Map construction method, terminal device, and storage medium

Abstract

Disclosed in the present application are a map construction method, a terminal device, and a storage medium. The map construction method comprises: acquiring sensing data of a vehicle; acquiring a navigation path on the basis of the sensing data, and generating a plurality of reference points on the navigation path; on the basis of the plurality of reference points and the sensing data, determining road boundary points respectively corresponding to the plurality of reference points; and on the basis of the road boundary points respectively corresponding to the plurality of reference points, constructing a map including a road topology. On the basis of the solution in the present application, stable reference points are generated by taking the advantage of the characteristic of a navigation path having few abrupt changes, and a map including a road topology is then constructed on the basis of the reference points and sensing data, thereby improving the stability of the constructed map.

Description

Map construction method, terminal device and storage medium

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on March 19, 2024, with application number 2024103092827 and application name “Map Construction Method, Terminal Device and Storage Medium”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application relates to the field of assisted driving technology, and in particular to a map construction method, terminal device, and storage medium. Background Art

[0003] High-precision maps (HD maps) are electronic maps that record detailed road information and are used for navigation-assisted driving. However, HD maps are expensive to maintain and struggle to adapt to the rapidly changing urban environment. Consequently, HD map-free navigation-assisted driving has become a key development direction. This HD map-free navigation-assisted driving relies on perception data to directly determine road topology and then construct a corresponding map for downstream intelligent driving programs to perform predictions, decision-making, and path planning. However, due to perception instability and occlusion, the constructed map is unstable.

[0004] Summary of the Invention

[0005] In order to solve or partially solve the problems existing in the related technology, the present application provides a map construction method, terminal device and storage medium, which can solve the problem of unstable maps constructed by navigation-assisted driving technology without high-precision maps.

[0006] To achieve the above objectives, the present application provides a map construction method, which includes:

[0007] Acquire vehicle perception data;

[0008] Acquire a navigation path based on the perception data, and generate a plurality of reference points on the navigation path;

[0009] Determining, based on the plurality of reference points and the perception data, road boundary points corresponding to each of the plurality of reference points;

[0010] A map including road topology is constructed based on the road boundary points corresponding to the plurality of reference points.

[0011] In one embodiment, the step of acquiring a navigation path based on the perception data includes:

[0012] Based on the perception data, obtaining a trajectory of the vehicle to be driven;

[0013] Based on a preset interface protocol, a navigation path corresponding to the to-be-traveled trajectory is generated.

[0014] In one embodiment, the step of obtaining the vehicle's trajectory to be driven based on the perception data includes:

[0015] Constructing road collision constraints based on the perception data;

[0016] Construct a cost function based on the preset navigation path points;

[0017] Based on the preset A-star algorithm, the road collision constraint, and the cost function, a trajectory of the vehicle to be driven is generated.

[0018] In one embodiment, the step of generating the navigation path corresponding to the to-be-traveled trajectory based on a preset interface protocol includes:

[0019] Based on the preset Advanced Driver Assistance System Interface Specification Version 2 ADASIS-V2 protocol, a navigation path corresponding to the to-be-traveled trajectory is generated.

[0020] In one embodiment, the step of determining the road boundary points corresponding to each of the plurality of reference points based on the plurality of reference points and the perception data includes:

[0021] determining, based on the perception data, at least one first side perception road boundary and at least one second side perception road boundary;

[0022] Based on the at least one first side perception road boundary and the at least one second side perception road boundary, first side road boundary points and second side road boundary points corresponding to the plurality of reference points are respectively determined.

[0023] In one embodiment, the step of determining the first side road boundary point and the second side road boundary point corresponding to each of the plurality of reference points based on the at least one first side perceived road boundary and the at least one second side perceived road boundary comprises:

[0024] Traversing the plurality of reference points, and for any target reference point traversed, generating a target normal of the navigation path with the target reference point as an intersection point;

[0025] Constructing a first point set based on the intersection of the target normal line and the at least one first side perception road boundary; and constructing a second point set based on the intersection of the target normal line and the at least one second side perception road boundary;

[0026] An intersection point in the first point set that is closest to the to-be-traveled trajectory is determined as a first side road boundary point corresponding to the target reference point; and an intersection point in the second point set that is closest to the to-be-traveled trajectory is determined as a second side road boundary point corresponding to the target reference point.

