CN116045995B

CN116045995B - Map generation system, method, vehicle and medium

Info

Publication number: CN116045995B
Application number: CN202310280527.3A
Authority: CN
Inventors: 南男
Original assignee: Beijing Jidu Technology Co Ltd
Current assignee: Beijing Jidu Technology Co Ltd
Priority date: 2023-03-21
Filing date: 2023-03-21
Publication date: 2023-06-20
Anticipated expiration: 2043-03-21
Also published as: CN116045995A

Abstract

The embodiment of the application provides a map generation system, a map generation method, a vehicle and a medium. The system comprises: the map topological structure generation module is used for generating road navigation information according to the positioning information and the navigation map; the road structure generation module is used for generating a lane-level road structure comprising a first lane where the vehicle is currently located and at least one second lane where the vehicle is to enter according to the lane lines acquired by the vision sensor; the lane navigation generation module is used for determining a second lane corresponding to the first lane and generating lane-level navigation information based on the lane-level road structure and the road navigation information; and the map generation module is used for establishing the connection relation between the first lane and the second lane based on the lane-level navigation information. When the road crossing is passed, the lane-level navigation information is generated based on the generated lane-level road structure and the current road navigation information, so that the lane-level reconstruction of the map with unsmooth road connection is realized, and the dependence on a high-precision map is avoided.

Description

Map generation system, method, vehicle and medium

Technical Field

The present disclosure relates to the field of vehicle control technologies, and in particular, to a map generation system, a map generation method, a vehicle, and a medium.

Background

With the development of vehicle technology, more and more vehicles begin to adopt functions such as automatic driving and auxiliary driving.

In the prior art, if an automatic driving mode of a vehicle needs to be started, high-precision map data support is required. However, the data collection and processing of the high-precision map consume a large amount of manpower and material resources, because the data collection is a huge intensive task, and hundreds of data collection vehicles are responsible for collecting source data for making the map so as to ensure that the map is updated quickly each time the road changes, which means that a large amount of manpower and material resources are required. In addition, when the vehicle-mounted system of the vehicle runs the high-precision map, a large amount of hardware resources are required to be occupied, and the requirement on the hardware resources of the vehicle is high, namely the hardware success is higher.

Disclosure of Invention

The embodiment of the application provides a map generation system, a map generation method, a vehicle and a map generation medium, which are used for improving the scheme of the success rate and the continuity of path planning.

In a first aspect, an embodiment of the present application provides a map generating system, including:

the road structure generation module is used for generating a lane-level road structure comprising a first lane where the vehicle is currently located and at least one second lane where the vehicle is to enter according to the lane lines acquired by the vision sensor; wherein the first lane and the second lane are different lanes connected with an intersection;

The map topological structure generation module is used for generating road navigation information according to the positioning information and the navigation map;

a lane navigation generation module for determining a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure and the road navigation information and generating lane-level navigation information;

and the map generation module is used for establishing a map containing the connection relation between the first lane and the target second lane based on the lane-level navigation information.

When a vehicle passes through an intersection, a plurality of roads are connected with the intersection. Each lane is divided by lane lines in each road connected with the intersection, however, because the widths of different roads are different, the number of the divided lanes and the widths of the lanes are not completely the same, and the corresponding relation of each lane in the different roads connected with the intersection cannot be or is difficult to be established one by one. Therefore, the road structure generating module can obtain the required lane-level road structure, and in particular, when constructing the lane-level road structure, roads in multiple directions connected with the intersection can be constructed, wherein the roads comprise a first lane, a second lane, a third lane, a fourth lane and the like. However, the fact that only the lane-level road structure has no current position information of the vehicle and no traveling direction information of the vehicle is insufficient for accurate navigation of the vehicle. Therefore, the map topological structure generation module is further utilized to generate road navigation information according to the positioning information and the conventional navigation map, so that a first lane where the vehicle is currently located and a target road which is to be driven into and comprises at least one second lane can be accurately known, and the target road is judged according to the driving direction. And further, a map containing lane-level navigation information of the connection relation between the first lane and the target second lane is generated by using the lane navigation generation module and the map generation module, so that lane-level navigation information with higher accuracy is obtained, and reliable high-accuracy navigation data can be provided for functions such as automatic driving and auxiliary driving without depending on the high-accuracy map.

Optionally, the road structure generating module is further configured to:

receiving a first current lane line and a second current lane line which are acquired by the vision sensor and are positioned on two sides of a first lane where a vehicle is currently positioned, and a first target lane line and a second target lane line which are positioned on two sides of the second lane where the vehicle is to be driven in;

reconstructing and generating a first lane based on the first current lane line and the second current lane line; the first lane comprises a first current lane extension line and a second current lane extension line;

and reconstructing and generating a second lane based on the first target lane line and the second target lane line.

In turn, a lane-level road structure may be generated based on the reconstructed first lane and the at least one second lane.

