CN117128996A

CN117128996A - Path determination method, path determination device, electronic equipment and storage medium

Info

Publication number: CN117128996A
Application number: CN202311160219.3A
Authority: CN
Inventors: 董雪; 余威; 裴新欣; 孟辉; 黎佳宜; 杨宇婷; 魏斯理
Original assignee: Apollo Zhilian Beijing Technology Co Ltd
Current assignee: Apollo Zhilian Beijing Technology Co Ltd
Priority date: 2023-09-08
Filing date: 2023-09-08
Publication date: 2023-11-28

Abstract

The disclosure provides a path determination method, a path determination device, electronic equipment and a storage medium. The disclosure relates to the technical field of automatic driving, in particular to the technical fields of map matching, path planning and the like. The specific scheme is as follows: determining a first key feature point of a navigation path in a navigation map; determining second key feature points corresponding to the first key feature points in the high-precision map based on attribute information corresponding to the first key feature points; determining a second sub-path matched with the first sub-path included in the navigation path in the high-precision map by combining attribute information corresponding to the second key feature points; and determining a target path corresponding to the navigation path in the high-precision map based on a second sub-path matched with the first sub-path in the high-precision map. According to the scheme, the navigation path under the navigation map can be converted into the high-precision path under the high-precision map without the need of local graphic point information, and the problem that the local graphic point information needs to be acquired when the navigation map and the high-precision map are heterogeneous is effectively solved.

Description

Path determination method, path determination device, electronic equipment and storage medium

Technical Field

The disclosure relates to the technical field of automatic driving, in particular to the technical fields of map matching, path planning and the like.

Background

In the automatic driving field, the vehicle automatic auxiliary navigation driving needs to map a navigation path, such as a standard (Standard Definition, SD) path, planned by a navigation map provided by an intelligent cabin onto a High Definition (HD) map of the intelligent driving field to obtain the HD path, so as to realize intelligent lane-level navigation based on the High Definition map. If the navigation map and the high-definition map belong to the homologous map, the SD path can be converted into the HD path by generating an SD and HD association table in advance. If the navigation map and the high-precision map belong to heterogeneous maps, the navigation map and the high-precision map are different in manufacturing process and the map model is very different, so that an SD and HD association table cannot be generated in advance. Therefore, how to convert the SD path under the heterogeneous map into the HD path is a technical problem to be solved.

Disclosure of Invention

The disclosure provides a path determination method, a path determination device, electronic equipment and a storage medium.

According to a first aspect of the present disclosure, there is provided a path determining method, including:

determining a first key feature point of a navigation path in a navigation map;

determining second key feature points corresponding to the first key feature points in the high-precision map based on attribute information corresponding to the first key feature points;

Determining a second sub-path matched with the first sub-path included in the navigation path in the high-precision map by combining attribute information corresponding to the second key feature points;

and determining a target path corresponding to the navigation path in the high-precision map based on a second sub-path matched with the first sub-path in the high-precision map.

According to a second aspect of the present disclosure, there is provided a path determining apparatus comprising:

the first determining module is used for determining a first key feature point of a navigation path in the navigation map;

the second determining module is used for determining second key feature points corresponding to the first key feature points in the high-precision map based on the attribute information corresponding to the first key feature points;

the third determining module is used for determining a second sub-path matched with the first sub-path included in the navigation path in the high-precision map by combining attribute information corresponding to the second key feature points;

and the fourth determining module is used for determining a target path corresponding to the navigation path in the high-precision map based on the second sub-path matched with the first sub-path in the high-precision map.

According to a third aspect of the present disclosure, there is provided an electronic device comprising:

at least one processor;

A memory communicatively coupled to the at least one processor;

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of the embodiments of the present disclosure.

According to a fourth aspect of the present disclosure, there is provided a non-transitory computer readable storage medium storing computer instructions for causing a computer to perform a method according to any one of the embodiments of the present disclosure.

According to a fifth aspect of the present disclosure, there is provided a computer program product comprising a computer program stored on a storage medium, which when executed by a processor, implements a method according to any of the embodiments of the present disclosure.

According to a sixth aspect of the present disclosure there is provided an autonomous vehicle comprising an electronic device as described in the third aspect.

According to the scheme, under the condition that the navigation map and the high-precision map are heterogeneous, the navigation path under the navigation map can be converted into the high-precision path under the high-precision map without local graphic point information, so that the problem that the local graphic point information needs to be acquired when the navigation map and the high-precision map are heterogeneous is effectively solved; meanwhile, the target path corresponding to the navigation path in the high-precision map is determined according to the first key feature point and the second key feature point, so that the speed of path conversion can be increased, the response time of a vehicle end is shortened, and the safety of an automatic driving vehicle is improved.

The foregoing summary is for the purpose of the specification only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features of the present application will become apparent by reference to the drawings and the following detailed description.

Drawings

In the drawings, the same reference numerals refer to the same or similar parts or elements throughout the several views unless otherwise specified. The figures are not necessarily drawn to scale. It is appreciated that these drawings depict only some embodiments according to the disclosure and are not therefore to be considered limiting of its scope.

FIG. 1 is a flow diagram of a path determination method according to an embodiment of the present disclosure;

FIG. 2 is a flow diagram of path conversion according to an embodiment of the present disclosure;

FIG. 3 is a first key feature point schematic diagram of a navigation map according to an embodiment of the present disclosure;

FIG. 4 is a second key feature point schematic of a high-precision map according to an embodiment of the present disclosure;

FIG. 5 is a first sub-path detail schematic of a navigation map according to an embodiment of the present disclosure;

FIG. 6 is a schematic diagram of a precursor path and a subsequent path according to an embodiment of the present disclosure;

FIG. 7 is a schematic illustration of double-sided and single-sided constraints according to an embodiment of the present disclosure;

fig. 8 is a schematic structural view of a path determining apparatus according to an embodiment of the present disclosure;

FIG. 9 is a schematic diagram of a scenario of a path determination method according to an embodiment of the present disclosure;

fig. 10 is a schematic structural diagram of an electronic device for implementing a path determining method of an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to facilitate understanding, and should be considered as merely exemplary. Accordingly, one of ordinary skill in the art will recognize that various changes and modifications of the embodiments described herein can be made without departing from the scope of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

The terms first, second, third and the like in the description and in the claims and in the above-described figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion, such as a series of steps or elements. The method, system, article, or apparatus is not necessarily limited to those explicitly listed but may include other steps or elements not explicitly listed or inherent to such process, method, article, or apparatus.

Before the technical scheme of the embodiments of the present disclosure is described, technical terms possibly used in the present disclosure are further described:

navigation map: real-time geographic data updates and network topology adjustments may be provided to support more efficient network planning, resource allocation, and service deployment.

High-precision map: the map is also called HD map, which is a digital map with high precision, high resolution and high update frequency and is used for positioning, navigation, decision-making and other tasks in automatic driving and intelligent traffic systems.

Shape point information: also called shape point information, a map has a plurality of LINKs (LINKs), each LINK is composed of a plurality of shape points, and the shape point information includes, but is not limited to, position information, angle information, and the like of the shape points.

