embodiment of the present disclosure, the map generation apparatus 500 can include a second obtaining module.

[0111] The second obtaining module is configured to determine a camera pose of each of a plurality of camera images according to a positioning track and a pose conversion relationship, to obtain the plurality of camera poses. The positioning track includes positioning information of each of a plurality of track points.

[0112] According to an embodiment of the present disclosure, the first determining sub-module can include a first determining unit, a second determining unit, and a first obtaining unit.

[0113] The first determining unit is configured to determine two-dimensional coordinate information of the target ground point according to a two-dimensional ground grid map.

[0114] The second determining unit is configured to determine height information of the target ground point from a target ground surface by using a three-dimensional ground grid model.

[0115] The first obtaining unit is configured to obtain position information of the target ground point according to the two-dimensional coordinate information and the height information.

[0116] According to an embodiment of the present disclosure, the second determining sub-module can include a third determining unit and a second obtaining unit.

[0117] The third determining unit is configured to determine a target position range according to a predetermined position offset range and the position information.

[0118] The second obtaining unit is configured to determine camera poses in the plurality of camera poses that match the target position range as candidate camera poses, to obtain a candidate camera pose set.

[0119] According to an embodiment of the present disclosure, the first obtaining module can include an obtaining sub-module.

[0120] The obtaining sub-module is configured to obtain a candidate projection pixel set according to the position information, the camera intrinsic parameter, and a candidate camera pose corresponding to each frame of candidate camera image included in the candidate camera image set, by using a projection equation.

[0121] According to an embodiment of the present disclosure, the second determining module can include a fourth determining sub-module and a fifth determining sub-module.

[0122] The fourth determining sub-module is configured to determine a target camera image corresponding to the target ground point from the candidate camera image set.

[0123] The fifth determining sub-module is configured to determine a candidate projection pixel corresponding to the target camera image as a target projection pixel corresponding to the target ground point.

[0124] According to an embodiment of the present disclosure, the fourth determining sub-module can include a fourth determining unit and a fifth determining unit.

[0125] The fourth determining unit is configured to determine a first adjacent ground point corresponding to the target ground point.

[0126] The fifth determining unit is configured to determine a target camera image corresponding to the target ground point from the candidate camera image set according to a target camera image corresponding to the first adjacent ground point.

[0127] According to an embodiment of the present disclosure, the fourth determining sub-module can include a sixth determining unit and a seventh determining unit.

[0128] The sixth determining unit is configured to, in a case where it is determined that the target ground point is a ground point on a one-way road, determine a target candidate camera image set from the candidate camera image set, wherein the target candidate camera image set is a candidate camera image set corresponding to a same single image track.

[0129] The seventh determining unit is configured to determine a target camera image corresponding to the target ground point from the target candidate camera image set.

[0130] According to an embodiment of the present disclosure, the seventh determining unit can include a first determining sub-unit and a second determining sub-unit.

[0131] The first determining sub-unit is configured to determine a second adjacent ground point corresponding to the target ground point from a set of other ground points corresponding to the target candidate camera image set.

[0132] The second determining sub-unit is configured to determine, from the target candidate camera image set, a target camera image corresponding to the target ground point according to the target camera image corresponding to the second adjacent ground point.

[0133] According to an embodiment of the present disclosure, the target projection pixel is a candidate projection pixel corresponding to a static object.

[0134] According to an embodiment of the present disclosure, the generating module 540 can include a fourth determining sub-module, a fifth determining sub-module, and a generating sub-module.

[0135] The fourth determining sub-module is configured to determine color information of the target projection pixel.

[0136] The fifth determining sub-module is configured to determine color information of the target ground point according to the color information of the target projection pixel.

[0137] The generating sub-module is configured to generate a projection map according to the color information of the target ground point.

[0138] According to an embodiment of the present disclosure, the present disclosure further provides an electronic device, a readable storage medium, and a computer program product.

[0139] According to an embodiment of the present disclosure, an electronic device includes at least one processor, and a memory connected with the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described above.