[0027] In one embodiment, the step of constructing a map including road topology based on the road boundary points corresponding to the plurality of reference points includes:

[0028] Constructing a first side road boundary topology based on the first side road boundary points corresponding to each of the plurality of reference points; and constructing a second side road boundary topology based on the second side road boundary points corresponding to each of the plurality of reference points;

[0029] constructing the road topology based on the first side road boundary topology and the second side road boundary topology;

[0030] Based on the road topology, a road-level map is constructed.

[0031] In one embodiment, the step of constructing the road topology based on the first side road boundary topology and the second side road boundary topology includes:

[0032] Constructing a first segment of road topology based on the first side road boundary topology and the second side road boundary topology;

[0033] constructing a second segment of road topology based on the navigation path;

[0034] The first segment of the road topology and the second segment of the road topology are spliced ​​together to obtain the road topology.

[0035] In one embodiment, the perception data includes perceived road boundaries.

[0036] In one embodiment, generating a plurality of reference points on the navigation path includes:

[0037] generating a plurality of reference points on the navigation path based on a preset reference point spacing; or,

[0038] A plurality of reference points are generated on the navigation path at unequal intervals.

[0039] In one embodiment, after constructing the map including the road topology based on the road boundary points corresponding to the plurality of reference points, the method further includes:

[0040] Obtaining the current position of the vehicle, uploading the plurality of reference points generated based on the navigation path and the current position of the vehicle to the cloud, and having the cloud analyze the plurality of reference points and the current position of the vehicle;

[0041] Receive navigation instructions issued by the cloud based on analysis results.

[0042] An embodiment of the present application further provides a terminal device, which includes a memory, a processor, and a map construction program stored in the memory and executable on the processor. When the map construction program is executed by the processor, the steps of the map construction method described above are implemented.

[0043] The embodiment of the present application further provides a computer-readable storage medium, on which a map construction program is stored. When the map construction program is executed by a processor, the steps of the map construction method described above are implemented.

[0044] The map construction method, terminal device, and storage medium proposed in the embodiments of this application obtain vehicle perception data; obtain a navigation path based on the perception data and generate a number of reference points on the navigation path; determine the road boundary points corresponding to each of the reference points based on the perception data; and construct a map containing road topology based on the road boundary points corresponding to each of the reference points. This solution utilizes the navigation path's less volatile nature to generate stable reference points, and then constructs a map containing road topology based on the reference points and perception data, improving the stability of the constructed map.

[0045] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.

[0047] FIG1 is a flowchart of a first exemplary embodiment of a map construction method of the present application;

[0048] FIG2 is a first view of the construction process involved in the map construction method of this application;

[0049] FIG3 is a second view of the construction process involved in the map construction method of the present application;

[0050] FIG4 is a third view of the construction process involved in the map construction method of this application;

[0051] FIG5 is a flowchart of a second exemplary embodiment of the map construction method of the present application;

[0052] FIG6 is a flowchart of a third exemplary embodiment of the map construction method of the present application;

[0053] FIG7 is a flowchart of a fourth exemplary embodiment of the map construction method of the present application;

[0054] FIG8 is a flowchart of a fifth exemplary embodiment of the map construction method of the present application;

[0055] FIG9 is a flowchart of a sixth exemplary embodiment of the map construction method of the present application;

[0056] FIG10 is a flowchart of a seventh exemplary embodiment of the map construction method of the present application;

[0057] FIG11 is a flow chart of an eighth exemplary embodiment of the map construction method of the present application. DETAILED DESCRIPTION

[0058] The following describes embodiments of the present application in more detail with reference to the accompanying drawings. Although the accompanying drawings illustrate embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. Rather, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0059] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. As used in this application and the appended claims, the singular forms "a," "an," "the," and "the" are intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more of the associated listed items.

[0060] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish information of the same type from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "plurality" is two or more, unless otherwise clearly and specifically defined.

[0061] The technical solution provided by the embodiments of this application includes: obtaining vehicle perception data; obtaining a navigation path based on the perception data and generating a number of reference points on the navigation path; determining the road boundary points corresponding to each of the reference points based on the perception data; and constructing a map containing road topology based on the road boundary points corresponding to each of the reference points. This solution utilizes the navigation path's less volatile nature to generate stable reference points, and then constructs a map containing road topology based on the reference points and perception data, improving the stability of the constructed map.