In the automatic driving field, the automatic auxiliary navigation driving (Navigate onAutopilot, NOA) of the vehicle needs to map a navigation path, such as an SD path, planned by a navigation map provided by an intelligent cabin onto a high-precision map of the intelligent driving field to obtain an HD path, so as to realize intelligent lane-level navigation of the vehicle based on the high-precision map. When the navigation map and the high-precision map are homologous (i.e. the navigation map and the high-precision map belong to the same map provider), an SD and HD association table is generally generated in advance offline by the cloud end, and the vehicle end/cloud end is matched with the SD and HD association table through table lookup based on the SD and HD association table to convert the SD path into the HD path. However, when the navigation map and the high-precision map are heterogeneous (i.e., the navigation map and the high-precision map belong to different map providers), the map model is very different due to different manufacturing processes of the navigation map and the high-precision map, especially because the urban road network is complex. Because the difference of the heterogeneous map is large, the method has higher requirement on algorithm robustness, and the heterogeneous map cannot generate the SD and HD association table in advance. In addition, since the SD path needs to be converted into the HD path in real time at the vehicle end, strict requirements are also imposed on response time and data volume.

At present, the SD path is converted into the HD path mainly by the following two implementation schemes:

off-line association table look-up mode: generating a static SD and HD association table through a navigation map and high-precision map matching algorithm, providing an offline SD and HD association table to the outside, and obtaining an HD path through the SD and HD association table lookup by a vehicle end according to the SD path;

hidden markov model (Hidden Markov Model, HMM) real-time matching approach: and matching the SD path with the HD path in real time through the HMM, and providing the HD path corresponding to the current SD path in real time.

The two schemes are more suitable for homology of the navigation map and the high-precision map and are not suitable for heterology of the high-precision map of the navigation map. The off-line association table look-up method requires that the navigation map and the high-precision map are homologous maps, otherwise off-line matching cannot be performed in advance. Therefore, it is not suitable for the application of the heterogeneous map. The HMM real-time matching method requires the high-precision map provider to provide the detailed data information of the high-precision map, especially the shape point information of the high-precision map, and when the navigation map and the high-precision map are different providers, the requirement cannot be met.

In order to at least partially solve one or more of the above problems and other potential problems, the present disclosure proposes a path determining method, in which, in the case where a navigation map and a high-precision map are heterogeneous, the navigation path under the navigation map can be converted into the high-precision path under the high-precision map without needing local graphic point information, so that the problem that the local graphic point information cannot be acquired when the navigation map and the high-precision map are heterogeneous is effectively solved; meanwhile, the target path corresponding to the navigation path in the high-precision map is determined according to the first key feature point and the second key feature point, so that the speed of path conversion can be increased, the response time of a vehicle end is shortened, and the safety of an automatic driving vehicle is improved.

The embodiment of the disclosure provides a path determining method, and fig. 1 is a schematic flow diagram of the path determining method according to the embodiment of the disclosure, and the path determining method can be applied to a path determining device. The path determining means is located in the electronic device. The electronic device includes, but is not limited to, a stationary device and/or a mobile device. For example, the fixed device includes, but is not limited to, a server, which may be a cloud server or a general server. For example, mobile devices include, but are not limited to, vehicle terminals. In some possible implementations, the path determination method may also be implemented by way of a processor invoking computer readable instructions stored in a memory. As shown in fig. 1, the path determining method includes:

s101: determining a first key feature point of a navigation path in a navigation map;

s102: determining second key feature points corresponding to the first key feature points in the high-precision map based on attribute information corresponding to the first key feature points;

s103: determining a second sub-path matched with the first sub-path included in the navigation path in the high-precision map by combining attribute information corresponding to the second key feature points;

s104: and determining a target path corresponding to the navigation path in the high-precision map based on a second sub-path matched with the first sub-path in the high-precision map.

In some embodiments, the navigation map may be a navigation map for human-machine interaction that may be deployed on a smart class of an autonomous vehicle. In the embodiment of the disclosure, the type of the navigation map is not necessarily limited, and the navigation map may be an SD map or other map that can be deployed on a smart cabin. The high-precision map may be a high-precision, high-resolution, and high-update-frequency digital map that may be deployed on the intelligent driving domain of an autonomous vehicle. Wherein the navigation map and the high-precision map belong to a heterogeneous map. Specifically, the navigation map and the high-precision map are different in provider, and the map making process may be different; and the accuracy of the high-precision map is generally higher than that of the navigation map.

In some implementations, acquiring data of the navigation map can include at least one of: acquiring a starting point, an ending point and key feature points (Turn-By-Turn, TBT) in the SD path; acquiring geometric information such as head and tail point coordinates, length, starting point angle, end point angle and the like of each path (marked as SDLINK) in the SD path; acquiring road grade information of each SDLINK in the SD path; and acquiring road type information such as a ramp, a main road, a secondary road, an intersection, an overhead road, a tunnel, a service area and the like of each SDLINK in the SD path.

In some implementations, acquiring data of the high-precision map can include at least one of: acquiring geometric information such as head and tail point coordinates, length, starting point angle, end point angle and the like of each path (marked as HDLINK) in a high-precision map; acquiring an identifier (Identity Document, ID) and a topology of each HDLINK; obtaining the road grade of each HDLINK; acquiring road type information such as a ramp, a main road, a secondary road, an intersection, an overhead road, a tunnel, a service area and the like of each HDLINK; and acquiring the spatial index data of each HDLINK.

In some implementations, the navigation path can include a plurality of first key feature points. In particular, the navigation path may comprise a start point, an end point and at least one first key feature point, which may be a TBT point.

In some embodiments, the attribute information corresponding to the first key feature point and the second key feature point may include: geometric information, road grade information of a precursor path, road grade information of a subsequent path, road type information of the precursor path, and road type information of the subsequent path. The above is merely exemplary and is not intended to limit all possible contents included in the attribute information, but is not intended to be exhaustive.

In some embodiments, whether the plurality of second key feature points corresponds to the plurality of first key feature points one-to-one exists in the following cases: (1) The X first key feature points correspond to the X second key feature points, namely the second key feature points and the first key feature points are in one-to-one correspondence; (2) X first key feature points correspond to Y second key feature points, and X is more than Y, namely one second key feature point corresponds to at least one first key feature point.

FIG. 2 shows a schematic flow diagram of path conversion, wherein the left side of the flow diagram is shown in FIG. 2, and a navigation map is shown, and a navigation path in the navigation map is determined according to a starting point and an ending point; and extracting relevant information of the navigation path, sending the relevant information of the navigation path to a high-precision map, and obtaining a target path corresponding to the navigation path in the high-precision map shown on the right side, wherein the target path can be applied to auxiliary driving of an automatic driving vehicle, and the auxiliary driving can be any driving mode of the automatic driving vehicle, such as a full-automatic driving mode, a man-machine co-driving mode and the like.