During the running process of the vehicle, the visual sensor is utilized to collect the current environment image, in particular to collect and identify the lane line image on the road surface. Of course, in order to reduce the occupation of computing resources, the acquisition of the images of each lane line of the road surface can be started according to the requirement of navigation information.

The definition of the lane lines in the images acquired by the vision sensor is uneven, and the lane lines in the intersections are interrupted, namely no lane line connection exists between lanes in roads in different directions. Therefore, it is necessary to reconstruct the lane lines acquired by the vision sensor. Specifically, a first lane is reconstructed based on a first current lane line and a second current lane line of a lane where the acquired vehicle is located currently. When reconstructing the first lane, it is necessary to simultaneously establish a first current lane extension line corresponding to the first current lane line and a second current lane extension line corresponding to the second current lane line, so that the relative positional relationship between the first lane and the second lane can be determined according to the extension lines. By the lane reconstruction mode, the position relationship and the corresponding relationship between the first lane and the second lane which do not have the lane line connection relationship are clearer.

Optionally, the lane navigation generating module is further configured to:

determining a target lane line opposite to the first lane according to the position relation between the first target lane line and the second target lane line and the first lane;

determining a first lane distance between the target lane line and the first current lane extension; and a second lane-distance between the target lane-line and the second current lane-extension;

if the first lane distance or the second lane distance is larger than a distance threshold, determining a lane corresponding to the first lane distance or the second lane distance as a target second lane;

the lane-level navigation information is generated based on the first lane, the target second lane, and the road navigation information.

In practical applications, the width of the road in different directions and the number of lanes contained in the road are not fixed, and it is impossible to ensure that the number of lanes and the width of the lanes contained in the different roads connected with the same intersection are completely the same. Thus, there is a need for an accurate match for a first lane in which a vehicle is currently located to a target second lane into which the vehicle is to be driven. In particular, the target second lane corresponding to the first lane may be determined in such a way that the minimum traverse distance is met and the desired width of the vehicle is met, so that the vehicle can achieve lane change simply and quickly. Therefore, the method for setting the distance threshold helps accurately select the target second lane with the minimum span distance from the roads comprising the second lanes, so that the vehicle can be successfully guided to drive into the target second lane from the first lane when passing through the intersection.

Optionally, the lane navigation generating module is further configured to:

comparing the first lane spacing to the second lane spacing;

according to the comparison result, determining the lane corresponding to the first lane distance or the second lane distance as a target second lane;

In practical applications, the width of the road in different directions and the number of lanes contained in the road are not fixed, and it is impossible to ensure that the number of lanes and the width of the lanes contained in the different roads connected with the same intersection are completely the same. Thus, there is a need for an accurate match for a first lane in which a vehicle is currently located to a target second lane into which the vehicle is to be driven. In particular, the target second lane corresponding to the first lane may be determined in such a way that the minimum traverse distance is met and the desired width of the vehicle is met, so that the vehicle can achieve lane change simply and quickly. The present application proposes a way to compare the first lane spacing with the second lane spacing to assist in accurately selecting a target second lane of minimum span spacing from a road containing a plurality of second lanes so as to be able to successfully guide a vehicle into the target second lane from the first lane when the vehicle passes the intersection.

Optionally, if the corresponding target second lane to be driven in is not determined according to the first lane distance and the second lane distance;

the lane navigation generating module is further configured to:

selecting an outermost lane or an innermost lane from among the roads including a plurality of lanes as the target second lane according to the position of the first lane in the current road;

When lane matching is performed, because the widths of different roads and the number of lanes included are different, each first lane cannot be ensured to be accurately matched with a proper target second lane sometimes according to a lane interval matching mode. When the first lane cannot be matched with the second lane in the two modes, according to the position of the first lane in the current road, selecting an outermost lane or an innermost lane from the roads which are to be driven into and comprise a plurality of lanes as a target second lane. Therefore, the target second vehicle road determined by the scheme can be used as a lane connected with the first vehicle road, so that the lane-level navigation information is generated by combining the road navigation information, and the lane-level navigation information can be simply and efficiently generated.

Optionally, the map generating module is further configured to: and establishing a map containing a lane reference line for guiding the vehicle to realize lane-level navigation based on the corresponding relation between the first lane and the second lane.

In practical applications, after determining the first lane and the target second lane, it is necessary to establish a lane reference line for guiding the vehicle to implement lane-level navigation, where two lanes connected by the lane reference line are lanes to be switched for driving when the vehicle passes through the intersection. By the method, the connection relation between the first lane and the second lane is established, and accurate connection between different lanes in the intersection is achieved under the condition that high-precision map data support is not available.

Optionally, the system further comprises: a controller; the controller is used for responding to an automatic driving mode starting request and generating a reconstruction instruction of the map of the intersection where the vehicle is located.

In practical application, in order to reduce occupation of hardware resources, lane-level map reconstruction work of the intersection can be omitted under the condition of manual driving. When receiving the automatic driving mode starting request, the intersection map reconstruction work is started to be executed.

