自动驾驶地图的构建方法及相关装置

Construction method of automatic driving map and related device

文档序号：5560 发布日期：2021-09-17 浏览：43次中文

1. A method for constructing an automatic driving map is characterized by comprising the following steps:

acquiring road information, intersection information and lane information of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

acquiring road traffic direction information according to the manual driving track data and the road information, and acquiring lane traffic direction information according to the lane information and the road traffic direction information;

acquiring intersection access point information according to the intersection information and the lane passing direction information;

and executing operation of generating a virtual topological center line according to the intersection access point information to obtain an automatic driving map of an area where the vehicle runs, wherein the virtual topological center line is a boundary line of the vehicle running in the intersection area.

2. The method according to claim 1, wherein the obtaining road information, intersection information and lane information of an area through which a vehicle travels according to the manual driving track data and/or the obstacle grid map comprises:

acquiring a road boundary of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

obtaining a road center line of an area where the vehicle runs according to the road boundary;

acquiring a crossing area of an area where the vehicle runs according to the road boundary and the road center line;

obtaining the lane boundary of the area where the vehicle runs according to the road boundary, the road center line and the intersection area;

and obtaining the lane central line of the area which is driven by the vehicle according to the lane boundary.

3. The method according to claim 2, wherein the obtaining of the road boundary of the area traveled by the vehicle from the manual driving trajectory data and/or the obstacle grid map comprises:

obtaining a plurality of track point buffer areas according to the manual driving track data and/or the barrier grid diagram;

performing fusion operation on the plurality of track point buffer areas to obtain a road buffer area;

and determining the boundary of the road buffer area as the road boundary of the area which is driven by the vehicle.

4. The method according to claim 2 or 3, wherein the obtaining a road center line of an area traveled by the vehicle according to the road boundary comprises:

performing equidistant shape point operation on the road boundary to obtain the road boundary after equidistant shape points;

obtaining a first Thiessen polygon edge set according to the road boundary after the shape points are equally spaced;

executing Thiessen polygon edge operation outside the road boundary after deleting the equidistant shape points on the first Thiessen polygon edge set to obtain a second Thiessen polygon edge set;

obtaining a buffer area of the road boundary according to the road boundary and a pre-stored first width;

obtaining a third Thiessen polygon edge set according to Thiessen polygon edges in the second Thiessen polygon edge set, wherein the Thiessen polygon edges are not intersected with the buffer area of the road boundary;

executing a suspension line deleting operation on the third Thiessen polygon edge set to obtain a fourth Thiessen polygon edge set;

and performing connection operation on the fourth Thiessen polygon edge set to obtain the road center line of the area where the vehicle runs.

5. The method of any of claims 2-4, wherein said obtaining an intersection area of an area traveled by said vehicle based on said roadway boundary and said roadway centerline comprises:

obtaining the central point of each intersection according to the road center line;

determining a circle with the center point of each intersection as a circle center and a pre-stored first length as a radius as a center point buffer area of each intersection;

executing operation of deleting sub-road center lines with lengths smaller than the pre-stored second length on each sub-road center line to obtain each road section center line, wherein each sub-road center line is the remaining road center line except the center point buffer area of each intersection in the road center line;

obtaining a buffer area of the center line of each road section according to the center line of each road section and a prestored second width;

and executing operation of deleting sub-intersection areas with the area smaller than a pre-stored first area for each sub-intersection area to obtain intersection areas of areas through which the vehicles drive, wherein each sub-intersection area is the remaining intersection area except the center line buffer area of each road section in the surface formed by the road boundary.

6. The method of any of claims 2-5, wherein said obtaining lane boundaries for an area traveled by said vehicle based on said road boundaries, said road centerline and said intersection area comprises:

obtaining each road section boundary except the intersection area in the road boundary;

obtaining the central line of each road section except the intersection area in the central line of the road;

judging whether each road section is a single-lane section or a double-lane section according to each road section boundary and each road section central line, wherein each road section is in one-to-one correspondence with each road section boundary and each road section central line;

if the road section A is a single-lane section, determining a road section boundary corresponding to the road section A as a lane boundary of the single-lane section which is driven by the vehicle, wherein the road section A is any one of all the single-lane sections included in each road section;

and if the road section B is a double-lane section, determining a road section boundary and a road section center line corresponding to the road section B as the lane boundary of the double-lane section traveled by the vehicle, wherein the road section B is any one of all the double-lane sections included in each road section.

7. The method according to any one of claims 2-6, wherein the obtaining road traffic direction information from the artificial driving trajectory data and the road information comprises:

obtaining each first relative angle difference set of the artificial driving track and each road section according to the artificial driving track data and the road center line, wherein each first relative angle difference set corresponds to each road section one by one;

if the variance of a first relative angle difference set C is smaller than a pre-stored first angle, determining that the road section corresponding to the first relative angle difference set C passes in one way, wherein the first relative angle difference set C is any one of all first relative angle difference sets of which the variance is smaller than the first angle and which are included in each first relative angle difference set;

if the mean value of the first relative angle difference set C is smaller than a prestored second angle, determining that the road passing direction of the road section corresponding to the first relative angle difference set C is forward passing;

and if the mean value of the first relative angle difference set C is larger than or equal to the second angle, determining that the road passing direction of the road section corresponding to the first relative angle difference set C is reverse passing.

8. The method of claim 7, further comprising:

if the variance of a first relative angle difference set D is larger than or equal to the first angle, determining that the road section corresponding to the first relative angle difference D passes in two directions, wherein the first relative angle difference set D is any one of all first relative angle difference sets of which the variance is larger than or equal to the first angle and included in each first relative angle difference set.

9. The method of claim 8, wherein obtaining lane traffic direction information from the lane information and the road traffic direction information comprises:

if the road section E obtained according to the lane information and the road traffic direction information is a bidirectional traffic double-lane section, according to two second relative angle difference sets of a road section center line corresponding to the road section E and two lane center lines of the road section E, the road section E is any one of all bidirectional traffic double-lane sections included in each road section;

if the mean value of a second relative angle difference set F is smaller than a prestored third angle, judging whether a lane central line corresponding to the second relative angle difference set F is intersected with a right buffer area of a road section central line corresponding to the road section E, wherein the second relative angle difference set F is any one of the two second relative angle difference sets; if so, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is a forward passing direction; if not, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is reverse passing;

if the mean value of a second relative angle difference set G is larger than or equal to the third angle, judging whether the lane central line corresponding to the second relative angle difference set G is intersected with the left buffer area of the road section central line corresponding to the road section E, wherein the second relative angle difference set G is any one of the two second relative angle difference sets; if so, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is a forward passing direction; if not, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is reverse passing.

10. The method of claim 9, further comprising:

if a road section H obtained according to the lane information and the road traffic direction information is a bidirectional traffic single-lane section, determining that the lane traffic direction of the road section H is bidirectional traffic, wherein the road section H is any one of all bidirectional traffic single-lane sections included in each road section;

and if the road section I obtained according to the lane information and the road passing direction information is a one-way passing single-lane section, determining the road passing direction of the road section I as the lane passing direction of the road section I, wherein the road section I is any one of all the one-way passing single-lane sections included in each road section.

11. The method of claim 10, wherein obtaining intersection access point information from the intersection information and the lane traffic direction information comprises:

obtaining a starting shape point and a stopping shape point of a lane central line J, wherein the lane central line J is any one of all lane central lines included in each road section;

obtaining a buffer area of each intersection area according to each intersection area and a prestored third width;

if a shape point K intersects with a buffer area of an intersection area L, determining that the shape point L belongs to the intersection area L, wherein the shape point K is any one of a starting shape point and a stopping shape point of the lane central line J, and the intersection area L is any one of the intersection areas;

if the lane passing direction of the lane where the lane central line J is located is forward passing, determining the initial shape point of the lane central line J as the ending shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the initial shape point of the virtual topological central line;

if the lane passing direction of the lane where the lane central line J is located is reverse passing, determining the starting shape point of the lane central line J as the starting shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line;

if the lane passing direction of the lane where the lane central line J is located is bidirectional passing, determining the starting shape point of the lane central line J as the starting shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line; or determining the starting shape point of the lane central line J as the ending shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the starting shape point of the virtual topological central line.

12. An automatic driving map construction device, comprising:

the acquiring unit is used for acquiring road information, intersection information and lane information of an area where the vehicle runs according to the manual driving track data and/or the barrier grid map;

the obtaining unit is also used for obtaining road traffic direction information according to the manual driving track data and the road information;

the obtaining unit is further used for obtaining lane passing direction information according to the lane information and the road passing direction information;

the obtaining unit is also used for obtaining the information of the access point of the intersection according to the intersection information and the lane passing direction information;

and the execution unit is used for executing the operation of generating a virtual topological center line according to the intersection access point information to obtain an automatic driving map of the area where the vehicle runs, wherein the virtual topological center line is a boundary line of the vehicle running in the intersection area.

13. The apparatus according to claim 12, wherein in obtaining road information, intersection information and lane information of an area through which the vehicle travels from the manual driving trajectory data and/or the obstacle grid map, the obtaining unit is specifically configured to:

acquiring a road boundary of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

obtaining a road center line of an area where the vehicle runs according to the road boundary;

acquiring a crossing area of an area where the vehicle runs according to the road boundary and the road center line;

obtaining the lane boundary of the area where the vehicle runs according to the road boundary, the road center line and the intersection area;

and obtaining the lane central line of the area which is driven by the vehicle according to the lane boundary.

14. The device according to claim 13, characterized in that, in obtaining the road boundary of the area traveled by the vehicle from the manual driving trajectory data and/or the obstacle grid map, the obtaining unit is specifically configured to:

obtaining a plurality of track point buffer areas according to the manual driving track data and/or the barrier grid diagram;

performing fusion operation on the plurality of track point buffer areas to obtain a road buffer area;

and determining the boundary of the road buffer area as the road boundary of the area which is driven by the vehicle.

