Decision environment construction method and device, vehicle, and storage medium

Abstract

The application relates to the technical field of automatic driving vehicle decision-making, in particular to a decision-making environment construction method and device, a vehicle and a storage medium, wherein the method comprises the following steps: acquiring a first center line set and a first boundary line set of a road around the vehicle; determining a center line of a current lane in which the vehicle is located from the first center line set according to a current position of the vehicle, and searching for a reference center line sequence of the vehicle according to the center line of the current lane; acquiring a lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set, associating the center line and the boundary line whose lateral distance meets an association condition, obtaining a reference boundary line sequence associated with the reference center line sequence, and constructing a decision-making environment of the vehicle based on the reference boundary line sequence. Therefore, the vehicle can quickly and accurately construct the lane boundary line and the road boundary associated with the center line based on the crowd-sourced map, the related operation of the automatic driving vehicle is prepared, and the problems of certain limitations and the like are solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving vehicle decision-making, and particularly relates to a decision-making environment construction method and device, a vehicle and a storage medium. BACKGROUND

[0002] Automatic driving of an intelligent vehicle needs to meet traffic rules for lane driving, and generally drives along a road center line. However, lane boundary lines and road boundaries need to be considered for lane changing, obstacle avoidance, on-ramp and off-ramp, etc. Existing high-precision map making needs to rely on high-precision acquisition equipment and a large amount of manual verification in the later stage, and has high mapping costs and a long cycle, thereby limiting the application scope of automatic driving. Therefore, a crowdsourcing map is used to obtain road related information to realize the reliability of automatic driving.

[0003] The crowdsourcing map is to use a large number of non-professional acquisition vehicles to detect changes in the surrounding environment in real time by using vehicle-mounted sensors, such as lane boundary lines, road boundaries, signboards, traffic lights, etc., and compare with a high-precision map. If a road change or other conditions are found, the data is uploaded to a cloud platform and then issued to other vehicles, thereby realizing rapid updating of map data.

[0004] The environment is constructed by a decision-making module of automatic driving. In some scenes of diverging and merging, lane number change, etc., the left and right boundary lines can be selected according to the driving requirements, the middle lane can be filled, and the connection relationship of the front and rear boundary lines can be determined. At the same time, the decision-making module only associates the lane boundary lines with the center line matched with navigation, which can reduce online consumption and realize online application.

[0005] In related technologies, most manufacturers add a part of manual verification to the processing of crowdsourcing mapping. The different information in the crowdsourcing map and the high-precision map is found by manual verification, which cannot be automatically generated and cannot be applied in a large range, has certain limitations, and wastes manpower and resources. The decision-making module of automatic driving only associates the lane boundary lines based on the high-precision map matched center line, cannot realize real-time understanding of road related information, reduces the safety of the vehicle, and has certain limitations. SUMMARY

[0006] The present application provides a decision-making environment construction method and device, a vehicle and a storage medium to solve the problem that a vehicle cannot quickly and accurately construct lane boundary lines and road boundaries associated with a center line based on a crowdsourcing map in related technologies, to prepare for related operations of an automatic driving vehicle, and to have certain limitations.

[0007] The first aspect embodiment of the present application provides a decision environment construction method, comprising the following steps: acquiring a first center line set and a first boundary line set of a road around the vehicle; determining a center line of a current lane in which the vehicle is located from the first center line set according to the current position of the vehicle, and searching for a reference center line sequence of the vehicle according to the center line of the current lane; acquiring a lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set, associating the center line and the boundary line whose associated lateral distance meets an association condition, obtaining a reference boundary line sequence associated with the reference center line sequence, and constructing a decision environment of the vehicle based on the reference center line sequence and the reference boundary line sequence.

[0008] According to the above technical means, the first center line set and the first boundary line set of the road around the vehicle are acquired, the center line of the current lane in which the vehicle is located is determined from the first center line set, and the reference center line sequence of the vehicle is searched; the lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set is acquired, the center line and the boundary line whose associated lateral distance meets the association condition are associated, and the decision environment of the vehicle is constructed based thereon. The decision environment model construction method based on the crowd-sourced map can quickly and accurately construct the center line and the corresponding lane boundary line sequence matching the navigation path required for decision, and prepare for subsequent path selection and planning of the vehicle, and has a wide range of applications.

[0009] Optionally, the acquiring of the lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set comprises: converting the first coordinates of each boundary line in the first boundary line set into second coordinates in a center line Frenet coordinate system, calculating the longitudinal distance and the lateral distance between the terminal point of the center line and the starting point and the ending point of each boundary line according to the second coordinates; searching for a second boundary line set in the first boundary line set which has a longitudinal overlapping area with the center line and is not intersected according to the longitudinal distance, wherein if the longitudinal distance corresponding to the starting point and the ending point of the boundary line is both outside the longitudinal coordinate range of the terminal point of the center line, it is determined that there is no longitudinal overlapping area between the boundary line and the center line; if the lateral coordinates corresponding to the starting point and the ending point are in different quadrants, it is determined that the boundary line intersects with the center line; identifying the minimum value of the lateral distance between the terminal point of the center line and the starting point and the ending point in the second boundary line set, and determining the lateral distance between each center line and each boundary line in the first boundary line set based on the minimum value.

[0010] According to the technical means, the first coordinate of each boundary line in the first boundary line set is converted into the second coordinate in the center line Frenet coordinate system, the longitudinal and transverse distances between the start and end points of each boundary line and the end point of the center line are calculated according to the second coordinate, the second boundary line set that is not intersected with the center line and has an overlapping area with the center line is searched based on the longitudinal distance, and the minimum value of the transverse distance between the start and end points of each boundary line and the end point of the center line is identified. The transverse distance between each center line and each boundary line is determined based on the minimum value. The Frenet coordinate system is used to determine the transverse and longitudinal distances, which can avoid the influence of the bending of the center line on the determination of the longitudinal and transverse distances, and ensure the accuracy of the path selection and planning of the vehicle.

[0011] Optionally, the center line and the boundary line associated with the transverse distance satisfying the association condition are used to obtain the reference boundary line sequence associated with the reference center line sequence, including: obtaining the longitudinal coordinates of the start and end points of the boundary line with the minimum transverse distance from the center line; determining whether the longitudinal coordinates of the start and end points are included in the longitudinal coordinate range of the center line; if yes, the boundary line is associated with the center line, otherwise, the longitudinal coordinates of the start and end points included in the longitudinal coordinate range of the center line are searched in the order of the transverse distance from small to large, and the boundary line with the minimum transverse distance under the same longitudinal coordinate of the center line is stored until the longitudinal coordinate search of all search positions of the center line is completed, and the reference boundary line sequence is obtained.

[0012] According to the technical means, in the embodiment of the application, the longitudinal coordinates of the start and end points of the boundary line with the minimum transverse distance from the center line are obtained, it is determined whether the longitudinal coordinates of the start and end points are included in the longitudinal coordinate range of the center line, if yes, the boundary line is associated with the center line, if not, the longitudinal coordinates of the start and end points included in the longitudinal coordinate range of the center line are searched in the order of the transverse distance from small to large, and the boundary line with the minimum transverse distance under the same longitudinal coordinate of the center line is stored until the longitudinal coordinate search of all search positions of the center line is completed, and the reference boundary line sequence is obtained. The lane boundary line and the road boundary near the center line of the vehicle are searched, so as to determine whether the left and right lanes are feasible, which can reduce the association search amount, reduce the demand for computing power, realize online deployment, and ensure the accuracy of the path selection and planning of the vehicle.