In a second aspect, an embodiment of the present application provides a map generating method, including:

Generating road navigation information according to the positioning information and the navigation map;

generating lane-level road structure information comprising a first lane where a vehicle is currently located and at least one second lane where the vehicle is to drive in according to lane lines acquired by a vision sensor; wherein the first lane and the second lane are different lanes connected with an intersection;

determining a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure information and the road navigation information and generating lane-level navigation information;

and establishing a map containing the connection relation between the first lane and the target second lane based on the lane-level navigation information.

In a third aspect, embodiments of the present application provide a vehicle, including: a vehicle body and a power source;

the vehicle body is provided with a memory and a processor;

the memory is used for storing one or more computer instructions;

the processor is configured to execute the one or more computer instructions for performing the steps in the method of the second aspect.

In a fourth aspect, embodiments of the present application provide a computer-readable storage medium, which when executed is capable of implementing the steps in the method of the second aspect.

In the map generation system, the method, the vehicle and the medium provided by the embodiment of the application, the visual sensor is configured in the vehicle, and in the running process of the vehicle, the lane line is collected and processed by the visual sensor to generate a lane-level road structure comprising a first lane where the vehicle is currently located and at least one second lane where the vehicle is to be driven in; and generating road navigation information according to the positioning information and the navigation map. Further, lane-level navigation information is generated in combination with the lane-level road structure and the road navigation information. After the lane-level navigation information is determined, a junction between the first lane and the target second lane is established, thereby enabling the vehicle to perform a driving task according to the junction between the first lane and the target second lane. Through the scheme, when the vehicle passes through the intersection, the lane-level navigation information can be generated based on the generated lane-level road structure and the current road navigation information, so that lane-level accurate navigation of the vehicle is realized, and the connection relationship between the first lane where the vehicle is currently located and the target second lane to be driven in is further established based on the lane-level navigation information. Under the condition that high-precision map support is not needed, lane-level reconstruction of a map with unsmooth road connection is realized, and dependence on the high-precision map and overhead of hardware resources caused by running the high-precision map are avoided.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:

FIG. 1 is a schematic diagram of a map generation system illustrated in the present application;

FIG. 2 is a schematic diagram illustrating a first lane and second lane position relationship according to an embodiment of the present disclosure;

FIG. 3 is a schematic diagram illustrating lane spacing of a first lane and a second lane according to an embodiment of the present application;

FIG. 4 is a schematic diagram illustrating another first lane and second lane position relationship according to an embodiment of the present disclosure;

fig. 5 is a schematic flow chart of a map generating method according to an embodiment of the present application;

FIG. 6 is a schematic diagram of map generation provided by an embodiment of the present application;

fig. 7 is a schematic structural diagram of a path optimization device according to an embodiment of the present application;

fig. 8 is a schematic structural diagram of a vehicle according to an embodiment of the present application.

Detailed Description

In order to enable those skilled in the art to better understand the present invention, the following description will make clear and complete descriptions of the technical solutions according to the embodiments of the present invention with reference to the accompanying drawings.

In some of the flows described in the description of the invention, the claims, and the figures described above, a number of operations occurring in a particular order are included, and the operations may be performed out of order or concurrently with respect to the order in which they occur. The sequence numbers of operations such as 101, 102, etc. are merely used to distinguish between the various operations, and the sequence numbers themselves do not represent any order of execution. In addition, the flows may include more or fewer operations, and the operations may be performed sequentially or in parallel. It should be noted that, the descriptions of "first" and "second" herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequence, and are not limited to the "first" and the "second" being different types.

In the prior art, in an automatic driving application scenario, a high-precision map is required to provide data support when a vehicle performs path planning. However, the data collection of the high-precision map consumes a large amount of manpower and material resources, the data collection is a huge intensive task, and hundreds of data collection vehicles are responsible for collecting source data for making the map so as to ensure that the map is updated quickly every time the road changes. The test vehicle uses various sensors such as global positioning system (Global Positioning System, GPS), inertial sensors (Inertial Measurement Unit, IMU), laser radar, cameras. And (5) fusing the data by collecting various data, and finally generating a high-precision map. The data processing of the high-precision map also consumes a great deal of manpower and material resources, and the data processing refers to how to sort, classify and streamline the collected data when the high-precision map is built offline so as to obtain an initial map template without any semantic information or comments. The high-precision map is operated in the vehicle system, a large amount of hardware resources are required to be occupied, the follow-up planning decision module has high dependence on the off-line high-precision map, calculation resources required by real-time calculation of the vehicle system cannot be met, and the vehicle system is not beneficial to the lightweight development of the vehicle system. Therefore, a map generation scheme capable of replacing a high-definition map and realizing lane-level guidance is required.

The embodiment of the application provides a map generation system. Fig. 1 is a schematic diagram of a map generating system illustrated in the present application. As can be seen from fig. 1, the system comprises: the system comprises a road structure generation module, a map topological structure generation module, a lane navigation generation module, a map generation module and the like. In particular the number of the elements,

the road structure generation module 11 is used for generating a lane-level road structure comprising a first lane where the vehicle is currently located and at least one second lane where the vehicle is to drive in according to the lane lines acquired by the vision sensor; the first lane and the second lane are different lanes connected with the intersection.

