WO2020215254A1

WO2020215254A1 - Lane line map maintenance method, electronic device and storage medium

Info

Publication number: WO2020215254A1
Application number: PCT/CN2019/084115
Authority: WO
Inventors: 崔健; 许睿; 刘晓洋; 唐蔚博
Original assignee: 深圳市大疆创新科技有限公司
Priority date: 2019-04-24
Filing date: 2019-04-24
Publication date: 2020-10-29
Also published as: CN111316328A

Abstract

Disclosed are a lane line map maintenance method, an electronic device and a storage medium. The method comprises: acquiring environmental data around a vehicle by means of a sensor, and recognizing a lane line region according to the environmental data; generating an overhead lane line map according to the environmental data and the lane line region; and updating a local lane line map within a preset range according to the overhead lane line map, wherein the preset range at least comprises a range of the overhead lane line map. Thus, a visual range of the updated local lane line map within the preset range is large, and when intelligent driving is performed on the basis of the local lane line map with the large visual range, accurate guidance of intelligent driving can be realized, thereby improving the safety of intelligent driving.

Description

Maintenance method, electronic equipment and storage medium of lane line map

Technical field

The embodiments of the present application relate to the field of automatic driving technology, and in particular, to a maintenance method, electronic equipment, and storage medium of a lane line map.

Background technique

With the continuous development of autonomous driving technology, the usual processing of sensor data can no longer meet the needs of autonomous driving. An important aspect of automatic driving is the recognition and detection of lane lines. Usually, the vehicle obtains the image of the environment around the vehicle through the equipped visual sensor, and uses methods such as image recognition to identify and detect the lane line in the image.

However, the lane line obtained by this method usually reflects a single line, and does not reflect the line shape of the actual lane line, such as single solid line, double yellow line, etc.; such lane lines with only lines cannot be provided for subsequent processing More accurate and intelligent decision-making basis. Moreover, since the image obtained by the vision sensor only covers a certain part of the environment around the vehicle, the lane line obtained by the recognition and detection is only the result of the lane line in the local area around the vehicle. The image of the changed area needs to be continuously changed during the driving of the vehicle. The data is processed, which will cause a large amount of calculation and will also consume more memory resources.

Summary of the invention

The embodiments of the present application provide a method for maintaining a lane line map, an electronic device, and a storage medium.

In the first aspect, an embodiment of the present application provides a method for maintaining a lane line map, including:

Acquire environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data;

According to the environmental data and the lane line area, generate an overlooked lane line map;

According to the overlooked lane line map, update the local lane line map of the preset range;

Wherein, the preset range at least includes the range of the overlooking lane line map.

In the second aspect, an embodiment of the present application provides a method for maintaining a lane line map, including:

According to the overlooked lane line map, a preset range of lane line partial map is updated, and the line shape of the lane line is determined according to the updated partial map of the preset range of lane lines.

In a third aspect, an embodiment of the present application provides a method for maintaining a lane line map, including:

Obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the shown overhead lane line map, and the lane line partial map of the preset range at the previous detection time is in line with the overlook A second area where the lane line map does not overlap, and a third area where the lane line partial map of the preset range at the previous detection time overlaps with the overhead lane line map;

The feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area, and the feature information of each lane line point in the third area is multiplexed.

In a fourth aspect, an embodiment of the present application provides an electronic device, including:

Memory, used to store computer programs;

The processor is used to execute the computer program, specifically to execute:

In a fifth aspect, an embodiment of the present application provides an electronic device, including:

Memory, used to store computer programs;

The processor is used to execute the computer program, specifically to execute:

In the sixth aspect, this application provides an electronic device according to an embodiment, including:

Memory, used to store computer programs;

The processor is used to execute the computer program, specifically to execute:

In a seventh aspect, an embodiment of the present application provides a vehicle, including a vehicle body and the electronic device according to any one of the fourth aspect to the sixth aspect installed on the vehicle body.

In an eighth aspect, an embodiment of the present application provides a vehicle, including: a vehicle body and the electronic device according to any one of the fourth to sixth aspects installed on the vehicle body.

In a ninth aspect, an embodiment of the present application provides a computer storage medium in which a computer program is stored, and the computer program, when executed, implements the lane marking according to any one of the first to third aspects. How to maintain the map.

The maintenance method, electronic equipment, and storage medium of the lane line map provided by the embodiments of the present application acquire environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data; according to the environmental data and the lane line Area, generating an overlooked lane line map; according to the overlooking lane line map, updating a preset range of the lane line partial map; wherein the preset range includes at least the range of the overlooking lane line map. In this way, the updated local map of the lane line with the preset range has a large visual range. When intelligent driving is performed based on the local map of the lane line with the large visual range, accurate guidance for intelligent driving can be realized, thereby improving the safety of intelligent driving. Sex.

Description of the drawings

In order to more clearly describe the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description These are some embodiments of the present application. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative work.

FIG. 1 is a schematic diagram of an application scenario involved in an embodiment of this application;

FIG. 2 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

3 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

4 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

FIG. 5 is a schematic diagram of a partial map of lane lines involved in an embodiment of the application;

6 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

FIG. 7 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

8 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

FIG. 9 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

10 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

11 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application;

FIG. 12 is a schematic diagram of an electronic device provided by an embodiment of this application;

FIG. 13 is a schematic diagram of an electronic device provided by an embodiment of the application;

FIG. 14 is a schematic diagram of an electronic device provided by an embodiment of this application;

15 is a schematic diagram of the structure of a vehicle provided by an embodiment of the application;

FIG. 16 is a schematic structural diagram of a transportation tool provided by an embodiment of the application.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present application clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments It is a part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the protection scope of this application.

The method of the embodiment of the present application is applicable to technical fields such as computer vision and intelligent driving, and can realize the maintenance of the lane line map, thereby improving the safety of intelligent driving.

At present, the image of the environment around the vehicle is usually obtained through the visual sensor mounted on the vehicle, and the lane line in the image is recognized and detected through methods such as image recognition. After the lane is detected, a lane line map is usually generated. The lane line map contains lane line information in the world coordinate system, which is used to provide a basis for decision-making for the driving of the vehicle.

However, because the image obtained by the vision sensor only covers a certain part of the environment around the vehicle, the lane line obtained by recognition and detection is only the result of the lane line within the visual range of the visual sensor around the vehicle. In some cases, the lane line map only includes the The detection result of the lane line within the visible range may be difficult to achieve accurate guidance to the intelligent driving vehicle, thereby reducing the safety of intelligent driving. In other cases, the lane line map may include accumulated lane line detection results to form a large-scale global lane line map. However, this will cause a great demand for vehicle calculation and storage resources, which may not be convenient for practical use.

FIG. 1 is a schematic diagram of an application scenario involved in an embodiment of this application. It should be noted that the application scenario of an embodiment of this application includes but is not limited to that shown in FIG. 1. As shown in Figure 1, the intelligent driving vehicle includes sensors. During the driving of the intelligent driving vehicle, the sensor can collect the surrounding environment data. The vehicle can update the local map of the preset range of lane lines in real time according to the collected environment data. The updated partial map of the lane line of the preset range is used to plan the state of intelligent driving of the vehicle, such as changing lanes, decelerating, or parking, etc., so as to realize accurate guidance for intelligent driving, thereby improving the safety of intelligent driving.

The technical solution of the present application will be described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments.

FIG. 2 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. As shown in FIG. 2, the method of the embodiment of the application includes:

S101: Acquire environmental data around the vehicle through sensors, and identify a lane line area according to the environmental data.

The execution subject of the embodiments of the present application is a device with a lane line maintenance function, such as a lane line maintenance device, hereinafter referred to as a maintenance device, which can be integrated in any electronic device as a part of the electronic device. Optionally, the maintenance device may also be a separate electronic device. The electronic device may be an in-vehicle device, for example, an auxiliary driving device or an automatic driving device installed in a vehicle.

Optionally, the maintenance device can also be directly integrated in the vehicle instead of in the form of a separate electronic device. For example, the maintenance device is a processing system inside an autonomous vehicle.

In the following, the execution subject is an electronic device as an example for description.

The electronic device of the embodiment of the present application is in communication connection with the sensor. The sensor is installed on the vehicle and can send collected environmental data around the vehicle to the electronic device. The electronic device processes the environmental data around the vehicle sent by the sensor to update A partial map of lane lines with a preset range.

Optionally, the electronic device is communicatively connected with the intelligent driving system, and the electronic device can send the updated partial map of the lane line of the preset range to the intelligent driving system, so that the intelligent driving system can follow the updated lane line of the preset range Local maps are used for intelligent driving control of the vehicle.

The embodiment of the present application does not limit the way of identifying the lane line area based on the environmental data. For example, the environmental data is input into a trained neural network, and the neural network is input into the lane line area around the vehicle.

S102. Generate an overlooked lane line map according to the environment data and the lane line area.

According to the above steps, the lane line area around the vehicle is identified, so that based on the environment data around the vehicle and the lane line area, an overlooked lane line map can be generated. The overhead lane line map is an overhead lane line map generated based on environmental data detected by the sensor once. For example, when the sensor is an imaging device, the overhead lane line map corresponds to a lane line with a single frame image range.

Exemplarily, this step may be that the electronic device projects the obtained environmental data to the bird's-eye view, and projects the lane line area obtained above to the bird's-eye view, and then generates an bird's-eye lane line map, which can be understood as a bird's-eye view Perspective.

Optionally, the sensor may specifically be a three-dimensional detection device, a vision sensor, etc., and a combination thereof. The vision sensor may be an imaging device, and the point cloud sensor includes lidar, Time Of Flight (TOF) ranging detection Equipment, depth vision sensor and high resolution millimeter wave radar, etc.

In a possible implementation manner, if the sensor is a point cloud sensor, the above S101 includes step A1, and the above step S102 includes step A2.

Step A1: Acquire environmental point cloud data around the vehicle through the point cloud sensor, and identify the lane line area according to the environmental point cloud data.

Step A2, according to the environmental point cloud data and the lane line area, generate an overlooked lane line map.

Specifically, in this implementation, if the sensor is a point cloud sensor, such as a lidar, the environmental data is environmental point cloud data around the vehicle collected by the point cloud sensor, and the environmental point cloud data is three-dimensional data. The device can identify the lane line area based on the environmental point cloud data. At the same time, the electronic device can generate an overlook lane map based on the environmental point cloud data and the identified lane line area.

In another possible implementation manner, if the sensor is a vision sensor and a point cloud sensor, the foregoing S101 includes steps B1 and B2, and the foregoing step S102 includes step B3.

Step B1: Obtain an image of the environment around the vehicle through the visual sensor, and obtain point cloud data of the environment around the vehicle through the three-dimensional detection data.

Step B2: Identify the lane line area according to the environmental point cloud data and the environmental image.

Step B3: Generate an overhead lane line map based on the environmental point cloud data, the environmental image, and the lane line area.

