WO2022237821A1

WO2022237821A1 - Method and device for generating traffic sign line map, and storage medium

Info

Publication number: WO2022237821A1
Application number: PCT/CN2022/092141
Authority: WO
Inventors: 何叶; 曾清喻; 李友浩; 王玉斌
Original assignee: 阿波罗智能技术(北京)有限公司
Priority date: 2021-05-14
Filing date: 2022-05-11
Publication date: 2022-11-17
Also published as: CN113298910A

Abstract

The present disclosure relates to the technical field of computers, and particularly relates to the fields of autonomous driving, autonomous parking, intelligent transportation, computer vision, and electronic maps. The present disclosure provides a method and device for generating a traffic sign line map, and a storage medium. A specific implementation solution comprises: on the basis of a grid map corresponding to a preset geographical region, according to a plurality of frames of point cloud data of a traffic sign line in the preset geographical region, mapping each frame of point cloud data into the grid map, and increasing a state value of each grid unit having the point cloud data, wherein the more the number of times of mapping the point cloud data into the grid unit is, the larger the state value of the grid unit is, and the larger the possibility that the traffic sign line passes through the grid unit is; and according to the grid unit having the state value greater than a preset threshold, generating a traffic sign line map of the preset geographical region. The present disclosure can reduce the amount of calculation of generating the traffic sign line map, and improve the generation efficiency, accuracy, and consistency of the map.

Description

Method, device and storage medium for generating traffic sign line map

This disclosure claims the priority of the Chinese patent application with the application number 202110528222.0 and the application name "Method, device and storage medium for generating traffic sign line maps" submitted to the China Patent Office on May 14, 2021, the entire contents of which are incorporated by reference incorporated in this disclosure.

technical field

The present disclosure relates to the fields of automatic driving, autonomous parking, intelligent transportation, computer vision, electronic maps and the like in computer technology, and in particular relates to a method, device and storage medium for generating a map of traffic sign lines.

Background technique

With the development of autonomous driving technology, sensors such as cameras and radars surrounding the body of autonomous driving vehicles are basically standard configurations. In order to maximize the use of the surrounding environment information obtained by these sensors, it is an effective means to restore and reconstruct the traffic sign lines (such as lane lines, ground arrows, sidewalks, etc.) in the environmental road to generate a traffic sign line map.

In the related art, most of the schemes for generating traffic sign line maps are based on the acquired vehicle position information and the point set data of traffic sign lines observed multiple times, and segmentally output the traffic sign line by curve clustering and fitting methods. Fragments, and splicing the fragments of traffic sign lines to get a complete map of traffic sign lines, the accuracy and consistency of traffic sign line maps are low.

Contents of the invention

The present disclosure provides a method, device and storage medium for generating a traffic sign line map.

According to a first aspect of the present disclosure, a method for generating a traffic sign line map is provided, including:

Obtain multi-frame point cloud data of traffic sign lines in a preset geographic area;

Mapping the point cloud data of each frame into a grid map corresponding to the preset geographical area, the grid map including a plurality of grid units;

Increase the state value of the grid unit corresponding to the point cloud data in each frame;

A traffic sign line map of the preset geographical area is generated according to the grid cells whose state values are greater than a preset threshold.

According to a second aspect of the present disclosure, a device for generating a traffic sign line map is provided, including:

The point cloud data acquisition module is used to acquire multi-frame point cloud data of traffic sign lines in the preset geographic area;

A data point mapping module, configured to map the point cloud data of each frame into a grid map corresponding to the preset geographic area, the grid map including a plurality of grid units;

The grid map update module is used to increase the state value of the grid unit corresponding to the point cloud data in each frame;

The traffic sign line map generation module is configured to generate the traffic sign line map of the preset geographical area according to the grid cells whose state values are greater than a preset threshold.

According to a third aspect of the present disclosure, an electronic device is provided, including:

at least one processor; and

a memory communicatively coupled 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, so that the at least one processor can execute the method described in the first aspect.

According to a fourth 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 cause the computer to execute the method described in the first aspect.

According to a fifth aspect of the present disclosure, there is provided a computer program product, the computer program product comprising: a computer program stored in a readable storage medium, at least one processor of an electronic device can read from the The computer program is read by reading the storage medium, and the at least one processor executes the computer program so that the electronic device executes the method described in the first aspect.

The method for generating the traffic sign line map provided by the present disclosure improves the accuracy of the traffic sign line map.

It should be understood that what is described in this section is not intended to identify key or important features of the embodiments of the present disclosure, nor is it intended to limit the scope of the present disclosure. Other features of the present disclosure will be readily understood through the following description.

Description of drawings

The accompanying drawings are used to better understand the present solution, and do not constitute a limitation to the present disclosure. in:

FIG. 1 is an example scene diagram that can implement a method for generating a traffic sign line map according to an embodiment of the present disclosure;

FIG. 2 is a flowchart of a method for generating a traffic sign line map provided by the first embodiment of the present disclosure;

Fig. 3 is a flow chart of a method for generating a traffic sign line map provided by the second embodiment of the present disclosure;

Fig. 4 is an example diagram of a traffic sign line map provided by a third embodiment of the present disclosure;

Fig. 5 is a schematic diagram of a device for generating a traffic sign line map provided by a third embodiment of the present disclosure;

FIG. 6 is a schematic diagram of a device for generating a traffic sign line map provided by a fourth embodiment of the present disclosure;

Fig. 7 is a block diagram of an electronic device used to implement the method for generating a traffic sign line map provided by an embodiment of the present disclosure.