The map topology generation module 12 is configured to generate road navigation information according to the positioning information and the navigation map.

A lane navigation generating module 13, configured to determine a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure and the road navigation information, and generate lane-level navigation information.

The map generating module 14 is configured to establish a map that includes a linking relationship between the first lane and the target second lane based on the lane-level navigation information.

In practical application, lanes are drawn on roads, but no lanes are drawn at intersections such as crossroads or T-junctions, and only individual intersections may have left turn guide lanes, but the left turn guide lanes extend to the center of the intersection and are not connected with any lanes to be driven. In addition, complex road conditions such as lane increase, decrease, widening, narrowing and the like can appear in some road junction links, and the corresponding relation between lanes in different roads which are simultaneously linked with the road junction is difficult to accurately determine by a vision sensor.

When a vehicle passes through an intersection, there may be multiple roads all engaged with the intersection. Each lane is divided by lane lines in each road connected with the intersection, however, because the widths of the different roads connected with the intersection are different, the number of the lanes and the widths of the lanes of the road are not completely the same, and the corresponding relation of each lane in the different roads connected with the intersection cannot be established or is difficult to establish one by one. Therefore, the road structure generating module can obtain a required lane-level road structure, specifically, when the lane-level road structure is constructed, roads in multiple directions connected with the intersection can be constructed, for example, when the road structure passes through the intersection, the collected lanes comprise a first lane, multiple second lanes, multiple third lanes and multiple fourth lanes; of course, some intersections are complex, not going right north and south, or there are more roads. However, without high-precision map support, only the lane-level road structure has no vehicle current position information and vehicle traveling direction information, which cannot meet the requirement of accurate navigation of the vehicle. Therefore, the map topological structure generation module is further utilized to generate road navigation information according to the positioning information and the conventional navigation map, so that a first lane where the vehicle is currently located and a target road which is to be driven into and comprises at least one second lane can be accurately known, and the target road is judged according to the driving direction. And further, the lane-level navigation information comprising the connection relation between the first lane and the target second lane is generated by utilizing the lane navigation generation module and the map generation module, so that the lane-level navigation information with higher accuracy is obtained, and reliable high-accuracy navigation data can be provided for functions such as automatic driving and auxiliary driving without depending on a high-accuracy map.

In one or more embodiments of the present application, the road structure generating module 11 is further configured to:

In turn, a lane-level road structure may be generated based on the reconstructed first lane and the at least one second lane. During the running process of the vehicle, the visual sensor is utilized to collect the current environment image, in particular to collect and identify the lane line image on the road surface. Of course, in order to reduce the occupation of computing resources, the acquisition of images of each lane line of the road surface can be started according to the navigation information. For example, when it is known that the vehicle will travel to the intersection according to the navigation information, an instruction for acquiring an intersection road surface image is sent to the vision sensor, and the calculation unit is controlled to perform calculation processing (for example, image recognition is performed by using a deep neural network model and reconstruction is performed) on the acquired intersection road surface image, so as to obtain a local map image of the intersection, wherein the map image not only includes a first lane and a plurality of second lanes, but also includes other roads and lanes, and then a target second lane corresponding to the first lane needs to be selected from the plurality of lanes, so as to provide basic data for creating a map including lane-level navigation information.

Fig. 2 is a schematic diagram illustrating a positional relationship between a first lane and a second lane according to an embodiment of the present application. As can be seen from fig. 2, the definition of the lane lines in the image acquired by the vision sensor is uneven, and the lane lines in the intersection are interrupted, that is, no lane line connection exists between lanes in different directions. Therefore, it is necessary to reconstruct the lane lines acquired by the vision sensor. Specifically, a first lane is reconstructed based on a first current lane line and a second current lane line of a lane where the acquired vehicle is located currently. When reconstructing the first lane, it is necessary to simultaneously establish a first current lane extension line corresponding to the first current lane line and a second current lane extension line corresponding to the second current lane line, so that the relative positional relationship between the first lane and the second lane can be determined according to the extension lines. As shown in fig. 2, the extension line of the lane L1 extends to the road including the lane L5, and one lane line of the lane L5 can be seen to be opposed to the lane L1. By the lane reconstruction mode, the position relationship and the opposite-entering relationship between the first lane and the second lane which do not have the lane line connection relationship are clearer, so that the first lane on which the vehicle is currently running and the target second lane to be driven in can be conveniently matched later.