[0013] Optionally, the reference boundary line sequence includes a road boundary line sequence and a lane boundary line sequence, and the center line and the boundary line associated with the transverse distance satisfying the association condition are used to obtain the reference boundary line sequence associated with the reference center line sequence, further including: counting the road boundary line sequence associated with the center line into the lane boundary line sequence that has an overlapping area with the center line and is not intersected, to obtain the lane boundary line sequence with the minimum transverse distance within the longitudinal distance of the center line.

[0014] According to the technical means, the road boundary line sequence associated with the center line is counted into the lane boundary line sequence which has an overlapping area with the center line and is not intersected, a lane boundary line sequence with the minimum transverse distance in the longitudinal distance of the center line is obtained, a general minimum transverse distance fast search method under multiple different longitudinal and transverse distance complex situations is proposed, the completeness and accuracy of the result are ensured, and thus the accuracy of path selection and planning of the subsequent vehicle is ensured.

[0015] Optionally, the boundary line includes a lane boundary line, and the searching the reference center line sequence of the vehicle according to the center line of the current lane includes: identifying an actual type of the lane boundary line associated with the center line of the current lane; if the actual type is an un-crossable type, performing subsequent center line searching according to the center line of the current lane and a current navigation path search to obtain the reference center line sequence; otherwise, performing subsequent center line searching according to the center line of the current lane and the center line of a neighboring lane associated with the center line of the current lane, and according to the center line of the current lane, the center line of the neighboring lane associated with the center line of the current lane, and the current navigation path search to obtain the reference center line sequence.

[0016] According to the technical means, the actual type of the lane boundary line associated with the center line of the current lane is identified, if the actual type is the un-crossable type, subsequent center line searching is performed according to the center line of the current lane and the current navigation path search to obtain the reference center line sequence, otherwise, the center line of the neighboring lane associated with the center line of the current lane is searched according to the center line of the current lane, and subsequent center line searching is performed based on the center line of the current lane, the center line of the neighboring lane associated with the center line of the current lane, and the current navigation path search to obtain the reference center line sequence, so as to determine the passable state of the current lane and each side lane, so that the subsequent decision module selects and plans the reference path based on the output reference center line sequence, and the safe driving of the vehicle is ensured.

[0017] Optionally, the searching the center line of the neighboring lane according to the center line of the current lane includes: converting a third coordinate of the center line of the neighboring lane into a fourth coordinate in a Frenet coordinate system of the center line of the current lane; calculating a longitudinal coordinate of the center line of the current lane according to the fourth coordinate, and searching a second center line set containing the position of the vehicle according to the longitudinal coordinates of the start point and the end point in the first center line set; and selecting the center line with the minimum transverse distance in the left and right directions in the second center line set to obtain the center line of the neighboring lane.

[0018] According to the technical means, the third coordinate of the center line of the adjacent lane is converted into the fourth coordinate in the Frenet coordinate system of the center line of the current lane, the longitudinal coordinate of the center line of the current lane is calculated according to the fourth coordinate, and the second center line set containing the position of the vehicle is searched according to the longitudinal coordinates of the start point and the end point in the first center line set. The center line of the adjacent lane is obtained by selecting the center line with the minimum lateral distance in the left and right directions in the second center line set, so as to facilitate the accuracy of path selection and planning of the subsequent vehicle.

[0019] Optionally, before the center line associated with the lateral distance and the boundary line meet the association condition, the method further includes: obtaining a longitudinal distance between the vehicle and the end point of the center line of the current lane; if the longitudinal distance is less than a lane-changing longitudinal length limit value, adding a subsequent center line according to the subsequent center line in the center line attribute, until the longitudinal distance is greater than or equal to the lane-changing longitudinal length limit value, obtaining a third center line set within the longitudinal distance limit value, and performing boundary line association based on the third center line set.

[0020] According to the technical means, the longitudinal distance between the vehicle and the end point of the center line of the current lane is obtained, if the longitudinal distance is less than a lane-changing longitudinal length limit value, a subsequent center line is added according to the subsequent center line in the center line attribute, until the longitudinal distance is greater than or equal to the lane-changing longitudinal length limit value, a third center line set within the longitudinal distance limit value is obtained, and boundary line association is performed based on this. According to the reference center line sequence of the current lane and the left and right drivable lanes, the association search amount can be reduced, the demand for computing power can be reduced, and online deployment can be realized.

[0021] Optionally, the center line of the current lane is determined from the first center line set according to the current position of the vehicle, including: performing navigation path selection according to the center line of the current lane in which the vehicle is located and the first center line set, to obtain a fourth center line set under the navigation path; calculating the deviation angle of the center line of the fourth center line set and the heading of the vehicle, selecting the center line with the heading deviation within a preset range, to obtain a fifth center line set; calculating the lateral distance from the current vehicle coordinate to the center line of the fifth center line set, and selecting the center line with the minimum lateral distance, to obtain the center line of the current lane.

[0022] According to the technical means, the fourth center line set under the navigation path is obtained according to the current lane center line and the first center line set, the center line with the heading deviation in the preset range is selected by calculating the deviation angle of the fourth center line set and the heading of the vehicle, the fifth center line set is obtained by calculating the lateral distance from the current vehicle coordinate to the center line of the fifth center line set, the center line of the current lane is obtained by selecting the center line with the minimum lateral distance, so that the vehicle can accurately judge the current lane center line, thereby ensuring the accuracy of the path selection and planning of the vehicle and ensuring the safe driving of the vehicle.

[0023] Optionally, the first center line set and the first boundary line set of the road around the vehicle are obtained by obtaining a crowd-sourced map of the vehicle, and reading the first center line set and the first boundary line set within the preset range of the vehicle from the crowd-sourced map according to the current position of the vehicle.

[0024] According to the technical means, the crowd-sourced map of the vehicle is obtained, the center line set, the road boundary line set and the lane boundary line set within the preset range of the vehicle are read from the crowd-sourced map according to the current position of the vehicle, so as to quickly and accurately construct the center line matched with the navigation path and the corresponding lane boundary line set required for decision-making, which has good universality and ensures the safe driving of the vehicle.

[0025] The second aspect embodiment of the application provides a decision environment construction device, which comprises: an acquisition module, configured to acquire a first center line set and a first boundary line set of a road around the vehicle; a search module, configured to determine a center line of a current lane from the first center line set according to a current position of the vehicle, and search a reference center line sequence of the vehicle according to the center line of the current lane; and a construction module, configured to acquire a lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set, associate the center line and the boundary line with a lateral distance meeting an association condition, obtain a reference boundary line sequence associated with the reference center line sequence, and construct a decision environment of the vehicle based on the reference center line sequence and the reference boundary line sequence.

[0026] The third aspect embodiment of the application provides a vehicle, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor executes the program to implement the decision environment construction method as described in the above embodiments.