[0140] According to an embodiment of the present disclosure, a non-transitory computer readable storage medium stores computer instructions, wherein the computer instructions are used to enable a computer to perform the method described above.

[0141] According to an embodiment of the present disclosure, a computer program product includes a computer program, and the computer program, when executed by a processor, implements the method described above.

[0142] Figure 6 A block diagram schematically illustrates an electronic device suitable for implementing the map generation method according to an embodiment of the present disclosure. The electronic device is intended to represent a variety of forms of digital computers, such as laptops, desktops, tablets, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device can also represent a variety of forms of mobile devices, such as personal digital assistants, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not intended to limit the implementations of the present disclosure described and / or claimed in this document.

[0143] AsFigure 6 As shown, the electronic device 600 includes a computing unit 601 that can perform various appropriate actions and processes in accordance with a computer program stored in a read-only memory (ROM) 602 or a computer program loaded into a random access memory (RAM) 603 from a storage unit 608. Various programs and data required for the operation of the electronic device 600 can also be stored in the RAM 603. The computing unit 601, the ROM 602, and the RAM 603 are connected to each other through a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0144] Various components in the electronic device 600 are connected to the I / O interface 605, including an input unit 606 such as a keyboard, a mouse, and the like, an output unit 607 such as various types of displays, a speaker, and the like, a storage unit 608 such as a magnetic disk, an optical disk, and the like, and a communication unit 609 such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 609 allows the electronic device 600 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0145] The computing unit 601 can be various general-purpose and / or special-purpose processing components having processing and computing capabilities. Some examples of the computing unit 601 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, and the like. The computing unit 601 performs various methods and processes described above, such as the map generation method. For example, in some embodiments, the map generation method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 608. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 600 via the ROM 602 and / or the communication unit 609. When the computer program is loaded into the RAM 603 and executed by the computing unit 601, one or more steps of the map generation method described above can be performed. Alternatively, in other embodiments, the computing unit 601 can be configured to perform the map generation method by any other appropriate means, such as by means of firmware.

[0146] Based on the foregoing electronic device, the present disclosure further provides an autonomous vehicle, which can include the electronic device, and can further include a communication component, a display screen for implementing a human-machine interface, and an information acquisition device for acquiring surrounding environment information, and the like, and the communication component, the display screen, and the information acquisition device are in communication connection with the electronic device. The electronic device included in the autonomous vehicle can implement the map generation method described in the embodiments of the present disclosure.

[0147] The electronic device can be integrally integrated with the communication component, the display screen, and the information collection device, or can be separately provided with the communication component, the display screen, and the information collection device.

[0148] Various implementations of the systems and techniques described above can be realized in digital electronic circuitry, integrated circuitry, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), a system on a chip (SOC), a complex programmable logic device (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0149] Program code for carrying out methods of the present disclosure can be written in any combination of one or more programming languages. The program code can be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus to produce a machine, such that the program code, when executed by the processor or controller, produces the functions / operations specified in the flowcharts and / or the block diagrams. The program code can be executed entirely on a machine, partially on a machine, partially on a machine as a stand-alone software package, partially on a machine and partially on a remote machine or entirely on a remote machine or server.

[0150] In the context of the present disclosure, a machine-readable medium can be a tangible medium that contains or stores a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include but is not limited to an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of the machine-readable storage medium will include one or more lines of electrical connections, portable computer disks, hard disk drives, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or Flash memory), optical fibers, portable compact disc read-only memories (CD-ROMs), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0151] To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0152] The systems and techniques described here can be implemented in a computing system that includes a back end component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front end component (e.g., a user computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network (LAN), a wide area network (WAN), and the Internet.

[0153] The computer system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server is generally established by computer programs running on the respective computers and having a client-server relationship to each other. The servers can be cloud servers, servers of a distributed system, or servers combined with a blockchain.

[0154] It should be understood that the various forms of flow shown above can be re-ordered, added to, or deleted from without departing from the spirit and scope of the present disclosure. For example, the steps recited in the present disclosure can be performed in parallel, in series, or in a different order, without departing from the desired results of the technology disclosed herein, which are not limited herein.