Detailed ways

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and they should be regarded as exemplary only. Accordingly, those of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope and spirit of the disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

First, the nouns involved in this disclosure are explained:

Odds: In statistics, the ratio of the probability of an event occurring to the probability of the event not occurring. If the probability of an event occurring is p, then the probability of the event occurring is p/(1-p).

The present disclosure provides a method, device and storage medium for generating a traffic sign line map. It is applied to the fields of automatic driving, autonomous parking, intelligent transportation, computer vision, electronic map and other fields in computer technology to improve the accuracy and consistency of traffic sign map.

The method for generating a map of traffic sign lines provided in the present disclosure can be specifically applied to any of the following scenarios of constructing a map of traffic sign lines in a certain geographical area:

One application scenario is: constructing a traffic sign map of roads in a certain city. The traffic sign map at least includes position information of traffic sign lines such as lane lines, sidewalks, and guiding arrows on the road. During the driving process of the vehicle, the position of the traffic marking line is determined according to the map of the traffic marking line, and the driving of the vehicle is controlled according to the position of the traffic marking line.

Another application scenario is: building a map of traffic sign lines on the ground in the parking lot. The traffic sign line map includes at least the location information of traffic sign lines such as parking space lines, lane lines, and guiding arrows in the parking lot. When applied to autonomous parking, in response to an autonomous parking instruction, the position of the parking space line is determined according to the traffic sign line map, and the vehicle is controlled to perform autonomous parking according to the position of the parking space line.

In addition, the method for generating the traffic sign line map can also be applied to other scenarios, which is not specifically limited here.

In the related art, most of the schemes for generating traffic sign line maps are based on the acquired vehicle position information and the point set data of traffic sign lines observed multiple times, and segmentally output the traffic sign line by curve clustering and fitting methods. Fragments, and splicing the fragments of traffic sign lines to get a complete map of traffic sign lines. Since the actual road contains various traffic sign lines corresponding to different curve types, only certain types or a few types of curve parameter models can not accurately fit traffic sign lines of other curve types, and there are cases of misidentification, resulting in traffic sign The error of the line map is large, and the splicing process will lead to poor consistency of the traffic sign line map (such as poor straightness of lane lines, poor smoothness of curved parts, etc.).

FIG. 1 is an example scene diagram that can implement the method for generating a traffic sign line map according to an embodiment of the present disclosure. Exemplarily, as shown in FIG. The image acquisition device 11 mounted on the vehicle 10 collects image data within a preset geographical area, and the positioning device 12 mounted on the vehicle 10 collects pose data of the vehicle 10 . In the data processing stage, the electronic device 20 obtains the traffic sign line point cloud data corresponding to each frame of image according to the collected image data and the pose data of the vehicle, and obtains multi-frame point cloud data of the traffic sign line; The multi-frame point cloud data generates a traffic sign line map of a preset geographic area.

Wherein, the image acquisition device may be a device on the vehicle for collecting images, such as a camera, a laser radar, etc., and the positioning device may be a device on the vehicle for locating the pose of the vehicle, which is not specifically limited here.

Fig. 2 is a flowchart of a method for generating a traffic sign line map provided by the first embodiment of the present disclosure. As shown in Figure 2, the specific steps of the method are as follows:

Step S201, acquiring multi-frame point cloud data of traffic sign lines in a preset geographical area.

In order to construct the traffic sign line map in the preset geographical area, it is necessary to obtain the point cloud data of the traffic sign line in the preset geographical area, and generate the traffic sign line map based on the point cloud data of the traffic sign line.

Wherein, each frame of point cloud data of the traffic sign line includes position information of multiple data points of the traffic sign line.

Exemplarily, based on the scene shown in FIG. 1 , in the data collection phase, when the vehicle is driving in a preset geographic area, the image acquisition device mounted on the vehicle collects image data in the preset geographic area, and the image data loaded on the vehicle The positioning device collects the pose data of the vehicle. The electronic device obtains the image data in the preset geographic area and the pose data of the vehicle, and determines the point cloud data of the traffic sign line corresponding to each frame of image according to the image data in the preset geographic area and the pose data of the vehicle.

Exemplarily, the electronic device may also acquire point cloud data in a preset geographic area scanned by the lidar, and extract point cloud data of traffic sign lines from the point cloud data in the preset geographic area.

Exemplarily, other devices may also pre-generate and store multi-frame point cloud data of traffic sign lines in a preset geographical area. The electronic device can also directly acquire multi-frame point cloud data of traffic sign lines in a preset geographical area from other devices.

Step S202 , mapping each frame of point cloud data to a grid map corresponding to a preset geographical area, the grid map including a plurality of grid units; increasing the state value of the grid unit corresponding to each frame of point cloud data.

Wherein, the grid map includes a plurality of grid units, and the state value of each grid unit is used to represent the probability information of the traffic sign line passing through the grid unit.

After obtaining the multi-frame point cloud data of traffic sign lines in the preset geographical area, traverse each frame of point cloud data in turn, update the state value of each grid cell in the grid map according to each frame of point cloud data, and traverse the point cloud After the data, the raster map is updated. For example, if there are N frames of point cloud data, update the grid map N times, where N is a positive integer.

For each frame of point cloud data, map the point cloud data to the grid map, that is, map each data point contained in the point cloud data to the grid map, and determine the grid cell where each data point falls .