15. The apparatus according to claim 13 or 14, characterized in that, in obtaining the road center line of the area traveled by the vehicle from the road boundary, the obtaining unit is specifically configured to:

performing equidistant shape point operation on the road boundary to obtain the road boundary after equidistant shape points;

obtaining a first Thiessen polygon edge set according to the road boundary after the shape points are equally spaced;

executing Thiessen polygon edge operation outside the road boundary after deleting the equidistant shape points on the first Thiessen polygon edge set to obtain a second Thiessen polygon edge set;

obtaining a buffer area of the road boundary according to the road boundary and a pre-stored first width;

executing a suspension line deleting operation on the third Thiessen polygon edge set to obtain a fourth Thiessen polygon edge set;

and performing connection operation on the fourth Thiessen polygon edge set to obtain the road center line of the area where the vehicle runs.

16. The apparatus according to any one of claims 13 to 15, characterized in that, in obtaining the intersection area of the area traveled by the vehicle from the road boundary and the road centerline, the obtaining unit is specifically configured to:

obtaining the central point of each intersection according to the road center line;

determining a circle with the center point of each intersection as a circle center and a pre-stored first length as a radius as a center point buffer area of each intersection;

obtaining a buffer area of the center line of each road section according to the center line of each road section and a prestored second width;

17. The apparatus according to any one of claims 13 to 16, characterized in that, in obtaining the lane boundary of the area through which the vehicle travels from the road boundary, the road center line, and the intersection area, the obtaining unit is specifically configured to:

obtaining each road section boundary except the intersection area in the road boundary;

obtaining the central line of each road section except the intersection area in the central line of the road;

18. The device according to claims 13-17, characterized in that, in obtaining road traffic direction information from the artificial driving trajectory data and the road information, the obtaining unit is specifically configured to:

19. The apparatus according to claim 18, wherein, in obtaining road traffic direction information from the artificial driving trajectory data and the road information, the obtaining unit is specifically configured to:

20. The apparatus according to claim 19, characterized in that, in obtaining lane traffic direction information from the lane information and the road traffic direction information, the obtaining unit is specifically configured to:

21. The apparatus according to claim 20, characterized in that, in obtaining lane traffic direction information from the lane information and the road traffic direction information, the obtaining unit is specifically configured to:

22. The apparatus according to claim 21, wherein in obtaining intersection entry and exit point information from the intersection information and the lane traffic direction information, the obtaining unit is specifically configured to:

obtaining a starting shape point and a stopping shape point of a lane central line J, wherein the lane central line J is any one of all lane central lines included in each road section;

obtaining a buffer area of each intersection area according to each intersection area and a prestored third width;

23. An apparatus for constructing an autopilot map, the apparatus comprising a processor, a communication interface, and a memory coupled to one another, wherein:

the processor is used for acquiring road information, intersection information and lane information of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

the processor is further used for obtaining road passing direction information according to the manual driving track data and the road information and obtaining lane passing direction information according to the lane information and the road passing direction information;

the processor is further used for obtaining intersection access point information according to the intersection information and the lane passing direction information;

the processor is further configured to execute an operation of generating a virtual topological center line according to the intersection access point information to obtain an automatic driving map of an area where the vehicle travels, where the virtual topological center line is a boundary line where the vehicle travels in the intersection area.

24. A computer-readable storage medium, characterized in that a computer program is stored, which is executed by hardware to implement the method of any one of claims 1 to 11 executed by the automatic driving map construction apparatus.

25. A chip system, characterized in that the chip system comprises a processor for supporting an automatic driving map construction apparatus to implement the method of any one of claims 1 to 11.

Background

At present, the process of constructing the barrier grid map is as follows: the moving posture of the vehicle and the vehicle periphery image are acquired and landmark information is extracted from the vehicle periphery image. And constructing an SLAM algorithm based on the instant positioning and the map, and generating a landmark map and a vehicle running track according to the motion attitude and the landmark information of the vehicle. And detecting a travelable area and constructing an obstacle grid map according to the vehicle travel track and the detected travelable area. However, the accuracy of the constructed barrier grid map is low, the automatic passenger-replacing parking scene with complex rules and extreme perception conditions cannot be met, and the guiding information meeting the driving rules cannot be provided for the vehicle.

Generally, the construction process of an indoor map of a building is: acquiring a moving track of a moving object; and constructing an indoor map of the building by recording and correcting the moving track of the moving object. However, the accuracy of the constructed indoor map of the building is low, the road boundary, the lane boundary, the road traffic direction information and the lane traffic direction information cannot be provided, and the grid map generated in the prior art cannot provide guidance information meeting the driving rule for the vehicle, and cannot meet the automatic pick-up parking scene.

The existing topological map construction process comprises the following steps: acquiring a vehicle running track; extracting key points of a vehicle running track; and constructing a topological map based on all the extracted key points. However, the topological map cannot meet various scenes due to the fact that the vehicle running track has large randomness, the accuracy of the constructed topological map is low, and road boundaries, lane boundaries, road traffic direction information and lane traffic direction information cannot be provided.

In order to solve the problems that a constructed map cannot meet an automatic passenger-replacing parking scene with complex rules and extreme perception conditions and cannot provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, the embodiment of the application provides a construction method and a related device of an automatic driving map.

Disclosure of Invention

The embodiment of the application provides a construction method of an automatic driving map and a related device, which are used for improving the accuracy of the constructed automatic driving map, so that an automatic driving vehicle can be driven automatically better by utilizing the automatic driving map.

In a first aspect, an embodiment of the present application provides a method for constructing an automatic driving map, including:

acquiring road information, intersection information and lane information of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

acquiring intersection access point information according to the intersection information and the lane passing direction information;

and executing the operation of generating a virtual topological center line according to the information of the access point of the intersection to obtain an automatic driving map of the area where the vehicle runs, wherein the virtual topological center line is a boundary line of the vehicle running in the intersection area.

Compared with the situation that the constructed map cannot meet the automatic passenger-replacing parking scene with complex rules and extreme perception conditions and cannot provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, in the embodiment of the application, the road information, the intersection information and the lane information of the area where the vehicle runs through are firstly obtained, then the road traffic direction information and the lane traffic direction information are respectively obtained, then the intersection access point information is obtained, and finally the operation of generating the virtual topological center line is executed based on the intersection access point information to obtain the automatic driving map of the area where the vehicle runs through. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining road information, intersection information and lane information of an area through which a vehicle travels according to the manual driving trajectory data and/or the obstacle grid map includes:

acquiring a road boundary of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

obtaining a road center line of an area where the vehicle runs according to the road boundary;

acquiring a crossing area of an area where a vehicle runs according to a road boundary and a road center line;

obtaining the lane boundary of the area where the vehicle runs according to the road boundary, the road center line and the intersection area;

a lane center line of an area through which the vehicle travels is obtained from the lane boundary.

As can be seen, in this example, the road boundary, the road center line, the intersection region, the lane boundary, and the lane center line of the region where the vehicle has traveled are sequentially obtained, the road traffic direction information and the lane traffic direction information are respectively obtained, the intersection access point information is obtained, the operation of generating the virtual topological center line is performed based on the intersection access point information, and the automatic driving map of the region where the vehicle has traveled is obtained. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining the road boundary of the area traveled by the vehicle from the manual driving trajectory data and/or the obstacle grid map comprises:

obtaining a plurality of track point buffer areas according to the manual driving track data and/or the barrier grid diagram;

performing fusion operation on the plurality of track point buffer areas to obtain a road buffer area;

the boundary of the road buffer area is determined as the road boundary of the area through which the vehicle travels.

It can be seen that, in this example, a plurality of track point buffer areas are obtained, the plurality of track point buffer areas are fused to obtain a road buffer area, the boundary of the road buffer area is determined as the road boundary of an area that a vehicle travels through, the road center line, the intersection area, the lane boundary and the lane center line of the area that the vehicle travels through are sequentially obtained, the road traffic direction information and the lane traffic direction information are respectively obtained, the intersection access point information is obtained, the operation of generating the virtual topological center line is executed based on the intersection access point information, and the automatic driving map of the area that the vehicle travels through is obtained. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining a road centerline of an area traveled by a vehicle based on road boundaries comprises:

performing equidistant shape point operation on the road boundary to obtain the road boundary after equidistant shape points;

obtaining a first Thiessen polygon edge set according to the road boundary after the shape points are equally spaced;

executing Thiessen polygon edge operation outside the road boundary after deleting the equally spaced shape points on the first Thiessen polygon edge set to obtain a second Thiessen polygon edge set;

obtaining a buffer area of the road boundary according to the road boundary and a pre-stored first width;

obtaining a third Thiessen polygon edge set according to the Thiessen polygon edge which does not intersect with the buffer area of the road boundary in the second Thiessen polygon edge set;

executing suspension line deletion operation on the third Thiessen polygon edge set to obtain a fourth Thiessen polygon edge set;

and performing connection operation on the fourth Thiessen polygon edge set to obtain the road center line of the area where the vehicle runs.

As can be seen, in this example, a first tesson polygon side set is obtained based on the road boundary after the equidistant shape points, tesson polygon sides outside the road boundary after the equidistant shape points are deleted from the first tesson polygon side set, a second tesson polygon side set is obtained, a buffer area of the road boundary is obtained, a third tesson polygon side set is obtained based on the tesson polygon sides where the second tesson polygon side set and the buffer area of the road boundary do not intersect, a suspension line is deleted from the third tesson polygon side set, a fourth tesson polygon side set is obtained, the fourth tesson polygon side sets are connected to obtain a road center line of the area through which the vehicle travels, an intersection area, a lane boundary and a lane center line of the area through which the vehicle travels are sequentially obtained, road traffic direction information and lane traffic direction information are respectively obtained, and intersection entrance and exit point information are obtained, and executing the operation of generating a virtual topological center line based on the information of the access point of the intersection to obtain an automatic driving map of the area through which the vehicle runs. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining the intersection region of the area traveled by the vehicle based on the road boundary and the road centerline comprises:

obtaining the central point of each intersection according to the central line of the road;

determining a circle with the center point of each intersection as the circle center and the prestored first length as the radius as a center point buffer area of each intersection;

executing operation of deleting the sub-road center lines with the lengths smaller than the pre-stored second length on each sub-road center line to obtain each road section center line, wherein each sub-road center line is the remaining road center line except the center point buffer area of each intersection in the road center line;

obtaining a buffer area of the center line of each road section according to the center line of each road section and a prestored second width;

and executing operation of deleting the sub-intersection areas with the area smaller than the pre-stored first area on each sub-intersection area to obtain the intersection area of the area through which the vehicle runs, wherein each sub-intersection area is the remaining intersection area except the center line buffer area of each road section in the surface formed by the road boundary.