[0155] The specific embodiments described above are not intended to be limiting, and persons skilled in the art will appreciate that various modifications, combinations, sub-combinations and alternatives can be made to the specific embodiments without departing from the spirit and scope of the disclosure. Any modifications, equivalent substitutions, improvements, and the like, made within the spirit and principles of the disclosure, are intended to be included in the scope of the disclosure.

Claims

1. A method for generating a map, comprising: determining a candidate camera image set according to position information of a target ground point, wherein the candidate camera image set comprises at least one frame of candidate camera image, and each frame of the candidate camera image is associated with the target ground point; obtaining a candidate projected pixel set according to the position information and a candidate camera pose corresponding to each frame of the candidate camera image, wherein at least one candidate projected pixel in the candidate projected pixel set is obtained by projecting the target ground point to the at least one frame of candidate camera image respectively; determining a target projected pixel corresponding to the target ground point from the candidate projected pixel set; and generating a projection map according to the target projected pixel, wherein the projection map is used to generate a map; wherein the determining the target projected pixel corresponding to the target ground point from the candidate projected pixel set comprises: determining a target camera image corresponding to the target ground point from the candidate camera image set; and determining the candidate projected pixel corresponding to the target camera image as the target projected pixel corresponding to the target ground point; wherein the determining the target camera image corresponding to the target ground point from the candidate camera image set comprises: determining a first neighboring ground point corresponding to the target ground point, wherein the first neighboring ground point comprises a ground point having a distance within a first predetermined distance range from the target ground point, and a target camera image corresponding to the first neighboring ground point has an association relationship with the target camera image; and determining the target camera image corresponding to the target ground point from the candidate camera image set according to the target camera image corresponding to the first neighboring ground point. 2.The method of claim 1, wherein the determining the position information of the target ground point comprises: determining two-dimensional coordinate information of the target ground point according to a two-dimensional ground grid map; determining height information of the target ground point from a target ground surface by using a three-dimensional ground grid model; and obtaining the position information of the target ground point according to the two-dimensional coordinate information and the height information. 3.The method of claim 2, further comprising: determining a camera pose of each of a plurality of camera images according to a positioning trajectory and a pose conversion relationship, to obtain the plurality of camera poses, wherein the positioning trajectory comprises positioning information of a plurality of trajectory points. 4.The method of claim 3, wherein the determining the position information of the target ground point comprises: determining the two-dimensional coordinate information of the target ground point according to a two-dimensional ground grid map; determining height information of the target ground point from a target ground surface by using a three-dimensional ground grid model; and obtaining the position information of the target ground point according to the two-dimensional coordinate information and the height information. 5.The method of claim 4, wherein the determining the candidate camera pose set from the plurality of camera poses according to the position information of the target ground point comprises: determining a target position range according to a predetermined position offset range and the position information; and determining a camera pose matching the target position range from the plurality of camera poses as the candidate camera pose, to obtain the candidate camera pose set. ​ ​ ​ ​ ​ ​ 2. The method of claim 1, wherein, ​ ​ ​ ​ ​ ​ 4. The method of claim 2, wherein, ​ ​ ​ ​ 5. The method of claim 2, wherein, ​ ​ ​ 6. The method of claim 1, wherein, The candidate projection pixel set is obtained according to the position information, camera intrinsic parameters and the candidate camera pose corresponding to each frame of the candidate camera image included in the candidate camera image set by using a projection equation. The candidate projection pixel set is obtained according to the position information, camera intrinsic parameters and the candidate camera pose corresponding to each frame of the candidate camera image included in the candidate camera image set by using a projection equation.

7. The method of claim 1, wherein, The target camera image corresponding to the target ground point is determined from the candidate camera image set, including: In a case where it is determined that the target ground point is a ground point on a single-lane road, a target candidate camera image set is determined from the candidate camera image set, wherein the target candidate camera image set is a candidate camera image set corresponding to a same single-lane track; and The target camera image corresponding to the target ground point is determined from the target candidate camera image set.