[0062] 1 , a first embodiment of a map construction method of the present application provides a flowchart, wherein the map construction method includes:

[0063] Step S10: Acquire vehicle perception data.

[0064] High-precision maps (HD maps) are electronic maps that record detailed road information and are used for navigation-assisted driving. However, HD maps are expensive to maintain and struggle to adapt to the rapidly changing urban environment. Consequently, HD map-free navigation-assisted driving has become a key development direction. This HD map-free navigation-assisted driving relies on perception data to directly determine road topology and then construct a corresponding map. However, due to perception instability and occlusion, the constructed map is unstable.

[0065] To solve the above problems, this embodiment proposes a method for realizing map construction in combination with a navigation path. First of all, the vehicle involved in this embodiment needs to have environmental perception capabilities. For example, the vehicle uses various perception devices (cameras, lidars, millimeter-wave radars, etc.) carried by it to obtain data on the surrounding environment. These perception devices are able to perceive information such as roads, vehicles, pedestrians, obstacles, etc. around the vehicle. Perception data may include images taken by cameras, intersection cloud data scanned by lidars, object positions detected by millimeter-wave radars, etc. In conjunction with Figure 2, the perception data includes the perceived road boundaries shown in Figure 2. The perceived road boundaries are the road edge positions determined by the vehicle based on perception. The perceived road boundaries are usually two-dimensional line segments with height attributes. Further, the perception data of the vehicle is obtained.

[0066] Step S20: acquiring a navigation path based on the perception data, and generating a plurality of reference points on the navigation path.

[0067] In one embodiment, the sensor data is processed and analyzed based on relevant protocols to identify suitable roads and plan a suitable driving path for the vehicle, thereby obtaining a navigation route. A navigation route typically has a starting point and an end point. The starting point is the vehicle's current location or starting point, and the end point is the vehicle's destination. The line between the starting and end points is formed by a line segment.

[0068] After obtaining the navigation path, it is necessary to further generate a number of reference points on the navigation path. The process of generating the number of reference points can be performed based on a preset reference point spacing, that is, the spacing between two adjacent reference points is fixed.

[0069] Alternatively, multiple reference points can be generated with unequal spacing. In this case, the reference point spacing needs to be adaptively set for each segment in the navigation path. Within the same segment of the navigation path, the reference points are spaced equally; within different segments of the navigation path, the reference points can be spaced differently.

[0070] Taking FIG3 as an example, a total of n reference points b1 to bn with equal intervals are generated in FIG3 , and the n reference points constitute a reference point set B.

[0071] Step S30: determining the road boundary points corresponding to the plurality of reference points based on the plurality of reference points and the perception data.

[0072] In one embodiment, since the navigation path rarely changes between consecutive frames, the navigation path has good stability. Each reference point is located at a specific position on the navigation path. Based on the stability of the navigation path, the reference point can be considered stable.

[0073] In real-world scenarios, there may be multiple perceived road boundaries on both sides of the navigation path, some of which may not belong to the road the vehicle is traveling on. For multiple perceived road boundaries on any side, it is necessary to determine the perceived road boundary corresponding to the road the vehicle is traveling on.

[0074] The trajectory to be driven lies within the road the vehicle intends to travel, and the perceived road boundary corresponding to the road the vehicle intends to travel is closest to the trajectory to be driven. Therefore, the perceived road boundaries on both sides of the navigation path can be determined based on the perception data. The intersection of the perceived road boundaries closest to the trajectory to be driven is further determined as the road boundary point corresponding to the reference point. It is understood that for any reference point, two corresponding road boundary points are determined, one for each side of the road.

[0075] Step S40: constructing a map including road topology based on the road boundary points corresponding to the plurality of reference points.

[0076] In one embodiment, after determining the road boundary points corresponding to each of the plurality of reference points, the road boundary points corresponding to the plurality of reference points that belong to a first side (e.g., the left side of the navigation path) can be included in a first point set, and the first point set can be used to construct a road boundary topology for the first side. The road boundary points corresponding to the plurality of reference points that belong to a second side (e.g., the right side of the navigation path) can be included in a second point set, and the second point set can be used to construct a road boundary topology for the second side. Furthermore, based on the first and second side road boundary topologies, a road topology is constructed, and a road-level map is constructed based on the road topology.