In some implementations, the navigation path can include a plurality of first sub-paths. Specifically, the navigation path includes at least two first sub-paths. For example, if the navigation path includes a first key feature point, the navigation path includes a first sub-path between the start point and the first key feature point and a first sub-path between the first key feature point and the end point. Still further exemplary, if the navigation path includes M first key feature points, the navigation path includes a first sub-path between the start point and the 1 st first key feature point, a first sub-path between the 1 st first key feature point and the 2 nd first key feature point, a first sub-path between the 2 nd first key feature point and the 3 rd first key feature point, …, and a first sub-path between the M first key feature point and the end point, M being an integer of 2 or more.

Fig. 3 shows a schematic diagram of first key feature points of a navigation map, as shown in fig. 3, in which one navigation path includes 4 first key feature points (Key Feature Points, KFP), denoted as KFP1, KFP2, KFP3 and KFP4, and the navigation path includes a plurality of first sub-paths, such as a first sub-path from a start point to KFP1, a first sub-path from KFP1 to KFP2, a first sub-path from KFP2 to KFP3, a first sub-path from KFP3 to KFP4, and a first sub-path from KFP4 to an end point.

According to the technical scheme, first key feature points of a navigation path in a navigation map are determined; determining second key feature points corresponding to the first key feature points in the high-precision map based on attribute information corresponding to the first key feature points; determining a second sub-path matched with the first sub-path included in the navigation path in the high-precision map by combining attribute information corresponding to the second key feature points; and determining a target path corresponding to the navigation path in the high-precision map based on a second sub-path matched with the first sub-path in the high-precision map. Under the condition that the navigation map and the high-precision map are heterogeneous, a target path matched with the navigation path can be determined by acquiring the first key feature point of the navigation map and the second key feature point of the high-precision map, and the problem that the local graphic point information is required to be acquired when the navigation map and the high-precision map are heterogeneous is effectively solved; meanwhile, the target path corresponding to the navigation path in the high-precision map is determined according to the first key feature point and the second key feature point, so that the speed of path conversion can be increased, the response time of a vehicle end is shortened, and the safety of an automatic driving vehicle is improved.

In some embodiments, determining, in the high-precision map, a second key feature point corresponding to the first key feature point based on attribute information corresponding to the first key feature point includes: acquiring path data in a preset range of a navigation path in the high-precision map based on attribute information corresponding to the first key feature points; determining candidate key feature points related to the navigation path based on the path data; and determining a second key feature point corresponding to the first key feature point in the high-precision map based on the candidate key feature points related to the navigation path.

In some embodiments, the attribute information corresponding to the first key feature point may include: geometric information of the first key feature point, attribute information of a precursor path of the first key feature point and attribute information of a subsequent path. Wherein the attribute information includes road class information and road type information.

In some embodiments, path data within a preset range of navigation paths in the high-precision map may be obtained by spatially indexing the high-precision map. Specifically, the spatial index is a data structure arranged according to a certain sequence according to spatial relations such as the position, the shape and the like of the HDLINK; the spatial index of the HDLINK may include summary information of the HDLINK, such as identification information of the HDLINK, an HDLINK circumscribed rectangle, and a pointer of the HDLINK to a spatial object entity. A spatial index (also known as a spatial data query) is used to describe the location information of data stored on the medium.

In some embodiments, the candidate key feature points may be path data within a preset range of navigation paths in the high-precision map; candidate key feature points associated with the navigation path are determined based on the path data. Wherein the number of candidate key feature points is greater than the number of second key feature points. Illustratively, a plurality of candidate key feature points within a range of 100 meters around a navigation path (SDpath) are obtained by spatial indexing according to HDLINK. The preset range can be set manually or can be set or adjusted automatically by the system. Here, the path data within the preset range is actually the data of all the second sub-paths within the preset range.

Fig. 4 shows a schematic diagram of second key feature points of the high-precision map, as shown in fig. 4, wherein 11 second key feature points are included in the high-precision map, and are denoted as KFP 1-KFP 11, and paths between adjacent second key feature points are called second sub-paths, such as second sub-paths between KFP1 and KFP2, second sub-paths between KFP2 and KFP3, second sub-paths between KFP3 and KFP4, and second sub-paths between KFP10 and KFP 11.

In some implementations, the target path in the high-definition map can include a plurality of second sub-paths. Specifically, a plurality of second sub-paths in the high-definition map, which are respectively matched with the plurality of first sub-paths, are sequentially connected according to the passing direction of the automatic driving vehicle, so as to obtain a target path (namely, an HD path). Here, the plurality of first sub-paths corresponds one-to-one to the plurality of second sub-paths.

According to the technical scheme, path data in a preset range of a navigation path in a high-precision map is obtained based on attribute information corresponding to a first key feature point; determining candidate key feature points related to the navigation path based on the path data; and determining a second key feature point corresponding to the first key feature point in the high-precision map based on the candidate key feature points related to the navigation path. Therefore, the second key feature points can be obtained from the candidate key feature points, so that the accuracy of determining the path under the heterogeneous map is improved, and the safety of the automatic driving vehicle is further improved.

In an embodiment of the present disclosure, determining, in a high-precision map, a second key feature point corresponding to a first key feature point based on candidate key feature points related to a navigation path includes: determining candidate key feature points corresponding to the first key feature points based on the candidate key feature points related to the navigation path; and determining a second key feature point corresponding to the first key feature point based on the candidate key feature point corresponding to the first key feature point.

Here, determining the candidate key feature point corresponding to the first key feature point includes: and taking the candidate key feature points in the preset range of each first key feature point in the high-precision map as candidate key feature points corresponding to the first key feature points.

The method comprises the steps of determining a first key feature point of a navigation path, performing spatial indexing according to an HDLINK, obtaining candidate key feature points in a range of 100 meters around the navigation path, obtaining 2 candidate key feature points corresponding to the first key feature point KFP1, and respectively calculating the similarity between the first key feature point and the 2 candidate key feature points; the similarity between the first key feature point and the first candidate key feature point is 0.3; the similarity between the first key feature point and the second candidate key feature point is 0.8; and determining the second candidate key feature point as a second key feature point corresponding to the first key feature point.

According to the technical scheme, candidate key feature points corresponding to the first key feature points are determined based on candidate key feature points related to the navigation path; and determining a second key feature point corresponding to the first key feature point based on the candidate key feature point corresponding to the first key feature point. Therefore, the second key feature points of the high-precision map can be accurately matched for the first key feature points in the navigation path, so that the high-precision path corresponding to the navigation path is obtained, the accuracy of path determination is improved, and the safety of the automatic driving vehicle is improved.

In an embodiment of the present disclosure, determining, based on candidate key feature points corresponding to the first key feature points, a second key feature point corresponding to the first key feature points includes: determining the similarity between the first key feature points and candidate key feature points corresponding to the first key feature points; and determining candidate key feature points with the similarity larger than a preset threshold corresponding to the first key feature points as second key feature points corresponding to the first key feature points.

In some embodiments, the preset threshold may be set manually or may be set automatically by the system. The preset threshold can be set or adjusted according to the precision requirement.