Specifically, in this implementation, if the sensor includes a vision sensor (such as an imaging device) and a point cloud sensor (such as a lidar), the environmental data obtained above includes two parts, which are the environmental images around the vehicle collected by the vision sensor and The point cloud sensor collects point cloud data around the vehicle. The electronic device recognizes the lane line area around the vehicle based on the point cloud data and the environment image. Since the point cloud data is three-dimensional data, the environment image includes the line type information of the lane line. In this way, the electronic device can accurately generate an overlooking lane line map based on the point cloud data, the environment image, and the lane line area.

In another possible implementation manner, if the sensor is a vision sensor, the above S101 includes step C1, and the above step S102 includes step C2.

Step C1: Obtain an image of the environment around the vehicle through the visual sensor, and identify the lane line area according to the environment image.

Step C2, according to the environment image and the lane line area, generate an overlooked lane line map.

Specifically, in this implementation manner, if the sensor is a visual sensor (for example, an imaging device), the environmental data obtained above is the environmental image around the vehicle collected by the visual sensor. The electronic device recognizes the lane line area in the environment image according to the environment image. Then, the electronic device can generate an overlook lane map based on the environment image and the lane line area.

The overhead lane line map generated above can be understood as the overhead lane line map at the current detection time, which is generated based on the environmental data collected by the sensor at the current detection time.

S103. Update a preset range of a partial lane line map according to the overlooked lane line map; wherein the preset range includes at least a range of the overlooked lane line map.

This step does not limit the specific size of the preset range, as long as it is ensured that the preset range at least includes the range of the overlook lane map, that is, the preset range is larger than the range of the overlook lane map. For example, the range of the overhead lane map is 100m in front of the vehicle, and the preset range includes 200m in front of the vehicle and 200m behind the vehicle.

In this way, the local lane line map of the preset range at the previous detection time is updated according to the overlooked lane line map at the current detection time obtained in the above steps, and the local lane line map of the preset range at the current detection time is obtained.

Optionally, if the current detection time is the initial detection time, the overhead lane line map at the initial detection time can be directly used as the initial lane line partial map of the preset range. Then, as time goes by, when it comes to the next detection time, use the overlooked lane line map at the next detection time to update the initial lane line partial map of the preset range, and proceed in sequence to achieve the preset range of the lane line partial map Real-time updates. In this way, the updated local map of the lane line of the preset range has a large visual range, including not only the information of the lane line at the current detection time, but also the information of the lane line at the historical time, based on the local map of the lane line with a large visible range When performing smart driving, accurate guidance for smart driving can be realized, thereby improving the safety of smart driving.

The method for maintaining the lane line map provided by the embodiment of the present application acquires environmental data around the vehicle through sensors, and recognizes the lane line area according to the environmental data; generates an overlook lane line based on the environmental data and the lane line area Map; according to the overlooked lane line map, update a preset range of lane line partial maps; wherein the preset range includes at least the range of the overlooked lane line map. In this way, the updated local map of the lane line with the preset range has a large visual range. When intelligent driving is performed based on the local map of the lane line with the large visual range, accurate guidance for intelligent driving can be realized, thereby improving the safety of intelligent driving. Sex.

3 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. On the basis of the foregoing embodiment, the embodiment of the present application relates to the foregoing step C1 according to the environment image and the image lane line area, The specific process of generating an overlook lane map, that is, the above step C1 may include:

S201: Determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view.

In the embodiment of the present application, if the sensor is a vision sensor, the vision sensor is, for example, an imaging device. The coordinate system of the environment image collected by the imaging device is not consistent with the world coordinate system of the bird's-eye view. Therefore, based on the environment image and the image lane line area, Before generating the overhead lane line map, the environment image needs to be projected to the overhead view world coordinate system. Therefore, the position conversion relationship between the environment image coordinate system and the overhead view world coordinate system needs to be determined. The position conversion relationship can include a conversion matrix, etc. It is no wonder that the coordinate system conversion can be realized.

In a possible implementation manner, the process of determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view in S201 may include the following steps:

S2011: Acquire calibration parameters of the vision sensor and posture information of the vehicle.

S2012: Determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view according to the calibration parameters of the vision sensor and the posture information of the vehicle.

In this implementation, since the vision sensor of the present application is installed on the vehicle, the posture information of the vehicle has an influence on the imaging result of the vision sensor. Therefore, it is necessary to obtain the posture information of the vehicle at the current detection time. At the same time, the calibration parameters of the vision sensor need to be obtained.

Optionally, the vehicle attitude information may include attitude information such as the pitch angle, roll angle, and yaw angle of the vehicle.

Optionally, assuming that the vision sensor is an imaging device, the calibration parameters of the imaging device include internal parameters and external parameters of the imaging device.

Next, the electronic device determines the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view based on the posture information of the vehicle at the current detection time and the calibration parameters of the visual sensor.

S202: Convert the environment image and the lane line area of the image to the bird's-eye view world coordinate system according to the position conversion relationship.

S203: Generate an overhead lane line map according to the environment image and the lane line area of the image in the overhead view world coordinate system.

Specifically, after determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view according to the above steps, the environment image and the lane line area of the image are converted to the world coordinate system of the bird's-eye view according to the position conversion relationship. In this way, an overlook lane line map can be generated based on the environment image in the bird's-eye view world coordinate system and the image lane line area in the bird's-eye view world coordinate system.

In the embodiment of this application, if the sensor is a vision sensor, the electronic device generates an overlook lane line map based on the environment image and the image lane line area collected by the vision sensor, first determining the coordinate system between the environment image's coordinate system and the bird's eye view world coordinate system Position conversion relationship; then, according to the position conversion relationship, the environment image and the lane line area of the image are converted to the bird's-eye view world coordinate system; then, based on the environment image and the image lane line area in the bird's-eye view world coordinate system, a bird’s eye view is generated Lane line map, to achieve accurate generation of the overlooked lane line map.

FIG. 4 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. Based on the above-mentioned embodiment, the embodiment of the present application relates to the above-mentioned update of the lane line of the preset range based on the above-mentioned overlooking the lane line map The specific process of the local map, the above S103 includes:

S300. Move the lane line partial map of the preset range at the previous detection time, so that the center of the lane line partial map of the preset range at the previous detection time and the center of the vehicle satisfy a preset correspondence relationship .

In the embodiment of the present application, before using the overhead lane line map at the current detection time to update the lane line partial map of the preset range at the previous detection time, and obtain the lane line partial map of the preset range at the current detection time, the above S300 is also included. , That is, move the lane line partial map of the preset range at the previous detection time, so that the center of the lane line partial map of the preset range at the previous detection time and the center of the vehicle meet the preset correspondence relationship, as shown in Figure 5 , So that the center of the local map of the lane line of the preset range at the previous detection time coincides with the center of the vehicle.

In a possible implementation manner, the foregoing S300 includes the following steps:

S3001. Obtain the displacement of the vehicle from the previous detection time to the current detection time.

S3002, according to the displacement, move the lane line local map of the preset range at the previous detection time so that the center of the lane line local map of the preset range at the previous detection time is the same as the center of the vehicle Meet the preset corresponding relationship.

Specifically, the electronic device is connected to the vehicle driving system and can obtain driving information such as the running speed of the vehicle from the vehicle driving system. In this way, the electronic device obtains the displacement of the vehicle from the vehicle driving system from the previous detection time to the current detection time. For example, the electronic device obtains the running speed of the vehicle at the previous detection time and the current detection time from the vehicle driving system. According to these two operating speeds, and the time difference between the previous detection time and the current detection time, The displacement of the vehicle from the previous detection time to the current detection time.

Then, according to the displacement of the vehicle from the previous detection time to the current detection time, move the lane line local map of the preset range at the previous detection time, so that the center of the lane line local map of the preset range at the previous detection time and the vehicle The center meets the preset correspondence relationship, that is, the center of the local map of the lane line of the preset range moves with the movement of the vehicle.

After moving the center of the lane line partial map of the preset range at the previous detection time with the movement of the vehicle, use the overlooked lane line map at the current detection time to update the preset range of the lane line partial at the previous detection time after the movement For the map, the specific process of obtaining the local map of the lane line of the preset range at the current detection time includes the following steps S301 to S303.

S301: Determine a mapping relationship between the world coordinates of the overlooked lane line map and the pixel coordinates of the predetermined range of the lane line local map at the previous detection time.

The coordinate system of each point on the map overlooking the lane line in the embodiment of the application is inconsistent with the coordinate system of each pixel on the local map of the lane line in the preset range at the previous detection time. The previous detection time is updated using the determined overlooking lane line map. Before the lane line partial map of the preset range, it is necessary to convert the overlooking lane line map and the lane line partial map of the preset range at the previous detection time to the same coordinate system. Therefore, first determine the world coordinates of the overlooking lane line map The mapping relationship with the pixel coordinates of the lane line partial map of the preset range at the previous detection time.

S302. According to the mapping relationship, map the world coordinates of each lane line point in the overlook lane line map to the pixel coordinates of the lane line partial map of the preset range at the previous detection time, to obtain each lane line in the overlook lane line map The coordinates of the point in the local map of the lane line within the preset range.

Specifically, according to the mapping relationship determined in S301, the world coordinates (x, y) of each lane line point in the overlooking lane line map are mapped to the pixel coordinates (i ,j), let i=x/si ₀ , j=y/sj ₀ , as shown in Figure 5, the local map of lane lines in the preset range includes multiple two-dimensional grids, each of which can be understood Is a data element, s represents the scale expressed by each data element in the real world, and (i ₀ , j ₀ ) is the center of the local map of the lane line of the preset range after the movement.

S303. According to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection time, using the feature information of each lane line point on the overlooking lane line map to update the previous detection time The feature information of each lane line point on the local map of the lane line of the preset range.

Specifically, according to the above steps, the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line partial map can be obtained, so that the lane line with each lane line point in the overlooking lane line map in the preset range can be used The coordinates in the local map are used to update the characteristic information of each lane line point on the local map of the preset range of the lane line at the previous detection time, and then obtain the lane line point on the local map of the preset range of the lane line at the current detection time Characteristic information.

Optionally, the characteristic information of the lane line point includes the two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and the probability table of the lane line point, and the probability table includes that the lane line point belongs to Probability values of different linear categories.

Optionally, the above-mentioned different linear types include at least one of solid lines, dashed lines, and diversion lines.

In a possible implementation manner, in S303, the feature information of each lane line point on the overlooked lane line map is used to update each lane line point on the lane line partial map of the preset range at the previous detection time. Characteristic information, including:

S3031. Obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overhead lane line map.

S3032. Acquire a second area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time.

S3033. Use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area.

Specifically, referring to FIG. 5, the dashed frame represents the range of the lane line partial map of the preset range at the previous detection time, and the realized frame represents the range of the lane line local map of the preset range at the current detection time. The range of the lane line partial map of the preset range at the current detection time at least includes the range of the overlooked lane line map. In this way, an area that does not overlap with the lane line partial map of the preset range at the previous detection time can be obtained, and it is recorded as the first area, such as the gray area in the solid line box in FIG. 5. At the same time, the area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time can be obtained, and it is recorded as the second area, such as the gray area in the dashed box in FIG. Next, the characteristic information of each lane line point in the second area is updated using the characteristic information of each lane line point in the first area.