As can be seen, in this example, the center point of each intersection is obtained, the center point buffer area of each intersection is obtained, the center line of each road section is obtained, the center line buffer area of each road section is obtained based on each center line of each road section, the sub-intersection area with the area smaller than the pre-stored first area is deleted from each sub-intersection area, the intersection area of the area where the vehicle travels is obtained, the lane boundary and the lane center line of the area where the vehicle travels are sequentially obtained, the road traffic direction information and the lane traffic direction information are respectively obtained, the intersection access point information is obtained, the operation of generating the virtual topological center line is executed based on the intersection access point information, and the automatic driving map of the area where the vehicle travels is obtained. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining lane boundaries of an area traveled by a vehicle based on a road boundary, a road centerline, and an intersection area comprises:

obtaining each road section boundary except for the intersection region in the road boundary;

obtaining the central line of each road section except the intersection area in the central line of the road;

and if the road section B is a double-lane section, determining the road section boundary and the road section center line corresponding to the road section B as the lane boundary of the double-lane section which is driven by the vehicle, wherein the road section B is any one of all the double-lane sections included in each road section.

It can be seen that, in this example, each road section boundary is obtained, each road section center line is obtained, whether each road section is a single-lane section or a double-lane section is judged based on each road section boundary and each road section center line, and if the road section a is a single-lane section, the road section boundary corresponding to the road section a is determined as the lane boundary of the single-lane section that the vehicle runs through; if the road section B is a dual-lane section, determining a road section boundary and a road section center line corresponding to the road section B as a lane boundary of the dual-lane section which is driven by the vehicle; the method comprises the steps of obtaining lane center lines of an area where a vehicle runs, respectively obtaining road traffic direction information and lane traffic direction information, obtaining intersection access point information, and executing operation of generating a virtual topological center line based on the intersection access point information to obtain an automatic driving map of the area where the vehicle runs. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining the road traffic direction information according to the manual driving track data and the road information comprises:

acquiring each first relative angle difference set of the artificial driving track and each road section according to the artificial driving track data and the road center line, wherein each first relative angle difference set corresponds to each road section one by one;

if the variance of the first relative angle difference set C is smaller than a pre-stored first angle, determining that the road section corresponding to the first relative angle difference set C passes in one way, wherein the first relative angle difference set C is any one of all first relative angle difference sets of which the variance is smaller than the first angle and which are included in each first relative angle difference set;

if the mean value of the first relative angle difference set C is smaller than a pre-stored second angle, determining that the road passing direction of the road section corresponding to the first relative angle difference set C is forward passing;

In some possible embodiments, the method further comprises:

and if the variance of the first relative angle difference set D is larger than or equal to the first angle, determining that the road section corresponding to the first relative angle difference D passes in two directions, wherein the first relative angle difference set D is any one of all first relative angle difference sets of which the variance is larger than or equal to the first angle and which are included in each first relative angle difference set.

It can be seen that, in this example, each first relative angle difference set of the artificial driving track and each road segment is obtained, and if the variance of the first relative angle difference set C is smaller than the pre-stored first angle, it is determined that the road segment corresponding to the first relative angle difference set C is one-way passing; if the mean value of the first relative angle difference set C is smaller than a pre-stored second angle, determining that the road passing direction of the road section corresponding to the first relative angle difference set C is forward passing; if the mean value of the first relative angle difference set C is larger than or equal to the second angle, determining that the road passing direction of the road section corresponding to the first relative angle difference set C is reverse passing; if the variance of the first relative angle difference set D is larger than or equal to the first angle, determining that the road section corresponding to the first relative angle difference D is in two-way traffic; obtaining lane passing direction information, obtaining intersection access point information, executing operation of generating a virtual topological center line based on the intersection access point information, and obtaining an automatic driving map of an area where vehicles pass. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining lane traffic direction information from the lane information and the road traffic direction information includes:

if the road section E obtained according to the lane information and the road traffic direction information is a two-way traffic two-lane section, according to two second relative angle difference sets of a road section center line corresponding to the road section E and two lane center lines of the road section E, the road section E is any one of all two-way traffic two-lane sections included in each road section;

if the mean value of the second relative angle difference set F is smaller than a prestored third angle, judging whether the lane central line corresponding to the second relative angle difference set F is intersected with the right buffer area of the road section central line corresponding to the road section E or not, wherein the second relative angle difference set F is any one of two second relative angle difference sets; if so, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is a forward passing direction; if not, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is reverse passing;

if the mean value of the second relative angle difference set G is larger than or equal to a third angle, judging whether the lane central line corresponding to the second relative angle difference set G is intersected with the left buffer area of the road section central line corresponding to the road section E, wherein the second relative angle difference set G is any one of two second relative angle difference sets; if so, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is a forward passing direction; if not, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is reverse passing.

In some possible embodiments, the method further comprises:

if the road section H obtained according to the lane information and the road traffic direction information is a bidirectional traffic single-lane section, determining that the lane traffic direction of the road section H is bidirectional traffic, wherein the road section H is any one of all bidirectional traffic single-lane sections included in each road section;

It can be seen that, in this example, if the road segment E is a two-way traffic two-lane segment, according to two second relative angle difference sets of the center line of the road segment corresponding to the road segment E and the center lines of the two lanes of the road segment E, if the mean value of the second relative angle difference set F is smaller than the pre-stored third angle, it is determined whether the center line of the lane corresponding to the second relative angle difference set F intersects with the right buffer area of the center line of the road segment corresponding to the road segment E; if so, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is a forward passing direction; if not, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is reverse passing; if the mean value of the second relative angle difference set G is larger than or equal to the third angle, judging whether the lane central line corresponding to the second relative angle difference set G is intersected with the left buffer area of the road section central line corresponding to the road section E; if so, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is a forward passing direction; if not, determining that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is reverse passing. If the road section H is a bidirectional single-lane passing section, determining that the lane passing direction of the road section H is bidirectional passing; if the road section I is a one-way passing single-lane section, determining the road passing direction of the road section I as the lane passing direction of the road section I; and acquiring intersection access point information, and executing the operation of generating a virtual topological center line based on the intersection access point information to obtain an automatic driving map of the area through which the vehicle passes. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In some possible embodiments, obtaining intersection access point information according to the intersection information and the lane traffic direction information includes:

obtaining a starting shape point and a stopping shape point of a lane central line J, wherein the lane central line J is any one of all lane central lines included in each road section;

obtaining a buffer area of each intersection area according to each intersection area and a prestored third width;

if the shape point K intersects with a buffer area of the intersection area L, determining that the shape point L belongs to the intersection area L, wherein the shape point K is any one of a starting shape point and a stopping shape point of a lane central line J, and the intersection area L is any one of the intersection areas;

if the lane passing direction of the lane where the lane central line J is located is reverse passing, determining the initial shape point of the lane central line J as the initial shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line;

if the lane passing direction of the lane where the lane central line J is located is bidirectional passing, determining the starting shape point of the lane central line J as the starting shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line; or, the starting shape point of the lane central line J is determined as the ending shape point of the virtual topological central line, and the ending shape point of the lane central line J is determined as the starting shape point of the virtual topological central line.

It can be seen that, in this example, a start shape point and an end shape point of a lane center line J are obtained, a buffer area of each intersection area is obtained based on each intersection area, if the shape point K intersects with the buffer area of the intersection area L, it is determined that the shape point L belongs to the intersection area L, if the lane passing direction of the lane where the lane center line J is located is forward passing, the start shape point of the lane center line J is determined as the end shape point of the virtual topological center line, and the end shape point of the lane center line J is determined as the start shape point of the virtual topological center line; if the lane passing direction of the lane where the lane central line J is located is reverse passing, determining the starting shape point of the lane central line J as the starting shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line; if the lane passing direction of the lane where the lane central line J is located is bidirectional passing, determining the starting shape point of the lane central line J as the starting shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line; or, the starting shape point of the lane central line J is determined as the ending shape point of the virtual topological central line, and the ending shape point of the lane central line J is determined as the starting shape point of the virtual topological central line. And executing the operation of generating a virtual topological center line based on the information of the access point of the intersection to obtain an automatic driving map of the area through which the vehicle runs. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

In a second aspect, an embodiment of the present application provides an automatic driving map building apparatus, including:

the obtaining unit is also used for obtaining road traffic direction information according to the manual driving track data and the road information;

the obtaining unit is also used for obtaining lane passing direction information according to the lane information and the road passing direction information;

the acquisition unit is also used for acquiring the information of the access point of the intersection according to the intersection information and the lane passing direction information;

and the execution unit is used for executing the operation of generating a virtual topological center line according to the information of the access point of the intersection to obtain an automatic driving map of the area where the vehicle runs, and the virtual topological center line is a boundary line of the vehicle running in the intersection area.

In a third aspect, the present application provides an automatic driving map construction apparatus, the apparatus comprising a processor, a communication interface and a memory coupled to each other, wherein:

the processor is used for acquiring road information, intersection information and lane information of an area where the vehicle runs according to the manual driving track data and/or the barrier grid map;

the processor is also used for obtaining road traffic direction information according to the manual driving track data and the road information and obtaining lane traffic direction information according to the lane information and the road traffic direction information;

the processor is also used for obtaining the information of the access point of the intersection according to the intersection information and the lane passing direction information;

and the processor is also used for executing the operation of generating a virtual topological center line according to the information of the access point of the intersection to obtain an automatic driving map of the area where the vehicle runs, wherein the virtual topological center line is a boundary line of the vehicle running in the intersection area.