In some embodiments, the similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point may include: geometric similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point, road grade similarity of a precursor path between the first key feature point and the candidate key feature point corresponding to the first key feature point, road grade similarity of a subsequent path between the first key feature point and the candidate key feature point corresponding to the first key feature point, road attribute similarity of the precursor path between the first key feature point and the candidate key feature point corresponding to the first key feature point, and road attribute similarity of the subsequent path. The similarity calculation formula is shown as formula (1):

Similarity＝geom_Similarity*Link_roadclass_Similarity*

Link_roadtype_similarity formula (1)

The Similarity represents Similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point; the geom_similarity represents the Similarity of geometric information between the first key feature point and the candidate key feature point corresponding to the first key feature point; link_roadclass_similarity represents the Similarity of the precursor subsequent path road type information between the first key feature point and the candidate key feature point corresponding to the first key feature point; link_roadtype_similarity represents a precursor subsequent path road class information Similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point.

In some implementations, the geometric information similarity may include: geometrical distance similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point, and precursor subsequent path angle similarity.

In some embodiments, the precursor subsequent path road type information similarity may include: and the similarity of the precursor subsequent path ramp, the main road, the auxiliary road, the intersection, the overhead road, the tunnel and the service area between the first key feature point and the candidate key feature point corresponding to the first key feature point.

Here, the precursor subsequent path is an abbreviation of a precursor path and a subsequent path.

According to the technical scheme, the similarity between the first key feature point and the candidate key feature points is calculated from the geometric information, the precursor path road grade, the subsequent path road grade, the precursor path road attribute, the subsequent path road attribute and the like, so that the accuracy of the second key feature point corresponding to the determined first key feature point can be improved, the accuracy of path determination is further improved, and the safety of an automatic driving vehicle is further improved.

In an embodiment of the present disclosure, determining a similarity between a first key feature point and a candidate key feature point corresponding to the first key feature point includes: and determining the similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point based on the attribute information corresponding to the first key feature point and the attribute information of the candidate key feature point corresponding to the first key feature point.

In an embodiment of the present disclosure, when determining the first key feature point, the method may further include: acquiring attribute information of a first key feature point; the attribute information may include geometric information of the first key feature point, road class information of the preceding subsequent path, and road type information of the preceding subsequent path.

In an embodiment of the present disclosure, when determining the second key feature point, the method may further include: acquiring attribute information of a second key feature point; the attribute information may include geometric information of the second key feature point, road class information of the preceding subsequent path, and road type information of the preceding subsequent path.

In an embodiment of the present disclosure, determining a similarity between a first key feature point and a candidate key feature point corresponding to the first key feature point includes: calculating the geometrical information similarity of the first key feature point and the candidate key feature points; calculating the precursor path road grade similarity of the first key feature point and the candidate key feature points; calculating the grade similarity of a subsequent path between the first key feature point and a plurality of candidate key feature points; calculating the precursor path road type similarity of the first key feature point and the candidate key feature points; calculating the similarity of the type of the subsequent path road between the first key feature point and the candidate key feature point; and multiplying the geometric similarity of each candidate key feature point and the first key feature point, the precursor path road class similarity, the subsequent path road class similarity, the precursor path road type similarity and the subsequent path road type similarity to obtain the similarity of each candidate key feature point and the first key feature point.

Illustratively, determining a first key feature point of the navigation path; obtaining two candidate key feature points based on the first key feature points; the geometrical similarity of the first key feature point and the first candidate key feature point is 0.5, the similarity of the road class of the precursor path is 0.1, the similarity of the road class of the subsequent path is 0.3, the similarity of the road type of the precursor path is 0.2 and the similarity of the road type of the subsequent path is 0.1; the geometrical similarity of the first key feature point and the second candidate key feature point is 0.6, the similarity of the road class of the precursor path is 0.7, the similarity of the road class of the subsequent path is 0.8, the similarity of the road type of the precursor path is 0.4 and the similarity of the road type of the subsequent path is 0.6; multiplying the geometric similarity of the two candidate key feature points, the road class similarity of the precursor path, the road class similarity of the subsequent path, the road type similarity of the precursor path and the road type similarity of the subsequent path to obtain a similarity of 0.0003 between the first key feature point and the first candidate key feature point and a similarity of 0.08064 between the first key feature point and the second candidate key feature point; and determining the second candidate key feature point as a second key feature point of the first key feature point.

According to the technical scheme, the similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point is determined based on the attribute information corresponding to the first key feature point and the attribute information of the candidate key feature point corresponding to the first key feature point. In this way, the similarity of the first key feature point and the candidate key feature points corresponding to the first key feature point can be considered from multiple aspects, so that the second key feature point with higher similarity with the first key feature point is obtained, and the accuracy of path determination is improved.

In an embodiment of the present disclosure, the attribute information includes at least one of: geometric information, road class information of a precursor path, road type information of a precursor path, road class information of a subsequent path, and road type information of a subsequent path.

In some embodiments, the geometric information may include geometric information such as head-to-tail coordinates, length, start angle, and end angle. The above is merely exemplary and is not intended to limit the total possible content of the geometric information included, but is not intended to be exhaustive.

In some embodiments, the road grade information may include expressways, arterial roads, major highways, minor highways, branches, rural highways, residential roads, and streets. The above is merely exemplary and is not intended to limit the total possible content included in the road class information, but is not intended to be exhaustive.

In some embodiments, the road type information may include urban roads, rural roads, and highways. The above is merely exemplary and is not intended to limit the total possible content included in the road type information, but is not intended to be exhaustive.

According to the technical scheme, the second sub-path matched with the first sub-path in the high-precision map is determined based on the attribute information of the first key feature point and the attribute information of the second key feature point, and data support can be provided for determining the target path of the high-precision map, so that the accuracy of path determination is improved, and further safety of an automatic driving vehicle is improved.

In an embodiment of the present disclosure, in combination with the second key feature point, determining, in the high-precision map, a second sub-path that matches the first sub-path included in the navigation path may include: for a first type first sub-path with two second key feature points matched, determining a first type second sub-path matched with the first type first sub-path in a high-precision map in a bilateral constraint mode, wherein the first type first sub-path comprises two adjacent first key feature points.

In an embodiment of the present disclosure, in combination with the second key feature point, determining, in the high-precision map, a second sub-path that matches the first sub-path included in the navigation path may include: for a second class first sub-path with a second key feature point, determining a second class second sub-path matched with the second class first sub-path in the high-precision map in a single-side constraint mode, wherein the second class first sub-path comprises the first key feature point.

In an embodiment of the present disclosure, in combination with the second key feature point, determining, in the high-precision map, a second sub-path that matches the first sub-path included in the navigation path may include: and determining a third type second sub-path matched with the third type first sub-path in the high-precision map based on the coordinate information and the attribute information of the third type first sub-path aiming at the third type first sub-path without the second key feature point matching, wherein the third type first sub-path comprises at least one first key feature point.

Fig. 5 shows a detailed schematic diagram of a first sub-path of the navigation path, as shown in fig. 5, with a solid circle representing a first key feature point and a bold solid line representing the navigation path, the navigation path comprising 3 first sub-paths, denoted LINK1, LINK2 and LINK3.