[0077] Taking Figure 4 as an example, we can see that there are two perceived road boundaries on both the left and right sides of the navigation path. For each of the reference points b1 to bn on the navigation path, we can determine the intersection point on the perceived road boundary closest to the desired trajectory. Based on these determined intersection points, we can then construct the road boundary topology for the road the vehicle will travel on. Finally, we use this road boundary topology to construct a road-level map.

[0078] In one possible implementation, the above-mentioned reference points can cooperate with the cloud to provide real-time navigation services for the vehicle. For example, the current position of the vehicle is obtained, and the several reference points generated based on the navigation path and the current position of the vehicle are uploaded to the cloud. The cloud accordingly analyzes the several reference points generated based on the navigation path and the current position of the vehicle. If the current position of the vehicle matches a specific reference point among the several reference points, the cloud can send a navigation instruction to the vehicle. The vehicle receives the navigation instruction and controls the vehicle's driving according to the navigation instruction. In this way, the problem of vehicles having difficulty obtaining navigation services in specific scenarios can be solved. Especially in some complex road sections, perception-based intelligent driving decisions may not be very effective, and cloud-assisted navigation can effectively cope with the challenges brought by complex road sections.

[0079] In this embodiment, the navigation path remains fixed in each frame without resetting the navigation. This utilizes the navigation path's inherently less volatile nature to generate stable reference points. The spatial positions of reference points in previous and subsequent frames remain stable. By recording the relative positions of reference points along the navigation path, a map containing road topology is constructed based on these reference points and perception data, improving the stability of the constructed map.

[0080] Further, referring to FIG5 , a second embodiment of the map construction method of the present application provides a flowchart. Based on the embodiment shown in FIG1 , the step S20 of “obtaining a navigation path based on the perception data” includes:

[0081] Step S21, obtaining a trajectory of the vehicle to be driven based on the perception data;

[0082] Step S22: generating a navigation path corresponding to the to-be-traveled trajectory based on a preset interface protocol.

[0083] In one embodiment, because perception data provides detailed information about the vehicle's surroundings, including road structure and obstacle locations, it serves as an important foundation for planning safe and efficient driving routes. Therefore, based on the perception data and a pre-set driving trajectory generation algorithm, a vehicle's intended driving trajectory can be generated. It will be appreciated that the road section within the intended driving trajectory can be characterized as a road section used to represent the vehicle's navigation needs and serve as the basis for the navigation route in subsequent steps.

[0084] Furthermore, a suitable interface protocol is pre-set, and the generated trajectory to be driven is converted into a data format that complies with the selected interface protocol. This may involve encoding and packaging the trajectory information to meet the message structure and communication requirements defined by the interface protocol. The converted trajectory is then processed using the interface protocol to generate a navigation path corresponding to the trajectory to be driven, which is used to guide the vehicle safely to its destination.

[0085] In one embodiment, the driving trajectory generation algorithm can be selected from the A* algorithm, the Dijkstra algorithm, the RRT (Rapidly-exploring Random Trees) algorithm, and the interface protocol can be ADASIS-V2 (Advanced Driver Assistance System Interface Specification Version 2), OpenDRIVE (Open Driving Scenario Description Language), ROS Navigation Stack, and other communication protocols used for vehicle navigation.

[0086] In this embodiment, a target trajectory is obtained through sensory data of the vehicle's surroundings. This target trajectory is then used to generate the vehicle's navigation path according to an interface protocol. This process effectively combines sensory data with a standardized interface protocol, ensuring the accuracy and reliability of navigation path planning.

[0087] Further, referring to FIG6 , a third embodiment of the map construction method of the present application provides a flowchart. Based on the embodiment shown in FIG5 , step S21 of “obtaining the vehicle’s trajectory to be driven based on the perception data” includes:

[0088] Step S211: constructing a road collision constraint based on the perception data;

[0089] Step S212: constructing a cost function based on the preset navigation path points;

[0090] Step S213 : generating a trajectory of the vehicle to be driven based on a preset A-star algorithm, the road collision constraint, and the cost function.

[0091] In one embodiment, to determine the vehicle's intended trajectory, the system first identifies potential obstacles in the vehicle's path, such as vehicles, pedestrians, and road construction, based on sensor data. The system then determines the effective driving area based on the location of the obstacles and the vehicle's size, and establishes collision constraints to prevent collisions. Notably, these constraints also meet kinematic constraints.