In a fourth aspect, the present application provides a computer-readable storage medium storing computer instructions, where the computer program is executed by hardware (for example, a processor, etc.) to implement part or all of the steps of any one of the methods performed by the automatic driving map construction apparatus in the embodiments of the present application.

In a fifth aspect, the present application provides a computer program product, which when run on a computer, causes the computer to execute some or all of the steps of the method for constructing an automatic driving map of the above aspects.

In a sixth aspect, an embodiment of the present application provides a chip system, where the chip system includes a processor, and is used to support an electronic device to implement part or all of the steps of the method for constructing an automatic driving map in the foregoing aspects.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments or the background art of the present application, the drawings required to be used in the embodiments or the background art of the present application will be described below.

Fig. 1A is a schematic flowchart of a first method for constructing an automatic driving map according to an embodiment of the present disclosure;

fig. 1B is a schematic flowchart of a process for obtaining road information, intersection information, and lane information of an area where a vehicle travels according to an embodiment of the present disclosure;

FIG. 1C is a schematic diagram of a process for obtaining a road boundary of an area traveled by a vehicle according to an embodiment of the present disclosure;

fig. 1D is a schematic flowchart of obtaining a plurality of trace point buffer areas according to an embodiment of the present application;

fig. 1E is a schematic view of a road buffer area provided in an embodiment of the present application;

FIG. 1F is a schematic diagram of a process for obtaining a road centerline of an area traveled by a vehicle according to an embodiment of the present disclosure;

FIG. 1G is a schematic diagram of a road boundary before and after an equidistant shape point provided by an embodiment of the present application;

FIG. 1H is a schematic view of a road centerline of an area traversed by a vehicle according to an embodiment of the present disclosure;

FIG. 1I is a schematic flow chart of obtaining a crossing area of an area traveled by a vehicle according to an embodiment of the present disclosure;

FIG. 1J is a schematic view of an intersection region of an area traversed by a vehicle according to an embodiment of the present disclosure;

FIG. 1K is a schematic diagram of a process for obtaining lane boundaries of an area traveled by a vehicle according to an embodiment of the present disclosure;

FIG. 1L is a schematic diagram illustrating a process for determining whether each road segment is a single-lane segment or a dual-lane segment according to an embodiment of the present disclosure;

fig. 1M is a schematic flowchart of a process for obtaining road traffic direction information according to an embodiment of the present application;

fig. 1N is a schematic flowchart of a process for obtaining a set of first relative angle differences between an artificial driving track and each road segment according to an embodiment of the present application;

fig. 1O is a schematic flowchart of a process for obtaining lane communication direction information according to an embodiment of the present disclosure;

fig. 1P is a schematic flow chart of obtaining information of an access point of an intersection according to an embodiment of the present application;

fig. 1Q is a schematic diagram of generating a virtual topological centerline according to an embodiment of the present application;

fig. 2 is a schematic flow chart of a second method for constructing an automatic driving map according to an embodiment of the present application;

fig. 3 is a schematic flowchart of a third method for constructing an automatic driving map according to an embodiment of the present application;

fig. 4 is a schematic flowchart of a method for implementing automatic valet parking based on an automatic driving map according to an embodiment of the present application;

FIG. 5 is a block diagram illustrating functional units of an apparatus for constructing an automatic driving map according to an embodiment of the present disclosure;

FIG. 6 is a schematic structural diagram of an apparatus for constructing an automatic driving map according to an embodiment of the present application;

fig. 7 is a schematic structural diagram of a chip system according to an embodiment of the present application.

Detailed Description

The embodiments of the present application will be described below with reference to the drawings. The terminology used in the description of the embodiments herein is for the purpose of describing particular embodiments herein only and is not intended to be limiting of the application.

Referring to fig. 1A, fig. 1A is a first method for constructing an automatic driving map provided in an embodiment of the present application, where the method for constructing an automatic driving map includes steps 101-104, which are as follows:

101. the automatic driving map construction device obtains road information, intersection information and lane information of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map.

The automatic driving map construction device obtains road information, intersection information and lane information of an area where a vehicle runs according to the manual driving track data; or the automatic driving map constructing device obtains road information, intersection information and lane information of the area where the vehicle runs according to the manual driving track data and the barrier grid map.

In some possible embodiments, as shown in fig. 1B, the step of obtaining road information, intersection information and lane information of an area where a vehicle travels according to the manual driving trajectory data and/or the obstacle grid map by the automatic driving map constructing apparatus includes steps S1-S5, which are as follows:

s1, the automatic driving map construction device obtains the road boundary of the area where the vehicle runs according to the manual driving track data and/or the barrier grid map;

s2, the automatic driving map construction device obtains the road center line of the area where the vehicle runs according to the road boundary;

s3, the automatic driving map construction device obtains the intersection area of the area where the vehicle runs according to the road boundary and the road center line;

s4, the automatic driving map construction device obtains the lane boundary of the area where the vehicle runs according to the road boundary, the road center line and the intersection area;

s5, the automatic driving map construction device obtains the lane center line of the area that the vehicle runs through according to the lane boundary.

The automatic driving map construction device obtains the road boundary of the area where the vehicle runs according to the manual driving track data; or, the automatic driving map construction device obtains the road boundary of the area where the vehicle runs according to the manual driving track data and the obstacle grid map.

In some possible embodiments, as shown in fig. 1C, the step of obtaining the road boundary of the area traveled by the vehicle according to the manual driving trajectory data and/or the obstacle grid map by the automatic driving map construction device comprises steps S11-S13, which are as follows:

s11, the automatic driving map construction device obtains a plurality of track point buffer areas according to the manual driving track data and/or the barrier grid map;

optionally, as shown in fig. 1D, the obtaining, by the automatic driving map construction device, a plurality of track point buffer areas according to the manual driving track data and/or the obstacle grid map includes steps S111 to S115, which are specifically as follows:

s111, the automatic driving map construction device sets i to be an increasing integer with an initial value of 0 and 1 as an interval, and judges whether i is smaller than N;

if so, the automatic driving map construction device executes the steps S112-S115.

And if not, the automatic driving map constructing device does not execute any operation.

And N is the total number of the first track points, and the total number of the first track points is obtained based on the manual driving track data.

And S112, the automatic driving map constructing device acquires the pose of the vehicle at the ith moment according to the manual driving track data, and converts the barrier grid map from the map coordinate system to the vehicle body coordinate system at the ith moment according to the pose of the vehicle at the ith moment.

The pose of the vehicle at the ith moment comprises an x-axis coordinate, a y-axis coordinate, a z-axis coordinate, a yaw angle, a pitch angle and a roll angle of the vehicle at the ith moment in a map coordinate system.

In the case where the construction apparatus of the automated driving map obtains a plurality of track point buffer areas from the manual driving trajectory data, the construction apparatus of the automated driving map skips step S112.

And S113, obtaining the travelable width of the vehicle at the ith moment by the automatic driving map constructing device.

Optionally, the obtaining, by the automatic driving map constructing device, a travelable width of the vehicle at the i-th time includes:

under the condition that the automatic driving map constructing device obtains a plurality of track point buffer areas according to the manual driving track data and the barrier grid map, the automatic driving map constructing device obtains the maximum y-axis coordinate (y1) of which the barrier probability is greater than the pre-stored first probability along the positive y-axis direction when the x-axis coordinate is the distance between the head of the vehicle and the origin of the vehicle body coordinate system at the ith moment; the construction device of the automatic driving map obtains the minimum y-axis coordinate (y2) that the obstacle probability is greater than the first probability in the negative direction of the y-axis when the size of the x-axis coordinate is the distance between the head of the vehicle and the origin of the vehicle body coordinate system at the i-th time; the construction device of the automatic driving map determines the sum of y1 and y2 as the travelable width of the vehicle at the ith time.

Or, in the case where the construction apparatus of the automated driving map obtains the plurality of track point buffer areas from the manual driving trajectory data, the construction apparatus of the automated driving map obtains the width of the vehicle, and determines the width of the vehicle as the travelable width of the vehicle at the i-th time.

The first probability may be 0.5 to 0.9.

And S114, the construction device of the automatic driving map obtains the radius of the ith track point buffer area according to the driving width of the vehicle at the ith moment.

Optionally, the method for constructing an automatic driving map may obtain the radius of the ith track point buffer area according to the travelable width of the vehicle at the ith time, where the method includes:

the method comprises the steps that a construction device of the automatic driving map obtains the nominal width of a vehicle;

the automatic driving map construction device judges whether the travelable width of the vehicle at the ith moment is larger than or equal to the nominal width of the vehicle;

if so, determining half of the driving width of the vehicle at the ith moment as the radius of the ith track point buffer area by the automatic driving map construction device;

and if not, the automatic driving map construction device determines half of the nominal width of the vehicle as the radius of the ith track point buffer area.

The nominal width of the vehicle is the distance between the left and right rear view mirrors of the vehicle.

And S115, the automatic driving map construction device obtains the ith track point buffer area according to the radius of the ith track point buffer area.

Optionally, the method for constructing an automatic driving map obtains the ith track point buffer area according to the radius of the ith track point buffer area, and includes:

the construction device of the automatic driving map determines an x-axis coordinate and a y-axis coordinate (an origin of a vehicle body coordinate system at the ith moment) of a vehicle in a map coordinate system as the circle center of an ith track point buffer area;

and the automatic driving map construction device makes a circle with the circle center and the radius of the ith track point buffer area to obtain the ith track point buffer area.

S12, executing fusion operation on the plurality of track point buffer areas by the automatic driving map construction device to obtain a road buffer area;

optionally, the constructing apparatus of the automatic driving map performs a fusion operation on the plurality of track point buffer areas to obtain a road buffer area, including:

the construction device of the automatic driving map obtains the intersection area of any two adjacent track point buffer areas in the plurality of track point buffer areas;

the construction device of the automatic driving map executes deletion operation on an intersection area of any two adjacent track point buffer areas included in one of the two adjacent track point buffer areas to obtain a road buffer area.