In some embodiments, matching a precursor path corresponding to a first key feature point with a precursor path of a second key feature point corresponding to the first key feature point; and matching the subsequent path corresponding to the first key feature point with the subsequent path of the second key feature point corresponding to the first key feature point.

Fig. 6 shows a schematic diagram of the precursor path and the subsequent path, as shown in fig. 6, the subsequent path of LINK1 being LINK2, LINK3 and LINK4; the precursor paths of LINK2, LINK3 and LINK4 are LINK1; the subsequent path for LINK2 is LINK5.

In some embodiments, the first class second sub-path may be derived by a double-sided constrained hidden Markov model; the first type second sub-path is a path corresponding to the first sub-path with two key feature point matches. The first class second sub-path may be derived by a double-sided constrained hidden Markov model. The second sub-path is a path corresponding to the first sub-path with a key feature point matching; specifically, the one key feature point may be a head end point or a tail end point. The third class of second sub-paths are paths corresponding to the first sub-paths having no key feature point matches.

In some embodiments, the first class second sub-path, the second class second sub-path, and the third class second sub-path, are sequentially less accurate.

FIG. 7 shows a two-sided constraint and a one-sided constraint schematic, as shown in FIG. 7 with the top side representing the navigation path and the bottom side representing the high-precision path. The navigation path comprises a first key feature point SD-KFP1 and a first key feature point SD-KFP2, and paths between the SD-KFP1 and the SD-KFP2 are called first sub-paths; in the high-precision map, a second key feature point corresponding to the first key feature point SD-KFP1 is HD-KFP1, and a second key feature point corresponding to the first key feature point SD-KFP2 is HD-KFP6; the path between HD-KFP1 and HD-KFP2 is referred to as the second sub-path. For SD-KFP1, the precursor path is SDLINK-1, and the subsequent path is SDLINK-2; for SD-KFP2, its precursor path is SDLINK-6. For the HD-KFP1, the precursor path is HDLINK-1, and the subsequent paths are HDLINK-2 and HDLINK-2'; for HD-KFP6, its precursor path is HDLINK-6. The path from the start point to SD-KFP1 is called a first sub path and is marked as subSDpath1; the path from the start point to HD-KFP1 is called a second sub path and is marked as subHDpath1; here, SDLINK-1 matches HDLINK-1, which belongs to the unilateral constraint where subSDPath1 matches subHDPath 1. Between SD-KFP1 and SD-KFP2, also referred to as a first sub-path, denoted as subSDpath2; the second sub-path between HD-KFP1 and HD-KFP6 is also called as subHDpath2; SDLINK-2 is matched with HDLINK-2, and SDLINK-6 is matched with HDLINK-6, belonging to the bilateral constraint that subSDpath2 is matched with subHDpath 2.

According to the technical scheme, the first sub-paths of the first type are respectively processed differently according to the number of the second key feature point matches, so that the response speed of path determination can be improved. Meanwhile, the accuracy of determining the HD path in the high-precision map corresponding to the SD path can be improved by carrying out bilateral constraint on the first sub-path of the first type and unilateral constraint on the first sub-path of the second type.

In the embodiment of the disclosure, determining a first type second sub-path matched with a first type first sub-path in a high-precision map by adopting a bilateral constraint mode comprises: matching a precursor path corresponding to a first key feature point in the first sub-path of the first type with a precursor path of a second key feature point corresponding to the first key feature point; matching a subsequent path corresponding to a second first key feature point in the first sub-path of the first type with a subsequent path of a second key feature point corresponding to the second first key feature point; the direction from the first key feature point to the second first key feature point is the direction of the first sub-path of the first type.

In some embodiments, determining the first type second sub-path that matches the first type first sub-path may incorporate bilateral constraints, and obtain the first type second sub-path through a pre-trained matching model. Here, the pre-trained matching model may be a hidden markov model.

In some embodiments, the head and tail points of each SDLINK in the first sub-path are spliced into a series of SD track points according to the traffic direction; if the head and tail SDLINKs are matched with the HDLINKs through the key feature points, the most similar second sub-path is matched through the HMM of the bilateral constraint.

In some embodiments, the HMM of the bilateral constraint is a statistical model for sequence modeling. It is a variant of HMM that is used to model and analyze sequences with certain constraints. In a conventional HMM, the sequence generation process is a one-way generation process from a hidden state to an observable state. While HMMs of bilateral constraints introduce bi-directional constraints, i.e. the sequence generation process takes into account both the generation from hidden state to observable state and the generation from observable state to hidden state. Specifically, the HMM of the bilateral constraint includes two HMMs: a forward HMM and a reverse HMM. The forward HMM is used for generation from the hidden state to the observable state and the reverse HMM is used for generation from the observable state to the hidden state. The two HMMs are mutually influenced by constraint conditions, so that bidirectional information transfer and constraint are realized. The HMM with the double-side constraint is introduced, so that the context information and the dependency relationship in the sequence can be better captured, and the modeling accuracy and generalization capability are improved.

According to the technical scheme, a precursor path corresponding to a first key feature point in a first sub-path of a first type is matched with a precursor path of a second key feature point corresponding to the first key feature point; and matching a subsequent path corresponding to a second first key feature point in the first sub-path of the first type with a subsequent path of a second key feature point corresponding to the second first key feature point. Therefore, the accuracy of converting the navigation path into the high-precision path can be improved through bilateral constraint under the condition of lacking shape point information, and the accuracy of path determination is improved.

In the embodiment of the disclosure, determining a second class second sub-path matched with the second class first sub-path in a high-precision map by adopting a single-side constraint mode includes: responding to the second type first sub-path to comprise a starting point of the navigation path and a first key feature point connected with the starting point, and matching a precursor path corresponding to the first key feature point connected with the starting point with a precursor path of a second key feature point corresponding to the first key feature point connected with the starting point; and responding to the second type first sub-path to comprise an end point of the navigation path and a first key feature point connected with the end point, and matching a subsequent path corresponding to the first key feature point connected with the end point with a subsequent path of a second key feature point corresponding to the first key feature point connected with the end point.

In some embodiments, the determining of the second-class second sub-path matching the second-class first sub-path may incorporate a single-side constraint, and the second-class second sub-path may be obtained through a pre-trained matching model. Here, the pre-trained matching model may be a hidden markov model.

In some embodiments, the head and tail points of each SDLINK in the first sub-path are spliced into a series of SD path track points according to the traffic direction; if the head/tail SDLINK is matched with the HDLINK through the key feature points, the most similar second sub-path is matched through a single-side constraint hidden Markov model.

In some embodiments, the single-sided constrained hidden Markov model is a time-series model for modeling a sequence of discrete events with hidden states. In this model, the observation sequence depends only on the sequence of hidden states, which are subject to a specific constraint. In conventional hidden markov models, transitions between hidden states are bi-directional, i.e., any one hidden state may transition to any other state. In a single-side constrained hidden markov model, however, transitions between hidden states are unidirectional and there is a specific order of transitions. Such constraints may be implemented by limiting the form of the transition probability matrix. The hidden Markov model with single side constraint can improve the fitting capacity and generalization capacity of the model by reasonably designing and applying the constraint, thereby improving the accuracy of path determination.