In one or more embodiments of the present application, the lane navigation generating module 13 is further configured to:

A schematic diagram of the lane separation of the first lane and the second lane as illustrated in the embodiment of fig. 3. As can be seen from fig. 3, in order to make the width of the road and the number of lanes included in the road in different directions not be fixed in practical application, it is impossible to ensure that the number of lanes and the width of the lanes included in the road joined to the same intersection are identical. Thus, there is a need for an accurate match for a first lane in which a vehicle is currently located to a target second lane into which the vehicle is to be driven. In particular, the target second lane corresponding to the first lane may be determined in such a way that the minimum traverse distance is met and the desired width of the vehicle is met, so that the vehicle can achieve lane change simply and quickly. Therefore, the method for setting the distance threshold helps accurately select the target second lane with the minimum span distance from the roads comprising the second lanes, so that the vehicle can be successfully guided to drive into the target second lane from the first lane when passing through the intersection.

For example, as shown in fig. 3, assuming that the first lane distance a1 and the second lane distance a2 are set to 1 meter. Assuming that a1 is greater than 1 meter and a2 is less than 1 meter, determining the lane corresponding to the first lane distance as a target second lane matching the first lane, in other words, determining the lane in which the lane line for measuring the first lane distance is located as the target second lane. Because the lateral movement distance is minimized when the vehicle is driven from the first lane into the second lane, a situation that straddles multiple lanes does not occur.

comparing the first lane spacing to the second lane spacing;

In practical applications, the width of the road in different directions and the number of lanes contained in the road are not fixed, and it is impossible to ensure that the number of lanes and the width of the lanes contained in the different roads connected with the same intersection are completely the same. Thus, there is a need for an accurate match for a first lane in which a vehicle is currently located to a target second lane into which the vehicle is to be driven. In particular, the target second lane corresponding to the first lane may be determined in such a way that the minimum traverse distance is met and the desired width of the vehicle is met, so that the vehicle can achieve lane change simply and quickly. The present application therefore proposes a way of comparing the first lane spacing with the second lane spacing to assist in accurately selecting a target second lane of minimum span spacing from a road containing multiple lanes so as to be able to successfully guide a vehicle into the second lane.

In one or more embodiments of the present application, if the corresponding target second lane to be driven in is not determined according to the first lane distance and the second lane distance;

The lane navigation generating module is further configured to:

When the lane matching is performed, because the widths of different roads and the number of the included lanes are different, each first lane cannot be ensured to be accurately matched with a proper second lane sometimes according to the lane interval matching mode. When the two ways cannot be matched with the target second lane, according to the position of the first lane in the current road, selecting an outermost or innermost lane from the roads which are to be driven into and comprise a plurality of second lanes as the target second lane. Therefore, the target second vehicle road determined by the scheme can be used as a lane connected with the first vehicle road, so that the lane-level navigation information is generated by combining the road navigation information, and the lane-level navigation information can be simply and efficiently generated.

For example, fig. 4 is a schematic diagram illustrating another positional relationship between a first lane and a second lane according to an embodiment of the present application. As can be seen from fig. 4, assuming that the lane L1 has been matched to the lanes L4, L5, when the lanes L2 are matched to the first lane L2, since the lanes L6, L7 have a vehicle occupation, the desired target second lane cannot be matched to the lane L2 in the above manner, and at this time, the outermost lane L8 or L9 may be matched to the lane L2 as the target second lane. Similarly, if the lane L1 does not match the lane, the innermost lane L4 is matched to the lane L1. By the method, each first lane can be matched with the second lane, and lane-level navigation information can be built efficiently.

The positioning information can be the position information of the current vehicle obtained based on GPRS or Beidou navigation and the like, and the position of the vehicle in the map can be known by further combining with the navigation map, so that road navigation information is generated based on the positioning information and the navigation map. However, the first lane where the vehicle is currently located cannot be accurately determined simply by the road navigation information, and the lane-level road structure provided by the road structure generation module needs to be further combined. From the lane-level road structure, the first lane in which the vehicle is currently located can be identified, and the lane-level road structure is combined with the road navigation information, so that the vehicle can be known on which lane presented in the navigation map. The road navigation information includes information such as the vehicle traveling speed and traveling direction, and further, it is possible to know which lane the vehicle is about to travel in. That is, the lane-level road structure is combined with the road navigation information by using the lane navigation generating module to generate lane-level navigation information, through which it is known which first lane the vehicle is currently on and which road containing the target second lane is to be driven on, and then the target second lane matching the first lane is selected from the roads containing at least one lane, so as to generate lane-level road navigation information.

In one or more embodiments of the present application, the map generation module is further configured to: and establishing a map containing a lane reference line for guiding the vehicle to realize lane-level navigation based on the corresponding relation between the first lane and the second lane.

In practical applications, after determining the first lane and the target second lane, it is necessary to establish a lane reference line for guiding the vehicle to implement lane-level navigation, where two lanes connected by the lane reference line are lanes on which the vehicle is to travel. Through the mode, the connection relation between the first lane and the target second lane is established, and accurate connection between different lanes in the intersection is achieved under the condition that high-precision map data support is not available.

Under some circumstances, when the accurate engagement relationship between different lanes is established through the lane reference line, the road condition on the current road and the positions and the running speeds of other vehicles are comprehensively considered. In particular, when the lane is reduced or the road is narrowed, there is often a situation that the vehicles are in parallel, that is, other vehicles can adjust the speed (for example, accelerate or decelerate) and reach the current first lane where the vehicles are located or the second lane where the vehicles are to drive in, so that the influence of the other vehicles is fully considered when selecting the second lane, and collision is avoided.