For example, as shown in fig. 1E, fig. 1E is a schematic view of a road buffer area provided in the embodiment of the present application, and an area surrounded by a road boundary is the road buffer area.

S13, the automatic driving map constructing apparatus determines the boundary of the road buffer area as the road boundary of the area through which the vehicle travels.

The automatic driving map construction device obtains a plurality of track point buffer areas according to the manual driving track data; or the automatic driving map constructing device obtains a plurality of track point buffer areas according to the manual driving track data and the barrier grid map.

In some possible embodiments, as shown in fig. 1F, the step of obtaining the road center line of the area traveled by the vehicle according to the road boundary by the automatic driving map construction device comprises steps S21-S27, which are as follows:

s21, the automatic driving map construction device performs equidistant shape point operation on the road boundary to obtain the road boundary after equidistant shape points;

optionally, the constructing apparatus of the automatic driving map performs shape and point operations on the road boundary at equal intervals to obtain the road boundary after the shape and point operations at equal intervals, and the method includes:

the method comprises the steps that a construction device of the automatic driving map obtains the distance between any adjacent former form point and any adjacent next form point in all form points included in a road boundary;

the automatic driving map construction device judges whether the distance between any adjacent previous shape point and any adjacent next shape point is greater than a pre-stored first distance;

and if so, inserting the shape point with the first distance from the last shape point between any adjacent last shape point and next shape point by the automatic driving map construction device, and if not, deleting the next shape point by the automatic driving map construction device (if the distance between any adjacent last shape point and next shape point is less than the first distance), so as to obtain the road boundary after the shape points are equally spaced.

The first distance may be 0.2 m.

For example, as shown in fig. 1G, fig. 1G is a schematic diagram of a road boundary before and after equally spaced shape points provided in an embodiment of the present application, distances between any two shape points adjacent to each other in all shape points included in the road boundary before equally spaced shape points are not necessarily the same, and distances between any two shape points adjacent to each other in all shape points included in the road boundary after equally spaced shape points are necessarily the same.

S22, the automatic driving map construction device obtains a first Thiessen polygon edge set according to the road boundary after the shape points are equally spaced;

s23, the automatic driving map construction device executes Thiessen polygon edge operation outside the road boundary after deleting equispaced shape points on the first Thiessen polygon edge set to obtain a second Thiessen polygon edge set;

s24, the automatic driving map construction device obtains a buffer area of the road boundary according to the road boundary and a pre-stored first width;

s25, the automatic driving map construction device obtains a third Thiessen polygon edge set according to the Thiessen polygon edge which does not intersect with the buffer area of the road boundary in the second Thiessen polygon edge set;

s26, the construction device of the automatic driving map executes suspension line deletion operation on the third Thiessen polygon side set to obtain a fourth Thiessen polygon side set;

a suspension line is a line segment in which one end point is not connected to other lines and the other end point is connected to two or more lines.

And S27, executing connection operation on the fourth Thiessen polygon edge set by the automatic driving map constructing device to obtain the road center line of the area through which the vehicle runs.

For example, as shown in fig. 1H, fig. 1H is a schematic diagram of a road centerline of an area where a vehicle travels, where lines between road boundaries are the road centerline according to an embodiment of the present application.

In some possible embodiments, as shown in fig. 1I, the apparatus for constructing an automatic driving map obtaining an intersection region of a region where a vehicle travels according to a road boundary and a road centerline includes steps S31-S35, which are as follows:

s31, the automatic driving map construction device obtains the central point of each intersection according to the road center line;

the center point of each intersection is the intersection of the center lines of the roads in the area that the vehicle travels through.

S32, determining a circle with the center point of each intersection as the center point and the prestored first length as the radius as a center point buffer area of each intersection by the automatic driving map construction device;

the first length may be 5 m.

S33, the automatic driving map construction device deletes sub-road center lines with lengths smaller than the pre-stored second length from each sub-road center line to obtain each road section center line, wherein each sub-road center line is the remaining road center line except the center point buffer area of each intersection in the road center line;

the second length may be 10 m.

S34, the automatic driving map construction device obtains each road section center line buffer area according to each road section center line and the pre-stored second width;

the second width may be 3 m.

Optionally, the constructing device of the automatic driving map obtains the buffer area of the center line of each road section according to the center line of each road section and the pre-stored second width, and the method includes:

and the automatic driving map constructing device respectively extends a second width to two sides along the vertical direction of the central line of each road section to obtain the buffer area of each road central line.

And S35, executing operation of deleting the sub-intersection areas with the area smaller than the pre-stored first area on each sub-intersection area by the automatic driving map construction device to obtain the intersection areas of the areas through which the vehicles run, wherein each sub-intersection area is the residual intersection area except the center line buffer area of each road section in the surface formed by the road boundary.

The first area may be 10m²。

For example, as shown in fig. 1J, fig. 1J is a schematic diagram of an intersection region of a region where a vehicle travels, where the region where the vehicle travels is composed of a road section region and an intersection region, according to an embodiment of the present application.

In some possible embodiments, as shown in fig. 1K, the apparatus for constructing an automatic driving map obtaining the lane boundary of the area where the vehicle travels according to the road boundary, the road centerline and the intersection area includes steps S41-S45, as follows:

s41, the automatic driving map construction device obtains each road section boundary except the intersection area in the road boundary;

s42, the automatic driving map construction device obtains the center lines of all road sections except the intersection area in the center lines of the roads;

s43, judging whether each road section is a single-lane section or a double-lane section according to each road section boundary and each road section central line by the automatic driving map constructing device, wherein each road section corresponds to each road section boundary and each road section central line one by one;

optionally, as shown in fig. 1L, the determining, by the automatic driving map constructing device, whether each road segment is a single-lane segment or a dual-lane segment according to each road segment boundary and each road segment center line includes steps S431 to S438, which are specifically as follows:

s431, the automatic driving map construction device sets j to be an increasing integer with an initial value of 0 and 1 as an interval, and judges whether j is smaller than M;

if so, the automatic driving map construction device executes the steps S432-S438.

And if not, the automatic driving map constructing device does not execute any operation.

M is the number of centerlines of each road segment.

And S432, extracting all shape points of the road section boundary corresponding to the jth road section central line by the automatic driving map constructing device.

And S433, extracting all shape points of the center line of the jth road section by the automatic driving map constructing device.

S434, the automatic driving map construction device sets k to be an increasing integer with an initial value of 0 and 1 as an interval, and judges whether k is smaller than P;

if so, the construction apparatus of the automatic driving map executes steps S435 to S436.

If not, the automatic driving map construction device executes step S437.

P is the number of all shape points of the jth road segment centerline.

And S435, selecting the figure point closest to the kth figure point from all the figure points of the road section boundary corresponding to the center line of the jth road section by the automatic driving map constructing device.

And S436, obtaining the road width of the position of the kth shape point according to the shape point which is closest to the kth shape point in all shape points of the road section boundary corresponding to the kth shape point and the centerline of the jth road section by the automatic driving map constructing device.

Optionally, the method for constructing an automatic driving map includes obtaining a road width of a k-th shape point position according to a shape point closest to a k-th shape point in all shape points of a road section boundary corresponding to the k-th shape point and a j-th road section center line, and includes:

the automatic driving map construction device acquires the coordinates of the kth form point;

the method comprises the steps that a construction device of the automatic driving map obtains coordinates of a shape point which is closest to a kth shape point in all shape points of a road section boundary corresponding to a jth road section center line;

and the automatic driving map construction device obtains the road width of the k-th shape point position according to the coordinates of the k-th shape point, the coordinates of the shape point closest to the k-th shape point in all shape points of the road section boundary corresponding to the j-th road section central line and a pre-stored road width formula.

The road width formula is pre-stored in the automatic driving map construction device, and the road width formula is as follows:

s＝[(x₃-x₄)²+(y₃-y₄)²]^1/2，

s is the road width at the k-th centroid position, (x)₃，y₃) Is the coordinates of the kth form point，(x₄，y₄) And the coordinates of the figure point which is closest to the k-th figure point in all the figure points of the road section boundary corresponding to the jth road section central line.

And S437, the automatic driving map construction device obtains the width of the jth road section according to the road width of each shape point position in the P shape points.

Optionally, the constructing device of the automatic driving map obtains the width of the jth road segment according to the road width of each shape point position in the P shape points, and includes:

the automatic driving map construction device obtains an average value of P road widths according to the road width of each shape point position in the P shape points, and the P road widths correspond to the P shape points one by one; the construction device of the automated driving map determines the average value of the P road widths as the width of the jth road segment.

S438, judging whether the width of the jth road section is smaller than a prestored fourth width by the automatic driving map constructing device;

and if so, the construction device of the automatic driving map determines that the jth road section is a single-lane section.

If not, the automatic driving map construction device determines that the jth road section is a double-lane section.

And S44, if the road section A is a single-lane section, determining the road section boundary corresponding to the road section A as the lane boundary of the single-lane section which is driven by the vehicle by the construction device of the automatic driving map, wherein the road section A is any one of all the single-lane sections included in each road section.

And S45, if the road section B is a double-lane section, determining the road section boundary and the road section center line corresponding to the road section B as the lane boundary of the double-lane section which is driven by the vehicle by the construction device of the automatic driving map, wherein the road section B is any one of all the double-lane sections included in each road section.

The construction device of the automatic driving map obtains the lane center line of the area that the vehicle runs through according to the lane boundary, and the construction device of the automatic driving map obtains the road center line of the area that the vehicle runs through according to the road boundary, which is not described herein.

102. The construction device of the automatic driving map obtains road passing direction information according to the manual driving track data and the road information, and obtains lane passing direction information according to the lane information and the road passing direction information.