According to the technical scheme, in response to the second type first sub-path comprising a starting point of a navigation path and a first key feature point connected with the starting point, a precursor path corresponding to the first key feature point connected with the starting point is matched with a precursor path of a second key feature point corresponding to the first key feature point connected with the starting point; and responding to the second type first sub-path to comprise an end point of the SD path and a first key feature point connected with the end point, and matching a subsequent path corresponding to the first key feature point connected with the end point with a subsequent path of a second key feature point corresponding to the first key feature point connected with the end point. Therefore, the accuracy of converting the navigation path under the navigation map into the high-precision path under the high-precision map can be improved through single-side constraint under the condition of lacking shape point information, so that the accuracy of path conversion is improved.

In some embodiments, deriving the third class second sub-path based on the coordinate information and the attribute information of the third class first sub-path includes: and obtaining a third type second sub-path through a pre-trained matching model based on the coordinate information and the attribute information of the third type first sub-path and the coordinate information and the attribute information of the third type second sub-path.

Here, the pre-trained matching model may be a hidden markov model.

In this way, even in the case of lack of shape point information, the navigation path under the navigation map can be converted into the high-precision path under the high-precision map by the hidden Markov model, so that the speed of path conversion is improved.

In an embodiment of the present disclosure, the path determining method may further include: the target path is transmitted to the autonomous vehicle to generate control information by the autonomous vehicle in conjunction with the target path, the control information being used to control the autonomous vehicle to travel.

In some embodiments, the target path is sent to a perception module of the autonomous vehicle, the perception module for generating control information of the autonomous vehicle based on the target path; the control information may include vehicle travel parameters, vehicle direction parameters, and the like.

In some embodiments, the target path is sent to a decision module of the autonomous vehicle for generating control information of the autonomous vehicle based on the target path; the control information may include vehicle travel parameters, vehicle direction parameters, whether to park, and the like.

Here, the target path can be applied to projects such as lane changing reminding of NOA, path calculation and guidance of man-machine co-driving, and the like.

According to the technical scheme, the target path is sent to the automatic driving vehicle, so that the automatic driving vehicle can combine with the target path to generate control information for controlling the automatic driving vehicle to run. Thus, the control information is generated through the target path in the high-precision map, so that the accuracy of the control information can be improved, and the safety of the automatic driving vehicle can be improved.

In the embodiments of the present disclosure, the sources of the navigation map and the high-precision map are different.

Illustratively, the navigation map is provided by a first object and the high-definition map is provided by a second object, the first object and the second object being two different objects.

According to the technical scheme, the map point information is not needed, the navigation path under the navigation map can be converted into the high-precision path under the high-precision map through the key feature points, and the problem that the heterogeneous navigation map and the high-precision map cannot obtain the map point information is effectively avoided. Because the target path corresponding to the navigation path in the high-precision map is determined according to the first key feature points and the second key feature points, the speed of path conversion can be improved, and the response time of a vehicle end is shortened, so that the safety of an automatic driving vehicle is improved.

The embodiment of the disclosure provides a path determining apparatus, as shown in fig. 8, which may include: a first determining module 801, configured to determine a first key feature point of a navigation path in a navigation map; a second determining module 802, configured to determine, in the high-precision map, a second key feature point corresponding to the first key feature point based on attribute information corresponding to the first key feature point; a third determining module 803, configured to determine, in combination with attribute information corresponding to the second key feature point, a second sub-path that matches the first sub-path included in the navigation path in the high-precision map; a fourth determining module 804, configured to determine a target path corresponding to the navigation path in the high-precision map based on the second sub-path in the high-precision map that matches the first sub-path.

In some embodiments, the second determining module 802 includes: the acquisition sub-module is used for acquiring path data in a preset range of a navigation path in the high-precision map based on attribute information corresponding to the first key feature points; a first determination sub-module for determining candidate key feature points related to the navigation path based on the path data; and the second determining submodule is used for determining a second key feature point corresponding to the first key feature point in the high-precision map based on the candidate key feature point related to the navigation path.

In some embodiments, the second determination submodule is configured to: determining candidate key feature points corresponding to the first key feature points based on the candidate key feature points related to the navigation path; and determining a second key feature point corresponding to the first key feature point based on the candidate key feature point corresponding to the first key feature point.

In some embodiments, the second determination submodule is further configured to: determining the similarity between the first key feature points and candidate key feature points corresponding to the first key feature points; and determining candidate key feature points with the similarity larger than a preset threshold corresponding to the first key feature points as second key feature points corresponding to the first key feature points.

In some embodiments, the second determination submodule is further configured to: and determining the similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point based on the attribute information corresponding to the first key feature point and the attribute information of the candidate key feature point corresponding to the first key feature point.

In some embodiments, the attribute information includes at least one of: geometric information, road grade information of a precursor path, road type information of the precursor path, road grade information of a subsequent path and road type information of the subsequent path.

In some embodiments, the third determining module 803 includes: and the third determining sub-module is used for determining a first type second sub-path matched with the first type first sub-path in the high-precision map by adopting a bilateral constraint mode aiming at the first type first sub-path matched with two second key feature points, wherein the first type first sub-path comprises two adjacent first key feature points. The third determining module 803 includes: and the fourth determining sub-module is used for determining a second class second sub-path matched with the second class first sub-path in the high-precision map by adopting a single-side constraint mode aiming at the second class first sub-path matched with a second key feature point, wherein the second class first sub-path comprises the first key feature point. The third determining module 803 includes: and a fifth determining sub-module, configured to determine, for a third type of first sub-path that does not have a second key feature point match, a third type of second sub-path that matches the third type of first sub-path in the high-precision map based on the coordinate information and the attribute information of the third type of first sub-path, where the third type of first sub-path includes at least one first key feature point.

In some embodiments, the third determination submodule is configured to: matching a precursor path corresponding to a first key feature point in the first sub-path of the first type with a precursor path of a second key feature point corresponding to the first key feature point; matching a subsequent path corresponding to a second first key feature point in the first sub-path of the first type with a subsequent path of a second key feature point corresponding to the second first key feature point; the direction from the first key feature point to the second first key feature point is the direction of the first sub-path of the first type.

In some embodiments, the fourth determination submodule is configured to: responding to the second type first sub-path to comprise a starting point of the navigation path and a first key feature point connected with the starting point, and matching a precursor path corresponding to the first key feature point connected with the starting point with a precursor path of a second key feature point corresponding to the first key feature point connected with the starting point; and responding to the second type first sub-path to comprise an end point of the navigation path and a first key feature point connected with the end point, and matching a subsequent path corresponding to the first key feature point connected with the end point with a subsequent path of a second key feature point corresponding to the first key feature point connected with the end point.

In some embodiments, the path determining apparatus further comprises: a transmitting module (not shown in fig. 8) for transmitting the target path to the autonomous vehicle to generate control information for controlling the autonomous vehicle to travel in conjunction with the target path by the autonomous vehicle.

In some embodiments, the navigation map and the high-precision map are of different sources.