[0027] The fourth aspect embodiment of the application provides a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the decision environment construction method as described in the above embodiments.

[0028] Therefore, the present application has at least the following beneficial effects:

[0029] (1) The embodiment of the present application obtains a first center line set and a first boundary line set of the road around the vehicle, determines the center line of the lane where the current vehicle is located according to the first center line set, and searches for a reference center line sequence of the vehicle; obtains the transverse distance between each center line in the reference center line sequence and each boundary line in the first boundary line set, associates the center line and the boundary line whose transverse distance meets the association condition, and constructs the decision environment of the vehicle based thereon; the method for constructing the decision environment model based on the crowd-sourced map can quickly and accurately construct the center line matching the navigation path and the corresponding lane boundary line sequence required for decision-making, thereby preparing for the subsequent path selection and planning of the vehicle, and having a wide range of applications.

[0030] (2) The embodiment of the present application converts the first coordinates of each boundary line in the first boundary line set into second coordinates in the center line Frenet coordinate system, calculates the longitudinal and transverse distances between the starting and ending points of each boundary line and the terminal point of the center line according to the second coordinates, searches for a second boundary line set in the first boundary line set that has an overlapping area with the center line in the longitudinal direction and does not intersect with the center line based on the longitudinal distance, and identifies the minimum value of the transverse distance between the starting and ending points and the terminal point of the center line. The transverse distance between each center line and each boundary line in the first boundary line is determined based on the minimum value. The Frenet coordinate system is used to judge the transverse and longitudinal distances, which can avoid the influence of the bending of the center line on the judgment of the longitudinal and transverse distances, and ensure the accuracy of the path selection and planning of the vehicle.

[0031] (3) In the embodiment of the present application, the longitudinal coordinates of the starting and ending points of the boundary line with the minimum transverse distance to the center line are obtained, and it is judged whether the longitudinal coordinates of the starting and ending points are contained in the longitudinal coordinate range of the center line. If yes, the boundary line is associated with the center line. If no, the starting and ending point longitudinal coordinates contained in the longitudinal coordinate range of the center line are searched in the order of the transverse distance from small to large, and the boundary line with the minimum transverse distance under the same longitudinal coordinate of the center line is stored. The reference boundary line sequence is obtained by searching the longitudinal coordinates of all search positions. The lane boundary lines and road boundaries near the center line of the vehicle are searched, so as to determine whether the left and right lanes are feasible, thereby reducing the association search amount, reducing the demand for computing power, realizing online deployment, and ensuring the accuracy of the path selection and planning of the vehicle.

[0032] (4) The embodiment of the present application counts the road boundary line sequence associated with the center line into the lane boundary line sequence that has an overlapping area with the center line in the longitudinal direction and does not intersect with the center line, and obtains the lane boundary line sequence with the minimum transverse distance in the longitudinal distance of the center line. A general minimum transverse distance fast search method under multiple different longitudinal and transverse distance complex situations is proposed, which ensures the completeness and accuracy of the results, thereby ensuring the accuracy of the subsequent path selection and planning of the vehicle.

[0033] (5) The embodiment of the present application identifies the actual type of the lane boundary line associated with the center line of the current lane, if the actual type is a non-crossable type, then the reference center line sequence is obtained by performing subsequent center line search according to the center line of the current lane and the current navigation path search, otherwise the center line of the adjacent lane is associated with the center line of the current lane, and the reference center line sequence is obtained by performing subsequent center line search based on the center line of the current lane, the associated center line of the adjacent lane and the current navigation path search, so as to determine the passable state of the current lane and each side lane, so that the subsequent decision module selects the reference path and plans the path based on the output reference center line sequence, and ensures the safe driving of the vehicle.

[0034] (6) The embodiment of the present application converts the third coordinates of the center line of the adjacent lane into the fourth coordinates in the Frenet coordinate system of the center line of the current lane, calculates the longitudinal coordinates of the center line of the current lane according to the fourth coordinates, and searches the second center line set containing the vehicle position according to the longitudinal coordinates of the first point and the end point in the first center line set, selects the center line with the minimum lateral distance in the left and right directions in the second center line set, and obtains the center line of the adjacent lane, so as to facilitate the accuracy of the subsequent vehicle path selection and planning.

[0035] (7) The embodiment of the present application obtains the longitudinal distance between the vehicle and the end point of the center line of the current lane, if the longitudinal distance is less than the lane changing longitudinal length limit value, then the subsequent center line is added according to the subsequent center line in the center line attribute, until the longitudinal distance is greater than or equal to the lane changing longitudinal length limit value, the third center line set within the longitudinal distance limit value is obtained, and the boundary line association is performed based on this, according to the reference center line sequence of the current lane and the left and right drivable lanes, the association search amount can be reduced, the demand for computing power can be reduced, and online deployment can be realized.

[0036] (8) The embodiment of the present application performs navigation path selection according to the center line of the current lane of the vehicle and the first center line set, obtains the fourth center line set under the navigation path, calculates the deviation angle of the center line with the heading of the vehicle, selects the center line with the heading deviation within the preset range, obtains the fifth center line set, calculates the lateral distance from the current vehicle coordinates to the center line of the fifth center line set, selects the center line with the minimum lateral distance, and obtains the center line of the current lane, so as to facilitate the vehicle to accurately judge the center line of the current lane, thereby ensuring the accuracy of the vehicle path selection and planning, and ensuring the safe driving of the vehicle.

[0037] (9) The embodiment of the application acquires a crowd-sourced map of a vehicle, reads a center line set, a road boundary line set and a lane boundary line set within a preset range of the vehicle from the crowd-sourced map according to a current position of the vehicle, so as to quickly and accurately construct a center line matched with a navigation path and a corresponding lane boundary line sequence required for decision-making, has good universality and guarantees safe driving of the vehicle.

[0038] Additional aspects and advantages of the application will be made apparent by the following description and the accompanying drawings. BRIEF DESCRIPTION OF DRAWINGS

[0039] The above and / or additional aspects and advantages of the application will become apparent and be readily understood by considering the following detailed description, including the accompanying drawings, in which:

[0040] Figure 1 A flowchart of a decision environment construction method according to an embodiment of the application is provided;

[0041] Figure 2 A decision environment model architecture diagram based on a crowd-sourced map according to an embodiment of the application is provided;

[0042] Figure 3 A current center line verification module logic block diagram according to an embodiment of the application is provided;

[0043] Figure 4 An associated center line verification schematic diagram according to an embodiment of the application is provided;

[0044] Figure 5 A center line and road boundary position relationship schematic diagram according to an embodiment of the application is provided;

[0045] Figure 6 A center line associated road boundary logic block diagram according to an embodiment of the application is provided;

[0046] Figure 7 A center line associated road boundary result schematic diagram according to an embodiment of the application is provided;

[0047] Figure 8 A center line and lane boundary line position relationship schematic diagram according to an embodiment of the application is provided;

[0048] Figure 9 A center line associated lane boundary line logic block diagram according to an embodiment of the application is provided;

[0049] Figure 10 A center line associated lane boundary line result schematic diagram according to an embodiment of the application is provided;

[0050] Figure 11A left-right associated center line search logic block diagram is provided according to an embodiment of the present application;

[0051] Figure 12 A left-right associated center line search result schematic diagram is provided according to an embodiment of the present application;

[0052] Figure 13 A block schematic diagram of a decision environment construction device is provided according to an embodiment of the present application;

[0053] Figure 14 A structure schematic diagram of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0054] Embodiments of the present application are described in detail below with reference to examples thereof illustrated in the accompanying drawings, in which the same or similar elements or elements having the same or similar functions are denoted throughout by the same or similar reference numerals. The embodiments described below by reference to the accompanying drawings are exemplary and are intended to explain the present application, and cannot be understood as limiting the present application.