Optionally, if the characteristic information of the lane line point includes the probability table of the lane line point, according to the above method, the probability table of each lane line point in the first area can be used to update the probability table of each lane line point in the second area, The probability table includes the probability values of lane line points belonging to different linear categories. In this way, the line shape of each lane line point in the lane line partial map of the preset range at the current detection time can be predicted, and then the line shape of the lane line can be accurately predicted. So that the vehicle can perform precise intelligent driving control based on the accurate lane line shape.

In a possible implementation manner, the above S3033 can be replaced by step D.

Step D: Based on the Bayesian update method, the feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area.

Continuing to refer to FIG. 5, FIG. 5 includes multiple grids, each network can be understood as a data element, and a data element can be understood as a lane line point. Assume that Z is the characteristic information of the lane line point in the first area, and A is the characteristic information of the lane line point in the second area.

In an example, according to the following formula (1), the characteristic information of each lane line point in the first area is used to update the characteristic information of each lane line point in the second area.

P(A|Z)=P(Z|A)*P(Z)/P(A), i=0,1,...,n (1)

Among them, P(Z|A) is the probability that Z is established when A is given, also called the posterior probability of Z, P(A|Z) is the probability that A is established when Z is given, also called the posterior probability of A The prior probability, P(Z) is the prior probability of Z, and P(A) is the prior probability of A.

In an example, the above step D can be replaced by step D1.

Step D1, based on the Bayesian update method and the negative observation method, use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area.

Exemplarily, based on the Bayesian update method and the negative observation method, for example, according to the following formula (2), the characteristic information of each lane line point in the first area is used to update the characteristic information of each lane line point in the second area.

among them,

Represents the log likelihood ratio, X is the norm, for example, it can be replaced by A, Z|Ai, etc. in the above formula,

Indicates that the observation is not in this state, that is, negative observation.

In the embodiment of the present application, through the negative observation, the problem of the false failure noise of the lane line and the accumulated blur of the lane line caused by the positioning offset can be eliminated from the time sequence.

In a possible implementation manner, if the characteristic information of the lane line point includes a probability table of the lane line point, the probability table includes the probability value of the lane line point belonging to each of the n preset line shapes. At this time, the lane line points in the second area include a set of linear states A0, A1,..., An, which are used to indicate that the location of the lane line point (that is, the data element) belongs to the line of different lanes. At the current detection time, the data A new lane line point observation value Z appears at the location of the yuan, so that formula (3) can be used to update the probability table of each lane line point in the second area using the probability table of each lane line point in the first area.

According to the above formula (3), the line shape of the lane line point at the current detection time can be obtained, and then the line shape of the lane line can be obtained, which provides a more reliable basis for intelligent driving.

In a possible implementation manner, the method in the embodiment of the present application further includes:

S3034. Obtain a third area that overlaps the overhead lane map in the lane line partial map of the preset range at the previous detection time.

S3035. Multiplex the characteristic information of each lane line point in the third area.

Continuing to refer to FIG. 5, where the local lane line map of the preset range at the previous detection time and the overhead lane line map at the current detection time have an overlapping area, which is recorded as the third area. In order to reduce the amount of updated data and save technical resources, when determining the lane line local map of the preset range at the current detection time, you can directly reuse the characteristic information of each lane line point corresponding to the third area at the previous detection time .

It should be noted that there is no sequence between the execution process of S3031 to S3033 and the execution process of S3034 to S3035.

The method of the embodiment of the present application determines the mapping relationship between the world coordinates of the overlooking lane line map and the pixel coordinates of the lane line local map of the preset range at the previous detection time; according to the mapping relationship, the The world coordinates of each lane line point in the overlooking lane line map are mapped to the pixel coordinates of the preset range of the lane line local map at the previous detection time, and each lane line point in the overlooking lane line map is obtained in the previous detection According to the coordinates of each lane line point in the lane line map overlooking the lane line in the lane line local map of the preset range at the previous detection time, use the coordinates on the lane line map overlooking the lane line The feature information of the lane line points is updated to update the feature information of each lane line point on the lane line local map of the preset range at the previous detection time, and then the lane line local map of the preset range at the current detection time is obtained.

Fig. 6 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. The technical problem to be solved in this embodiment is that the lane line obtained by the existing method is usually only a single line and cannot reflect the actual lane. The line shape of the line, such as single solid line, double yellow line, etc., such a lane line with only lines cannot provide more accurate and intelligent decision-making basis for subsequent processing. In order to solve this technical problem, as shown in FIG. 6, the method of the embodiment of the present application includes:

S401: Acquire environmental data around the vehicle through sensors, and identify a lane line area according to the environmental data.

S402: Generate an overlooked lane line map based on the environment data and the lane line area.

Optionally, the above-mentioned sensor may specifically be a three-dimensional detection device and a vision sensor, etc., and a combination thereof. The vision sensor may be an imaging device. The point cloud sensor includes a lidar, an ultrasonic detection device, and a Time Of Flight (TOF) method. ) Ranging detection equipment, vision sensors and laser detection equipment, etc.

In a possible implementation, if the sensor is a point cloud sensor, the above S401 includes step A1, and the above step S402 includes step A2.

In another possible implementation manner, if the sensors are vision sensors and point cloud sensors, the above S401 includes steps B1 and B2, and the above step S402 includes step B3.

Step B3: Generate an overlook lane map based on the environmental point cloud data, the environmental image and the lane line area.

In another possible implementation manner, if the sensor is a vision sensor, the above S401 includes step C1, and the above step S402 includes step C2.

For the specific execution process of the foregoing steps S401 to S402, reference may be made to the description of the foregoing S101 to S102, which will not be repeated here.

S403: According to the overlooked lane line map, update a preset range of the lane line partial map.

For the specific execution process of the foregoing step S403, reference may be made to the description of the foregoing S103, which will not be repeated here.

In an example, as shown in Figure 5, both the overlooking lane line map and the lane line partial map of the preset range include multiple lane line points, and each lane line point can be understood as a data element, the characteristics of the lane line point The information includes a probability table of lane line points, and the probability table includes the probability values of lane line points belonging to different linear categories. Optionally, the above-mentioned different linear types include at least one of solid lines, dashed lines, and diversion lines.

Here, using the probability table of the lane line points, the process of updating the partial map of the lane line of the preset range based on the overlooked lane line map is introduced.

It is assumed that the probability table of the lane line point above includes the probability value that the lane line point belongs to each of the n preset line shapes. At this time, the lane line points at the previous detection time include a set of linear states A0, A1,..., An, which are used to indicate that the location of the lane line point (ie, the data element) belongs to the line of different lanes. At the current detection time, A new lane line point observation value Z appears at the location of the data element, so that formula (3) can be used to update each lane line local map of the preset range at the previous detection time using the probability table of each lane line point in the overlooking lane line map The probability table of lane line points obtains the probability table of each lane line point in the preset range of the lane line local map at the current detection time.

among them,

According to the above formula (3), the probability table of the lane line point at the current detection time can be obtained.

S404: Determine the line shape of the lane line according to the updated partial map of the lane line in the preset range.

According to the above steps, each lane line point in the local map of the lane line of the preset range after the update includes a probability table belonging to a different line shape. In this way, the lane line at the current detection time can be determined according to the probability table of each lane line point. Linear. For example, it can be determined that the lane line is a solid line, a dashed line, etc., because different lane lines have different driving meanings. This can provide more accurate decision-making basis for intelligent driving based on the detected line shape of the lane line, thereby improving Improve the reliability and safety of intelligent driving.

In a possible implementation manner, determining the line shape of the lane line according to the updated local map of the lane line in the preset range in S404 may include:

S4041. Determine the line shape of each lane line point according to the updated probability table of each lane line point in the lane line partial map of the preset range.

Since the probability table of each lane line point includes the probability values that the lane line point belongs to different linear categories, for example, the preset linear category is n categories, and the probability table includes the lane line point belonging to each of the n categories The probability values of are, for example, P1, P2,..., Pn, and these probability values are processed to obtain a line shape corresponding to a probability value as the line shape of the lane line point. For example, take the line shape corresponding to the intermediate value of these probability values as the line shape of the lane line point, or compare the above-mentioned probability values with the preset value, and obtain the line shape corresponding to the probability value satisfying the preset value as the lane line The line shape of the point.

Optionally, the line shape corresponding to the maximum probability value among the above probability values is determined as the line shape of the lane line point. For example, in P1, P2,..., Pn, the maximum probability value is P2, then the line shape corresponding to P2 is determined as The line shape of the lane line point.

S4042. Determine the line shape of the lane line according to the line shape of each lane line point.

According to the above steps, after determining the line shape of the lane line point, according to the line shape of the lane line point, the line shape of each lane line constituting the lane line can be used as the line shape of the lane line.

In the method of the embodiment of the present application, the environmental data around the vehicle is acquired through sensors, and the lane line area is identified according to the environmental data; an overlook lane line map is generated according to the environmental data and the lane line area; The lane line map updates the local map of the lane line of the preset range, and accurately determines the line shape of the lane line according to the updated local map of the lane line of the preset range. In this way, a more accurate decision-making basis can be provided for intelligent driving according to the line shape of the detected lane line, thereby improving the reliability and safety of intelligent driving.

FIG. 7 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. Based on the embodiment shown in FIG. 6, the embodiment of the application relates to the foregoing step C1 according to the environment image and the image For the lane line area, the specific process of generating an overlooked lane line map, that is, the above step C1 may include:

S501: Determine a position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view.

In a possible implementation manner, the process of determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view in S501 may include the following steps:

S5011: Acquire calibration parameters of the vision sensor and posture information of the vehicle.

S5012: Determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view according to the calibration parameters of the vision sensor and the posture information of the vehicle.

For the specific execution process of the foregoing step S501, reference may be made to the description of the foregoing S201, which will not be repeated here.

S502: Convert the environment image and the lane line area of the image to the world coordinate system of the bird's-eye view according to the position conversion relationship.

S503: Generate an overlook lane line map according to the environment image and the lane line area of the image in the overlook map world coordinate system.

For the specific execution process of the foregoing steps S502 and S503, reference may be made to the description of the foregoing S202 and S203, which will not be repeated here.

FIG. 8 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. On the basis of the foregoing embodiment, the embodiment of the present application relates to the foregoing update of the lane line of the preset range based on the above-mentioned overlooking the lane line map The specific process of the local map, the above S403 includes:

S600. Move the lane line partial map of the preset range at the previous detection time, so that the center of the lane line partial map of the preset range at the previous detection time and the center of the vehicle satisfy a preset correspondence relationship .

In a possible implementation manner, the foregoing S600 includes the following steps:

S6001. Obtain the displacement of the vehicle from the previous detection time to the current detection time.

S6002, according to the displacement, move the lane line local map of the preset range at the previous detection time so that the center of the lane line local map of the preset range at the previous detection time is the same as the center of the vehicle Meet the preset corresponding relationship.