It should be understood by those skilled in the art that the functions of each processing module in the path determining apparatus according to the embodiments of the present disclosure may be understood by referring to the foregoing description of the path determining method, and each processing module in the path determining apparatus according to the embodiments of the present disclosure may be implemented by using an analog circuit that implements the functions of the embodiments of the present disclosure, or may be implemented by running software that implements the functions of the embodiments of the present disclosure on an electronic device.

According to the path determining device disclosed by the embodiment of the disclosure, under the condition that the navigation map and the high-precision map are heterogeneous, the navigation path under the navigation map can be converted into the high-precision path under the high-precision map without the need of local graphic point information, so that the problem that the local graphic point information is acquired when the navigation map and the high-precision map are heterogeneous is effectively avoided, the speed and the accuracy of path matching can be improved, and the safety of an automatic driving vehicle is improved.

The embodiment of the disclosure provides a scene diagram of path determination, as shown in fig. 9.

As described above, the path determining method provided by the embodiment of the present disclosure is applied to an electronic device. Electronic devices are intended to represent various forms of digital computers, such as servers, blade servers, mainframes, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as a car-end digital assistant, a car-end telephone, and other similar computing devices.

In particular, the electronic device may specifically perform the following operations:

determining a first key feature point of a navigation path in a navigation map;

The first key feature points of the navigation path can be obtained from a navigation map of the automatic driving vehicle. Navigation maps and high-precision maps may be obtained from map data sources connected by autonomous vehicles. The map data source may be various forms of data storage devices, such as a laptop computer, desktop computer, workstation, personal digital assistant, server, blade server, mainframe computer, and other suitable computers. The map data source may also represent various forms of mobile devices, such as a car-end digital assistant, a car-end telephone, and other similar computing devices. In addition, the map data source may be located at a vehicle end or at a cloud end.

It should be understood that the scene diagram shown in fig. 9 is merely illustrative and not restrictive, and that various obvious changes and/or substitutions may be made by one skilled in the art based on the example of fig. 9, and the resulting technical solutions still fall within the scope of the disclosure of the embodiments of the present disclosure.

In the technical scheme of the disclosure, the acquisition, storage, application and the like of the related user personal information all conform to the regulations of related laws and regulations, and the public sequence is not violated.

According to embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, a computer program product, and an autonomous vehicle.

Fig. 10 shows a schematic block diagram of an example electronic device 1000 that may be used to implement embodiments of the present disclosure. Electronic devices are intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. The electronic device may also represent various forms of mobile apparatuses, such as personal digital assistants, cellular telephones, smartphones, wearable devices, and other similar computing apparatuses. The components shown herein, their connections and relationships, and their functions, are meant to be exemplary only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 10, the apparatus 1000 includes a computing unit 1001 that can perform various appropriate actions and processes according to a computer program stored in a Read-Only Memory (ROM) 1002 or a computer program loaded from a storage unit 1008 into a random access Memory (Random Access Memory, RAM) 1003. In the RAM 1003, various programs and data required for the operation of the device 1000 can also be stored. The computing unit 1001, the ROM 1002, and the RAM 1003 are connected to each other by a bus 1004. An Input/Output (I/O) interface 1005 is also connected to bus 1004.

Various components in device 1000 are connected to I/O interface 1005, including: an input unit 1006 such as a keyboard, a mouse, and the like; an output unit 1007 such as various types of displays, speakers, and the like; a storage unit 1008 such as a magnetic disk, an optical disk, or the like; and communication unit 1009 such as a network card, modem, wireless communication transceiver, etc. Communication unit 1009 allows device 1000 to exchange information/data with other devices via a computer network, such as the internet, and/or various telecommunications networks.

The computing unit 1001 may be a variety of general and/or special purpose processing components having processing and computing capabilities. Some examples of computing unit 1001 include, but are not limited to, a central processing unit CPU, a graphics processing unit (Graphics Processing Unit, GPU), various dedicated artificial intelligence (Artificial Intelligence, AI) computing chips, various computing units running machine learning model algorithms, digital signal processors (Digital Signal Processor, DSP), and any suitable processors, controllers, microcontrollers, and the like. The computing unit 1001 performs the respective methods and processes described above, for example, a path determination method. For example, in some embodiments, the path determination method may be implemented as a computer software program tangibly embodied on a machine-readable medium, such as the storage unit 1008. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 1000 via ROM 1002 and/or communication unit 1009. When the computer program is loaded into the RAM 1003 and executed by the computing unit 1001, one or more steps of the path determination method described above may be performed. Alternatively, in other embodiments, the computing unit 1001 may be configured to perform the path determination method in any other suitable way (e.g., by means of firmware).

Various implementations of the systems and techniques described here above can be implemented in digital electronic circuitry, integrated circuitry, field programmable gate arrays (Field Programmable Gate Array, FPGAs), application specific integrated circuits (Application Specific Integrated Circuit, ASICs), application-specific standard products (ASSPs), system On Chip (SOC), complex programmable logic devices (Complex Programmable Logic Device, CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs, the one or more computer programs may be executed and/or interpreted on a programmable system including at least one programmable processor, which may be a special purpose or general-purpose programmable processor, that may receive data and instructions from, and transmit data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for carrying out methods of the present disclosure may be written in any combination of one or more programming languages. These program code may be provided to a processor or controller of a general purpose computer, special purpose computer, or other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions/operations specified in the flowchart and/or block diagram to be implemented. The program code may execute entirely on the machine, partly on the machine, as a stand-alone software package, partly on the machine and partly on a remote machine or entirely on the remote machine or server.

In the context of this disclosure, a machine-readable medium may be a tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. The machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. The machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access Memory, a read-Only Memory, an erasable programmable read-Only Memory (EPROM), a flash Memory, an optical fiber, a portable compact disc read-Only Memory (Compact Disk Read Only Memory, CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having: a display device (e.g., cathode Ray Tube (CRT) or liquid crystal display (Liquid Crystal Display, LCD) monitor) for displaying information to a user; and a keyboard and pointing device (e.g., a mouse or trackball) by which a user can provide input to the computer. Other kinds of devices may also be used to provide for interaction with a user; for example, feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form, including acoustic input, speech input, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a background component (e.g., as a data server), or that includes a middleware component (e.g., an application server), or that includes a front-end component (e.g., a user computer having a graphical user interface or a web browser through which a user can interact with an implementation of the systems and techniques described here), or any combination of such background, middleware, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (Local Area Network, LAN), wide area network (Wide Area Network, WAN) and the internet.

The computer system may include a client and a server. The client and server are typically remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. The server may be a cloud server, a server of a distributed system, or a server incorporating a blockchain.

It should be appreciated that various forms of the flows shown above may be used to reorder, add, or delete steps. For example, the steps recited in the present disclosure may be performed in parallel, sequentially, or in a different order, provided that the desired results of the disclosed aspects are achieved, and are not limited herein.