[0055] Currently, high-precision map making needs to rely on high-precision collection equipment and a large amount of manual verification in the later stage, and the mapping cost is high and the cycle is long, which limits the application range of automatic driving. The crowdsourcing map making relies on the lane boundary line, road boundary, signboard, traffic light and the like obtained by the camera on the mass-produced vehicle multiple times, and the lane boundary line, road boundary, signboard and traffic light position and the like are automatically clustered offline to construct the map elements, and the vehicle driving track is inferred according to the vehicle inertial navigation to construct the lane center line, and the whole process does not need manual verification, which greatly reduces the mapping cost, the map freshness is high, and the use range of automatic driving is expanded, and intelligent driving can be realized according to the daily driving route.

[0056] The output result of the crowdsourcing map in the embodiment of the present application includes a center line sequence, a lane boundary line sequence, a road boundary sequence, and the attributes of a road and a lane group. The center line of the crowdsourcing map is inconsistent with the lane boundary line and the road boundary in terms of source and length, and has no associated relationship. Each center line comes from a different track inference result, and the center line has no left-right associated relationship.

[0057] The automatic driving of the intelligent automobile needs to meet the traffic rules of lane driving, and generally drives according to the road center line. However, the lane boundary line and the road boundary need to be considered when changing lanes, bypassing obstacles, and going up and down ramps, and therefore, in order to realize reliable driving under the crowdsourcing map, the lane boundary line and the road boundary associated with the center line need to be constructed based on the crowdsourcing map.

[0058] Therefore, the embodiment of the present application uses the crowdsourcing map online to construct the decision environment, which can quickly and completely determine the lane boundary line and the road boundary near the center line, and prepare for the subsequent path selection and planning.

[0059] The decision environment construction method, device, vehicle, and storage medium are described below with reference to the accompanying drawings. Specifically, Figure 1 A flowchart of a decision environment construction method provided by an embodiment of the present application is shown in FIG. 1.

[0060] As shown in FIG. 1, the decision environment construction method includes the following steps: Figure 1

[0061] In step S101, a first center line set and a first boundary line set of a road around the vehicle are acquired.

[0062] The first center line set can be a set of center lines of a lane where the vehicle is currently located, which is not specifically limited herein.

[0063] The boundary line includes a lane boundary line and a road boundary line. Therefore, the first boundary line set can be a sequence of lane boundary lines and road boundary lines of the lane where the vehicle is currently located, which is not specifically limited herein.

[0064] It can be understood that the first center line set and the first boundary line set of the road around the vehicle are acquired, so as to facilitate subsequent determination of the center line of the lane where the vehicle is currently located.

[0065] In the embodiment of the present application, the first center line set and the first boundary line set of the road around the vehicle are acquired, including: acquiring a crowd-sourced map of the vehicle; reading the first center line set and the first boundary line set within a preset range of the vehicle from the crowd-sourced map according to the current position of the vehicle.

[0066] The preset range can be a range set by the user, for example, a range of 2 km with the vehicle as the center. The range can be set or adjusted according to the actual intention of the user, which is not specifically limited herein.

[0067] It can be understood that the crowd-sourced map of the vehicle is acquired, and the center line set, the road boundary line set, and the lane boundary line set within the range of the vehicle can be read from the crowd-sourced map according to the current position of the vehicle, so as to quickly and accurately construct the center line matched with the navigation path and the corresponding road and lane boundary line set required for decision, which has good universality and ensures the safe driving of the vehicle.

[0068] In step S102, the center line of the lane where the vehicle is currently located is determined from the first center line set according to the current position of the vehicle, and a reference center line sequence of the vehicle is searched according to the center line of the lane where the vehicle is currently located.

[0069] ​It can be understood that the embodiment of the application determines the center line of the current lane from the first center line set according to the current position of the vehicle, and searches for the reference center line sequence of the vehicle, so as to facilitate subsequent construction of the decision environment of the vehicle.

[0070] In the embodiment of the application, the center line of the current lane is determined from the first center line set according to the current position of the vehicle, comprising: performing navigation path pruning according to the current lane center line of the vehicle and the first center line set to obtain a fourth center line set under the navigation path; calculating the deviation angle of the center line of the fourth center line set from the heading of the vehicle; selecting the center line with a heading deviation within a preset range to obtain a fifth center line set; calculating the lateral distance of the current vehicle coordinate to the center line of the fifth center line set; and pruning the center line with the minimum lateral distance to obtain the center line of the current lane.

[0071] The preset range can be a range set by the user, for example, the center line with a navigation deviation of 0.5 m, which is not limited herein.

[0072] It can be understood that the embodiment of the application performs navigation path pruning according to the current lane center line of the vehicle and the first center line set to obtain a fourth center line set under the navigation path, calculates the deviation angle of the center line of the fourth center line set from the heading of the vehicle, selects the center line with a heading deviation within a preset range to obtain a fifth center line set, calculates the lateral distance of the current vehicle coordinate to the center line of the fifth center line set, and prunes the center line with the minimum lateral distance to obtain the center line of the current lane, so as to facilitate accurate judgment of the current lane center line by the vehicle, thereby ensuring the accuracy of path selection and planning by the vehicle and ensuring safe driving of the vehicle.

[0073] In the embodiment of the application, the reference center line sequence of the vehicle is searched according to the center line of the current lane, comprising: identifying the actual type of the lane boundary line associated with the center line of the current lane; if the actual type is an un-crossable type, performing subsequent center line search according to the center line of the current lane and the current navigation path search to obtain the reference center line sequence; otherwise, according to the center line of the current lane and the center line of the adjacent lane associated with the center line of the current lane, performing subsequent center line search according to the center line of the current lane, the center line of the adjacent lane associated with the center line of the current lane and the current navigation path search to obtain the reference center line sequence.

[0074] The un-crossable type can be a double yellow line, a yellow line or a white solid line, indicating that the side lane is not drivable, which is not limited herein.

[0075] It can be understood that, by identifying the actual type of the lane boundary line associated with the center line of the current lane, if the actual type is the non-crossable type, the subsequent center line search is performed according to the center line of the current lane and the current navigation path search, to obtain a reference center line sequence, otherwise, the center line of the adjacent lane is associated according to the center line of the current lane, and the subsequent center line search is performed based on the center line of the current lane, the associated center line of the adjacent lane and the current navigation path search, to obtain a reference center line sequence, so as to determine the passable state of the current lane and each side lane, so that the subsequent decision module selects and plans the reference path based on the output reference center line sequence, and ensures the safe driving of the vehicle.