In one or more embodiments of the present application, the system further comprises: a controller; the controller is used for responding to an automatic driving mode starting request and generating a reconstruction instruction of the map of the intersection where the vehicle is located.

Based on the same thought, the embodiment of the application provides a map generation method. Fig. 5 is a schematic flow chart of a map generating method according to an embodiment of the present application. As can be seen from fig. 5, the method comprises the steps of:

step 501: and generating road navigation information according to the positioning information and the navigation map.

Step 502: generating lane-level road structure information comprising a first lane where a vehicle is currently located and at least one second lane where the vehicle is to drive in according to lane lines acquired by a vision sensor; the first lane and the second lane are different lanes connected with the intersection.

Step 503: and determining a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure information and the road navigation information and generating lane-level navigation information.

Step 504: and establishing a map containing the connection relation between the first lane and the target second lane based on the lane-level navigation information.

Through the scheme, when the vehicle passes through the intersection, the lane-level navigation information can be generated based on the generated lane-level road structure and the current road navigation information, so that lane-level accurate navigation of the vehicle is realized, and the connection relationship between the first lane where the vehicle is currently located and the target second lane to be driven in is further established based on the lane-level navigation information. Under the condition that high-precision map support is not needed, lane-level reconstruction of a map with unsmooth road connection is realized, and dependence on the high-precision map and overhead of hardware resources caused by running the high-precision map are avoided.

Optionally, the method further comprises: receiving a first current lane line and a second current lane line which are acquired by the vision sensor and are positioned on two sides of a lane where a vehicle is currently positioned, and a first target lane line and a second target lane line which are positioned on two sides of the lane where the vehicle is to drive in;

Optionally, the method further comprises: determining a target lane line opposite to the first lane according to the position relation between the first target lane line and the second target lane line and the first lane;

Comparing the first lane spacing to the second lane spacing;

If the corresponding target second vehicle road to be driven in is not determined according to the first vehicle road distance and the second vehicle road distance; optionally, the method further comprises: selecting an outermost lane or an innermost lane from among the roads including a plurality of lanes as the target second lane according to the position of the first lane in the current road;

Optionally, the method further comprises: and establishing a map containing a lane reference line for guiding the vehicle to realize lane-level navigation based on the corresponding relation between the first lane and the second lane.

Optionally, the method further comprises: and responding to the automatic driving mode starting request, and generating a reconstruction instruction of the map of the intersection where the vehicle is located.

In order to facilitate understanding, the following description will exemplify the present application through specific examples. Fig. 6 is a schematic diagram of map generation according to an embodiment of the present application.

The visual sensor senses the road environment by using a camera, and outputs lane lines to the road structure generation module after calculation by using a neural network; the road structure generation module can reconstruct the distribution situation of the local lanes of the front 200m and the rear 50m of the vehicle at the current moment according to the reference line.

When passing through an intersection, the difficulty is that a road meeting the navigation requirement is selected from a plurality of subsequent roads behind the intersection, and an automatic driving automobile is expected to walk along a navigation path, so that navigation information is expected to be acquired to generate a lane reference line consistent with navigation.

The map topological structure generation module combines a global navigation map and positioning information, can provide road-level navigation information, and can guide an automatic driving automobile to turn left or right or go straight.

For an automatic driving car, it is far from enough to know which direction to drive next, and for better experience, a lane navigation generation module is required to generate lane-level navigation information, i.e. which lane to drive next at an intersection, from the lane-level road structure and the road navigation information.

The reference line in the automatic driving system provides a driving path expected by the automatic driving automobile and is critical to the planning and control module; after the road level navigation is obtained, since there is no lane line at the intersection, a local reference line generation module is required to generate a smooth lane reference line usable for vehicle control according to the upstream road level navigation information.

Specifically, when the lane-level road structure is generated, it is possible to confirm which road the vehicle needs to travel to the right and left at the intersection, based on the matching with the lane-level navigation information.

However, since there are often multiple lanes on a road, in order to get a better driving experience, it is necessary to find the lane on which driving is expected to follow according to the following rules:

(1) If the width of the overlapping area of the current lane extension surface of the vehicle and the lane to be driven in is more than or equal to 1m, establishing the lane connection relation.

(2) After the mode (1), the exit lanes without lanes are left, and the lane connection relation is established with the outermost side or the innermost side.

After being processed by the lane navigation generating module, the target lane expected to be driven is obtained, but the lane reference line at the intersection is not obtained; the map generation module is required to generate a smooth lane reference line usable for vehicle control from the upstream information.

In the map generation module, a smooth, controllable reference line is generated using a Bezier curve. For example, an intersection reference line is generated using a 3-order bezier curve, and a control point in the middle of the road is generated according to the end point of the current lane, the position of the start point of the target lane, and the angle.