[0092] Furthermore, a cost function is constructed based on the preset navigation path points (including the vehicle's starting and ending points). This cost function takes into account factors such as the smoothness, safety, and efficiency of the vehicle's route, defining cost factors for different route segments. Based on the vehicle's current position and the destination, these cost factors are weighed to determine the optimal route. Furthermore, a search is performed using the preset A-star algorithm, combined with road collision constraints and the cost function. During the search process, a potential trajectory for the vehicle is generated based on the collision constraints and the cost function, ensuring a safe and smooth journey to the destination.

[0093] The A-star algorithm is a heuristic search algorithm for finding the shortest path in a graph or network. It uses a heuristic evaluation function in the graph, taking into account the lengths of previously found paths and the estimated distance to the goal, to select the most promising path to explore. By continuously updating a priority queue of paths, the A-star algorithm efficiently finds the optimal path from a starting point to a destination.

[0094] Taking Figures 2, 3, and 4 as examples, the trajectories to be driven are obtained based on the A-star algorithm search. It can be seen that the trajectories to be driven meet the requirements of smoothness, safety, and efficiency.

[0095] In this embodiment, by constructing road collision constraints and cost functions, combined with the A* algorithm to generate the desired trajectory, we ensure the safety and efficiency of the route. Furthermore, the desired trajectory also provides the data foundation for subsequent navigation path planning and road-level map construction.

[0096] Further, referring to FIG7 , a fourth embodiment of the map construction method of the present application provides a flowchart. Based on the embodiment shown in FIG5 , step S22 of “generating a navigation path corresponding to the to-be-traveled trajectory based on a preset interface protocol” includes:

[0097] Step S221 : generating a navigation path corresponding to the to-be-traveled trajectory based on a preset Advanced Driver Assistance System Interface Specification Version 2 ADASIS-V2 protocol.

[0098] In one embodiment, the ADASIS-V2 (Advanced Driver Assistance Systems Interface Specifications Version 2) protocol is a communication protocol for vehicle driver assistance systems. It defines the interface specifications between vehicles and map data providers, including message formats and data exchange methods, to enable functions such as vehicle navigation and autonomous driving. The ADASIS-V2 protocol facilitates efficient interaction between vehicles and map data through standardized communication methods, supporting vehicle safety and a better driving experience.

[0099] To obtain a navigation path, the trajectory to be driven obtained based on the perception data can be processed and analyzed based on the ADASIS-V2 protocol to generate a navigation path corresponding to the trajectory to be driven, ensuring that the navigation path is smooth, safe, and efficient, thereby providing accurate navigation guidance.

[0100] This embodiment only uses the ADASIS-V2 protocol as an example to illustrate the process of obtaining the navigation path. In theory, as long as there is a relevant protocol that can process and analyze the driving trajectory to generate a navigation path corresponding to the driving trajectory, then the protocol can replace the ADASIS-V2 protocol in this embodiment to achieve a similar effect.

[0101] In this embodiment, the ADASIS-V2 protocol is used to generate a navigation path corresponding to the trajectory to be driven, ensuring standardization and compatibility of path planning. Furthermore, without resetting the navigation, the navigation path generated using the ADASIS-V2 protocol remains fixed in each frame, thereby ensuring that subsequent reference points generated based on the navigation path remain stable.

[0102] Further, referring to FIG8 , a fifth embodiment of the map construction method of the present application provides a flowchart. Based on the embodiment shown in FIG5 , step S30 of “determining the road boundary points corresponding to the plurality of reference points based on the plurality of reference points and the perception data” includes:

[0103] Step S31, determining at least one first-side perception road boundary and at least one second-side perception road boundary based on the perception data;

[0104] Step S32: Based on the at least one first side perceived road boundary and the at least one second side perceived road boundary, respectively determine the first side road boundary points and the second side road boundary points corresponding to the plurality of reference points.

[0105] In one embodiment, in an actual scenario, multiple perceived road boundaries may exist on both sides of the navigation path, and some of these road boundaries may not belong to the road the vehicle is traveling on. In this case, at least one first-side perceived road boundary and at least one second-side perceived road boundary may be determined based on the perception data.