[0076] In the embodiment of the application, the center line of the adjacent lane is associated according to the center line of the current lane, including: converting the third coordinates of the center line of the adjacent lane into fourth coordinates in the Frenet coordinate system of the center line of the current lane; calculating the longitudinal coordinates of the center line of the current lane according to the fourth coordinates, and searching for a second center line set containing the position of the vehicle according to the longitudinal coordinates of the first point and the last point in the first center line set; and selecting the center line with the minimum lateral distance in the left and right directions in the second center line set to obtain the center line of the adjacent lane.

[0077] The third coordinates can be global coordinates of the adjacent center line, and the fourth coordinates can be the Frenet coordinates of the adjacent lane center line converted to the current center line, which are not limited here.

[0078] It can be understood that, by converting the third coordinates of the center line of the adjacent lane into fourth coordinates in the Frenet coordinate system of the center line of the current lane, calculating the longitudinal coordinates of the center line of the current lane according to the fourth coordinates, and searching for a second center line set containing the position of the vehicle according to the longitudinal coordinates of the first point and the last point in the first center line set, the center line of the adjacent lane is selected in the second center line set with the minimum lateral distance in the left and right directions, so as to facilitate the accuracy of subsequent path selection and planning of the vehicle.

[0079] In step S103, the lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set is obtained, the center line and the boundary line associated with the lateral distance satisfying the association condition are associated, to obtain a reference boundary line sequence associated with the reference center line sequence, and the decision environment of the vehicle is constructed based on the reference center line sequence and the reference boundary line sequence.

[0080] It can be understood that the embodiments of the present application obtain the lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set, associate the center line and the boundary line whose lateral distance meets the association condition, and construct the decision environment of the vehicle based thereon. The method for constructing the decision environment model based on the crowd-sourced map can quickly and accurately construct the center line matched with the navigation path and the corresponding lane boundary line sequence required for decision, and prepare for the subsequent path selection and planning of the vehicle, and has a wide range of application.

[0081] In the embodiments of the present application, obtaining the lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set includes: converting the first coordinates of each boundary line in the first boundary line set into second coordinates in the center line Frenet coordinate system, calculating the longitudinal distance and the lateral distance between the starting point and the ending point of each boundary line and the terminal point of the center line according to the second coordinates; searching for a second boundary line set in the first boundary line set that has a longitudinal overlap region with the center line and does not intersect with the center line according to the longitudinal distance, wherein if the longitudinal distance corresponding to the starting point and the ending point of the boundary line is both outside the longitudinal coordinate range of the terminal point of the center line, it is determined that there is no longitudinal overlap region between the boundary line and the center line; if the lateral coordinates corresponding to the starting point and the ending point are in different quadrants, it is determined that the boundary line intersects with the center line; identifying the minimum value of the lateral distance between the starting point and the ending point of each center line and the terminal point of the center line, and determining the lateral distance between each center line and each boundary line in the first boundary line set based on the minimum value.

[0082] It can be understood that the embodiments of the present application convert the first coordinates of each boundary line in the first boundary line set into second coordinates in the center line Frenet coordinate system, calculate the longitudinal distance and the lateral distance between the starting point and the ending point of each boundary line and the terminal point of the center line according to the second coordinates, search for a second boundary line set in the first boundary line set that has a longitudinal overlap region with the center line and does not intersect with the center line based on the longitudinal distance, identify the minimum value of the lateral distance between the starting point and the ending point of the center line and the terminal point of the center line, and determine the lateral distance between each center line and each boundary line in the first boundary line set based on the minimum value. The Frenet coordinate system is used to judge the lateral and longitudinal distances, which can avoid the influence of the bending of the center line on the judgment of the longitudinal and lateral distances, and ensure the accuracy of the path selection and planning of the vehicle.

[0083] In the embodiment of the present application, the reference boundary line sequence associated with the reference center line sequence is obtained by associating the center line and the boundary line with a lateral distance meeting the association condition, comprising: obtaining the start point and end point longitudinal coordinates of the boundary line with the minimum lateral distance from the center line; determining whether the start point and end point longitudinal coordinates are contained in the longitudinal coordinate range of the center line; if yes, associating the boundary line with the center line, otherwise, searching the start point and end point longitudinal coordinates contained in the longitudinal coordinate range of the center line in the order of lateral distance from small to large, and storing the boundary line with the minimum lateral distance under the same longitudinal coordinate of the center line until the search of the longitudinal coordinates of all search positions is completed to obtain the reference boundary line sequence.

[0084] It can be understood that in the embodiment of the present application, the start and end point longitudinal coordinates of the boundary line with the minimum lateral distance from the center line are obtained, it is determined whether the start and end point longitudinal coordinates are contained in the longitudinal coordinate range of the center line, if yes, the boundary line is associated with the center line, if not, the start and end point longitudinal coordinates contained in the longitudinal coordinate range of the center line are searched in the order of lateral distance from small to large, and the boundary line with the minimum lateral distance under the same longitudinal coordinate of the center line is stored until the search of the longitudinal coordinates of all search positions is completed to obtain the reference boundary line sequence. The lane boundary line and road boundary near the center line of the vehicle are searched to determine whether the left and right lanes are feasible, which can reduce the association search amount, reduce the demand for computing power, realize online deployment, and ensure the accuracy of path selection and planning of the vehicle.

[0085] In the embodiment of the present application, before the center line and the boundary line with a lateral distance meeting the association condition are associated, the longitudinal distance between the end point of the center line of the current lane where the vehicle is located and the vehicle is obtained; if the longitudinal distance is less than the lane change longitudinal length limit value, the subsequent center line is added according to the subsequent center line in the center line attribute until the longitudinal distance is greater than or equal to the lane change longitudinal length limit value, to obtain a third center line set within the longitudinal distance limit value, and the boundary line association is performed based on the third center line set.

[0086] The longitudinal length limit value can be 2s time distance, which is not limited here.

[0087] It can be understood that in the embodiment of the present application, the longitudinal distance between the end point of the center line of the current lane where the vehicle is located and the vehicle is obtained, if the longitudinal distance is less than the lane change longitudinal length limit value, the subsequent center line is added according to the subsequent center line in the center line attribute until the longitudinal distance is greater than or equal to the lane change longitudinal length limit value, to obtain a third center line set within the longitudinal distance limit value, and the boundary line association is performed based on this. According to the reference center line sequence of the current lane and the left and right feasible lanes, the association search amount can be reduced, the demand for computing power can be reduced, and online deployment can be realized.

[0088] According to the decision environment construction method proposed in this application, a first set of center lines and a first set of boundary lines of the roads surrounding the vehicle are obtained. The center line of the lane currently occupied by the vehicle is determined based on the first set of center lines, and a reference center line sequence of the vehicle is searched. The lateral distance between each center line in the reference center line sequence and each boundary line in the first set of boundary lines is obtained. Center lines and boundary lines whose lateral distances satisfy the association condition are associated, and the vehicle's decision environment is constructed based on this. This decision environment model construction method based on a crowdsourced map can quickly and accurately construct the center lines and corresponding lane boundary line sequences that match the navigation path required for decision-making, preparing for the vehicle's subsequent path selection and planning, and has a wide range of applications. Therefore, it solves the problem in related technologies where the vehicle does not quickly and accurately construct lane boundary lines and road boundaries associated with the center line based on a crowdsourced map, thus limiting its operation.