For the specific execution process of the foregoing step S600, reference may be made to the description of the foregoing S300, which will not be repeated here.

After moving the center of the lane line partial map of the preset range at the previous detection time with the movement of the vehicle, use the overlooked lane line map at the current detection time to update the preset range of the lane line partial at the previous detection time after the movement For the map, the specific process of obtaining the local map of the lane line of the preset range at the current detection time includes the following steps S601 to S603.

S601: Determine a mapping relationship between the world coordinates of the overlooked lane line map and the pixel coordinates of the preset range of the lane line local map at the previous detection time.

S602. According to the mapping relationship, map the world coordinates of each lane line point in the overlooking lane line map to the pixel coordinates of the lane line partial map of the preset range at the previous detection time, to obtain each lane line in the overlooking lane line map The coordinates of the point in the local map of the lane line within the preset range.

S603. According to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection time, using the feature information of each lane line point on the overlooking lane line map to update the previous detection time The feature information of each lane line point on the local map of the lane line of the preset range.

For the specific execution process of the foregoing steps S601 to S603, reference may be made to the description of the foregoing S301 to S303, which will not be repeated here.

In a possible implementation manner, in S603, the feature information of each lane line point on the overlooked lane line map is used to update each lane line point on the lane line partial map of the preset range at the previous detection time. Characteristic information, including:

S6031. Obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overhead lane line map.

S6032. Acquire a second area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time.

S6033: Use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area.

For the specific execution process of the foregoing steps S6031 to S6033, reference may be made to the description of the foregoing S3031 to S3033, which will not be repeated here.

In a possible implementation manner, the above S6033 can be replaced by step D.

In an example, the above step D can be replaced by step D1.

For the specific execution process of the foregoing step S6033, reference may be made to the description of the foregoing S3033, which will not be repeated here.

S6034. Acquire a third area that overlaps with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time.

S6035. Multiplex the characteristic information of each lane line point in the third area.

It should be noted that the specific execution process of the foregoing S6031 to S6033 can refer to the description of the foregoing S3031 and S3033, and there is no sequence between the foregoing execution process of S6031 to S6033 and the foregoing execution process of S6034 to S6035.

Fig. 9 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. The technical problem to be solved by the embodiment of the application is that the existing method needs to continuously process the image data of the transformed area during the driving of the vehicle , This will cause a large amount of calculation and will also consume more memory resources. In order to solve this technical problem, as shown in FIG. 9, the method of the embodiment of the present application includes:

S701: Acquire environmental data around the vehicle through sensors, and identify a lane line area according to the environmental data.

S702: Generate an overlooked lane line map based on the environment data and the lane line area.

Optionally, the sensor may specifically be a three-dimensional detection device, a vision sensor, etc., and a combination thereof. The vision sensor may be an imaging device. The point cloud sensor includes lidar, ultrasonic detection equipment, and Time Of Flight (TOF). ) Ranging detection equipment, vision sensors and laser detection equipment, etc.

In a possible implementation, if the sensor is a point cloud sensor, the above S701 includes step A1, and the above step S702 includes step A2.

In another possible implementation manner, if the sensors are vision sensors and point cloud sensors, the foregoing S701 includes steps B1 and B2, and the foregoing step S702 includes step B3.

In another possible implementation manner, if the sensor is a vision sensor, the above S701 includes step C1, and the above step S702 includes step C2.

For the specific execution process of the foregoing S701 to S702, reference may be made to the description of the foregoing S101 and S102, which will not be repeated here.

S703. Obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overlooking lane line map, and the overlooking lane line in the preset range of the lane line partial map at the previous detection time A second area where the map does not overlap, and a third area where the lane line partial map of the preset range at the previous detection time overlaps with the overhead lane line map.

Referring to FIG. 5, referring to FIG. 5, the dashed frame represents the range of the lane line partial map of the preset range at the previous detection time, and the realization box represents the range of the lane line local map of the preset range at the current detection time. The range of the lane line partial map of the preset range at the current detection time at least includes the range of the overlooked lane line map. In this way, an area that does not overlap with the lane line partial map of the preset range at the previous detection time can be obtained, and it is recorded as the first area, such as the gray area in the solid line box in FIG. 5. At the same time, the area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time can be obtained, and it is recorded as the second area, such as the gray area in the dashed box in FIG.

Continuing to refer to FIG. 5, the local map of the lane line of the preset range at the previous detection time and the map of the overlooking lane line at the current detection time have an overlapping area, which is recorded as the third area.

S704. Use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area, and reuse the characteristic information of each lane line point in the third area.

Specifically, the feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area. In order to reduce the amount of updated data and save technical resources and memory resources, when determining the lane line local map of the preset range at the current detection time, you can directly reuse the lane line points corresponding to the third area at the previous detection time Characteristic information.

The method of the embodiment of the present application obtains the first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overlooking lane line map, and the lane line partial of the preset range at the previous detection time The second area on the map that does not overlap with the overlooking lane line map, and the third area where the lane line partial map of the preset range at the previous detection time overlaps with the overlooking lane line map; each lane line point in the first area is used The feature information updates the feature information of each lane line point in the second area, and reuses the feature information of each lane line point in the third area. This not only realizes the update of the local map of the lane line in the preset range, and obtains the local map of the lane line in the preset range at the current detection time. At the same time, in the update process, reuse the local lane line of the preset range at the previous detection time. The third area where the map overlaps with the map overlooking the lane line, thereby reducing the amount of updated data and saving technical resources and memory resources.

FIG. 10 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. On the basis of the foregoing embodiment, the embodiment of the present application relates to the foregoing step C1 according to the environment image and the image lane line area, The specific process of generating an overlook lane map, that is, the above step C1 may include:

S801: Determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view.

In a possible implementation manner, the process of determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view in S801 may include the following steps:

S8011, acquire calibration parameters of the vision sensor and posture information of the vehicle.

S8012, according to the calibration parameters of the vision sensor and the posture information of the vehicle, determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view.

For the specific execution process of the foregoing S8011 to S8012, reference may be made to the description of the foregoing S2011 and S2012, and details are not described herein again.

S802: According to the position conversion relationship, convert the environment image and the lane line area of the image to the bird's-eye view world coordinate system.

S803: Generate an overhead lane line map according to the environment image and the lane line area of the image in the overhead view world coordinate system.

For the specific execution process of the foregoing S802 to S803, reference may be made to the description of the foregoing S202 and S203, which will not be repeated here.

FIG. 11 is a flowchart of a method for maintaining a lane line map provided by an embodiment of the application. On the basis of the above-mentioned embodiment, the embodiment of this application relates to the above-mentioned updating the second by using the characteristic information of each lane line point in the first area The feature information of each lane line point in the area, and the multiplexing of the feature information of each lane line point in the third area, the above S703 includes:

S900. Move the lane line partial map of the preset range at the previous detection time, so that the center of the lane line partial map of the preset range at the previous detection time and the center of the vehicle satisfy a preset correspondence relationship .

In a possible implementation manner, the foregoing S900 includes the following steps:

S9001. Acquire the displacement of the vehicle from the previous detection time to the current detection time;

S9002, according to the displacement, move the lane line local map of the preset range at the previous detection time, so that the center of the lane line local map of the preset range at the previous detection time is the same as the center of the vehicle Meet the preset corresponding relationship.

For the specific execution process of the foregoing S9001 to S9002, reference may be made to the description of the foregoing S3001 and S3002, which will not be repeated here.

After the center of the lane line partial map of the preset range at the previous detection time is moved with the movement of the vehicle, the feature information of each lane line point in the first area is used to update the specific information of each lane line point in the second area. The process includes the following steps S901 to S903.

S901. Determine a mapping relationship between the world coordinates of the overlooked lane line map and the pixel coordinates of the predetermined range of the lane line local map at the previous detection time.

S902. According to the mapping relationship, map the world coordinates of each lane line point in the overlooking lane line map to the pixel coordinates of the lane line partial map of the preset range at the previous detection time, to obtain each lane line in the overlooking lane line map The coordinates of the point in the local map of the lane line within the preset range.

S903. According to the coordinates of each lane line point in the overlooking lane line map in the lane line local map of the preset range at the previous detection time, update the second area using the characteristic information of each lane line point in the first area The characteristic information of each lane line point in.

In a possible implementation manner, using the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area in S903 can be replaced by step D.

In an example, the above step D can be replaced by step D1.

For the steps of the foregoing embodiment, reference may be made to the description of the foregoing embodiment shown in FIG. 4, which is not repeated here.

The method of the embodiment of the present application determines the mapping relationship between the world coordinates of the overlooking lane line map and the pixel coordinates of the lane line local map of the preset range at the previous detection time; according to the mapping relationship, the The world coordinates of each lane line point in the overlooking lane line map are mapped to the pixel coordinates of the preset range of the lane line local map at the previous detection time, and each lane line point in the overlooking lane line map is obtained in the previous detection The coordinates in the preset range of the lane line local map; according to the coordinates of each lane line point in the overlooked lane line map in the preset range of the lane line local map at the previous detection time, use the first area The feature information of each lane line point updates the feature information of each lane line point in the second area, so as to obtain a local lane line map of a preset range at the current detection time.

FIG. 12 is a schematic diagram of an electronic device provided by an embodiment of this application. As shown in FIG. 12, an electronic device 200 of an embodiment of this application includes at least one memory 210 and at least one processor 220. Among them, the memory 210 is used to store a computer program; the processor 220 is used to execute the computer program.

The processor 220, when executing the computer program, acquires environmental data around the vehicle through sensors, and recognizes the lane line area according to the environmental data; generates an overlook lane line map based on the environmental data and the lane line area; The overlooked lane line map updates a preset range of the lane line partial map; wherein the preset range includes at least the range of the overlooked lane line map.

Optionally, the above-mentioned sensor may be arranged on the electronic device 200 or outside the electronic device 200, and the sensor is in communication connection with the electronic device.

The electronic device of the embodiment of the present application may be used to execute the technical solution of the method embodiment shown above, and its implementation principles and technical effects are similar, and will not be repeated here.

In a possible implementation manner, the sensor includes a vision sensor,

The processor 220 is specifically configured to obtain an image of the environment around the vehicle through the visual sensor, and recognize the lane line area according to the environment image; the environment data is the environment image.

In a possible implementation manner, the processor 220 is specifically configured to determine a position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view; according to the position conversion relationship, convert the environment The image and the lane line area of the image are converted to the bird's-eye view world coordinate system; and the bird's eye view lane line map is generated according to the environment image and the image lane line area in the bird's eye view world coordinate system.

In a possible implementation, the processor 220 is specifically configured to obtain the calibration parameters of the vision sensor and the posture information of the vehicle; according to the calibration parameters of the vision sensor and the posture information of the vehicle, Determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view.