[0106] It can be understood that the perceived road boundary corresponding to the road the vehicle intends to travel is closest to the trajectory to be traveled. Therefore, for any reference point, the distance from the trajectory to be traveled to at least one first-side perceived road boundary can be analyzed based on that reference point, and the intersection point on the first-side perceived road boundary closest to the trajectory to be traveled can be determined as the first-side road boundary point corresponding to that reference point. Similarly, the distance from the trajectory to be traveled to at least one second-side perceived road boundary can be analyzed based on that reference point, and the intersection point on the second-side perceived road boundary closest to the trajectory to be traveled can be determined as the second-side road boundary point corresponding to that reference point. In other words, for each reference point, there is a corresponding first-side road boundary point and a corresponding second-side road boundary point.

[0107] Since there are several reference points, a traversal method can be used to determine the first side road boundary points and the second side road boundary points corresponding to each of the reference points, so as to ensure that the road shape and structure can be accurately described on both sides of the road.

[0108] In a possible implementation, by analyzing the first side road boundary point and the second side road boundary point corresponding to the same reference point in the previous and next frames, the jump caused by the change between the previous and next frames can be stably perceived.

[0109] In this embodiment, based on the perception data and a number of reference points, the perception road boundary corresponding to each reference point is determined for subsequent construction of a map including road topology, effectively improving the stability of the constructed map.

[0110] Further, referring to FIG9 , a sixth embodiment of the map construction method of the present application provides a flowchart. Based on the embodiment shown in FIG8 , step S32 of "determining, based on the at least one first side perceived road boundary and the at least one second side perceived road boundary, the first side road boundary points and the second side road boundary points corresponding to the plurality of reference points" includes:

[0111] Step S321, traversing the plurality of reference points, and for any target reference point traversed, generating a target normal of the navigation path with the target reference point as an intersection point;

[0112] Step S322: constructing a first point set based on the intersection of the target normal line and the at least one first side perception road boundary; and constructing a second point set based on the intersection of the target normal line and the at least one second side perception road boundary;

[0113] Step S323: Determine the intersection point of the first point set that is closest to the to-be-traveled trajectory as the first side road boundary point corresponding to the target reference point; and determine the intersection point of the second point set that is closest to the to-be-traveled trajectory as the second side road boundary point corresponding to the target reference point.

[0114] In one embodiment, a plurality of reference points are traversed, and for any target reference point found during the traversal, a target normal line for the navigation path is generated with the target reference point as the intersection point. The target normal line passes through the reference point and is perpendicular to the navigation path, pointing outside the navigation path, and intersecting the path to be driven, all first-side perceived road boundaries, and all second-side perceived road boundaries.

[0115] Then, a first point set is constructed based on the intersection of the target normal line and at least one first-side perceived road boundary; and a second point set is constructed based on the intersection of the target normal line and at least one second-side perceived road boundary. The intersection point in the first point set that is closest to the path to be traveled is determined as the first-side road boundary point corresponding to the target reference point; and the intersection point in the second point set that is closest to the path to be traveled is determined as the second-side road boundary point corresponding to the target reference point. In one possible implementation, the distance between each point in the first and second point sets and the path to be traveled is calculated along the corresponding normal line.

[0116] Taking Figure 4 as an example, for the traversed reference point bx, the corresponding normal of the navigation path is generated with reference point bx as the intersection point. A first point set is constructed based on the intersection points h1 and h2 of the normal corresponding to reference point bx with the left perceived road boundary. A second point set is constructed based on the intersection points h3 and h4 of the normal corresponding to reference point bx with the right second perceived road boundary. The intersection point in the first point set closest to the path to be traveled is then determined as the first side road boundary point corresponding to reference point bx. Furthermore, the intersection point in the second point set closest to the path to be traveled is determined as the second side road boundary point corresponding to reference point bx. Due to the traversal method employed, the first side road boundary point and the second side road boundary point corresponding to each of base points b1 through bn are ultimately determined.

[0117] In this embodiment, by traversing the reference points and generating the normal of the navigation path, the first point set and the second point set corresponding to the reference points are constructed in combination with the perceived road boundaries on both sides. The intersection point corresponding to the road on which the vehicle is to travel is found from the first point set and the second point set corresponding to the reference point as the road boundary point corresponding to the reference point, so that a stable road-level map can be subsequently constructed based on the road boundary points corresponding to each reference point, thereby effectively improving the stability of the constructed map.