[0089] The following will combine Figures 2 to 12 The architecture of the decision-making environment model based on crowdsourced maps is described in detail below:

[0090] Crowdsourced maps, after passing through map middleware, output centerline sets, boundary line sequences, road boundary line sequences, the current centerline, whether the centerline matches the navigation, and subsequent centerlines, among other attributes. The decision environment model construction architecture diagram is shown below. Figure 2 As shown, the embodiment specifically includes the following six modules:

[0091] 1. Current centerline verification

[0092] Since autonomous vehicles need to comply with traffic rules regarding lane travel, the location of the vehicle on the center line is fundamental to autonomous driving. Subsequent trajectory planning, lane changing operations, and other actions are all based on the vehicle's position. Therefore, it is necessary to verify the current center line location.

[0093] If the current centerline verification result is to select the centerline with the smallest lateral distance within a certain limit on the navigation path, then the specific logic is as follows: Figure 3 As shown: First, given the current lane centerline and the set of centerlines, the navigation path is filtered to obtain the centerline set L1 under the navigation path; Second, for the centerlines in the centerline set L1, the deviation angle between the centerline and the vehicle's heading is calculated, and the centerlines with heading deviation within the set range are selected to obtain the centerline set L2; Finally, the lateral distance from the current vehicle coordinates to the centerlines in the centerline set L2 is calculated, and the centerline with the smallest lateral distance is selected and output as the current centerline.

[0094] The diagram illustrating the verification results of the associated centerline ID is as follows: Figure 4 As shown: After obtaining the current centerline, you can search for the boundary lines and road boundaries associated with the centerline.

[0095] 2. In-lane longitudinal distance limit value centerline determination

[0096] The centerline set, road boundary set and boundary line set within a 2 km range from the ego vehicle are issued by the map middleware. To reduce the calculation amount of the subsequent centerline and lane boundary line association, the lane boundary line of the lane where the ego vehicle is located is first judged. If the lane boundary line is a solid line within the set lane change longitudinal length limit value, the corresponding direction centerline search and association are not required, and the search and association calculation amount is reduced. The set lane change longitudinal length limit value in this example is 2s.

[0097] First, according to the current centerline associated with the ego vehicle, if the longitudinal distance from the ego vehicle position to the end point of the current centerline is less than the set lane change longitudinal length limit value, the successor centerline is added according to the successor centerline in the centerline attribute, until the length of the centerline set where the ego vehicle starts exceeds the set lane change longitudinal length limit value. Thus, the centerline set L3 within the longitudinal distance limit value is obtained. After obtaining the centerline set L3 within the lane change longitudinal distance limit value, the next step is to search for the road boundary and lane boundary line near the centerline.

[0098] 3. Centerline associated road boundary search

[0099] (1) The actual road boundary and the centerline cannot be completely parallel, and may even intersect, have different lengths, and exist in states of complete overlap, partial overlap and complete non-overlap, as shown in Figure 5 .

[0100] The purpose of the road boundary search is to obtain the road boundary coordinate point with the minimum transverse distance within the longitudinal distance range of the centerline. To facilitate the calculation and representation of the relationship between the centerline and the road boundary, and to reduce the influence of curves on distance judgment, the global coordinates of the road boundary are changed to Frenet coordinates under the centerline, and the longitudinal S and transverse L coordinates of each coordinate point of the road boundary are calculated.

[0101] (2) The centerline associated road boundary logic block diagram is shown in Figure 6 , and specifically:

[0102] First, the global coordinates of the road boundary are converted to the longitudinal S and transverse L coordinates under the centerline Frenet coordinate system, the longitudinal S and transverse L distances of the first and last points of the road boundary are calculated, if the longitudinal S value of the first point and the longitudinal S value of the last point are both outside the longitudinal S coordinate range of the end point of the centerline, there is no longitudinal overlap region between the road boundary and the centerline; if the positive and negative of the transverse L values of the first and last points are not the same, the boundary line intersects the centerline. The road boundary set B1 that has a longitudinal overlap region with the centerline and does not intersect the centerline can be obtained, as shown in Figure 5 (1), (2), (3) and (4).

[0103] In the road boundary set B1, the minimum value of the lateral distance between the first and last points of each road boundary is taken as the lateral distance of the road boundary from the center line.

[0104] First, the road boundary with the minimum lateral distance from the center line is selected. If the longitudinal S coordinates of the first and last points of the road boundary include the longitudinal S range of the center line, the road boundary coordinates within the longitudinal range of the center line are directly output. Otherwise, the road boundary with the second minimum lateral distance is searched, the relationship between the longitudinal S coordinates of the first and last points of the road boundary and the longitudinal S coordinates of the center line and the S coordinates of the already associated road boundary first and last points is judged, and the road boundary coordinate points with the minimum lateral distance at the same longitudinal S coordinate are stored. Until the search of the longitudinal S coordinates of the center line is completed, the road boundary search is exited, and the road boundary coordinate points associated with the center line are output. The result of the association of the center line with the road boundary is shown in Figure 7 .

[0105] 4. Center line associated lane boundary line search

[0106] (1) Since there are cases where there are road boundaries but no lane boundary lines in actual scenarios, it is necessary to use the road boundaries as virtual lane boundary lines and jointly search with the remaining lane boundary lines to ensure the constraints of the vehicle driving lane.

[0107] The relationship between the actual lane boundary line and the center line can also have a longitudinal complete overlap, partial overlap, and complete non-overlap, or even a phase state, as shown in Figure 8 .

[0108] The purpose of the lane boundary line search is to obtain the lane boundary line coordinate points with the minimum lateral distance within the longitudinal distance range of the center line. Similarly, in order to avoid the influence of curved lines on distance judgment, the global coordinates are changed to Frenet coordinates, and the longitudinal S and lateral L coordinates of each coordinate point of the lane boundary line in the Frenet coordinate system of the center line are calculated.

[0109] (2) The logical block diagram of the center line associated road boundary is shown in Figure 9 .

[0110] The steps of the center line associated lane boundary line are consistent with the steps of the center line associated road boundary, and are performed according to the steps in 3. The difference between the associated road boundary is that the road boundary associated with the center line needs to be added to the lane line sequence Li1 that has an overlapping area with the center line and is not intersected, and finally the lane boundary line coordinate points with the minimum lateral distance within the longitudinal S distance of the center line are output.

[0111] The output is the lane boundary line coordinate points associated with the center line, and the association result is shown in Figure 10 .

[0112] 5. Left and right associated centerline search

[0113] The left and right associated centerline search logic block diagram is shown in Figure 11

[0114] If the centerline one side lane boundary line in the centerline set L3 is a solid line or other type of line that cannot be crossed, the side lane is not drivable, and there is no need to search for the corresponding side centerline, thereby obtaining the drivable state of each side lane.

[0115] According to the drivable state of the left and right lanes, search for the corresponding direction centerline. Convert the global coordinates of the adjacent centerline to the Frenet coordinates of the current centerline, calculate the longitudinal S coordinate of the ego vehicle on the centerline, then in the centerline set L1, search for the centerline set L4 containing the ego vehicle position according to the longitudinal S coordinate of the centerline start and end points, and finally select the centerline with the minimum lateral distance in the left and right directions in the centerline set L4 to obtain the left and right direction drivable centerline.