In some possible embodiments, as shown in fig. 1M, the step of obtaining the road traffic direction information according to the manual driving trajectory data and the road information by the automatic driving map construction device includes steps S61-S65, which are as follows:

s61, the automatic driving map construction device obtains each first relative angle difference set of the manual driving track and each road section according to the manual driving track data and the road center line, and each first relative angle difference set corresponds to each road section one by one;

optionally, as shown in fig. 1N, the step of obtaining, by the automatic driving map constructing device, each first relative angle difference set between the artificial driving trajectory and each road segment according to the artificial driving trajectory data and the road centerline includes steps S611 to S618, which are specifically as follows:

s611, extracting the shape points of the road center line by the automatic driving map constructing device to obtain M first shape point sets.

M is the number of individual road segments.

S612, the automatic driving map construction device sets M to be an increasing integer with an initial value of 0 and an interval of 1, and judges whether M is smaller than M;

if so, the automatic driving map construction device executes steps S613 to S617.

If not, the automatic driving map construction device executes step S618.

S613, the automatic driving map construction device obtains an mth rectangular buffer area according to the mth shape point and the (m +1) th shape point.

Optionally, the constructing apparatus of the automatic driving map obtains an mth rectangular buffer area according to the mth shape point and the (m +1) th shape point, and includes:

the automatic driving map construction device respectively extends and prestores fifth widths to two sides along a direction which passes through the m-th form point and is vertical to the central line of the road to obtain two first end points;

the construction device of the automatic driving map respectively extends a fifth width to two sides along the direction which passes through the (m +1) th form point and is vertical to the central line of the road to obtain two second end points;

and the automatic driving map constructing device connects the two first end points and the two second end points to obtain the m-th matrix buffer area.

S614, the automatic driving map construction device sets n as an increasing integer with an initial value of 0 and 1 as an interval, and judges whether n is smaller than Q;

if so, the construction apparatus of the automatic driving map executes step S615.

If not, the automatic driving map construction device executes step S617.

And Q is the total number of the second track points, and the total number of the second track points is obtained based on the manual driving track data.

S615, judging whether the nth track point is intersected with the mth track point buffer area by the automatic driving map construction device;

if yes, the automatic driving map construction device executes step S616.

If not, the automatic driving map construction device executes step S614.

And S616, obtaining the relative angle difference of the nth track point by the automatic driving map constructing device.

Optionally, the obtaining, by the automatic driving map constructing device, a relative angle difference of the nth track point includes:

the automatic driving map construction device calculates the course angle of the nth track point;

the construction device of the automatic driving map obtains a course angle of the mth figure point (an included angle between a line segment formed by connecting the mth figure point and the (m +1) th figure point and the true north direction);

and the construction device of the automatic driving map determines the difference value between the course angle of the nth track point and the course angle of the mth form point as the relative angle difference of the nth track point.

S617, determining an average value of a plurality of relative angle differences of a plurality of track points intersected with the mth track point buffer area as a first relative angle difference of the mth track point buffer area by the automatic driving map construction device, wherein the plurality of relative angle differences correspond to the plurality of track points one to one.

And S618, determining a set formed by a plurality of first relative angle differences of a plurality of track point buffer areas included in each road section as each first relative angle difference set of each road section by the construction device of the automatic driving map, wherein the plurality of first relative angle differences correspond to the plurality of track point buffer areas one by one.

S62, if the variance of the first relative angle difference set C is smaller than a pre-stored first angle, the automatic driving map construction device determines that the road section corresponding to the first relative angle difference set C is one-way passing, and the first relative angle difference set C is any one of all first relative angle difference sets of which the variance is smaller than the first angle and which are included in each first relative angle difference set;

s63, if the mean value of the first relative angle difference set C is smaller than a pre-stored second angle, determining that the road passing direction of the road section corresponding to the first relative angle difference set C is forward passing by the automatic driving map construction device;

and S64, if the mean value of the first relative angle difference set C is larger than or equal to the second angle, the automatic driving map construction device determines that the road passing direction of the road section corresponding to the first relative angle difference set C is reverse passing.

In some possible embodiments, the method further comprises:

s65, if the variance of the first set of relative angle differences D is equal to or greater than the first angle, the device for constructing an automatic driving map determines that the road segment corresponding to the first relative angle difference D is for two-way traffic, and the first set of relative angle differences D is any one of all the first sets of relative angle differences whose variances included in the first set of relative angle differences are equal to or greater than the first angle.

In some possible embodiments, as shown in fig. 1O, the obtaining of the lane passing direction information by the construction apparatus of the automatic driving map according to the lane information and the road passing direction information includes steps S71-S75, which are as follows:

s71, if the road section E obtained according to the lane information and the road traffic direction information is a two-way traffic two-lane section, the automatic driving map construction device gathers two second relative angle differences of the center line of the road section corresponding to the road section E and the center lines of the two lanes of the road section E, and the road section E is any one of all the two-way traffic two-lane sections included in each road section;

the construction device of the automatic driving map refers to the construction device of the automatic driving map according to the two second relative angle difference sets of the road section center line corresponding to the road section E and the two lane center lines of the road section E, and obtains each first relative angle difference set of the manual driving track and each road section according to the manual driving track data and the road center line, which is not described herein.

S72, if the mean value of the second relative angle difference set F is smaller than a prestored third angle, the automatic driving map construction device judges whether the lane central line corresponding to the second relative angle difference set F is intersected with the right buffer area of the road section central line corresponding to the road section E, and the second relative angle difference set F is any one of two second relative angle difference sets; if so, the construction device of the automatic driving map determines that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is the forward passing; if not, the automatic driving map construction device determines that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set F is located is reverse passing;

s73, if the mean value of the second relative angle difference set G is larger than or equal to a third angle, the automatic driving map construction device judges whether the lane central line corresponding to the second relative angle difference set G is intersected with the left buffer area of the road section central line corresponding to the road section E, and the second relative angle difference set G is any one of two second relative angle difference sets; if so, the construction device of the automatic driving map determines that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is the forward passing; if not, the automatic driving map construction device determines that the lane passing direction of the lane where the lane center line corresponding to the second relative angle difference set G is located is reverse passing.

In some possible embodiments, the method further comprises:

s74, if the road section H obtained according to the lane information and the road traffic direction information is a bidirectional traffic single-lane section, the automatic driving map construction device determines that the lane traffic direction of the road section H is bidirectional traffic, and the road section H is any one of all bidirectional traffic single-lane sections included in each road section;

and S75, if the road section I obtained according to the lane information and the road traffic direction information is a one-way traffic single-lane section, determining the road traffic direction of the road section I as the lane traffic direction of the road section I by the automatic driving map construction device, wherein the road section I is any one of all the one-way traffic single-lane sections included in each road section.

103. And the automatic driving map construction device obtains the information of the access point of the intersection according to the intersection information and the lane passing direction information.

In some possible embodiments, as shown in fig. 1P, the step of obtaining the intersection access point information according to the intersection information and the lane passing direction information by the construction apparatus of the automatic driving map includes steps S81-S86, which are as follows:

s81, obtaining a starting shape point and a finishing shape point of a lane central line J by the automatic driving map construction device, wherein the lane central line J is any one of all lane central lines included in each road section;

s82, the automatic driving map construction device obtains the buffer area of each intersection area according to each intersection area and the prestored third width;

s83, if the shape point K intersects with the buffer area of the intersection area L, the automatic driving map construction device determines that the shape point L belongs to the intersection area L, the shape point K is any one of the starting shape point and the ending shape point of the lane center line J, and the intersection area L is any one of the intersection areas;

s84, if the lane passing direction of the lane where the lane central line J is located is the forward passing, the construction device of the automatic driving map determines the initial shape point of the lane central line J as the ending shape point of the virtual topological central line, and determines the ending shape point of the lane central line J as the initial shape point of the virtual topological central line;

s85, if the lane passing direction of the lane where the lane central line J is located is reverse passing, the construction device of the automatic driving map determines the initial shape point of the lane central line J as the initial shape point of the virtual topological central line, and determines the ending shape point of the lane central line J as the ending shape point of the virtual topological central line;

s86, if the lane passing direction of the lane where the lane central line J is located is bidirectional passing, the construction device of the automatic driving map determines the initial shape point of the lane central line J as the initial shape point of the virtual topological central line and determines the ending shape point of the lane central line J as the ending shape point of the virtual topological central line; or, the construction device of the automatic driving map determines the start shape point of the lane central line J as the end shape point of the virtual topological central line, and determines the end shape point of the lane central line J as the start shape point of the virtual topological central line.

104. The automatic driving map construction device executes operation of generating a virtual topological center line according to the intersection access point information to obtain the automatic driving map of the area where the vehicle runs, and the virtual topological center line is a boundary line of the vehicle running in the intersection area.

Optionally, the automatic driving map constructing device executes an operation of generating a virtual topological center line according to the intersection access point information to obtain the automatic driving map of the area through which the vehicle travels, including:

the method comprises the steps that a construction device of the automatic driving map obtains all starting shape points and all termination nodes of a virtual topological center line of the same intersection area;

and connecting any one starting shape point in all the starting shape points with all the rest termination nodes by the automatic driving map constructing device to obtain the automatic driving map of the area where the vehicle runs.

All virtual topological center lines of the intersection region can be Bezier curves or other curves meeting vehicle kinematics, and are not limited herein.

For example, as shown in fig. 1Q, fig. 1Q is a schematic diagram of generating virtual topological centerlines provided by an embodiment of the present application, and all virtual topological centerlines of each intersection area in four intersection areas are bezier curves.

Referring to fig. 2, fig. 2 is a second method for constructing an automatic driving map according to an embodiment of the present application, where the method for constructing an automatic driving map includes steps 201 and 210, which are as follows:

201. the construction device of the automatic driving map obtains sensor data.

Optionally, the device for constructing an automatic driving map obtains sensor data, and includes:

the frame device of the automatic driving map respectively acquires camera data, laser radar data, millimeter wave radar data, combined inertial navigation data and wheel speed data through a camera, a laser radar, a millimeter wave radar, a combined inertial navigation sensor and a wheel speed sensor; the automatic driving map construction device determines camera data, laser radar data, millimeter wave radar data, combined inertial navigation data and wheel speed data as sensor data. Wherein, camera, laser radar, millimeter wave radar, combination are used to lead, the fast sensor of wheel all integrates in the vehicle.