[0118] Further, referring to FIG10 , a seventh embodiment of the map construction method of the present application provides a flowchart. Based on the embodiment shown in FIG8 , step S40 of “constructing a map including road topology based on the road boundary points corresponding to the plurality of reference points” includes:

[0119] Step S41: constructing a first side road boundary topology based on the first side road boundary points corresponding to the plurality of reference points; and constructing a second side road boundary topology based on the second side road boundary points corresponding to the plurality of reference points;

[0120] Step S42: constructing the road topology based on the first side road boundary topology and the second side road boundary topology;

[0121] Step S43: constructing a road-level map based on the road topology.

[0122] In one embodiment, the first side road boundary points corresponding to the multiple reference points are connected in the order of the multiple reference points to construct a first side road boundary topology; and the second side road boundary points corresponding to the multiple reference points are connected in the order of the multiple reference points to construct a second side road boundary topology.

[0123] Furthermore, a road topology can be constructed using the first side road boundary topology and the second side road boundary topology. A road-level map can then be constructed using the road topology. It is understood that the road-level map includes the road topology, which corresponds to the road that the vehicle is traveling on.

[0124] In this embodiment, by constructing the road boundary topology on both sides based on the road boundary points corresponding to the reference point, and combining the two road boundary topologies to construct the road topology, this helps to fully describe the road structure and improve the reliability and consistency of the map. Finally, a road-level map is constructed based on the road topology, making the road-level map more stable.

[0125] Further, referring to FIG11 , an eighth embodiment of the map construction method of the present application provides a flowchart. Based on the embodiment shown in FIG10 , step S42 of “constructing the road topology based on the first side road boundary topology and the second side road boundary topology” includes:

[0126] Step S421: constructing a first segment of road topology based on the first side road boundary topology and the second side road boundary topology;

[0127] Step S422: constructing a second road topology segment based on the navigation path;

[0128] Step S423: splicing the first segment of the road topology and the second segment of the road topology to obtain the road topology.

[0129] In one embodiment, current navigation-assisted driving technologies without HD maps rely heavily on sensory data and are unable to provide road topology information beyond visual range. To implement navigation-assisted driving without HD maps, this embodiment uses the navigation path obtained in the previous embodiment and constructs the first road topology segment using the first and second road boundary topologies determined based on sensory data and reference points.

[0130] Then, the second road topology segment is constructed using the navigation path. This process first requires obtaining the road topology information of the precise map following the first road topology segment, generating a number of beyond-horizon reference points on the navigation path following the first road topology segment, and determining the first side road boundary points and second side road boundary points corresponding to each of the several beyond-horizon reference points in a manner similar to the sixth embodiment of the present application. The second road topology segment is constructed based on the first side road boundary points and second side road boundary points corresponding to each of the several beyond-horizon reference points. Finally, the first road topology segment and the second road topology segment are spliced ​​together to construct the road topology.

[0131] In this embodiment, the first road topology segment is determined based on sensory data and offers high accuracy. The second road topology segment, however, is determined based on road topology information from a precise map. While less accurate than the first segment, it provides valuable insights for beyond-visual-range (BLOS) road sections, where sensory data is difficult to obtain. The road topology constructed using the first and second road topology segments can, to a certain extent, address BOS issues for navigation-assisted driving, improving the accuracy and reliability of predictions, decision-making, and path planning in downstream intelligent driving programs.

[0132] In addition, an embodiment of the present application also proposes a terminal device, which includes a memory, a processor, and a map construction program stored in the memory and executable on the processor. When the map construction program is executed by the processor, the steps of the map construction method described above are implemented.

[0133] Since all the technical solutions of all the aforementioned embodiments are adopted when this map construction program is executed by the processor, it has at least all the beneficial effects brought by all the technical solutions of all the aforementioned embodiments, which will not be described one by one here.

[0134] In addition, an embodiment of the present application further provides a computer-readable storage medium, on which a map construction program is stored. When the map construction program is executed by a processor, the steps of the map construction method described above are implemented.

[0135] Since all the technical solutions of all the aforementioned embodiments are adopted when this map construction program is executed by the processor, it has at least all the beneficial effects brought by all the technical solutions of all the aforementioned embodiments, which will not be described one by one here.

[0136] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0137] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0138] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, and of course can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as mentioned above, and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, controlled terminal, or network device, etc.) to execute the method of each embodiment of the present application.

[0139] The embodiments of the present application have been described above. The above description is illustrative and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the described embodiments. The terminology used herein is selected to best explain the principles of the embodiments, their practical applications, or improvements to the technology in the market, or to enable other persons skilled in the art to understand the embodiments disclosed herein.