[0116] The left and right associated centerline search result diagram is shown in Figure 12

[0117] 6. Reference centerline sequence search

[0118] After obtaining the centerline where the ego vehicle is located and the left and right drivable centerlines, search for the reference centerline sequence according to the successor centerline in the centerline attribute and whether it is on the navigation path. When there are multiple successor centerlines for a centerline, there are also multiple groups of corresponding reference centerline sequences.

[0119] After obtaining the reference centerline sequence, search for the road boundary and lane boundary line for each centerline, and finally output the road boundary sequence and lane boundary line sequence corresponding to the centerline set.

[0120] At this point, the decision environment model construction based on the crowd-sourced map is completed, and the subsequent decision module selects the reference path and path planning based on the output reference centerline sequence and the corresponding road boundary and lane boundary line.

[0121] ​​In summary, the embodiment of the present application determines whether the left and right lanes are feasible by searching the lane boundary lines near the center line where the ego vehicle is located and the road boundary, only searches the left, middle and right feasible lanes backward, and can reduce the matching search amount of the center line correlation; converts the global coordinates into the Frenet coordinate system of the center line for judgment, which can avoid the influence of the center line bending on the distance judgment; selects the lateral distance as the basis for the correlation of the road boundary and the lane center line, realizes the general search in the complex situation of multiple different longitudinal and lateral distance lane lines, guarantees the completeness and correctness of the key lateral constraints, and the algorithm has universality and is beneficial to modular development. The decision environment model construction method based on the crowd-sourced map provided in the present application can quickly and accurately construct the center line matched with the navigation path and the corresponding lane boundary line sequence required for decision.

[0122] Secondly, the decision environment construction device according to the embodiment of the present application is described with reference to the accompanying drawings.

[0123] Figure 13 FIG. 1 is a block schematic diagram of the decision environment construction device of the embodiment of the present application.

[0124] As shown in FIG. 1, the decision environment construction device 10 includes an acquisition module 100, a search module 200 and a construction module 300. Figure 13

[0125] The acquisition module 100 is configured to acquire a first center line set and a first boundary line set of the road around the vehicle; the search module 200 is configured to determine the center line of the current lane from the first center line set according to the current position of the vehicle, and search the reference center line sequence of the vehicle according to the center line of the current lane; and the construction module 300 is configured to acquire the lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set, correlate the center line and the boundary line whose lateral distance satisfies the correlation condition, obtain the reference boundary line sequence correlated with the reference center line sequence, and construct the decision environment of the vehicle based on the reference center line sequence and the reference boundary line sequence.

[0126] It should be noted that the foregoing explanation and description of the decision environment construction method embodiment are also applicable to the decision environment construction device of the embodiment, which will not be described here again.

[0127] ​According to the decision environment construction device provided by the embodiment of the application, the first center line set and the first boundary line set of the road around the vehicle are acquired, the center line of the lane where the current vehicle is located is determined according to the first center line set, and the reference center line sequence of the vehicle is searched; the lateral distance between each center line in the reference center line sequence and each boundary line in the first boundary line set is acquired, the center line and the boundary line whose associated lateral distance meets the association condition are associated, and the decision environment of the vehicle is constructed based on the associated center line and boundary line. The decision environment model construction method based on the crowd-sourced map can quickly and accurately construct the center line required for decision and matched with the navigation path and the corresponding lane boundary line sequence, and prepares for the subsequent path selection and planning of the vehicle, and has a wide range of application. Therefore, the problems that the vehicle in the related art cannot quickly and accurately construct the lane boundary line associated with the center line and the road boundary based on the crowd-sourced map, and that the related operation of the autonomous vehicle is prepared, and that there is a certain limitation are solved.

[0128] Figure 14 The vehicle provided by the embodiment of the application is shown in the structural schematic diagram. The vehicle can include:

[0129] The memory 1401, the processor 1402, and the computer program stored in the memory 1401 and executable on the processor 1402.

[0130] The processor 1402 implements the decision environment construction method provided in the above embodiment when executing the program.

[0131] Further, the vehicle further includes:

[0132] The communication interface 1403 is used for communication between the memory 1401 and the processor 1402.

[0133] The memory 1401 is used to store the computer program executable on the processor 1402.

[0134] The memory 1401 can include a high-speed RAM (Random Access Memory, random access memory) memory, and can also include a non-volatile memory, for example, at least one disk memory.

[0135] If the memory 1401, the processor 1402 and the communication interface 1403 are implemented independently, the communication interface 1403, the memory 1401 and the processor 1402 can be connected with each other through a bus and complete communication between each other. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into an address bus, a data bus, a control bus, etc. For convenience of representation, Figure 14 Only one thick line is used to represent the bus in the figure, but it does not mean that there is only one bus or only one type of bus.

[0136] Optionally, in a specific implementation, if the memory 1401, the processor 1402 and the communication interface 1403 are integrated on a chip, the memory 1401, the processor 1402 and the communication interface 1403 can complete communication between each other through an internal interface.

[0137] The processor 1402 can be a CPU (Central Processing Unit), or an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the embodiments of the present application.

[0138] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, and the program is executed by a processor to implement the decision environment construction method as above.

[0139] In the description of the present specification, the description of the terms "one embodiment", "some embodiments", "an example", "a specific example" or "some examples" means that the specific features, structures, materials or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present application. In the present specification, the illustrative description of the above terms is not necessarily for the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or N embodiments or examples in a suitable manner. In addition, the person skilled in the art can combine and combine the different embodiments or examples described in the present specification and the features of the different embodiments or examples without contradiction.

[0140] In addition, the terms "first", "second", etc. are used only for descriptive purposes and do not connote or imply relative importance or a quantity of the indicated technical features. Thus, a feature defined with "first", "second", etc. can include at least one of the feature, explicitly or implicitly. In the description of the present application, the meaning of "N" is at least two, such as two, three, etc., unless otherwise explicitly and specifically limited.

[0141] Any process or method descriptions or blocks in flow charts or otherwise described herein can be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps in the process, and alternate implementations are possible. In some embodiments, the processes or methods described in this application can be tailored or adapted by, for example, varying or omitting certain steps or methods described, varying the ordering of certain steps or methods described, or employing additional or alternative steps or methods described, as would be understood by one of ordinary skill in the art.

[0142] It should be understood that portions of the present application can be realized with hardware, software, firmware or a combination thereof. In the above embodiments, the N steps or methods can be realized with software or firmware stored in a memory and executed by a suitable instruction execution system. As in another embodiment, if realized with hardware, any one or a combination of the following technologies known in the art can be used: discrete logic circuit with logic gates for implementing logical functions of data signals, application specific integrated circuit with suitable combination logic gates, programmable gate array, field programmable gate array, etc.

[0143] Those skilled in the art can understand that all or part of the steps carried out by the above-mentioned embodiment method can be completed by a program instructing the relevant hardware, and the program can be stored in a computer readable storage medium. When the program is executed, it includes one of the steps of the method embodiment or a combination thereof.

