CN115585816B

CN115585816B - Lane-level map matching method and device

Info

Publication number: CN115585816B
Application number: CN202211469907.3A
Authority: CN
Inventors: 杨宁; 张传明; 王亦乐
Original assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Current assignee: Beijing Baidu Netcom Science and Technology Co Ltd
Priority date: 2022-11-23
Filing date: 2022-11-23
Publication date: 2023-03-21
Anticipated expiration: 2042-11-23
Also published as: CN115585816A

Abstract

The disclosure provides a lane-level map matching method and device, relates to a computer technology and a data processing technology, in particular to a high-precision map technology, and can be applied to automatic driving, autonomous parking and intelligent transportation. The specific implementation scheme is as follows: the method comprises the steps of obtaining an observation sequence of a vehicle running in a preset high-precision road network, wherein the preset high-precision road network comprises a plurality of road sections, each road section comprises a lane group, each lane group comprises a plurality of lanes, obtaining switching types of two adjacent observation points in the observation sequence, wherein the switching types are used for representing whether the two adjacent observation points are the observation points of the same lane group, sampling a calculation method corresponding to the switching types, calculating to obtain the transition probability between the two adjacent observation points, determining a target lane-level path matched with the observation sequence according to the transition probability and the obtained emission probability, and realizing lane-level map matching due to different calculation methods corresponding to different switching types.

Description

Lane-level map matching method and device

Technical Field

The present disclosure relates to computer technologies and data processing technologies, and in particular, to a high-precision map technology, which can be applied to autonomous driving, autonomous parking, and intelligent transportation, and in particular, to a lane-level map matching method and apparatus.

Background

Hidden markov models are widely used in various fields, for example, hidden markov models can be used in the fields of speech recognition and optical character recognition, hidden markov models can be used in the field of machine translation, and hidden markov models can be used in the field of map matching.

When the hidden markov model is applied to the field of map matching, it can be based on the main flow of the hidden markov model: 1) Searching candidate route points; 2) Calculating the emission probability and the transition probability; 3) The calculation result is realized by map matching at a road level.

Disclosure of Invention

The disclosure provides a lane-level map matching method and a lane-level map matching device, which are used for realizing lane-level map matching.

According to a first aspect of the present disclosure, there is provided a lane-level map matching method, comprising:

acquiring an observation sequence of a vehicle running in a preset high-precision road network, wherein the preset high-precision road network comprises a plurality of road sections, each road section comprises a lane group, and each lane group comprises a plurality of lanes;

acquiring a switching type of two adjacent observation points in the observation sequence, wherein the switching type is used for representing whether the two adjacent observation points are observation points of the same lane group;

sampling a calculation method corresponding to the switching type, calculating to obtain a transition probability between the two adjacent observation points, and determining a target lane level path matched with the observation sequence according to the transition probability and the obtained emission probability, wherein the calculation methods corresponding to different switching types are different.

According to a second aspect of the present disclosure, there is provided a lane-level map matching apparatus including:

the vehicle monitoring system comprises a first acquisition unit, a second acquisition unit and a monitoring unit, wherein the first acquisition unit is used for acquiring an observation sequence of a vehicle running in a preset high-precision road network, the preset high-precision road network comprises a plurality of road sections, each road section comprises a lane group, and the lane group comprises a plurality of lanes;

the second acquisition unit is used for acquiring a switching type of two adjacent observation points in the observation sequence, wherein the switching type is used for representing whether the two adjacent observation points are observation points of the same lane group;

the calculation unit is used for sampling a calculation method corresponding to the switching type and calculating to obtain the transition probability between the two adjacent observation points;

and the first determining unit is used for determining a target lane-level path matched with the observation sequence according to the transition probability and the acquired emission probability, wherein the calculation methods corresponding to different switching types are different.

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

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein the content of the first and second substances,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the method of the first aspect.

According to a fourth aspect of the present disclosure, there is provided a non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method according to the first aspect.

According to a fifth aspect of the present disclosure, there is provided a computer program product comprising: a computer program, stored in a readable storage medium, from which at least one processor of an electronic device can read the computer program, execution of the computer program by the at least one processor causing the electronic device to perform the method of the first aspect.