For any grid unit, if there is a data point in the frame of point cloud data falling into the grid unit, it means that in the image corresponding to the frame of point cloud data, there is a pixel point of a traffic sign line at the corresponding position of the grid unit, It can be considered that based on the frame image, it is more likely that the traffic sign line passes through the grid unit, so the state value of the grid unit is increased.

In this step, the grid unit with point cloud data refers to the grid unit in which at least one data point falls in the frame of point cloud data. A grid cell without point cloud data refers to a grid cell in which no data point falls in the point cloud data of the frame. For a grid cell with point cloud data, increase the state value of the grid cell, and for a grid cell without point cloud data, keep the state value of the grid cell unchanged.

When traversing each frame of point cloud data and updating the grid map according to each frame of point cloud data, the state value of the grid unit with point cloud data will increase, and the state value of the grid unit without point cloud data will remain unchanged. After all the point cloud data, the more times the point cloud data is mapped to the grid unit, the greater the state value of the grid unit, which means the greater the possibility of the traffic sign line passing through the grid unit.

Step S203 , generating a traffic sign line map of a preset geographical area according to the grid cells whose state values are greater than a preset threshold.

In this step, grid cells whose status values are greater than the preset threshold are extracted from the grid map according to the preset threshold, and a traffic sign line map of the preset geographic area is generated according to the grid cells whose status values are greater than the preset threshold.

Wherein, the preset threshold can be set and adjusted according to the needs of actual application scenarios, and is not specifically limited here.

In this embodiment, based on the grid map corresponding to the preset geographic area and the multi-frame point cloud data of traffic sign lines in the preset geographic area, each frame of point cloud data is mapped to the grid map by traversing each frame of point cloud data In , the state value of the grid cells with point cloud data is increased, while the state value of the grid cells without point cloud data remains unchanged. After traversing all the point cloud data, the more times the point cloud data is mapped to the grid unit, the greater the state value of the grid unit, and the greater the possibility of the traffic sign passing through the grid unit. According to the grid cells whose state value is greater than the preset threshold, the traffic sign line map of the preset geographical area is generated, and the traffic sign line map can be generated through a series of addition operations, which greatly reduces the amount of calculation and improves the map generation efficiency. At the same time, the overall map is generated at one time, avoiding the misidentification and stitching process caused by curve fitting and stitching, adapting to various map construction scenarios, and improving the accuracy and consistency of the map.

Fig. 3 is a flow chart of a method for generating a traffic sign line map provided by the second embodiment of the present disclosure. On the basis of the first embodiment above, in this embodiment, the state value of the grid unit can be the probability information of the traffic sign line passing through the grid unit, and the initial value of the state value of the grid unit is 0.

As shown in Figure 3, the specific steps of the method are as follows:

Step S301 , acquiring images within a preset geographical area collected by the vehicle, and obtaining pose data when the vehicle collects the images.

In this embodiment, based on the scene shown in FIG. 1 , during the data collection phase, the image acquisition device mounted on the vehicle collects image data in the preset geographic area while the vehicle is driving in the preset geographic area. Wherein, the image data may include multiple frames of images.

The loaded positioning device on the vehicle collects the pose data of the vehicle. Wherein, the pose data includes pose data of the vehicle collected at multiple times. According to the time stamp of image collection and the time stamp of the pose data of the vehicle, the pose data of the vehicle when each frame of image is collected can be determined.

The electronic device acquires multiple frames of images in the preset geographical area collected by the vehicle, and pose data when the vehicle collects the images.

Wherein, there are traffic sign lines in the preset geographical area, and the collected multiple frames of images can cover the traffic sign lines in the preset geographical area.

In this embodiment, the preset geographic area is the geographic range of the traffic sign line map that needs to be acquired according to the requirements of the business scenario. For example, if a city map is established, the preset geographical area may be an area where a city is located; if a parking lot map needs to be established, the preset geographical area may be an area where a parking lot is located. The preset geographical area can be set and adjusted according to the needs of the actual application scenario, which is not specifically limited in this embodiment.

Step S302. Determine the coordinates of the mapping points of the pixels of the traffic sign lines in the image in the vehicle body coordinate system.

In this step, each frame of image can be processed separately, and according to each frame of image, the three-dimensional coordinates of the mapping point of the pixel point of the traffic sign line in the image in the vehicle body coordinate system can be determined.

In this embodiment, this step can be implemented in the following manner:

Determine the pixel points of the traffic sign lines in the image; map the pixel points to the mapping points in the vehicle body coordinate system according to the camera model corresponding to the image, and obtain the coordinates of the mapping points.

Exemplarily, according to each frame of image, determine the pixel coordinates of the pixel points of the traffic sign lines in the image in the image coordinate system; according to the camera model corresponding to the image, map the pixel points to the mapping points in the vehicle body coordinate system, and obtain The 3D coordinates of the mapped point.

Specifically, by performing image processing on each frame of image, the traffic sign line information in the image is extracted, the pixel points of the traffic sign line in the image are determined, and the pixel coordinates of the pixel points of the traffic sign line are further determined. According to the camera model, the pixels are back-projected into the physical world, so that the pixels are mapped to the mapping points in the vehicle body coordinate system, and the three-dimensional coordinates of the mapping points of the pixels of the traffic sign lines in the vehicle body coordinate system are obtained. The advantage of this setting is that by transforming the traffic sign line from the image coordinate system to the vehicle body coordinate system, the processing of the traffic sign line can be realized, thereby improving the efficiency of updating the traffic sign line map.