In a possible implementation manner, the processor 220 uses the overlooked lane line map at the current detection time to update the preset range of the lane line local map at the previous detection time to obtain the current detection time Before the partial map of the lane line of the preset range,

The processor 220 is specifically configured to move the lane line local map of the preset range at the previous detection time, so that the center of the lane line local map of the preset range at the previous detection time is equal to that of the vehicle. The center meets the preset correspondence relationship.

In a possible implementation, the processor 220 is specifically configured to obtain the displacement of the vehicle from the previous detection moment to the current detection moment; according to the displacement, move the preset value at the previous detection moment The local lane line map of the range, so that the center of the local lane line map of the preset range at the previous detection time and the center of the vehicle satisfy the preset correspondence relationship.

In a possible implementation manner, the processor 220 is further configured to determine the mapping relationship between the world coordinates of the overlooking lane line map and the pixel coordinates of the preset range of the lane line local map at the previous detection time According to the mapping relationship, map the world coordinates of each lane line point in the overlook lane line map to the pixel coordinates of the preset range of the lane line local map at the previous detection time to obtain the overlook lane line The coordinates of each lane line point in the map in the previously detected local map of the lane line of the preset range.

In a possible implementation manner, the processor 220 is specifically configured to, according to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection moment, Using the feature information of each lane line point on the overlooking lane line map, update the feature information of each lane line point on the lane line partial map of the preset range at the previous detection time.

In a possible implementation manner, the processor 220 is specifically configured to obtain a first area in the overlooking lane line map that does not overlap with the lane line partial map of the preset range at the previous detection time; The second area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time; the second area is updated using the characteristic information of each lane line point in the first area The characteristic information of each lane line point in.

In a possible implementation manner, the processor 220 is further configured to obtain a third area that overlaps the overlooked lane line map in the preset range of the lane line partial map at the previous detection time; Characteristic information of each lane line point in the third area.

In a possible implementation manner, the processor 220 is specifically configured to use the characteristic information of each lane line point in the first area to update each lane line point in the second area based on a Bayesian update method. Characteristic information.

In a possible implementation manner, the processor 220 is specifically configured to use the characteristic information of each lane line point in the first area to update the first area based on the Bayesian update method and the negative observation method. 2. The characteristic information of each lane line point in the area.

In a possible implementation manner, the characteristic information of the lane line point includes two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and a probability table of the lane line point, and the probability table includes The probability values of the lane line points belonging to different linear categories.

Optionally, the different linear categories include at least one of a solid line, a dashed line, and a guide line.

In a possible implementation manner, the sensor includes a point cloud sensor,

The processor 220 is specifically configured to obtain environmental point cloud data around the vehicle through the point cloud sensor, and identify the lane line area according to the environmental point cloud data; the environmental data is the environment Point cloud data.

In a possible implementation, the sensor includes a vision sensor and a point cloud sensor,

The processor 220 is specifically configured to obtain an image of the environment around the vehicle through the vision sensor, and obtain environmental point cloud data around the vehicle through the point cloud sensor; according to the environmental point cloud data and the The environment image recognizes the lane line area; according to the environment point cloud data, the environment image, and the lane line area, an overhead lane line map is generated.

FIG. 13 is a schematic diagram of an electronic device provided by an embodiment of the application. As shown in FIG. 13, the electronic device 300 of the embodiment of the application includes at least one memory 310 and at least one processor 320. Among them, the memory 310 is used to store a computer program; the processor 320 is used to execute the computer program.

The processor 230 is configured to execute the computer program, specifically for acquiring environmental data around the vehicle through sensors, and identifying a lane line area based on the environmental data; generating a bird’s eye view based on the environmental data and the lane line area Lane line map; according to the overlooked lane line map, update the lane line partial map of the preset range, and determine the line shape of the lane line according to the updated partial lane line map of the preset range.

Optionally, the above-mentioned sensor may be arranged on the electronic device 300 or outside the electronic device 300, and the sensor is in communication connection with the electronic device.

In a possible implementation manner, the sensor includes a vision sensor,

The processor 320 is specifically configured to obtain an image of the environment around the vehicle through the visual sensor, and recognize the lane line area according to the environment image; the environment data is the environment image.

In a possible implementation manner, the processor 320 is specifically configured to determine a position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view; according to the position conversion relationship, convert the environment The image and the lane line area of the image are converted to the bird's-eye view world coordinate system; and the bird's eye view lane line map is generated according to the environment image and the image lane line area in the bird's eye view world coordinate system.

In a possible implementation, the processor 320 is specifically configured to obtain the calibration parameters of the vision sensor and the posture information of the vehicle; according to the calibration parameters of the vision sensor and the posture information of the vehicle, Determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view.

In a possible implementation manner, the processor 320 uses the overlooked lane line map at the current detection time to update the preset range of the lane line local map at the previous detection time to obtain the current detection time Before the partial map of the lane line of the preset range,

The processor 320 is further configured to move the local lane line map of the preset range at the previous detection time, so that the center of the lane line local map of the preset range at the previous detection time is equal to that of the vehicle. The center meets the preset correspondence relationship.

In a possible implementation manner, the processor 320 is specifically configured to obtain the displacement of the vehicle from the previous detection time to the current detection time; according to the displacement, move the preset value at the previous detection time The local lane line map of the range, so that the center of the local lane line map of the preset range at the previous detection time and the center of the vehicle satisfy the preset correspondence relationship.

In a possible implementation, the processor 320 is further configured to determine the mapping relationship between the world coordinates of the overlooked lane line map and the pixel coordinates of the preset range of the lane line local map at the previous detection time According to the mapping relationship, map the world coordinates of each lane line point in the overlook lane line map to the pixel coordinates of the preset range of the lane line local map at the previous detection time to obtain the overlook lane line The coordinates of each lane line point in the map in the lane line local map of the preset range at the previous detection time.

In a possible implementation, the processor 320 is specifically configured to, according to the coordinates of each lane line point in the overlook lane line map in the preset range of the lane line local map at the previous detection time, Using the feature information of each lane line point on the overlooking lane line map, update the feature information of each lane line point on the lane line partial map of the preset range at the previous detection time.

In a possible implementation manner, the processor 320 is specifically configured to obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overhead lane line map; The second area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time; the second area is updated using the characteristic information of each lane line point in the first area The characteristic information of each lane line point in.

In a possible implementation manner, the processor 320 is further configured to obtain a third area that overlaps the overlooked lane line map in the lane line partial map of the preset range at the previous detection time; Characteristic information of each lane line point in the third area.

In a possible implementation manner, the processor 320 is specifically configured to use the characteristic information of each lane line point in the first area to update each lane line point in the second area based on a Bayesian update method Characteristic information.

In a possible implementation manner, the processor 320 is specifically configured to use the characteristic information of each lane line point in the first area to update the first area based on the Bayesian update method and the negative observation method. 2. The characteristic information of each lane line point in the area.

In a possible implementation manner, the characteristic information of the lane line point includes two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and a probability table of the lane line point, and the probability table includes The probability values that the lane line points belong to different line shapes.

Optionally, the different line shapes include at least one of a solid line, a dashed line, and a guide line.

In a possible implementation, the processor 320 is specifically configured to determine each lane line point according to the updated probability table of each lane line point in the lane line local map of the preset range The line shape of the lane line; the line shape of the lane line is determined according to the line shape of each lane line point.

In a possible implementation, the processor 320 is specifically configured to, for each lane line point in the updated lane line local map of the preset range, set the maximum probability table of the lane line point The line shape corresponding to the probability value is determined as the line shape of the lane line point.

In a possible implementation manner, the sensor includes a point cloud sensor,

The processor 320 is specifically configured to obtain environmental point cloud data around the vehicle through the point cloud sensor, and identify the lane line area according to the environmental point cloud data; the environmental data is the environment Point cloud data.

The processor 320 is specifically configured to obtain an image of the environment around the vehicle through the vision sensor, and obtain environmental point cloud data around the vehicle through the point cloud sensor; according to the environmental point cloud data and the The environment image recognizes the lane line area; according to the environment point cloud data, the environment image, and the lane line area, an overhead lane line map is generated.

FIG. 14 is a schematic diagram of an electronic device provided by an embodiment of this application. As shown in FIG. 14, an electronic device 400 in an embodiment of this application includes at least one memory 410 and at least one processor 420. Wherein, the memory 410 is used to store a computer program; the processor 420 is used to execute the computer program.

The processor 430 is configured to execute the computer program, and is specifically configured to obtain environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data; generate a bird's eye view based on the environmental data and the lane line area Lane line map; acquiring the first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overlooked lane line map, and the lane line partial map of the preset range at the previous detection time A second area that does not overlap with the overhead lane line map, and a third area where the preset range of the lane line partial map at the previous detection time overlaps the overhead lane line map; use the first area The feature information of each lane line point updates the feature information of each lane line point in the second area, and the feature information of each lane line point in the third area is multiplexed.

Optionally, the above-mentioned sensor may be arranged on the electronic device 400 or outside the electronic device 400, and the sensor is in communication connection with the electronic device.

In a possible implementation manner, the sensor includes a vision sensor,

The processor 430 is specifically configured to obtain an image of the environment around the vehicle through the visual sensor, and recognize the lane line area according to the environment image; the environment data is the environment image.

In a possible implementation manner, the processor 430 is specifically configured to determine a position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view; according to the position conversion relationship, convert the environment The image and the lane line area of the image are converted to the bird's-eye view world coordinate system; and the bird's eye view lane line map is generated according to the environment image and the image lane line area in the bird's eye view world coordinate system.

In a possible implementation, the processor 430 is specifically configured to obtain the calibration parameters of the vision sensor and the posture information of the vehicle; according to the calibration parameters of the vision sensor and the posture information of the vehicle, Determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view.

In a possible implementation manner, the processor 430 uses the overlooked lane line map at the current detection time to update the preset range of the lane line local map at the previous detection time to obtain the current detection time Before the partial map of the lane line of the preset range,

The processor 430 is further configured to move the local lane line map of the preset range at the previous detection time, so that the center of the lane line local map of the preset range at the previous detection time is equal to that of the vehicle. The center meets the preset correspondence relationship.

In a possible implementation, the processor 430 is specifically configured to obtain the displacement of the vehicle from the previous detection moment to the current detection moment; according to the displacement, move the preset value at the previous detection moment The local lane line map of the range, so that the center of the local lane line map of the preset range at the previous detection time and the center of the vehicle satisfy the preset correspondence relationship.

In a possible implementation, the processor 430 is further configured to determine the mapping relationship between the world coordinates of the overlooking lane line map and the pixel coordinates of the preset range of the lane line local map at the previous detection time According to the mapping relationship, map the world coordinates of each lane line point in the overlook lane line map to the pixel coordinates of the preset range of the lane line local map at the previous detection time to obtain the overlook lane line The coordinates of each lane line point in the map in the lane line local map of the preset range at the previous detection time.

In a possible implementation, the processor 430 is specifically configured to, according to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection moment, The feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area.

In a possible implementation manner, the processor 430 is specifically configured to use the characteristic information of each lane line point in the first area to update each lane line point in the second area based on a Bayesian update method Characteristic information.