The lane-level map matching method and device provided by the disclosure comprise the following steps: the method comprises the steps of obtaining an observation sequence of a vehicle running on a preset high-precision road network, wherein the preset high-precision road network comprises a plurality of road sections, each road section comprises a lane group, each lane group comprises a plurality of lanes, obtaining a switching type of two adjacent observation points in the observation sequence, wherein the switching type is used for representing whether the two adjacent observation points are observation points of the same lane group, sampling a calculation method corresponding to the switching type, calculating to obtain a transition probability between the two adjacent observation points, determining a target lane-level path matched with the observation sequence according to the transition probability and the obtained emission probability, wherein the calculation methods corresponding to different switching types are different, determining the switching type of the two adjacent observation points by combining the preset high-precision road network, calculating to obtain the transition probability between the two adjacent observation points by adopting the calculation method corresponding to the switching type, determining the technical characteristics of the target lane-level path by combining the transition probability and the obtained emission probability, and realizing lane-level map matching, thereby improving the accuracy and reliability of map matching.

It should be understood that the statements in this section do not necessarily identify key or critical features of the embodiments of the present disclosure, nor do they limit the scope of the present disclosure. Other features of the present disclosure will become apparent from the following description.

Drawings

The drawings are included to provide a better understanding of the present solution and are not to be construed as limiting the present disclosure. Wherein:

FIG. 1 is a schematic diagram according to a first embodiment of the present disclosure;

FIG. 2 is a schematic illustration of a pre-programmed road network and a pre-programmed road network according to the present disclosure;

FIG. 3 is a schematic diagram according to a second embodiment of the present disclosure;

FIG. 4 is a comparison schematic of an implementation of road level map matching, and an implementation of lane level map matching according to the present disclosure;

FIG. 5 is a schematic illustration of a third embodiment according to the present disclosure;

FIG. 6 is a schematic diagram according to a fourth embodiment of the present disclosure;

FIG. 7 is a schematic diagram according to a fifth embodiment of the present disclosure;

FIG. 8 is a schematic diagram of a lane-level map matching method according to an embodiment of the present disclosure;

FIG. 9 is a comparison of road-level matching results and target lane-level paths according to the present disclosure;

fig. 10 is a schematic diagram of a service system of a lane-level map matching method according to the present disclosure;

FIG. 11 is a schematic diagram according to a sixth embodiment of the present disclosure;

FIG. 12 is a schematic diagram according to a seventh embodiment of the present disclosure;

FIG. 13 is a schematic diagram according to an eighth embodiment of the present disclosure;

fig. 14 is a block diagram of an electronic device for implementing a lane-level map matching method of an embodiment of the present disclosure.

Detailed Description

Exemplary embodiments of the present disclosure are described below with reference to the accompanying drawings, in which various details of the embodiments of the disclosure are included to assist understanding, and which are to be considered as merely exemplary. Accordingly, those 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 and spirit of the present disclosure. Also, descriptions of well-known functions and constructions are omitted in the following description for clarity and conciseness.

To facilitate the reader's understanding of the disclosure, at least some of the technical terms of the disclosure are now set forth as follows:

the road network is a road system in which various roads are interconnected and meshed in a certain area.

Hidden Markov Models (HMMs) are widely used in various fields, for example, hidden Markov models can be used in the fields of speech recognition and optical character recognition, hidden Markov models can be used in the field of machine translation, and Hidden Markov models can be used in the field of map matching.

For example, when the hidden markov model is used in the field of map matching, the hidden markov model may be understood as a model for determining a matching path (which may also be referred to as an observation path) of an observation sequence, i.e., the hidden markov model has an input of the observation sequence and an output of the observation sequence and the output of the observation sequence is the matching path.

The observation sequence is a sequence of sampling points, for example, the observation sequence is a sequence of sampling points of a Global Positioning System (GPS) System for a vehicle to travel on a road, and may also be referred to as trajectory data.

For example, a global positioning system is deployed on a vehicle, and when the vehicle travels along a road, the global positioning system can sample the position of the vehicle to obtain a sampling point, and since the vehicle travels continuously, i.e., in a time sequence, a sequence of sampling points (i.e., an observation sequence) including a plurality of sampling points can be obtained through sampling.

The main process of the hidden Markov model can be divided into three steps, which are respectively as follows: 1) Searching candidate route points; 2) Calculating the emission probability and the transition probability; 3) And calculating a result.

The following is described with respect to finding candidate route sites:

and searching for a candidate route point means that for each observation point in the observation sequence, a road section within a preset distance range from the observation point is determined in the road network, the position closest to the observation point on the determined road section is calculated, and the point of the position on the determined road section is the candidate route point.