Based on the same thought, the embodiment of the application also provides a map generation device. Fig. 7 is a schematic structural diagram of a path optimization device according to an embodiment of the present application. As can be seen from fig. 7, the device comprises:

the first generation module 71 is configured to generate road navigation information according to the positioning information and the navigation map.

A second generation module 72, configured to generate lane-level road structure information including a first lane in which the vehicle is currently located and at least one second lane into which the vehicle is to travel according to the lane lines acquired by the vision sensor; the first lane and the second lane are different lanes connected with the intersection.

The third generating module 73 is configured to determine a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure information and the road navigation information, and generate lane-level navigation information.

The relationship establishing module 74 is configured to establish a map that includes a joining relationship between the first lane and the target second lane based on the lane-level navigation information.

Fig. 8 is a schematic structural diagram of a vehicle according to an embodiment of the present application, where, as shown in fig. 8, a vehicle device is configured on the vehicle, and the vehicle device includes: a memory 801 and a controller 802.

The memory 801 is used for storing a computer program and may be configured to store other various data to support operations on the vehicle device. Examples of such data include instructions for any application or method operating on the vehicular device, contact data, phonebook data, messages, pictures, videos, and the like.

The Memory 801 may be implemented by any type or combination of volatile or non-volatile Memory devices, such as Static Random-Access Memory (SRAM), electrically erasable programmable Read-Only Memory (Electrically Erasable Programmable Read Only Memory, EEPROM), erasable programmable Read-Only Memory (Electrical Programmable Read Only Memory, EPROM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk, or optical disk.

The vehicle apparatus further includes: and a display device 803. A controller 802 coupled to the memory 801 for executing a computer program in the memory 801 for:

and generating road navigation information according to the positioning information and the navigation map.

Generating lane-level road structure information comprising a first lane where a vehicle is currently located and at least one second lane where the vehicle is to drive in according to lane lines acquired by a vision sensor; the first lane and the second lane are different lanes connected with the intersection.

And determining a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure information and the road navigation information and generating lane-level navigation information.

Optionally, the controller 802 is further configured to receive the first current lane line and the second current lane line, which are acquired by the visual sensor and are located on two sides of the lane where the vehicle is currently located, and the first target lane line and the second target lane line, which are located on two sides of the lane where the vehicle is to be driven in;

Optionally, the controller 802 is further configured to determine a target lane line opposite to the first lane according to the positional relationship between the first target lane line and the second target lane line and the first lane;

Comparing the first lane spacing to the second lane spacing;

Optionally, the controller 802 is further configured to establish a map including a lane reference line for guiding the vehicle to implement lane-level navigation based on the correspondence between the first lane and the second lane.

Optionally, the controller 802 is further configured to generate a reconstruction instruction for the map of the intersection where the vehicle is located in response to the automatic driving mode initiation request.

The display device 803 in fig. 8 described above includes a screen, which may include a Liquid Crystal Display (LCD) and a Touch Panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from a user. The touch panel includes one or more touch sensors to sense touches, swipes, and gestures on the touch panel. The touch sensor may sense not only the boundary of a touch or slide action, but also the duration and pressure associated with the touch or slide operation.

The audio component 804 of fig. 8 above may be configured to output and/or input audio signals. For example, the audio component includes a Microphone (MIC) configured to receive external audio signals when the device in which the audio component is located is in an operational mode, such as a call mode, a recording mode, and a speech recognition mode. The received audio signal may be further stored in a memory or transmitted via a communication component. In some embodiments, the audio assembly further comprises a speaker for outputting audio signals.

Further, as shown in fig. 8, the vehicle apparatus further includes: communication component 805, power component 806, and the like. Only part of the components are schematically shown in fig. 8, which does not mean that the vehicle device only comprises the components shown in fig. 3.

The communication component 805 in fig. 8 described above is configured to facilitate communication between the device in which the communication component is located and other devices, either in a wired or wireless manner. The device in which the communication component is located may access a wireless network based on a communication standard, such as WiFi,2G, 3G, 4G, or 5G, or a combination thereof. In one exemplary embodiment, the communication component may be implemented based on near field communication (Near Field Communication, NFC) technology, radio frequency identification (Radio Frequency Identification, RFID) technology, infrared data association (Infrared Data Association, irDA) technology, ultra Wideband (UWB) technology, bluetooth technology, and other technologies.

Wherein the power supply module 806 provides power to various components of the device in which the power supply module is located. The power components may include a power management system, one or more power sources, and other components associated with generating, managing, and distributing power for the devices in which the power components are located.

Accordingly, embodiments of the present application also provide a computer readable storage medium, which when executed is capable of implementing the steps in the method embodiment of fig. 5.