[0144] Although the embodiments of the present application have been shown and described above, it should be understood that the above-mentioned embodiments are exemplary and cannot be understood as limiting the present application, and those skilled in the art can make changes, modifications, replacements and variations to the above-mentioned embodiments within the scope of the present application.

Claims

1. A method for constructing a decision-making environment, characterized in that, Includes the following steps: Obtain the first set of centerlines and the first set of boundarylines of the roads surrounding this vehicle; The centerline of the current lane is determined from the first set of centerlines based on the vehicle's current position, and the reference centerline sequence of the vehicle is searched based on the centerline of the current lane. Obtain the lateral distance between each centerline in the reference centerline sequence and each boundary line in the first boundary line set, associate centerlines and boundary lines whose lateral distances satisfy the association condition, and obtain the reference boundary line sequence associated with the reference centerline sequence. Construct the decision environment of the vehicle based on the reference centerline sequence and the reference boundary line sequence. Wherein, the boundary line includes the lane boundary line, and the step of searching the reference centerline sequence of the vehicle based on the centerline of the currently occupied lane includes: Identify the actual type of lane boundary line associated with the centerline of the currently occupied lane; If the actual type is an uncrossable type, then a subsequent centerline search is performed based on the centerline of the current lane and the current navigation path to obtain the reference centerline sequence; Otherwise, the center lines of adjacent lanes are associated with the center line of the current lane, and a subsequent center line search is performed based on the center line of the current lane, the associated center lines of adjacent lanes, and the current navigation path search to obtain the reference center line sequence.

2. The method according to claim 1, characterized in that, The step of obtaining the lateral distance between each centerline in the reference centerline sequence and each boundary line in the first boundary line set includes: The first coordinates of each boundary line in the first set of boundary lines are converted into second coordinates in the Frenet coordinate system of the center line. Based on the second coordinates, the longitudinal distance and lateral distance between the first point and the last point of each boundary line and the end point of the center line are calculated respectively. Search the first set of boundary lines for a second set of boundary lines that have a longitudinal overlap with the center line but do not intersect it, based on the longitudinal distance. If the longitudinal distances corresponding to the first and last points of the boundary line are both outside the longitudinal coordinate range of the end point of the center line, it is determined that there is no longitudinal overlap between the boundary line and the center line. If the lateral coordinates corresponding to the first and last points are in different quadrants, it is determined that the boundary line intersects the center line. Identify the minimum lateral distance between the first and last points of the second set of boundary lines and the end point of the center line, and determine the lateral distance between each center line and each boundary line in the first set of boundary lines based on the minimum value.

3. The method according to claim 2, characterized in that, The associated lateral distances satisfy the association conditions for centerlines and boundary lines, resulting in a reference boundary line sequence associated with the reference centerline sequence, including: Obtain the longitudinal coordinates of the first and last points of the boundary line with the smallest lateral distance from the center line; Determine whether the longitudinal coordinates of the first and last points are contained within the longitudinal coordinate range of the centerline; If so, then associate the boundary line with the center line; otherwise, search for the first and last longitudinal coordinates within the longitudinal coordinate range of the center line in ascending order of the lateral distance, and store the boundary line with the smallest lateral distance under the same longitudinal coordinate of the center line, until the longitudinal coordinate search of all search positions of the center line is completed, and obtain the reference boundary line sequence.

4. The method according to claim 3, characterized in that, The reference boundary line sequence includes a road boundary line sequence and a lane boundary line sequence. The reference boundary line sequence associated with the reference centerline sequence, obtained by associating centerlines and boundary lines whose lateral distances satisfy the association conditions, further includes: The road boundary line sequence associated with the centerline is included in the lane boundary line sequence that overlaps with the centerline but does not intersect, to obtain the lane boundary line sequence with the smallest lateral distance within the longitudinal distance of the centerline.

5. The method according to claim 1, characterized in that, The step of associating the centerlines of adjacent lanes based on the centerline of the currently occupied lane includes: Transform the third coordinate of the centerline of the adjacent lane to the fourth coordinate in the Frenet coordinate system of the centerline of the current lane; Calculate the longitudinal coordinate of the centerline of the current lane based on the fourth coordinate, and search for a second centerline set containing the position of the vehicle based on the longitudinal coordinates of the first and last points in the first centerline set. Select the centerline with the smallest lateral distance in the left and right directions from the second set of centerlines to obtain the centerlines of adjacent lanes.

6. The method according to claim 1, characterized in that, Before associating the centerline and boundary line whose lateral distance satisfies the association condition, the following is also included: Obtain the longitudinal distance between the vehicle and the end point of the centerline of the current lane; If the longitudinal distance is less than the lane change longitudinal length limit, then a successor centerline is added according to the successor centerline in the centerline attribute until the longitudinal distance is greater than or equal to the lane change longitudinal length limit, thus obtaining a third centerline set within the longitudinal distance limit, and boundary line association is performed based on the third centerline set.

7. The method according to claim 1, characterized in that, Determining the centerline of the lane currently in which the vehicle is located from the first set of centerlines based on the vehicle's current position includes: Based on the center line of the lane where the vehicle is currently located and the first set of center lines, the navigation path is filtered to obtain the fourth set of center lines under the navigation path; Calculate the deviation angle between the centerline of the fourth centerline set and the vehicle's heading, select the centerline whose heading deviation is within a preset range, and obtain the fifth centerline set; Calculate the lateral distance from the current vehicle coordinates to the centerline of the fifth centerline set, and select the centerline with the smallest lateral distance to obtain the centerline of the current lane.

8. The method according to claim 1, characterized in that, The acquisition of the first centerline set and the first boundary line set of the roads surrounding the vehicle includes: Obtain the crowdsourced map of the vehicle; Based on the vehicle's current location, the first centerline set and the first boundary line set within the vehicle's preset range are read from the crowdsourced map.

9. A decision-making environment construction device, characterized in that, include: The acquisition module is used to acquire the first set of center lines and the first set of boundary lines of the roads surrounding the vehicle; The search module is used to determine the centerline of the lane currently occupied by the vehicle from the first centerline set based on the vehicle's current position, and to search for a reference centerline sequence of the vehicle based on the centerline of the lane currently occupied. A construction module is used to obtain the lateral distance between each centerline in the reference centerline sequence and each boundary line in the first boundary line set, associate centerlines and boundary lines whose lateral distances satisfy the association conditions, obtain the reference boundary line sequence associated with the reference centerline sequence, and construct the decision environment of the vehicle based on the reference centerline sequence and the reference boundary line sequence. Wherein, the boundary line includes the lane boundary line, and the step of searching the reference centerline sequence of the vehicle based on the centerline of the currently occupied lane includes: Identify the actual type of lane boundary line associated with the centerline of the currently occupied lane; If the actual type is an uncrossable type, then a subsequent centerline search is performed based on the centerline of the current lane and the current navigation path to obtain the reference centerline sequence; Otherwise, the center lines of adjacent lanes are associated with the center line of the current lane, and a subsequent center line search is performed based on the center line of the current lane, the associated center lines of adjacent lanes, and the current navigation path search to obtain the reference center line sequence.

10. A vehicle, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the decision environment construction method as described in any one of claims 1-8.

11. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the decision environment construction method as described in any one of claims 1-8.

Images