202. The automatic driving map construction device constructs a positioning map according to the sensor data.

203. And the automatic driving map constructing device stores the positioning map to the positioning map layer.

204. In the process of storing the positioning map to the positioning map layer, the automatic driving map constructing device obtains the pose data of the vehicle and stores the pose data of the vehicle to the manual driving track database.

Optionally, the obtaining, by the automatic driving map constructing device, pose data of the vehicle includes:

the automatic driving map construction device obtains pose data of a vehicle through a vehicle-mounted sensor. The vehicle-mounted sensor comprises a camera, a laser radar, a millimeter wave radar, a combined inertial navigation sensor and a wheel speed sensor. The pose data of the vehicle comprise an x-axis coordinate, a y-axis coordinate, a z-axis coordinate, a yaw angle, a pitch angle and a roll angle of the vehicle at a plurality of moments in a map coordinate system.

205. The automatic driving map construction device constructs an obstacle grid map according to the sensor data.

206. The automatic driving map construction device stores the obstacle grid map into the obstacle grid map library.

Step 205 and step 206 are optional steps.

207. The automatic driving map construction device obtains road information, intersection information and lane information of an area where a vehicle runs according to manual driving track data stored in a manual driving track database and/or an obstacle grid map stored in an obstacle grid map library.

Step 207 is described with reference to step 101 and will not be described further herein.

208. The construction device of the automatic driving map obtains road passing direction information according to the manual driving track data and the road information, and obtains lane passing direction information according to the lane information and the road passing direction information.

Step 208 is described with reference to step 102 and will not be described further herein.

209. And the automatic driving map construction device obtains the information of the access point of the intersection according to the intersection information and the lane passing direction information.

Step 209 is described with reference to step 103 and will not be described further herein.

210. The automatic driving map construction device executes operation of generating a virtual topological center line according to the intersection access point information to obtain the automatic driving map of the area where the vehicle runs, and the virtual topological center line is a boundary line of the vehicle running in the intersection area.

Step 210 is described with reference to step 104 and will not be described further herein.

Compared with the situation that the constructed map cannot meet the automatic passenger-replacing parking scene with complex rules and extreme perception conditions and cannot provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, in the embodiment of the application, a user drives a vehicle to enter a parking lot for the first time, and the construction device of the automatic driving map constructs the automatic driving map. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy of the constructed automatic driving map is improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

Referring to fig. 3, fig. 3 is a third method for constructing an automatic driving map according to an embodiment of the present application, where the method for constructing an automatic driving map includes steps 301 and 310, which are as follows:

301. the construction device of the automatic driving map obtains sensor data.

Step 301 is described with reference to step 201 and will not be described here.

302. The automatic driving map construction device extracts and processes the sensor data to obtain the characteristics of the matched positioning map.

303. And the automatic driving map constructing device matches the characteristics of the matched positioning map with the characteristics in the positioning map layer to obtain the pose data of the vehicle.

304. The construction device of the automatic driving map stores the pose data of the vehicle to the manual driving track database.

305. The automatic driving map construction device constructs an obstacle grid map according to the sensor data.

306. The automatic driving map construction device stores the obstacle grid map into the obstacle grid map library.

Step 305 and step 306 are optional steps.

307. The automatic driving map construction device obtains road information, intersection information and lane information of an area where a vehicle runs according to manual driving track data stored in a manual driving track database and/or an obstacle grid map stored in an obstacle grid map library.

Step 307 is described with reference to step 101 and will not be described further herein.

308. The construction device of the automatic driving map obtains road passing direction information according to the manual driving track data and the road information, and obtains lane passing direction information according to the lane information and the road passing direction information.

Step 308 is described with reference to step 102 and will not be described further herein.

309. The automatic driving map construction device obtains intersection access point information according to intersection information and lane traffic direction information.

Step 309 is described with reference to step 103 and will not be described further herein.

310: the automatic driving map construction device executes the operation of generating a virtual topological central line according to the information of the access point of the intersection to obtain the automatic driving map of the area where the vehicle runs, and the virtual topological central line is the boundary line of the vehicle running in the intersection area.

Step 310 is described with reference to step 104 and will not be described further herein.

Compared with the situation that the constructed map cannot meet the automatic passenger-replacing parking scene with complex rules and extreme perception conditions and cannot provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, in the embodiment of the application, the user drives the vehicle again to enter the parking lot, and the construction device of the automatic driving map updates the constructed automatic driving map. The constructed automatic driving map can meet the automatic passenger parking scene with complex rules and extreme perception conditions, and can provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, so that the accuracy and richness of the constructed automatic driving map are improved, and the automatic driving vehicle can be driven automatically better by using the automatic driving map.

Referring to fig. 4, fig. 4 is a schematic flow chart of a method for implementing automatic passenger transportation based on an automatic driving map according to an embodiment of the present application, where the method for implementing automatic passenger transportation based on an automatic driving map includes steps 401 and 410, which are as follows:

401. the construction device of the automatic driving map obtains sensor data.

402. The automatic driving map construction device extracts and processes the sensor data to obtain the characteristics for matching the positioning map.

403. And the automatic driving map constructing device matches the characteristics of the matched positioning map with the characteristics in the positioning map layer to obtain the pose data of the vehicle.

404. And the construction device of the automatic driving map sends the pose data of the vehicle to the automatic driving module.

405. The map query module acquires the current position of the vehicle and judges whether the vehicle enters an area capable of automatic driving according to the current position of the vehicle.

406. If so, the map query module sends the current position of the vehicle and the area where the vehicle enters and can be automatically driven to the automatic driving module.

407. The automatic driving module receives the current position of the vehicle and the area where the vehicle enters the automatic driving sent by the map query module, and obtains the target parking space selected by the user.

408. And the automatic driving module sends the current position of the vehicle and the target parking space to the path planning module.

409. The path planning module receives the current position and the target parking space of the vehicle sent by the automatic driving module, and plans the driving route of the vehicle according to the automatic driving map of the area where the vehicle passes, the current position of the vehicle and the target parking space of the vehicle.

410. The path planning module sends the driving route of the vehicle to the planning control module, so that the planning control module can complete the searching of the target parking space, obstacle avoidance and parking and warehousing.

Compared with the situation that the constructed map cannot meet the automatic passenger-replacing parking scene with complex rules and extreme sensing conditions and cannot provide road boundaries, lane boundaries, road traffic direction information and lane traffic direction information, in the embodiment of the application, the constructed automatic driving map can meet the automatic passenger-replacing parking scene with complex rules and extreme sensing conditions and can provide the road boundaries, the lane boundaries, the road traffic direction information and the lane traffic direction information, so that a user drives a vehicle to enter an area where the automatic driving map is constructed, and can drive the vehicle to a target parking space and finish parking through automatic driving.

Referring to fig. 5, fig. 5 is a block diagram illustrating functional units of an apparatus for constructing an automatic driving map according to an embodiment of the present application, where the apparatus 500 for constructing an automatic driving map includes:

an obtaining unit 501, configured to obtain road information, intersection information, and lane information of an area where a vehicle travels according to manual driving trajectory data and/or an obstacle grid map;

the obtaining unit 501 is further configured to obtain road traffic direction information according to the manual driving trajectory data and the road information;

the obtaining unit 501 is further configured to obtain lane passing direction information according to the lane information and the road passing direction information;

the obtaining unit 501 is further configured to obtain intersection access point information according to the intersection information and the lane traffic direction information;

the executing unit 502 is configured to execute an operation of generating a virtual topological center line according to the intersection access point information, so as to obtain an automatic driving map of an area where the vehicle travels, where the virtual topological center line is a boundary line where the vehicle travels in the intersection area.

In some possible embodiments, in obtaining road information, intersection information, and lane information of an area through which a vehicle travels from manual driving trajectory data and/or an obstacle grid map, the obtaining unit 501 is a specific unit:

acquiring a road boundary of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

obtaining a road center line of an area where the vehicle runs according to the road boundary;

acquiring a crossing area of an area where a vehicle runs according to a road boundary and a road center line;

obtaining the lane boundary of the area where the vehicle runs according to the road boundary, the road center line and the intersection area;

a lane center line of an area through which the vehicle travels is obtained from the lane boundary.

In some possible embodiments, in obtaining the road boundary of the area traveled by the vehicle according to the manual driving trajectory data and/or the obstacle grid map, the obtaining unit 501 is specifically configured to:

obtaining a plurality of track point buffer areas according to the manual driving track data and/or the barrier grid diagram;

performing fusion operation on the plurality of track point buffer areas to obtain a road buffer area;

the boundary of the road buffer area is determined as the road boundary of the area through which the vehicle travels.

In some possible embodiments, in obtaining the road center line of the area traveled by the vehicle according to the road boundary, the obtaining unit 501 is specifically configured to:

performing equidistant shape point operation on the road boundary to obtain the road boundary after equidistant shape points;

obtaining a first Thiessen polygon edge set according to the road boundary after the shape points are equally spaced;

executing Thiessen polygon edge operation outside the road boundary after deleting the equally spaced shape points on the first Thiessen polygon edge set to obtain a second Thiessen polygon edge set;

obtaining a buffer area of the road boundary according to the road boundary and a pre-stored first width;

obtaining a third Thiessen polygon edge set according to the Thiessen polygon edge which does not intersect with the buffer area of the road boundary in the second Thiessen polygon edge set;

executing suspension line deletion operation on the third Thiessen polygon edge set to obtain a fourth Thiessen polygon edge set;

and performing connection operation on the fourth Thiessen polygon edge set to obtain the road center line of the area where the vehicle runs.