In addition, in this step, any method in the prior art that converts the two-dimensional coordinates of the pixel points in the image into three-dimensional coordinates in the vehicle body coordinate system can also be used, and this embodiment will not repeat them here.

Step S303, according to the pose data when the vehicle collects images, project the mapping points to the world coordinate system to obtain the coordinates of the corresponding projection points; generate a frame of point cloud data according to the three-dimensional coordinates of the projection points determined for each frame of image.

In this step, project the mapping points to the world coordinate system according to the pose data when the vehicle collects images, and obtain the corresponding projection points and the three-dimensional coordinates of the projection points, and form a frame according to the three-dimensional coordinates of the projection points determined for each frame of image point cloud data.

After obtaining the three-dimensional coordinates of the mapping points of the pixels of the traffic sign lines in the image in the vehicle body coordinate system, further, according to the pose of the vehicle when the image is collected, the pixels of the traffic sign lines in the image are placed in the vehicle body coordinate system The mapping points below are projected to the world coordinate system, and the projection points of the pixels of the traffic sign lines in the image in the world coordinate system and the three-dimensional coordinates of the projection points are obtained.

In this embodiment, the three-dimensional coordinates of the projected points determined according to each frame of image constitute a frame of point cloud data.

In addition, in this step, any method in the prior art for converting the three-dimensional coordinates in the vehicle body coordinate system into the three-dimensional coordinates in the world coordinate system can also be used, and this embodiment will not repeat it here.

The above-mentioned steps S301-S303 are an optional implementation mode for obtaining multiple frames of point cloud data corresponding to traffic sign lines provided by this embodiment, by identifying the position and posture of the vehicle in each frame of image and the image when the image is collected, The pixel points of the traffic sign line, and transform the coordinates of the pixel point to the world coordinate system, and obtain the three-dimensional coordinates of multiple data points corresponding to the pixel points of the traffic sign line mapped to the world coordinate system, as a part of the traffic sign line Frame point cloud data can obtain accurate location cloud data of traffic marking lines, thus providing a data basis for generating accurate traffic marking line maps.

In another optional implementation of this embodiment, the electronic device may also acquire point cloud data in a preset geographic area scanned by the lidar, and extract traffic sign lines from the point cloud data in the preset geographic area The point cloud data of the traffic sign line can be obtained from various point cloud data.

In another optional implementation manner of this embodiment, other devices may also pre-generate and store multi-frame point cloud data of traffic sign lines in a preset geographic area. The electronic device can also directly acquire multi-frame point cloud data of traffic sign lines in a preset geographical area from other devices.

Step S304, constructing a grid map of a preset geographical area, and initializing the state values of each grid unit in the grid map.

Among them, the state value of the grid unit can be the probability information of the traffic sign line passing through the grid unit, and the initial value of the state value of the grid unit is 0. The advantage of this setting is that the state value can reflect that the grid unit has a lane The probability of the line passing through.

In this embodiment, m represents the grid map, and m _i represents the ith grid unit in the grid map. m _i has the following two states: the traffic sign line passes through the grid unit m _i (that is, the data points of the traffic sign line fall into m _i , which can be represented by m _i = 1), and the traffic sign line does not pass through the grid unit m _i (that is, none of the data points of the traffic marking line falls into m _i , which can be represented by m _i =0).

Use p(m _i =1) or p(m _i ) to indicate the probability that the traffic sign line passes through the grid unit _mi , and use p(m _i =0) or p(~m _i ) to indicate that the traffic sign line does not pass through the grid _Probability of unit mi.

Use x _t to represent the pose of the vehicle at time t, use z to represent the point cloud data corresponding to a frame of image collected by the vehicle at time t, use z ^k _t to represent the kth data point in z _t , and use n to represent the data point in z _t Total number of data points, k=1, 2, . . . , n.

Then, updating the grid map can be modeled as the problem of solving the probability that all grid cells of the grid map have traffic sign lines passing by (which can be expressed as p(m|x _0:t ,z _0:t )). Assuming that each grid cell is independent of each other, there is the following formula (1):

Among them, p(m _i |x _0:t , z _0:t ) represents the probability value of the traffic sign line passing through the grid unit m _i determined based on the point cloud data and pose data at time 0-t.

Based on formula (1), using Bayesian rule and Markov assumption, the following formulas (2) and (3) can be obtained:

Among them, p(~m _i |x _0:t ,z _0:t ) represents the probability value that the traffic sign line determined based on the point cloud data and pose data at time 0-t does not pass through the grid cell _mi .

Considering the stability of the probability value data calculation, in order to improve the possibility that the state value of the grid cell in the grid map can accurately and stably represent the possibility of the traffic sign line passing the grid cell, the logarithm of both sides can be used to calculate the traffic sign The probability of the line passing through the grid unit (expressed by p) is transformed as follows to obtain the probability information of the traffic sign line passing through the grid unit:

l represents the probability information of the traffic sign line passing through the grid cell.

The above (2) is divided by (3), and the logarithm is taken on both sides at the same time, then the following formula (4) is obtained:

Among them, p(z _t |m _i ,x _t ) is the likelihood probability of the observation model, p(z _t |～m _i ,x _t )=1-p(z _t |m _i ,x _t ). If the pose x _t of the vehicle at time t and whether there is a traffic sign passing in the grid unit m _i can be deduced from the observed value (collected point cloud data) at time t.