8. The method of claim 7, wherein, The target camera image corresponding to the target ground point is determined from the target candidate camera image set, including: A second adjacent ground point corresponding to the target ground point is determined from a set of other ground points corresponding to the target candidate camera image set; and The target camera image corresponding to the target ground point is determined from the target candidate camera image set according to a target camera image corresponding to the second adjacent ground point.

9. The method of any one of claims 1-8, wherein, The target projection pixel is a candidate projection pixel corresponding to a static object.

10. The method of any one of claims 1-8, wherein, The projection map is generated according to the target projection pixel, including: Color information of the target projection pixel is determined; Color information of the target ground point is determined according to the color information of the target projection pixel; and The projection map is generated according to the color information of the target ground point.

11. A map generation apparatus, comprising: a first determination module configured to determine a candidate camera image set according to position information of a target ground point, wherein the candidate camera image set includes at least one frame of candidate camera image, and each frame of the candidate camera image is associated with the target ground point; a first obtaining module configured to obtain a candidate projection pixel set according to the position information and a candidate camera pose corresponding to each frame of the candidate camera image, wherein at least one candidate projection pixel in the candidate projection pixel set is obtained by projecting the target ground point to the at least one frame of candidate camera image respectively; a second determination module configured to determine a target projection pixel corresponding to the target ground point from the candidate projection pixel set; and a generation module configured to generate a projection map according to the target projection pixel, wherein the projection map is used to generate a map; wherein the second determination module includes: a fourth determination submodule configured to determine a target camera image corresponding to the target ground point from the candidate camera image set; a fifth determination submodule configured to determine a candidate projection pixel corresponding to the target camera image as a target projection pixel corresponding to the target ground point; the fourth determination submodule includes: The fourth determining unit is configured to determine a first neighboring ground point corresponding to the target ground point, the first neighboring ground point including a ground point having a distance from the target ground point within a first predetermined distance range, and a target camera image corresponding to the target ground point and the first neighboring ground point having an association relationship. The fifth determining unit is configured to determine, according to the target camera image corresponding to the first neighboring ground point, the target camera image corresponding to the target ground point from the candidate camera image set.

12. The apparatus of claim 11, wherein, The first determining module comprises: The first determining sub-module is configured to determine position information of the target ground point. The second determining sub-module is configured to determine a candidate camera pose set from a plurality of camera poses according to the position information of the target ground point. The third determining sub-module is configured to determine, as the candidate camera image, a camera image corresponding to each of at least one candidate camera pose included in the candidate camera pose set, to obtain the candidate camera image set.

13. The apparatus of claim 12, further comprising: The second obtaining module is configured to determine a camera pose of each of a plurality of camera images according to a positioning track and a pose conversion relationship, to obtain the plurality of camera poses, wherein the positioning track includes positioning information of each of a plurality of track points.

14. The apparatus of claim 12, wherein, The first determining sub-module comprises: The first determining unit is configured to determine two-dimensional coordinate information of the target ground point according to a two-dimensional ground grid map. The second determining unit is configured to determine height information of the target ground point from a target ground surface by using a three-dimensional ground grid model. The first obtaining unit is configured to obtain position information of the target ground point according to the two-dimensional coordinate information and the height information.

15. The apparatus of claim 12, wherein, The second determining sub-module comprises: The third determining unit is configured to determine a target position range according to a predetermined position offset range and the position information. The second obtaining unit is configured to determine, as the candidate camera pose, a camera pose of the plurality of camera poses that matches the target position range, to obtain the candidate camera pose set.

16. An electronic device, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-10.

17. A non-transitory computer readable storage medium having stored thereon computer instructions, wherein, The computer instructions are used to enable the computer to perform the method of any one of claims 1-10.

18. A computer program product comprising a computer program which, when executed by a processor, implements the method of any one of claims 1-10.

19. An autonomous vehicle comprising the electronic device of claim 16.

Images