In a possible implementation manner, the processor 430 is specifically configured to use the characteristic information of each lane line point in the first area to update the first area based on the Bayesian update method and the negative observation method. 2. The characteristic information of each lane line point in the area.

In a possible implementation manner, the sensor includes a point cloud sensor,

The processor 430 is specifically configured to obtain environmental point cloud data around the vehicle through the point cloud sensor, and identify the lane line area according to the environmental point cloud data; the environmental data is the environment Point cloud data.

The processor 430 is specifically configured to obtain an image of the environment around the vehicle through the visual sensor, and obtain environmental point cloud data around the vehicle through the point cloud sensor; according to the environmental point cloud data and the The environment image recognizes the lane line area; according to the environment point cloud data, the environment image, and the lane line area, an overhead lane line map is generated.

FIG. 15 is a schematic structural diagram of a vehicle provided by an embodiment of the application. As shown in FIG. 15, a vehicle 50 in this embodiment includes a body 51 and an electronic device 52 installed on the body 51.

Wherein, the electronic device 52 is the electronic device of any one of FIGS. 12 to 14, and the electronic device 52 is used for the maintenance of the lane line map.

Optionally, the electronic device 52 is installed on the roof of the vehicle body 51, and the sensor is installed on the vehicle body to collect environmental data around the vehicle.

Optionally, the electronic device 52 is installed on the front windshield of the vehicle body 51, or the electronic device 52 is installed on the rear windshield of the vehicle body 51.

Optionally, the electronic device 52 is installed on the front of the vehicle body 51, or the electronic device 52 is installed on the rear of the vehicle body 51.

The embodiment of the present application does not limit the installation position of the electronic device 52 on the body 51, which is specifically determined according to actual needs.

The vehicle of the embodiment of the present application can be used to implement the technical solution of the above-mentioned embodiment of the method for maintaining a lane line map, and its implementation principles and technical effects are similar, and will not be repeated here.

FIG. 16 is a schematic structural diagram of a vehicle provided by an embodiment of the application. As shown in FIG. 16, the vehicle 60 of this embodiment includes: a vehicle body 61 and an electronic device 62 installed on the vehicle body 61.

Wherein, the electronic device 62 is the electronic device shown in any one of FIGS. 12 to 14, and the electronic device 62 is used for maintaining lane lines.

Optionally, the vehicle 60 in this embodiment may be a ship, automobile, bus, railway vehicle, aircraft, railway locomotive, scooter, bicycle, etc.

Optionally, the electronic device 62 can be installed on the front, rear, or middle of the vehicle body 61, etc. The embodiment of the present application does not limit the installation position of the electronic device 62 on the vehicle body 61, and is specifically determined according to actual needs.

The transportation tool of the embodiment of the present application can be used to implement the technical solutions of the above-mentioned maintenance method embodiment of the lane line map, and its implementation principles and technical effects are similar, and will not be repeated here.

Further, when at least a part of the functions of the lane line map maintenance method in the embodiment of the present application is realized by software, the embodiment of the present application also provides a computer storage medium, and the computer storage medium is used to store the computer software for the lane line maintenance. The instructions, when run on the computer, enable the computer to execute various possible lane line map maintenance methods in the foregoing method embodiments. When the computer-executable instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application can be generated in whole or in part. The computer instructions can be stored in a computer storage medium, or transmitted from one computer storage medium to another computer storage medium, and the transmission can be transmitted to another by wireless (such as cellular communication, infrared, short-range wireless, microwave, etc.) Website site, computer, server or data center for transmission. The computer storage medium may be any available medium that can be accessed by a computer or a data storage device such as a server or data center integrated with one or more available media. The usable medium may be a magnetic medium (for example, a floppy disk, a hard disk, a magnetic tape), an optical medium (for example, a DVD), or a semiconductor medium (for example, an SSD).

The embodiments of the present application also provide a computer storage medium in which program instructions are stored. The program execution may include part or all of the steps of the lane line map maintenance method in the foregoing embodiments.

A person of ordinary skill in the art can understand that all or part of the steps in the above method embodiments can be implemented by a program instructing relevant hardware. The foregoing program can be stored in a computer readable storage medium. When the program is executed, it is executed. Including the steps of the foregoing method embodiment; and the foregoing storage medium includes: read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disks or optical disks, etc., which can store program codes Medium.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the application, not to limit them; although the application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand: It is still possible to modify the technical solutions described in the foregoing embodiments, or equivalently replace some or all of the technical features; these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the application range.

Claims

A method for maintaining a lane line map is characterized in that it includes:

Acquire environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data;

According to the environmental data and the lane line area, generate an overlooked lane line map;

According to the overlooked lane line map, update the local lane line map of the preset range;

Wherein, the preset range at least includes the range of the overlooking lane line map.
The method of claim 1, wherein the sensor comprises a vision sensor,

The acquiring environmental data around the vehicle through a sensor and identifying the lane line area according to the environmental data includes: acquiring an environmental image around the vehicle through the visual sensor, and identifying the lane based on the environmental image Line area

The environmental data is the environmental image.
The method according to claim 2, wherein the generating an overlooked lane line map according to the environment image and the lane line area of the image comprises:

Determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view;

According to the position conversion relationship, converting the environment image and the lane line area of the image to the bird's-eye view world coordinate system;

According to the environment image in the world coordinate system of the bird's-eye view map and the lane line area of the image, a bird's eye view lane line map is generated.
The method according to claim 3, wherein the determining the position conversion relationship between the coordinate system of the environmental image and the world coordinate system of the bird's-eye view comprises:

Acquiring the calibration parameters of the vision sensor and the posture information of the vehicle;

According to the calibration parameters of the vision sensor and the posture information of the vehicle, the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's eye view is determined.
The method according to claim 3, characterized in that, using the overlooked lane line map at the current detection time, the local lane line map of the preset range at the previous detection time is updated to obtain the preview at the current detection time. Before setting the local map of the lane line of the range, the method includes:

The local map of the lane line of the preset range at the previous detection time is moved so that the center of the local map of the lane line of the preset range at the previous detection time meets the preset correspondence relationship with the center of the vehicle.
The method according to claim 5, wherein the local map of the preset range of the lane line at the previous detection time is moved so that the local map of the lane line of the preset range at the previous detection time is The center and the center of the vehicle meet the preset correspondence relationship, including:

Acquiring the displacement of the vehicle from the previous detection moment to the current detection moment;

According to the displacement amount, move the lane line local map of the preset range at the previous detection time so that the center of the lane line local map of the preset range at the previous detection time and the center of the vehicle meet all requirements. Describe the preset correspondence.
The method according to claim 6, wherein the method further comprises:

Determining the mapping relationship between the world coordinates of the overlooking lane line map and the pixel coordinates of the lane line partial map of the preset range at the previous detection time;

According to the mapping relationship, map the world coordinates of each lane line point in the overlook lane line map to the pixel coordinates of the preset range of the lane line local map at the previous detection time to obtain the overlook lane line map The coordinates of each lane line point in the previously detected lane line local map of the preset range.
The method according to claim 7, wherein the updating a local map of a predetermined range of lane lines according to the overlooked lane line map comprises:

According to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection time, the feature information of each lane line point on the overlooking lane line map is used to update the previous 1. The characteristic information of each lane line point on the lane line partial map of the preset range at the detection time.
The method according to claim 8, wherein the feature information of each lane line point on the overlooked lane line map is used to update each lane line local map of the preset range at the previous detection time. The characteristic information of lane line points, including:

Acquiring a first area in the overlooking lane line map that does not overlap with the lane line partial map of the preset range at the previous detection time;

Acquiring a second area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time;

The feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area.
The method according to claim 8 or 9, further comprising:

Acquiring a third area overlapping with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time;

Multiplexing the characteristic information of each lane line point in the third area.
The method according to claim 9, wherein the using the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area comprises:

Based on the Bayesian update method, the feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area.
The method according to claim 11, characterized in that the characteristic information of each lane line point in the first area is used to update the characteristic information of each lane line point in the second area based on the Bayesian update method ,include:

Based on the Bayesian updating method and the negative observation method, the characteristic information of each lane line point in the first area is used to update the characteristic information of each lane line point in the second area.
The method according to any one of claims 8-12, wherein the characteristic information of the lane line point includes the two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and the lane line point The probability table includes the probability values of the lane line points belonging to different linear categories.
The method according to claim 13, wherein the different linear categories include at least one of a solid line, a dashed line, and a diversion line.
The method according to claim 1, wherein the sensor comprises a point cloud sensor,

The acquiring environmental data around the vehicle through sensors, and identifying the lane line area according to the environmental data includes: acquiring environmental point cloud data around the vehicle through the point cloud sensor, and according to the environmental point cloud data Identify the lane line area;

The environmental data is the environmental point cloud data.
The method according to claim 1, wherein the sensor includes a vision sensor and a point cloud sensor,

The acquiring environmental data around the vehicle through a sensor, and identifying the lane line area according to the environmental data includes: acquiring an environmental image around the vehicle through the visual sensor, and acquiring the surrounding environment of the vehicle through the point cloud sensor Environmental point cloud data;

Identifying the lane line area according to the environmental point cloud data and the environmental image;

The generating an overlooked lane line map according to the environmental data and the lane line area includes: generating an overlooked lane line map according to the environmental point cloud data, the environmental image, and the lane line area.
A method for maintaining a lane line map is characterized in that it includes:

Acquire environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data;

According to the environmental data and the lane line area, generate an overlooked lane line map;

According to the overlooked lane line map, a preset range of lane line partial map is updated, and the line shape of the lane line is determined according to the updated partial map of the preset range of lane lines.
The method of claim 17, wherein the sensor comprises a vision sensor,

The acquiring environmental data around the vehicle through a sensor and identifying the lane line area according to the environmental data includes: acquiring an environmental image around the vehicle through the visual sensor, and identifying the lane based on the environmental image Line area

The environmental data is the environmental image.
The method according to claim 18, wherein the generating an overlook lane line map according to the environment image and the lane line area of the image comprises:

Determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view;

According to the position conversion relationship, converting the environment image and the lane line area of the image to the bird's-eye view world coordinate system;

According to the environment image in the world coordinate system of the bird's-eye view map and the lane line area of the image, a bird's eye view lane line map is generated.
The method according to claim 19, wherein the determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view comprises:

Acquiring the calibration parameters of the vision sensor and the posture information of the vehicle;

According to the calibration parameters of the vision sensor and the posture information of the vehicle, the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's eye view is determined.
The method according to claim 19, characterized in that, using the overlooked lane line map at the current detection time, the local lane line map of the preset range at the previous detection time is updated to obtain the preview at the current detection time. Before setting the local map of the lane line of the range, the method includes:

The local map of the lane line of the preset range at the previous detection time is moved so that the center of the local map of the lane line of the preset range at the previous detection time meets the preset correspondence relationship with the center of the vehicle.
22. The method according to claim 21, wherein the local map of the preset range of the lane line at the previous detection time is moved so that the local map of the lane line of the preset range at the previous detection time is The center and the center of the vehicle meet the preset correspondence relationship, including:

Acquiring the displacement of the vehicle from the previous detection moment to the current detection moment;

According to the displacement amount, move the lane line local map of the preset range at the previous detection time so that the center of the lane line local map of the preset range at the previous detection time and the center of the vehicle meet all requirements. Describe the preset correspondence.
The method according to claim 22, wherein the method further comprises:

Determining the mapping relationship between the world coordinates of the overlooking lane line map and the pixel coordinates of the lane line local map of the preset range at the previous detection time;

According to the mapping relationship, map the world coordinates of each lane line point in the overlook lane line map to the pixel coordinates of the preset range of the lane line local map at the previous detection time to obtain the overlook lane line map The coordinates of each lane line point in the lane line local map of the preset range at the previous detection time.
22. The method according to claim 23, wherein the updating a local lane line map of a preset range according to the overlooked lane line map comprises:

According to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection time, the feature information of each lane line point on the overlooking lane line map is used to update the previous 1. The characteristic information of each lane line point on the lane line partial map of the preset range at the detection time.
The method according to claim 24, wherein the feature information of each lane line point on the overlooked lane line map is used to update each lane line local map of the preset range at the previous detection time. The characteristic information of lane line points, including:

Acquiring a first area in the overlooking lane line map that does not overlap with the lane line partial map of the preset range at the previous detection time;

Acquiring a second area that does not overlap with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time;

The feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area.
The method according to claim 24 or 25, further comprising:

Acquiring a third area overlapping with the overlooked lane line map in the lane line partial map of the preset range at the previous detection time;

Multiplexing the characteristic information of each lane line point in the third area.
The method according to claim 25, wherein the using the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area comprises:

Based on the Bayesian update method, the feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area.
28. The method according to claim 27, wherein the Bayesian updating method is used to update the characteristic information of each lane line point in the second area using the characteristic information of each lane line point in the first area ,include:

Based on the Bayesian updating method and the negative observation method, the characteristic information of each lane line point in the first area is used to update the characteristic information of each lane line point in the second area.
The method according to any one of claims 24-28, wherein the characteristic information of the lane line point includes the two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and the lane line point The probability table includes the probability values of the lane line points belonging to different linear shapes.
The method according to claim 29, wherein the different line shapes include at least one of a solid line, a dashed line, and a diversion line.
The method according to claim 29, wherein the determining the line shape of the lane line according to the updated local map of the lane line of the preset range comprises:

Determine the line shape of each lane line point according to the updated probability table of each lane line point in the lane line partial map of the preset range;

Determine the line shape of the lane line according to the line shape of each lane line point.
The method according to claim 31, wherein the line shape of each lane line point is determined according to the updated probability table of each lane line point in the lane line partial map of the preset range, include:

For each lane line point in the updated lane line partial map of the preset range, the line shape corresponding to the maximum probability value in the probability table of the lane line point is determined as the line shape of the lane line point.
The method according to claim 17, wherein the sensor comprises a point cloud sensor,

The acquiring environmental data around the vehicle through sensors, and identifying the lane line area according to the environmental data includes: acquiring environmental point cloud data around the vehicle through the point cloud sensor, and according to the environmental point cloud data Identify the lane line area;

The environmental data is the environmental point cloud data.
The method according to claim 17, wherein the sensor includes a vision sensor and a point cloud sensor,

The acquiring environmental data around the vehicle through a sensor, and identifying the lane line area according to the environmental data includes: acquiring an environmental image around the vehicle through the visual sensor, and acquiring the surrounding environment of the vehicle through the point cloud sensor Environmental point cloud data;

Identifying the lane line area according to the environmental point cloud data and the environmental image;

The generating an overlooked lane line map according to the environmental data and the lane line area includes: generating an overlooked lane line map according to the environmental point cloud data, the environmental image, and the lane line area.
A method for maintaining a lane line map is characterized in that it includes:

Acquire environmental data around the vehicle through sensors, and identify the lane line area based on the environmental data;

According to the environmental data and the lane line area, generate an overlooked lane line map;

Obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the shown overhead lane line map, and the lane line partial map of the preset range at the previous detection time is in line with the overlook A second area where the lane line map does not overlap, and a third area where the lane line partial map of the preset range at the previous detection time overlaps with the overhead lane line map;

The feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area, and the feature information of each lane line point in the third area is multiplexed.
The method of claim 35, wherein the sensor comprises a vision sensor,

The acquiring environmental data around the vehicle through a sensor and identifying the lane line area according to the environmental data includes: acquiring an environmental image around the vehicle through the visual sensor, and identifying the lane based on the environmental image Line area

The environmental data is the environmental image.
The method according to claim 36, wherein said generating an overlook lane line map according to said environment image and said image lane line area comprises:

Determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view;

According to the position conversion relationship, converting the environment image and the lane line area of the image to the bird's-eye view world coordinate system;

According to the environment image in the world coordinate system of the bird's-eye view map and the lane line area of the image, a bird's eye view lane line map is generated.
The method according to claim 37, wherein the determining the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view comprises:

Acquiring the calibration parameters of the vision sensor and the posture information of the vehicle;

According to the calibration parameters of the vision sensor and the posture information of the vehicle, the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's eye view is determined.
The method of claim 37, wherein the method further comprises:

The local map of the lane line of the preset range at the previous detection time is moved so that the center of the local map of the lane line of the preset range at the previous detection time meets the preset correspondence relationship with the center of the vehicle.
The method according to claim 39, wherein the local map of the preset range of the lane line at the previous detection time is moved so that the local map of the lane line of the preset range at the previous detection time is The center and the center of the vehicle meet the preset correspondence relationship, including:

Acquiring the displacement of the vehicle from the previous detection moment to the current detection moment;

According to the displacement amount, move the lane line local map of the preset range at the previous detection time so that the center of the lane line local map of the preset range at the previous detection time and the center of the vehicle meet all requirements. Describe the preset correspondence.
The method according to claim 40, wherein the method further comprises:

Determining the mapping relationship between the world coordinates of the overlooking lane line map and the pixel coordinates of the lane line partial map of the preset range at the previous detection time;

According to the mapping relationship, map the world coordinates of each lane line point in the overlook lane line map to the pixel coordinates of the preset range of the lane line local map at the previous detection time to obtain the overlook lane line map The coordinates of each lane line point in the lane line local map of the preset range at the previous detection time.
The method according to claim 41, wherein the using the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area comprises:

According to the coordinates of each lane line point in the overlooked lane line map in the preset range of the lane line local map at the previous detection time, the first area is updated using the characteristic information of each lane line point 2. The characteristic information of each lane line point in the area.
The method according to claim 42, wherein the using the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area comprises:

Based on the Bayesian update method, the feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area.
The method according to claim 43, wherein the Bayesian update method is used to update the characteristic information of each lane line point in the second area using the characteristic information of each lane line point in the first area ,include:

Based on the Bayesian updating method and the negative observation method, the characteristic information of each lane line point in the first area is used to update the characteristic information of each lane line point in the second area.
The method according to any one of claims 42-44, wherein the characteristic information of the lane line point includes the two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and the lane line point The probability table includes the probability values of the lane line points belonging to different linear shapes.
The method according to claim 45, wherein the different line shapes include at least one of a solid line, a dashed line, and a diversion line.
The method of claim 35, wherein the sensor comprises a point cloud sensor,

The acquiring environmental data around the vehicle through sensors, and identifying the lane line area according to the environmental data includes: acquiring environmental point cloud data around the vehicle through the point cloud sensor, and according to the environmental point cloud data Identify the lane line area;

The environmental data is the environmental point cloud data.
The method according to claim 35, wherein the sensor comprises a vision sensor and a point cloud sensor,

The acquiring environmental data around the vehicle through a sensor, and identifying the lane line area according to the environmental data includes: acquiring an environmental image around the vehicle through the visual sensor, and acquiring the surrounding environment of the vehicle through the point cloud sensor Environmental point cloud data;

Identifying the lane line area according to the environmental point cloud data and the environmental image;

The generating an overlooked lane line map according to the environmental data and the lane line area includes: generating an overlooked lane line map according to the environmental point cloud data, the environmental image, and the lane line area.
An electronic device, characterized in that it comprises:

Memory, used to store computer programs;

The processor is used to execute the computer program, specifically to execute:

Acquire environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data;

According to the environmental data and the lane line area, generate an overlooked lane line map;

According to the overlooked lane line map, update the local lane line map of the preset range;

Wherein, the preset range at least includes the range of the overlooking lane line map.
The electronic device of claim 49, wherein the sensor comprises a visual sensor,

The processor is specifically configured to obtain an image of the environment around the vehicle through the visual sensor, and recognize the lane line area according to the environment image; the environment data is the environment image.
The electronic device according to claim 50, wherein:

The processor is specifically configured to determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view; and convert the environment image and the lane line area of the image to Under the bird's-eye view world coordinate system; generate an bird's-eye lane line map based on the environment image and the image lane line area in the bird's eye view world coordinate system.
The electronic device according to claim 51, wherein:

The processor is specifically configured to acquire the calibration parameters of the vision sensor and the posture information of the vehicle; determine the coordinate system and the bird's eye view of the environment image according to the calibration parameters of the vision sensor and the posture information of the vehicle The position conversion relationship between the world coordinate system of the graph.
The electronic device according to claim 52, wherein the processor uses the overlooked lane line map at the current detection time to update the preset range of the lane line local map at the previous detection time to obtain the current detection time Before the local map of the lane line of the preset range at the time,

The processor is specifically configured to move the local lane line map of the preset range at the previous detection time, so that the center of the lane line local map of the preset range at the previous detection time is the same as the center of the vehicle Meet the preset correspondence relationship.
The electronic device according to claim 53, wherein:

The processor is specifically configured to acquire the displacement amount of the vehicle from the previous detection moment to the current detection moment; according to the displacement amount, move the local lane line map of the preset range at the previous detection moment to make the previous The center of the lane line partial map of the preset range and the center of the vehicle at the detection time meet the preset correspondence relationship.
The electronic device according to claim 54, wherein:

The processor is further configured to determine the mapping relationship between the world coordinates of the overlooked lane line map and the pixel coordinates of the lane line local map of the preset range at the previous detection time; according to the mapping relationship, the The world coordinates of each lane line point in the overlooking lane line map are mapped to the pixel coordinates of the preset range of the lane line local map at the previous detection time, and each lane line point in the overlooking lane line map is obtained in the previous detection The coordinates in the local map of the lane line of the preset range.
The electronic device according to claim 55, wherein:

The processor is specifically configured to use each lane on the overlooking lane line map according to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection time The feature information of the line point updates the feature information of each lane line point on the local map of the lane line of the preset range at the previous detection time.
The electronic device according to claim 56, wherein:

The processor is specifically configured to obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overlooking lane line map; obtain the preset range at the previous detection time A second area in the local lane line map that does not overlap with the overlooked lane line map; using the feature information of each lane line point in the first area to update the feature information of each lane line point in the second area.
The electronic device according to claim 56 or 57, wherein:

The processor is further configured to obtain a third area that overlaps the overlooked lane line map in the lane line partial map of the preset range at the previous detection time; multiplex each lane line point in the third area Characteristic information.
The electronic device according to claim 57, wherein:

The processor is specifically configured to use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area based on a Bayesian update method.
The electronic device according to claim 59, wherein

The processor is specifically configured to use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area based on the Bayesian update mode and the negative observation mode .
The electronic device according to any one of claims 56 to 60, wherein the characteristic information of the lane line point includes the two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and the lane line A point probability table, the probability table including the probability values of the lane line points belonging to different linear categories.
The electronic device according to claim 61, wherein the different linear types include at least one of a solid line, a dashed line, and a diversion line.
The electronic device of claim 49, wherein the sensor comprises a point cloud sensor,

The processor is specifically configured to obtain environmental point cloud data around the vehicle through the point cloud sensor, and identify the lane line area according to the environmental point cloud data; the environmental data is the environmental point Cloud data.
The electronic device of claim 49, wherein the sensor comprises a vision sensor and a point cloud sensor,

The processor is specifically configured to obtain an image of the environment around the vehicle through the vision sensor, and obtain environmental point cloud data around the vehicle through the point cloud sensor; according to the environmental point cloud data and the environment The image identifies the lane line area; according to the environmental point cloud data, the environment image, and the lane line area, an overlooked lane line map is generated.
An electronic device, characterized in that it comprises:

Memory, used to store computer programs;

The processor is used to execute the computer program, specifically to execute:

Acquire environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data;

According to the environmental data and the lane line area, generate an overlooked lane line map;

According to the overlooked lane line map, a preset range of lane line partial map is updated, and the line shape of the lane line is determined according to the updated partial map of the preset range of lane lines.
The electronic device of claim 65, wherein the sensor comprises a vision sensor,

The processor is specifically configured to obtain an image of the environment around the vehicle through the visual sensor, and recognize the lane line area according to the environment image; the environment data is the environment image.
The electronic device according to claim 66, wherein:

The processor is specifically configured to determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view; and convert the environment image and the lane line area of the image to Under the bird's-eye view world coordinate system; generate an bird's-eye lane line map based on the environment image and the image lane line area in the bird's eye view world coordinate system.
The electronic device according to claim 67, wherein:

The processor is specifically configured to acquire the calibration parameters of the vision sensor and the posture information of the vehicle; determine the coordinate system and the bird's eye view of the environment image according to the calibration parameters of the vision sensor and the posture information of the vehicle The position conversion relationship between the world coordinate system of the graph.
The electronic device according to claim 67, wherein the processor uses the overlooked lane line map at the current detection time to update the preset range of the lane line local map at the previous detection time to obtain the current detection time Before the local map of the lane line of the preset range at the time,

The processor is further configured to move the lane line partial map of the preset range at the previous detection time, so that the center of the lane line partial map of the preset range at the previous detection time is the same as the center of the vehicle Meet the preset correspondence relationship.
The electronic device according to claim 69, wherein:

The processor is specifically configured to acquire the displacement amount of the vehicle from the previous detection moment to the current detection moment; according to the displacement amount, move the local lane line map of the preset range at the previous detection moment to make the previous The center of the lane line partial map of the preset range and the center of the vehicle at the detection time meet the preset correspondence relationship.
The electronic device according to claim 70, wherein:

The processor is further configured to determine the mapping relationship between the world coordinates of the overlooked lane line map and the pixel coordinates of the lane line local map of the preset range at the previous detection time; according to the mapping relationship, the The world coordinates of each lane line point in the overlooking lane line map are mapped to the pixel coordinates of the preset range of the lane line local map at the previous detection time, and each lane line point in the overlooking lane line map is obtained in the previous detection The coordinates in the local map of the lane line of the preset range at the time.
The electronic device according to claim 71, wherein:

The processor is specifically configured to use each lane on the overlooking lane line map according to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection time The feature information of the line point updates the feature information of each lane line point on the local map of the lane line of the preset range at the previous detection time.
The electronic device according to claim 72, wherein:

The processor is specifically configured to obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the overlooking lane line map; obtain the preset range at the previous detection time A second area in the local lane line map that does not overlap with the overlooked lane line map; using the feature information of each lane line point in the first area to update the feature information of each lane line point in the second area.
The electronic device according to claim 72 or 73, wherein:

The processor is further configured to obtain a third area that overlaps the overlooked lane line map in the lane line partial map of the preset range at the previous detection time; multiplex each lane line point in the third area Characteristic information.
The electronic device according to claim 73, wherein:

The processor is specifically configured to use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area based on a Bayesian update method.
The electronic device according to claim 75, wherein:

The processor is specifically configured to use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area based on the Bayesian update mode and the negative observation mode .
The electronic device according to any one of claims 72-76, wherein the characteristic information of the lane line point includes the two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and the lane line A probability table of points, the probability table including probability values that the lane line points belong to different linear shapes.
The electronic device according to claim 77, wherein the different line shapes include at least one of a solid line, a dashed line, and a diversion line.
The electronic device according to claim 77, wherein:

The processor is specifically configured to determine the line shape of each lane line point according to the updated probability table of each lane line point in the lane line partial map of the preset range; according to each lane line The line shape of the point determines the line shape of the lane line.
The electronic device according to claim 79, wherein:

The processor is specifically configured to determine the line shape corresponding to the maximum probability value in the probability table of the lane line point for each lane line point in the local map of the lane line of the preset range after the update is the The line shape of the lane line point.
The electronic device according to claim 65, wherein the sensor comprises a point cloud sensor,

The processor is specifically configured to obtain environmental point cloud data around the vehicle through the point cloud sensor, and identify the lane line area according to the environmental point cloud data; the environmental data is the environmental point Cloud data.
The electronic device according to claim 65, wherein the sensor comprises a vision sensor and a point cloud sensor,

The processor is specifically configured to obtain an image of the environment around the vehicle through the vision sensor, and obtain environmental point cloud data around the vehicle through the point cloud sensor; according to the environmental point cloud data and the environment The image identifies the lane line area; according to the environmental point cloud data, the environment image, and the lane line area, an overlooked lane line map is generated.
An electronic device, characterized in that it comprises:

Memory, used to store computer programs;

The processor is used to execute the computer program, specifically to execute:

Acquire environmental data around the vehicle through sensors, and identify the lane line area according to the environmental data;

According to the environmental data and the lane line area, generate an overlooked lane line map;

Obtain a first area that does not overlap with the lane line partial map of the preset range at the previous detection time in the shown overhead lane line map, and the lane line partial map of the preset range at the previous detection time is in line with the overlook A second area where the lane line map does not overlap, and a third area where the lane line partial map of the preset range at the previous detection time overlaps with the overhead lane line map;

The feature information of each lane line point in the first area is used to update the feature information of each lane line point in the second area, and the feature information of each lane line point in the third area is multiplexed.
The electronic device of claim 83, wherein the sensor comprises a vision sensor,

The processor is specifically configured to obtain an image of the environment around the vehicle through the visual sensor, and recognize the lane line area according to the environment image; the environment data is the environment image.
The electronic device according to claim 84, wherein:

The processor is specifically configured to determine the position conversion relationship between the coordinate system of the environment image and the world coordinate system of the bird's-eye view; and convert the environment image and the lane line area of the image to Under the bird's-eye view world coordinate system; generate an bird's-eye lane line map based on the environment image and the image lane line area in the bird's eye view world coordinate system.
The electronic device according to claim 85, wherein:

The processor is specifically configured to acquire the calibration parameters of the vision sensor and the posture information of the vehicle; determine the coordinate system and the bird's eye view of the environment image according to the calibration parameters of the vision sensor and the posture information of the vehicle The position conversion relationship between the world coordinate system of the graph.
The electronic device of claim 85, wherein the processor uses the overlooked lane line map at the current detection time to update the preset range of the lane line local map at the previous detection time to obtain the current detection time Before the local map of the lane line of the preset range at the time,

The processor is further configured to move the lane line partial map of the preset range at the previous detection time, so that the center of the lane line partial map of the preset range at the previous detection time is the same as the center of the vehicle Meet the preset correspondence relationship.
The electronic device according to claim 87, wherein:

The processor is specifically configured to acquire the displacement amount of the vehicle from the previous detection moment to the current detection moment; according to the displacement amount, move the local lane line map of the preset range at the previous detection moment to make the previous The center of the lane line partial map of the preset range and the center of the vehicle at the detection time meet the preset correspondence relationship.
The electronic device according to claim 88, wherein:

The processor is further configured to determine the mapping relationship between the world coordinates of the overlooked lane line map and the pixel coordinates of the lane line local map of the preset range at the previous detection time; according to the mapping relationship, the The world coordinates of each lane line point in the overlooking lane line map are mapped to the pixel coordinates of the preset range of the lane line local map at the previous detection time, and each lane line point in the overlooking lane line map is obtained in the previous detection The coordinates in the local map of the lane line of the preset range at the time.
The electronic device according to claim 89, wherein:

The processor is specifically configured to use each lane line in the first area according to the coordinates of each lane line point in the overlooking lane line map in the preset range of the lane line local map at the previous detection time The feature information of the point updates the feature information of each lane line point in the second area.
The electronic device according to claim 90, wherein:

The processor is specifically configured to use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area based on a Bayesian update method.
The electronic device according to claim 91, wherein:

The processor is specifically configured to use the characteristic information of each lane line point in the first area to update the characteristic information of each lane line point in the second area based on the Bayesian update mode and the negative observation mode .
The electronic device according to any one of claims 90-92, wherein the characteristic information of the lane line point includes the two-dimensional coordinates of the lane line point in the bird's-eye view coordinate system and the lane line A probability table of points, the probability table including probability values that the lane line points belong to different linear shapes.
The electronic device according to claim 93, wherein the different line shapes include at least one of a solid line, a dashed line, and a diversion line.
The electronic device of claim 83, wherein the sensor comprises a point cloud sensor,

The processor is specifically configured to obtain environmental point cloud data around the vehicle through the point cloud sensor, and identify the lane line area according to the environmental point cloud data; the environmental data is the environmental point Cloud data.
The electronic device according to claim 83, wherein the sensor comprises a vision sensor and a point cloud sensor,

The processor is specifically configured to obtain an image of the environment around the vehicle through the vision sensor, and obtain environmental point cloud data around the vehicle through the point cloud sensor; according to the environmental point cloud data and the environment The image identifies the lane line area; according to the environmental point cloud data, the environment image, and the lane line area, an overlooked lane line map is generated.
A vehicle, characterized by comprising: a vehicle body and the electronic device according to any one of claims 49-96 installed on the vehicle body.
A vehicle, characterized by comprising: a vehicle body and the electronic device according to any one of claims 49-96 installed on the vehicle body.
A computer storage medium, characterized in that a computer program is stored in the storage medium, and the computer program, when executed, realizes the lane line map maintenance method according to any one of claims 1-48.