Applying Bayes' rule, there are the following formulas (5) and (6):

Substituting formulas (5) and (6) into formula (4), the following formula (7) can be obtained:

l(m _i |x _0:t ,z _0:t )＝l(m _i |x _0:t-1 ,z _0:t-1 )+l(m _i |z _t ,x _t )-l( m _i x _t ) (7)

Wherein, l(m _i |x _t )=l(m _i ), represents the prior initial probability value of the grid unit m _i . l(m _i |z _t , x _t ) represents the probability value of the inverse observation model, which is a fixed value.

l(m _i |x _0:t , z _0:t ) represents the probability information of the traffic sign line passing through the grid unit m _i determined according to the multi-frame point cloud data and the pose data of the vehicle at time 0-t. l(m _i |x _0:t-1 , z _0:t-1 ) means that the traffic sign line determined according to the multi-frame point cloud data and the pose data of the vehicle at time 0-t-1 passes through the grid unit m _i probability information.

In this embodiment, the state value of the grid unit is used to represent the probability information of the traffic sign line passing through the grid unit, and the initial state value of the grid unit is 0 (that is, the initial value of the probability information of the traffic sign line passing through the grid unit is 0, the probability that the traffic sign line passes through the grid cell is 0.5).

Based on the above formula (7), taking l(m _i |z _t , x _t ) as the incremental value, for the observation data at time t (a frame of point cloud data at time t), if there is a data point of the traffic sign line falling into the grid unit m _i , the state value of the grid unit _mi updated based on the point cloud data at time 0-t-1 can be increased by an incremental value, and the grid unit m after this update can be obtained The state value of _i .

In this way, the initial state values of all grid cells in the grid map are set to 0, indicating that the probability of the traffic sign line passing the grid unit is 0.5, and then according to the multi-frame point cloud data of the traffic sign line, each frame point cloud The data is mapped into a raster map. With the increase of the point cloud data frame mapped to the grid unit, the state value of the grid unit increases. After all the point cloud data mapping is completed, it is possible to accurately determine the possibility that each grid unit has a traffic sign line passing by. Therefore, the traffic sign line map can be accurately generated.

It should be noted that, in another implementation of this embodiment, before acquiring the point cloud data of traffic sign lines, a grid map of a preset geographical area may be constructed in advance, and the values of each grid unit in the grid map may be initialized. status value. That is, step S304 may be performed before steps S301-S303.

After obtaining the multi-frame point cloud data of traffic sign lines, and constructing and initializing the grid map, through steps S305-S306, traverse each frame of point cloud data, and map each frame of point cloud data to the corresponding preset geographical area In the grid map, the state value of the grid unit corresponding to each frame of point cloud data is added, so that the possibility of traffic sign lines passing by each grid unit can be accurately determined, so that the traffic sign line map can be accurately generated.

Step S305 , each frame of point cloud data includes position information of multiple data points of the traffic sign line, and according to the position information of each data point in each frame of point cloud data, determine the grid unit corresponding to each data point.

In this step, according to the position information of each data point in each frame of point cloud data, the grid unit where each data point falls is determined, and the grid unit where the data point falls is used as the grid unit corresponding to the data point.

In this embodiment, the grid unit with point cloud data refers to the grid unit in which at least one data point falls in the frame of point cloud data. A grid cell without point cloud data refers to a grid cell that does not have any data points in the point cloud data of the frame.

In addition, for all data points and all grid cells, if a data point falls on the boundary of a certain grid cell, it is determined that the data point falls into the grid cell; or, for all data points And all grid cells, if a data point falls on the boundary of a certain grid cell, it is determined that the data point does not fall into any grid cell. For all data points and all grid cells, for the case of falling on the boundary, the same processing can be done uniformly.

Step S306, increasing the state value of the grid unit corresponding to each data point by a preset increment.

In this step, the state value of the grid cell with the corresponding data point is increased by a preset increment.

If one or more data points fall into a certain grid unit in a frame of point cloud data, the state value of the grid unit is increased by a preset increment. For a frame of point cloud data, if there are data points in the frame of point cloud data that fall into the grid unit, no matter whether the number of data points falling into the grid unit is one or more, the state value of the grid unit will be increased A preset increment.

If there is no data point falling into a certain grid unit in a frame of point cloud data, keep the state value of the grid unit unchanged.

Wherein, the preset increment can be a preset probability value of the inverse observation model, which can be set and adjusted according to the needs of actual application scenarios, and is not specifically limited here.

Step S307, generating a traffic sign line map of a preset geographical area according to the grid cells whose state values are greater than a preset threshold.

In this step, grid cells whose status values are greater than the preset threshold are extracted from the grid map according to the preset threshold, and a traffic sign line map of the preset geographic area is generated according to the grid cells whose status values are greater than the preset threshold. The advantage of this setting is that the traffic sign line map of the preset geographical area can be generated more accurately, and the error rate of the traffic sign line map can be reduced.

In an optional implementation manner, this step can be implemented in the following manner:

According to the state value of each grid cell in the grid map, the grid cells whose state value is greater than the preset threshold are extracted, and the area covered by the grid cells whose state value is greater than the preset threshold is taken as the area covered by the traffic sign line, and we get A map of traffic sign lines for preset geographic areas.