In some possible embodiments, in obtaining 501 intersection regions of an area traveled by a vehicle from road boundaries and road centerlines, the obtaining is specifically for:

obtaining the central point of each intersection according to the central line of the road;

determining a circle with the center point of each intersection as the circle center and the prestored first length as the radius as a center point buffer area of each intersection;

obtaining a buffer area of the center line of each road section according to the center line of each road section and a prestored second width;

In some possible embodiments, in obtaining the lane boundary of the area through which the vehicle travels according to the road boundary, the road centerline and the intersection area, the obtaining unit 501 is specifically configured to:

obtaining each road section boundary except for the intersection region in the road boundary;

obtaining the central line of each road section except the intersection area in the central line of the road;

In some possible embodiments, in obtaining the road traffic direction information according to the manual driving trajectory data and the road information, the obtaining unit 501 is specifically configured to:

In some possible embodiments, in obtaining lane traffic direction information from lane information and road traffic direction information, the obtaining unit 501 is specifically configured to:

In some possible embodiments, in obtaining the intersection entry and exit point information according to the intersection information and the lane traffic direction information, the obtaining unit 501 is specifically configured to:

obtaining a starting shape point and a stopping shape point of a lane central line J, wherein the lane central line J is any one of all lane central lines included in each road section;

obtaining a buffer area of each intersection area according to each intersection area and a prestored third width;

if the lane passing direction of the lane where the lane central line J is located is bidirectional passing, determining the starting shape point of the lane central line J as the starting shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line; or, the starting shape point of the lane central line J is determined as the ending shape point of the virtual topological central line, and the ending shape point of the lane central line J is determined as the starting shape point of the virtual topological central line.

Referring to fig. 6, fig. 6 is a schematic structural diagram of an apparatus for constructing an automatic driving map according to an embodiment of the present application, where the apparatus 600 for constructing an automatic driving map includes a memory 601, a communication interface 602, and a processor 603, which are coupled to each other; such as memory 601, communication interface 602, and processor 603, are coupled by bus 604.

The Memory 601 may include, but is not limited to, a Random Access Memory (RAM), an Erasable Programmable Read Only Memory (EPROM), a Read-Only Memory (ROM), or a portable Read-Only Memory (CD-ROM), and the like, and the Memory 601 is used for related instructions and data.

The processor 603 may be one or more Central Processing Units (CPUs), and in the case that the processor 603 is one CPU, the CPU may be a single-core CPU or a multi-core CPU.

The processor 603 is configured to read the program code stored in the memory 601, and cooperate with the communication interface 602 to execute part or all of the steps of the method performed by the automatic driving map construction apparatus 600 in the above-described embodiment of the present application.

The processor 603 is used for acquiring road information, intersection information and lane information of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

the processor 603 is further configured to obtain road traffic direction information according to the manual driving trajectory data and the road information, and obtain lane traffic direction information according to the lane information and the road traffic direction information;

the processor 603 is further configured to obtain intersection access point information according to the intersection information and the lane traffic direction information;

the processor 603 is further configured to perform an operation of generating a virtual topological center line according to the intersection access point information, so as to obtain an automatic driving map of an area where the vehicle travels, where the virtual topological center line is a boundary line where the vehicle travels in the intersection area.

In some possible embodiments, in obtaining road information, intersection information, and lane information of an area through which the vehicle travels from the manual driving trajectory data and/or the obstacle raster map, the processor 603 is specifically configured to:

acquiring a road boundary of an area where a vehicle runs according to the manual driving track data and/or the barrier grid map;

obtaining a road center line of an area where the vehicle runs according to the road boundary;

acquiring a crossing area of an area where a vehicle runs according to a road boundary and a road center line;

obtaining the lane boundary of the area where the vehicle runs according to the road boundary, the road center line and the intersection area;

a lane center line of an area through which the vehicle travels is obtained from the lane boundary.

In some possible embodiments, in obtaining the road boundary of the area traveled by the vehicle from the manual driving trajectory data and/or the obstacle grid map, the processor 603 is specifically configured to:

obtaining a plurality of track point buffer areas according to the manual driving track data and/or the barrier grid diagram;

performing fusion operation on the plurality of track point buffer areas to obtain a road buffer area;

the boundary of the road buffer area is determined as the road boundary of the area through which the vehicle travels.

In some possible embodiments, in obtaining the road-center line of the area traveled by the vehicle according to the road boundary, the processor 603 is specifically configured to:

performing equidistant shape point operation on the road boundary to obtain the road boundary after equidistant shape points;

obtaining a first Thiessen polygon edge set according to the road boundary after the shape points are equally spaced;

executing Thiessen polygon edge operation outside the road boundary after deleting the equally spaced shape points on the first Thiessen polygon edge set to obtain a second Thiessen polygon edge set;

obtaining a buffer area of the road boundary according to the road boundary and a pre-stored first width;

obtaining a third Thiessen polygon edge set according to the Thiessen polygon edge which does not intersect with the buffer area of the road boundary in the second Thiessen polygon edge set;

executing suspension line deletion operation on the third Thiessen polygon edge set to obtain a fourth Thiessen polygon edge set;

and performing connection operation on the fourth Thiessen polygon edge set to obtain the road center line of the area where the vehicle runs.

In some possible embodiments, in obtaining the intersection region of the area traveled by the vehicle based on the road boundary and the road centerline, the processor 603 is specifically configured to:

obtaining the central point of each intersection according to the central line of the road;

determining a circle with the center point of each intersection as the circle center and the prestored first length as the radius as a center point buffer area of each intersection;

obtaining a buffer area of the center line of each road section according to the center line of each road section and a prestored second width;

In some possible embodiments, in obtaining lane boundaries of an area traveled by a vehicle based on road boundaries, road centerlines, and intersection areas, the processor 603 is specifically configured to:

obtaining each road section boundary except for the intersection region in the road boundary;

obtaining the central line of each road section except the intersection area in the central line of the road;

In some possible embodiments, in obtaining the road traffic direction information according to the manual driving trajectory data and the road information, the processor 603 is specifically configured to:

In some possible embodiments, in obtaining lane traffic direction information from the lane information and the road traffic direction information, the processor 603 is specifically configured to:

obtaining a starting shape point and a stopping shape point of a lane central line J, wherein the lane central line J is any one of all lane central lines included in each road section;

obtaining a buffer area of each intersection area according to each intersection area and a prestored third width;

if the lane passing direction of the lane where the lane central line J is located is bidirectional passing, determining the starting shape point of the lane central line J as the starting shape point of the virtual topological central line, and determining the ending shape point of the lane central line J as the ending shape point of the virtual topological central line; or, the starting shape point of the lane central line J is determined as the ending shape point of the virtual topological central line, and the ending shape point of the lane central line J is determined as the starting shape point of the virtual topological central line.

The present application also provides a computer-readable storage medium storing computer instructions, where the computer program is executed by hardware (for example, a processor, etc.) to implement part or all of the steps of any one of the methods performed by the automatic driving map forming apparatus in the present application.

Embodiments of the present application also provide a computer program product, which when run on a computer or a processor, causes the computer or the processor to execute some or all of the steps of the method for constructing an autopilot map according to the above aspects.

Referring to fig. 7, fig. 7 is a schematic structural diagram of a chip system according to an embodiment of the present disclosure, where the chip system 700 may include: a processor 701, and one or more interfaces 702 coupled to the processor 701. The following are exemplary:

the processor 701 may be used to read and execute computer readable instructions. In particular implementations, the processor 701 may mainly include a controller, an operator, and a register. Illustratively, the controller is mainly responsible for instruction decoding and sending out control signals for operations corresponding to the instructions. The arithmetic unit is mainly responsible for executing fixed-point or floating-point arithmetic operation, shift operation, logic operation and the like, and can also execute address operation and conversion. The register is mainly responsible for storing register operands, intermediate operation results and the like temporarily stored in the instruction execution process. In a specific implementation, the hardware architecture of the processor 701 may be an Application Specific Integrated Circuit (ASIC) architecture, a microprocessor without interlocked pipeline stage architecture (MIPS) architecture, an advanced reduced instruction set machine (ARM) architecture, or an NP architecture. The processors 701 may be single core or multicore.

Illustratively, the interface 702 may be used to input data to be processed to the processor 701, and may output a processing result of the processor 701 to the outside. In a specific implementation, the interface 702 may be a general purpose input/output (GPIO) interface, and may be connected to a plurality of peripheral devices (e.g., a display (LCD), a camera (camara), a Radio Frequency (RF) module, etc.). The interface 702 is connected to the processor 701 via a bus 703.

In some possible embodiments, the processor 701 may be configured to call, from the memory, an implementation program or data of the method for constructing an automatic driving map provided in one or more embodiments of the present application on a network device or a terminal device side, so that the chip may implement the method for constructing an automatic driving map shown in fig. 1A, fig. 2, and fig. 3, and the method for implementing an automatic valet parking method based on an automatic driving map shown in fig. 4. The memory may be integrated with the processor 701 or may be coupled to the system-on-chip 700 via the interface 702, i.e. the memory may be part of the system-on-chip 700 or may be separate from the system-on-chip 700. The interface 702 may be used to output the results of the execution by the processor 701. In this application, the interface 702 may be specifically configured to output the decoding result of the processor 701. Reference may be made to the foregoing embodiments for a method for constructing an automatic driving map provided in one or more embodiments of the present application, which are not described herein again.

In the above-described embodiments, all or part of the functions may be implemented by software, hardware, or a combination of software and hardware. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. The procedures or functions according to the embodiments of the present application are all or partially generated when the computer program instructions are loaded and executed on a computer. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a magnetic tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a Solid State Disk (SSD)), among others.

One of ordinary skill in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by hardware related to instructions of a computer program, which may be stored in a computer-readable storage medium, and when executed, may include the processes of the above method embodiments. And the aforementioned storage medium includes: various media capable of storing program codes, such as ROM or RAM, magnetic or optical disks, etc.

完整详细技术资料下载

上一篇：石墨接头机器人自动装卡簧、装栓机

下一篇：移动服务共享车辆的路径决策系统和方法

Construction method of automatic driving map and related device

相关技术

网友询问留言