Claims

1. A map construction method, characterized in that: The map construction method comprises: Acquire vehicle perception data; Acquire a navigation path based on the perception data, and generate a plurality of reference points on the navigation path; Determining, based on the plurality of reference points and the perception data, road boundary points corresponding to each of the plurality of reference points; A map including road topology is constructed based on the road boundary points corresponding to the plurality of reference points.

2. The map construction method according to claim 1, wherein: The acquiring of a navigation path based on the perception data includes: Based on the perception data, obtaining a trajectory of the vehicle to be driven; Based on a preset interface protocol, a navigation path corresponding to the to-be-traveled trajectory is generated.

3. The map construction method according to claim 2, wherein: The acquiring, based on the perception data, a trajectory of the vehicle to be driven, includes: Constructing road collision constraints based on the perception data; Construct a cost function based on the preset navigation path points; Based on the preset A-star algorithm, the road collision constraint, and the cost function, a trajectory of the vehicle to be driven is generated.

4. The map construction method according to claim 2, wherein: The generating of the navigation path corresponding to the to-be-traveled trajectory based on the preset interface protocol includes: Based on the preset Advanced Driver Assistance System Interface Specification Version 2 ADASIS-V2 protocol, a navigation path corresponding to the to-be-traveled trajectory is generated.

5. The map construction method according to claim 2, wherein: The determining, based on the plurality of reference points and the perception data, road boundary points corresponding to the plurality of reference points includes: determining, based on the perception data, at least one first side perception road boundary and at least one second side perception road boundary; Based on the at least one first side perception road boundary and the at least one second side perception road boundary, the first side road boundary point and the second side road boundary point corresponding to each of the plurality of reference points are respectively determined. Side road boundary point.

6. The map construction method according to claim 5, wherein: The determining, based on the at least one first side perceived road boundary and the at least one second side perceived road boundary, respectively corresponding to the plurality of reference points, of the first side road boundary points and the second side road boundary points includes: Traversing the plurality of reference points, and for any target reference point traversed, generating a target normal of the navigation path with the target reference point as an intersection point; Constructing a first point set based on the intersection of the target normal line and the at least one first side perception road boundary; and constructing a second point set based on the intersection of the target normal line and the at least one second side perception road boundary; An intersection point in the first point set that is closest to the to-be-traveled trajectory is determined as a first side road boundary point corresponding to the target reference point; and an intersection point in the second point set that is closest to the to-be-traveled trajectory is determined as a second side road boundary point corresponding to the target reference point.

7. The map construction method according to claim 5, wherein: The step of constructing a map including road topology based on the road boundary points corresponding to the plurality of reference points includes: Constructing a first side road boundary topology based on the first side road boundary points corresponding to each of the plurality of reference points; and constructing a second side road boundary topology based on the second side road boundary points corresponding to each of the plurality of reference points; constructing the road topology based on the first side road boundary topology and the second side road boundary topology; Based on the road topology, a road-level map is constructed.

8. The map construction method according to claim 7, wherein: The constructing the road topology based on the first side road boundary topology and the second side road boundary topology includes: Constructing a first segment of road topology based on the first side road boundary topology and the second side road boundary topology; constructing a second segment of road topology based on the navigation path; The first segment of the road topology and the second segment of the road topology are spliced ​​together to obtain the road topology.

9. The map construction method according to claim 1, wherein: The perception data includes perceived road boundaries.

10. The map construction method according to claim 1, wherein: The generating of a plurality of reference points on the navigation path comprises: generating a plurality of reference points on the navigation path based on a preset reference point spacing; or, A plurality of reference points are generated on the navigation path at unequal intervals.

11. The map construction method according to claim 1, wherein: After constructing a map including road topology based on the road boundary points corresponding to the plurality of reference points, the method further includes: Obtaining the current position of the vehicle, uploading the plurality of reference points generated based on the navigation path and the current position of the vehicle to the cloud, and having the cloud analyze the plurality of reference points and the current position of the vehicle; Receive navigation instructions issued by the cloud based on analysis results.

12. A terminal device, characterized in that: The terminal device includes a memory, a processor, and a map construction program stored in the memory and executable on the processor. When the map construction program is executed by the processor, the map construction method according to any one of claims 1 to 8 is implemented.

13. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a map construction program, which, when executed by a processor, implements the map construction method according to any one of claims 1 to 8.

Images