In another optional implementation manner, this step can be implemented in the following manner:

According to the status value of each grid unit in the grid map, extract the grid unit whose status value is greater than the preset threshold; use the data point corresponding to the grid unit whose status value is greater than the preset threshold as the trajectory point of the traffic sign line, according to the status The location information of the data points corresponding to the grid cells whose value is greater than the preset threshold value is used to generate the traffic sign line map of the preset geographic area.

In this embodiment, by updating the probability information of the grid cells through a series of addition operations, the constructed traffic sign line map data can be obtained, and the probability information can be discretized and expressed as an 8-bit (bit) or 16-bit integer (int type), so it is also very convenient for computer operations. Compared with curve fitting, it is easier to implement, because curve fitting needs to use nonlinear optimization to iteratively solve the operation. It involves the derivation of the cost function and the selection of parameters of the curve model. The constructed traffic sign line map can effectively resist the influence of data noise, and maintain the consistency of the map (such as the straightness of lane lines, the smoothness of curved parts, etc.). Exemplarily, an example of a traffic sign line map generated by the method provided in this embodiment is shown in FIG. 4 , and the straightness of the lane line and the smoothness of the curved part are good.

In this embodiment, based on the grid map corresponding to the preset geographical area, the initial state values of all grid cells in the grid map are set to 0, which means that the probability of traffic sign lines passing through the grid cells is 0.5. Further, according to the multi-frame point cloud data of traffic sign lines in the preset geographical area, by traversing each frame of point cloud data, each frame of point cloud data is mapped to a grid map, and the grid unit with point cloud data is increased The state value of the raster cells without point cloud data remains unchanged. After traversing all the point cloud data, the more times the point cloud data is mapped to the grid unit, the greater the state value of the grid unit, and the greater the possibility that the traffic sign line passes through the grid unit. According to the state value greater than the preset Threshold grid cells to generate a traffic sign line map of a preset geographic area. The traffic sign line map can be generated through a series of addition operations, which greatly reduces the amount of calculation and improves the efficiency of map generation. At the same time, the overall map is generated at one time, avoiding the misidentification and stitching process caused by curve fitting and stitching, adapting to various map construction scenarios, and improving the accuracy and consistency of the map.

Fig. 5 is a schematic diagram of a device for generating a traffic sign line map provided by a third embodiment of the present disclosure. The device for generating a traffic sign line map provided in an embodiment of the present disclosure may execute the processing procedure provided in the method embodiment for generating a traffic sign line map. As shown in FIG. 5 , the device 50 for generating a traffic sign line map includes: a point cloud data acquisition module 501 , a data point mapping module 502 , a grid map update module 503 and a traffic sign line map generation module 504 .

Specifically, the point cloud data acquisition module 501 is configured to acquire multi-frame point cloud data of traffic sign lines in a preset geographical area.

The data point mapping module 502 is configured to map each frame of point cloud data into a grid map corresponding to a preset geographical area, and the grid map includes a plurality of grid units.

The grid map update module 503 is used to increase the state value of the grid unit corresponding to each frame of point cloud data.

The traffic sign line map generating module 504 is configured to generate a traffic sign line map of a preset geographical area according to the grid cells whose state values are greater than a preset threshold.

The device provided in the embodiments of the present disclosure may be specifically configured to execute the method embodiment provided in the above-mentioned first embodiment, and the specific functions will not be repeated here.

Fig. 6 is a schematic diagram of a device for generating a traffic sign line map provided by a fourth embodiment of the present disclosure. The device for generating a traffic sign line map provided in an embodiment of the present disclosure may execute the processing procedure provided in the method embodiment for generating a traffic sign line map. As shown in FIG. 6 , the device 60 for generating a traffic sign line map includes: a point cloud data acquisition module 601 , a data point mapping module 502 , a grid map update module 603 and a traffic sign line map generation module 604 .

Specifically, the point cloud data acquisition module 601 is configured to acquire multi-frame point cloud data of traffic sign lines in a preset geographic area.

The data point mapping module 602 is configured to map each frame of point cloud data into a grid map corresponding to a preset geographical area, and the grid map includes a plurality of grid units.

The grid map updating module 603 is used to increase the state value of the grid unit corresponding to each frame of point cloud data.

The traffic sign line map generation module 604 is configured to generate a traffic sign line map of a preset geographical area according to the grid cells whose state values are greater than a preset threshold.

Optionally, the status value of the grid unit is the probability information of the traffic sign line passing through the grid unit, and the initial value of the status value of the grid unit is 0.

Optionally, each frame of point cloud data includes position information of multiple data points of the traffic sign line. The data point mapping module 602 in FIG. 6 is also used to determine the grid unit corresponding to each data point according to the position information of each data point in each frame of point cloud data.

The grid map update module 603 is further configured to increase the state value of the grid unit corresponding to each data point by a preset increment.

Optionally, as shown in Figure 6, the point cloud data acquisition module 601 includes:

The data acquisition sub-module 6011 is used to acquire the images in the preset geographical area collected by the vehicle, and obtain the pose data when the vehicle collects the images;

The first mapping sub-module 6012 is used to determine the mapping points of the pixels of the traffic sign lines in the image in the vehicle body coordinate system;

The second mapping sub-module 6013 is used to project the mapping points to the world coordinate system according to the pose data, obtain the coordinates of the corresponding projection points, and generate a frame of point cloud data according to the coordinates of the projection points.

Optionally, the first mapping submodule is also used for:

Optionally, as shown in Figure 6, the traffic sign line map generating module 604 includes:

The first generation sub-module 6041 is configured to use the area covered by the grid cells whose state value is greater than the preset threshold as the area covered by the traffic sign line to obtain the traffic sign line map of the preset geographical area.