In the embodiment of the application, a visual sensor is configured in a vehicle, and a lane-level road structure comprising a first lane and a second lane is generated after a lane line is acquired and processed by the visual sensor in the running process of the vehicle; and generating road navigation information according to the positioning information and the navigation map. Further, lane-level navigation information is generated in combination with the lane-level road structure and the road navigation information. After the lane-level navigation information is determined, a junction between the first lane and the second lane is established, thereby enabling the vehicle to perform a driving task according to the junction between the first lane and the second lane. Through the scheme, when the vehicle passes through the intersection, the lane-level navigation information can be generated based on the generated lane-level road structure and the current road navigation information, so that lane-level accurate navigation of the vehicle is realized, and the connection relationship between the first lane where the vehicle is currently located and the second lane to be driven in is further established based on the lane-level navigation information. Under the condition that high-precision map support is not needed, lane-level reconstruction of a map with unsmooth road connection is realized, and dependence on the high-precision map and overhead of hardware resources caused by running the high-precision map are avoided.

It should be appreciated by those skilled in the art that embodiments of the invention may be provided as a method, system, or computer readable storage medium. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the invention may take the form of a computer-readable storage medium embodied in one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.

These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.

These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.

In one typical configuration, a computing device includes one or more processors (CPUs), input/output interfaces, network interfaces, and memory. The memory may include volatile memory in a computer-readable medium, random Access Memory (RAM) and/or nonvolatile memory, such as Read Only Memory (ROM) or flash memory (flash RAM). Memory is an example of computer-readable media.

The foregoing is merely exemplary of the present application and is not intended to limit the present application. Various modifications and changes may be made to the present application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc. which are within the spirit and principles of the present application are intended to be included within the scope of the claims of the present application.

Claims

1. A map generation system, the system comprising:

the road structure generation module is used for generating a lane-level road structure comprising a first lane where the vehicle is currently located and at least one second lane where the vehicle is to enter according to the lane lines acquired by the vision sensor; wherein the first lane and the second lane are different lanes connected with an intersection; also used for: receiving a first current lane line and a second current lane line which are acquired by the vision sensor and are positioned on two sides of a lane where a vehicle is currently positioned, and a first target lane line and a second target lane line which are positioned on two sides of the lane where the vehicle is to drive in; reconstructing and generating a first lane based on the first current lane line and the second current lane line; the first lane comprises a first current lane extension line and a second current lane extension line; reconstructing and generating a second lane based on the first target lane line and the second target lane line;

A lane navigation generation module for determining a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure and the road navigation information and generating lane-level navigation information; also used for: determining a target lane line opposite to the first lane according to the position relation between the first target lane line and the second target lane line and the first lane; determining a first lane distance between the target lane line and the first current lane extension; and a second lane-distance between the target lane-line and the second current lane-extension; if the first lane distance or the second lane distance is larger than a distance threshold, determining a lane corresponding to the first lane distance or the second lane distance as a target second lane; generating the lane-level navigation information based on the first lane, the target second lane, and the road navigation information;

2. The system of claim 1, wherein if the corresponding target second lane to be driven in is not determined based on the first lane spacing and the second lane spacing;

The lane navigation generating module is further configured to:

3. The system of claim 2, wherein the map generation module is further configured to:

and establishing a map containing a lane reference line for guiding the vehicle to realize lane-level navigation based on the corresponding relation between the first lane and the second lane.

4. The system of claim 1, wherein the system further comprises: a controller;

the controller is used for responding to an automatic driving mode starting request and generating a reconstruction instruction of the map of the intersection where the vehicle is located.

5. A map generation method, the method comprising:

generating lane-level road structure information comprising a first lane where a vehicle is currently located and at least one second lane where the vehicle is to drive in according to lane lines acquired by a vision sensor; wherein the first lane and the second lane are different lanes connected with an intersection; further comprises: receiving a first current lane line and a second current lane line which are acquired by the vision sensor and are positioned on two sides of a lane where a vehicle is currently positioned, and a first target lane line and a second target lane line which are positioned on two sides of the lane where the vehicle is to drive in; reconstructing and generating a first lane based on the first current lane line and the second current lane line; the first lane comprises a first current lane extension line and a second current lane extension line; reconstructing and generating a second lane based on the first target lane line and the second target lane line;

Determining a target second lane corresponding to the first lane from the at least one second lane based on the lane-level road structure information and the road navigation information and generating lane-level navigation information; the method specifically comprises the following steps: determining a target lane line opposite to the first lane according to the position relation between the first target lane line and the second target lane line and the first lane; determining a first lane distance between the target lane line and the first current lane extension; and a second lane-distance between the target lane-line and the second current lane-extension; if the first lane distance or the second lane distance is larger than a distance threshold, determining a lane corresponding to the first lane distance or the second lane distance as a target second lane; generating the lane-level navigation information based on the first lane, the target second lane, and the road navigation information;

6. A vehicle, characterized by comprising: a vehicle body and a steer-by-wire system;

The vehicle body is provided with a memory and a processor;

the memory is used for storing one or more computer instructions;

the processor is configured to execute the one or more computer instructions for performing the steps in the method of claim 5.

7. A computer readable storage medium, which when executed is capable of implementing the steps in the method of claim 5.