The above detailed description should not be taken as limiting the scope of the present disclosure. It will be apparent to those skilled in the art that various modifications, combinations, sub-combinations and alternatives are possible, depending on design requirements and other factors. Any modifications, equivalent substitutions, improvements, etc. that are within the principles of the present disclosure are intended to be included within the scope of the present disclosure.

Claims

1. A path determination method, comprising:

determining a first key feature point of a navigation path in a navigation map;

determining a second key feature point corresponding to the first key feature point in a high-precision map based on attribute information corresponding to the first key feature point;

And determining a target path corresponding to the navigation path in the high-precision map based on the second sub-path matched with the first sub-path in the high-precision map.

2. The method of claim 1, wherein the determining, in a high-precision map, a second key feature point corresponding to the first key feature point based on the attribute information corresponding to the first key feature point, includes:

acquiring path data in a preset range of the navigation path in the high-precision map based on attribute information corresponding to the first key feature points;

determining candidate key feature points related to the navigation path based on the path data;

and determining a second key feature point corresponding to the first key feature point in a high-precision map based on the candidate key feature point related to the navigation path.

3. The method of claim 2, wherein the determining, in a high-precision map, a second key feature point corresponding to the first key feature point based on the candidate key feature points related to the navigation path comprises:

determining candidate key feature points corresponding to the first key feature points based on the candidate key feature points related to the navigation path;

Determining the similarity between the first key feature point and the candidate key feature point corresponding to the first key feature point based on the attribute information corresponding to the first key feature point and the attribute information of the candidate key feature point corresponding to the first key feature point;

and determining the candidate key feature points, corresponding to the first key feature points, with the similarity larger than a preset threshold value as second key feature points corresponding to the first key feature points.

4. The method of claim 1, wherein the attribute information comprises at least one of:

geometric information, road grade information of a precursor path, road type information of the precursor path, road grade information of a subsequent path and road type information of the subsequent path.

5. The method of claim 1, wherein the determining, in the high-precision map, a second sub-path that matches the first sub-path included in the navigation path in combination with the second key feature point comprises:

for a first sub-path of a first type with two second key feature points matched, determining a second sub-path of the first type matched with the first sub-path of the first type in the high-precision map by adopting a bilateral constraint mode, wherein the first sub-path of the first type comprises two adjacent first key feature points.

6. The method of claim 1, wherein the determining, in the high-precision map, a second sub-path that matches the first sub-path included in the navigation path in combination with the second key feature point comprises:

for a second class first sub-path with a second key feature point, determining a second class second sub-path matched with the second class first sub-path in the high-precision map in a single-side constraint mode, wherein the second class first sub-path comprises a first key feature point.

7. The method of claim 1, wherein the determining, in the high-precision map, a second sub-path that matches the first sub-path included in the navigation path in combination with the second key feature point comprises:

and determining a third type second sub-path matched with the third type first sub-path in the high-precision map based on the coordinate information and the attribute information of the third type first sub-path aiming at the third type first sub-path without second key feature point matching, wherein the third type first sub-path comprises at least one first key feature point.

8. The method of claim 5, wherein the determining a first type of second sub-path in the high-precision map that matches the first type of first sub-path using bilateral constraint comprises:

Matching a precursor path corresponding to a first key feature point in the first sub-path of the first type with a precursor path of a second key feature point corresponding to the first key feature point;

matching a subsequent path corresponding to a second first key feature point in the first sub-path of the first type with a subsequent path of a second key feature point corresponding to the second first key feature point; the direction from the first key feature point to the second first key feature point is the direction of the first sub-path of the first type.

9. The method of claim 6, wherein the determining a second class second sub-path in the high-precision map that matches the second class first sub-path using a single-side constraint approach comprises:

responding to the second type first sub-path to comprise a starting point of a navigation path and a first key feature point connected with the starting point, and matching a precursor path corresponding to the first key feature point connected with the starting point with a precursor path of a second key feature point corresponding to the first key feature point connected with the starting point;

and responding to the second type first sub-path to comprise an end point of the navigation path and a first key feature point connected with the end point, and matching a subsequent path corresponding to the first key feature point connected with the end point with the subsequent path of a second key feature point corresponding to the first key feature point connected with the end point.

10. The method of claim 1, further comprising:

and sending the target path to an automatic driving vehicle so as to generate control information by the automatic driving vehicle in combination with the target path, wherein the control information is used for controlling the automatic driving vehicle to run.

11. The method of any of claims 1 to 10, wherein the navigation map and the high-precision map are of different sources.

12. A path determining apparatus comprising:

the third determining module is used for determining a second sub-path matched with the first sub-path included in the navigation path in the high-precision map by combining the attribute information corresponding to the second key feature point;

and a fourth determining module, configured to determine a target path corresponding to the navigation path in the high-precision map based on the second sub-path matched with the first sub-path in the high-precision map.

13. The apparatus of claim 12, wherein the second determination module comprises:

the acquisition sub-module is used for acquiring path data in a preset range of the navigation path in the high-precision map based on the attribute information corresponding to the first key feature points;

a first determining sub-module for determining candidate key feature points related to the navigation path based on the path data;

and the second determining submodule is used for determining a second key feature point corresponding to the first key feature point in the high-precision map based on the candidate key feature point related to the navigation path.

14. The apparatus of claim 13, wherein the second determination submodule is configured to:

15. The apparatus of claim 12, wherein the attribute information comprises at least one of:

16. The apparatus of claim 12, wherein the third determination module comprises:

and the third determining sub-module is used for determining a first type second sub-path matched with the first type first sub-path in the high-precision map by adopting a bilateral constraint mode aiming at the first type first sub-path matched with two second key feature points, wherein the first type first sub-path comprises two adjacent first key feature points.

17. The apparatus of claim 12, wherein the third determination module comprises:

and the fourth determining sub-module is used for determining a second class second sub-path matched with the second class first sub-path in the high-precision map by adopting a single-side constraint mode aiming at the second class first sub-path matched with a second key feature point, wherein the second class first sub-path comprises the first key feature point.

18. The apparatus of claim 12, wherein the third determination module comprises:

And a fifth determining sub-module, configured to determine, for a third type of first sub-path that does not have second key feature point matching, a third type of second sub-path that matches the third type of first sub-path in the high-precision map based on coordinate information and attribute information of the third type of first sub-path, where the third type of first sub-path includes at least one first key feature point.

19. The apparatus of claim 16, wherein the third determination submodule is configured to:

20. The apparatus of claim 17, wherein the fourth determination submodule is configured to:

21. The apparatus of claim 12, further comprising:

and the sending module is used for sending the target path to an automatic driving vehicle so as to generate control information by the automatic driving vehicle in combination with the target path, wherein the control information is used for controlling the automatic driving vehicle to run.

22. The apparatus of any of claims 12 to 21, wherein the navigation map and the high-precision map are of different sources.

23. An electronic device, comprising:

at least one processor; and

A memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-11.

24. A non-transitory computer readable storage medium storing computer instructions for causing a computer to perform the method according to any one of claims 1-11.

25. A computer program product comprising a computer program stored on a storage medium, which, when executed by a processor, implements the method according to any of claims 1-11.

26. An autonomous vehicle comprising the electronic device of claim 23.