The second generation sub-module 6042 is used for:

The data points corresponding to the grid cells whose state value is greater than the preset threshold are used as the trajectory points of the traffic sign line, and the traffic signs of the preset geographical area are generated according to the position information of the data points corresponding to the grid cells whose state value is greater than the preset threshold line map.

Optionally, as shown in Figure 6, the device 60 for generating the traffic sign line map also includes:

A raster map building block 605 for:

Construct the grid map corresponding to the preset geographical area, and initialize the state value of each grid unit.

The device provided in the embodiment of the present disclosure may be specifically used to execute the method embodiment provided in the above-mentioned second embodiment, and specific functions will not be repeated here.

According to the embodiments of the present disclosure, the present disclosure also provides an electronic device and a readable storage medium.

According to an embodiment of the present disclosure, the present disclosure also provides a computer program product. The computer program product includes: a computer program, the computer program is stored in a readable storage medium, and at least one processor of an electronic device can read the program from the readable storage medium. Taking a computer program, at least one processor executes the computer program so that the electronic device executes the solution provided by any one of the above embodiments.

FIG. 7 shows a schematic block diagram of an example electronic device 700 that may be used to implement embodiments of the present disclosure. Electronic device is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Electronic devices may also represent various forms of mobile devices, such as personal digital processing, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are by way of example only, and are not intended to limit implementations of the disclosure described and/or claimed herein.

As shown in FIG. 7 , an electronic device 700 includes a computing unit 701, which can perform calculations according to a computer program stored in a read-only memory (ROM) 702 or a computer program loaded from a storage unit 708 into a random access memory (RAM) 703. Various appropriate actions and processes are performed. In the RAM 703, various programs and data necessary for the operation of the device 700 can also be stored. The computing unit 701, ROM 702, and RAM 703 are connected to each other through a bus 704. An input/output (I/O) interface 705 is also connected to the bus 704 .

Multiple components in the device 700 are connected to the I/O interface 705, including: an input unit 706, such as a keyboard, a mouse, etc.; an output unit 707, such as various types of displays, speakers, etc.; a storage unit 708, such as a magnetic disk, an optical disk, etc. ; and a communication unit 709, such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 709 allows the device 700 to exchange information/data with other devices over a computer network such as the Internet and/or various telecommunication networks.

The computing unit 701 may be various general-purpose and/or special-purpose processing components having processing and computing capabilities. Some examples of computing units 701 include, but are not limited to, central processing units (CPUs), graphics processing units (GPUs), various dedicated artificial intelligence (AI) computing chips, various computing units that run machine learning model algorithms, digital signal processing processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 701 executes various methods and processes described above, such as a method of generating a traffic sign line map. For example, in some embodiments, a method of generating a traffic sign map may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as storage unit 708. In some embodiments, part or all of the computer program may be loaded and/or installed on the device 700 via the ROM 702 and/or the communication unit 709. When the computer program is loaded into the RAM 703 and executed by the computing unit 701, one or more steps of the method for generating a traffic sign line map described above can be performed. Alternatively, in other embodiments, the computing unit 701 may be configured in any other appropriate way (for example, by means of firmware) to execute the method for generating a traffic sign line map.

Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips Implemented in a system of systems (SOC), load programmable logic device (CPLD), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include being implemented in one or more computer programs executable and/or interpreted on a programmable system including at least one programmable processor, the programmable processor Can be special-purpose or general-purpose programmable processor, can receive data and instruction from storage system, at least one input device, and at least one output device, and transmit data and instruction to this storage system, this at least one input device, and this at least one output device an output device.

Program codes for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes may be provided to a processor or controller of a general-purpose computer, a special purpose computer, or other programmable data processing devices, so that the program codes, when executed by the processor or controller, make the functions/functions specified in the flow diagrams and/or block diagrams Action is implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media would include one or more wire-based electrical connections, portable computer discs, hard drives, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, compact disk read only memory (CD-ROM), optical storage, magnetic storage, or any suitable combination of the foregoing.

To provide for interaction with the user, the systems and techniques described herein 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 pointing device (eg, a mouse or a trackball) through which a user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and can be in any form (including Acoustic input, speech input or, tactile input) to receive input from the user.

The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., as a a user computer having a graphical user interface or web browser through which a user can interact with embodiments of the systems and techniques described herein), or including such backend components, middleware components, Or any combination of front-end components in a computing system. The components of the system can be interconnected by any form or medium of digital data communication, eg, a communication network. Examples of communication networks include: Local Area Network (LAN), Wide Area Network (WAN) and the Internet.

A computer system may include clients and servers. Clients and servers are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by computer programs running on the respective computers and having a client-server relationship to each other. The server can be a cloud server, also known as cloud computing server or cloud host, which is a host product in the cloud computing service system to solve the problem of traditional physical host and VPS service ("Virtual Private Server", or "VPS") Among them, there are defects such as difficult management and weak business scalability. The server can also be a server of a distributed system, or a server combined with a blockchain.

It should be understood that steps may be reordered, added or deleted using the various forms of flow shown above. For example, each step described in the present application may be executed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in the present disclosure can be achieved, no limitation is imposed herein.

The specific implementation manners described above do not limit the protection scope of the present disclosure. It should be apparent to those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made depending on design requirements and other factors. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present disclosure shall be included within the protection scope of the present disclosure.

Claims

A method for generating a map of traffic sign lines, comprising:

Obtain multi-frame point cloud data of traffic sign lines in a preset geographic area;

Mapping the point cloud data of each frame into a grid map corresponding to the preset geographical area, the grid map including a plurality of grid units;

Increase the state value of the grid unit corresponding to the point cloud data in each frame;

A traffic sign line map of the preset geographical area is generated according to the grid cells whose state values are greater than a preset threshold.
The method according to claim 1, wherein the state value of the grid cell is the probability information of the traffic sign line passing the grid cell, and the initial value of the state value of the grid cell is 0.
The method according to claim 1 or 2, wherein the point cloud data of each frame includes position information of a plurality of data points of traffic marking lines;

The mapping of each frame of point cloud data to the grid map corresponding to the preset geographical area includes: determining the position information of each data point in the point cloud data of each frame. The grid cell corresponding to the point;

The increasing the state value of the grid unit corresponding to the point cloud data in each frame includes: increasing the state value of the grid unit corresponding to each data point by a preset increment.
The method according to any one of claims 1-3, wherein the point cloud data of each frame of the traffic marking line in the preset geographic area is acquired by the following method:

Acquiring the images in the preset geographical area collected by the vehicle, and obtaining the pose data when the vehicle collects the images;

Determining the mapping points of the pixels of the traffic sign lines in the image in the vehicle body coordinate system;

According to the pose data, project the mapping point to the world coordinate system to obtain the coordinates of the corresponding projection point;

A frame of point cloud data is generated according to the coordinates of the projected points.
The method according to claim 4, wherein said determining the mapping points of the pixels of the traffic sign lines in the image in the vehicle body coordinate system comprises:

Determine the pixel points of the traffic sign lines in the image;

According to the camera model corresponding to the image, the pixel points are mapped to mapping points in the vehicle body coordinate system to obtain the coordinates of the mapping points.
The method according to any one of claims 1-5, wherein said generating the traffic sign line map of said preset geographic area according to the grid cells whose status values are greater than a preset threshold includes:

Taking the area covered by the grid cells whose state values are greater than the preset threshold value as the area covered by the traffic sign line to obtain the traffic sign line map of the preset geographical area.
The method according to any one of claims 1-5, wherein said generating a traffic sign line map according to a grid unit whose state value is greater than a preset threshold includes:

The data points in the point cloud data corresponding to the grid cells whose state value is greater than the preset threshold are used as the trajectory points of the traffic marking line;

A traffic sign line map of the preset geographical area is generated according to the location information of the track point.
The method according to any one of claims 1-7, further comprising:

A grid map corresponding to the preset geographic area is constructed, and a state value of each grid unit is initialized.
A device for generating a map of traffic sign lines, comprising:

The point cloud data acquisition module is used to acquire multi-frame point cloud data of traffic sign lines in the preset geographic area;

A data point mapping module, configured to map the point cloud data of each frame into a grid map corresponding to the preset geographic area, the grid map including a plurality of grid units;

The grid map update module is used to increase the state value of the grid unit corresponding to the point cloud data in each frame;

The traffic sign line map generation module is configured to generate the traffic sign line map of the preset geographical area according to the grid cells whose state values are greater than a preset threshold.
The device according to claim 9, wherein the state value of the grid cell is the probability information of the traffic sign line passing through the grid cell, and the initial value of the state value of the grid cell is 0.
The device according to claim 9 or 10, wherein the point cloud data of each frame includes position information of a plurality of data points of traffic marking lines;

The data point mapping module is also used to:

According to the position information of each data point in the point cloud data of each frame, determine the grid unit corresponding to each data point;

The raster map update module is also used for:

The state value of the grid cell corresponding to each data point is increased by a preset increment.
The device according to any one of claims 9-11, wherein the point cloud data acquisition module includes:

The data acquisition sub-module is used to acquire images in the preset geographical area collected by the vehicle, and obtain pose data when the vehicle collects the images;

The first mapping submodule is used to determine the mapping points of the pixels of the traffic sign lines in the image in the vehicle body coordinate system;

The second mapping sub-module is used to project the mapping points into the world coordinate system according to the pose data, obtain the coordinates of the corresponding projection points, and generate a frame of point cloud data according to the coordinates of the projection points.
The device according to claim 12, wherein the first mapping submodule is further used for:

Determine the pixel points of the traffic sign lines in the image;

According to the camera model corresponding to the image, the pixel points are mapped to mapping points in the vehicle body coordinate system to obtain the coordinates of the mapping points.
The device according to any one of claims 9-13, wherein the traffic sign line map generation module includes:

The first generation sub-module is used to use the area covered by the grid cells whose state value is greater than the preset threshold as the area covered by the traffic sign line, and obtain the traffic sign line map of the preset geographical area.
The device according to any one of claims 9-13, wherein the traffic sign line map generation module includes:

The second generates submodules for:

The data points in the point cloud data corresponding to the grid cells whose state value is greater than the preset threshold are used as the trajectory points of the traffic marking line;

A traffic sign line map of the preset geographical area is generated according to the location information of the track point.
The apparatus according to any one of claims 9-15, further comprising:

Raster map building blocks for:

A grid map corresponding to the preset geographic area is constructed, and a state value of each grid unit is initialized.
An electronic device comprising:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions are executed by the at least one processor, so that the at least one processor can perform any one of claims 1-8. Methods.
A non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute the method according to any one of claims 1-8.
A computer program product comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1-8.