The number of the candidate mesh points is at least one, that is, the number of the candidate mesh points may be one or more. The preset distance range may be determined based on a demand, a history, a test, and the like, and the present embodiment is not limited thereto, for example, the preset distance range may be 50 meters.

For example, for a matching scenario with relatively high matching accuracy, the preset distance range may be set to a relatively small value; conversely, for a matching scene with relatively low matching accuracy, the preset distance range may be set to a relatively large value.

The calculation of the emission probability is described as follows:

the emission probability may also be referred to as a radial probability, and refers to the probability that the observation sequence corresponds to the road segment. The method for calculating the emission probability comprises the following steps: for each observation point, the distance (which may be referred to as a distance factor) and the angle (which may be referred to as an angle factor) between the observation point and the candidate road network point are calculated, and the emission probability (which may be understood as the probability that the observation point is a point of the road segment where the candidate road network point is located) is calculated according to the distance factor and the angle factor.

The calculation of transition probabilities is described as follows:

the transition probability refers to the probability that the vehicle travels from one road segment to another road segment, such as the probability that the vehicle travels from the road segment where one of two adjacent observation points in the observation sequence is located to the road segment where the other of the two adjacent observation points is located.

For example, two adjacent observation points are a first observation point and a second observation point, the road segment where the first observation point is located is a first road segment, the road segment where the second observation point is located is a second road segment, and the probability that the vehicle travels from the first road segment to the second road segment may be referred to as transition probability. The first road section and the second road section may be the same road section or different road sections.

The method for calculating the transition probability comprises the following steps: and calculating the transition probability among the candidate points according to the distance and the distance among the candidate points in each group of candidate points in the candidate point set of the last observation point and the candidate point set of the current observation point.

The calculation results are described below:

after the transmission probability and the transition probability are calculated, a Viterbi probability is calculated by combining the transmission probability and the transition probability, for example, an optimal road-level path corresponding to the observation sequence is obtained by a Viterbi (Viterbi) algorithm.

However, in conjunction with the above analysis, when the hidden markov model is applied to the map matching field, the path corresponding to the observation sequence calculated based on the hidden markov model is a road-level path. That is, in the above-described embodiment, only the road-level map matching can be achieved based on the hidden markov model.

However, with the development of the automatic driving technology and the continuous improvement of the requirements of the user on driving safety and driving precision, the road-level map matching cannot meet the map matching requirement, and how to realize the lane-level map matching is an urgent problem to be solved.

In order to realize lane-level map matching, the present disclosure provides a technical idea based on creative labor: the method comprises the steps that a high-precision road network is obtained through pre-compiling, the high-precision road network is a lane-level road network and comprises a plurality of road sections, each road section comprises a lane group (for example, one road section is one lane group), and one lane group comprises a plurality of lanes, and adjustment is carried out on the basis of a calculation mode of the high-precision road network on the transfer probability and/or the emission probability so as to realize lane-level map matching.

Based on the technical concept, the present disclosure provides a lane-level map matching method and device, which relates to a computer technology and a data processing technology, in particular to a high-precision map technology, and can be applied to automatic driving, autonomous parking and intelligent transportation to achieve lane-level map matching.

Fig. 1 is a schematic diagram according to a first embodiment of the present disclosure, and as shown in fig. 1, a lane-level map matching method of an embodiment of the present disclosure includes:

s101: and acquiring an observation sequence of the vehicle running in a preset high-precision road network.

The preset high-precision road network comprises a plurality of road sections, each road section comprises a lane group, and each lane group comprises a plurality of lanes.

For example, the executing subject of this embodiment may be a lane-level map matching device (hereinafter, referred to as a matching device), the matching device may be a server, a roadside device, a terminal device, a processor, a chip, and the like, and this embodiment is not limited thereto.

In a system architecture of intelligent transportation vehicle-road cooperation, the road side equipment comprises road side sensing equipment with a computing function and road side computing equipment connected with the road side sensing equipment, the road side sensing equipment (such as a road side camera) is connected to the road side computing equipment (such as a Road Side Computing Unit (RSCU)), the road side computing equipment is connected to a server, and the server can communicate with an automatic driving vehicle or an auxiliary driving vehicle in various modes; or the roadside sensing device comprises a calculation function, and the roadside sensing device is directly connected to the server. The above connections may be wired or wireless.

If the matching device is a server, the server may be a local server and a cloud server, the server may be a cloud control platform, a vehicle-road cooperative management platform, a central subsystem, an edge computing platform, a cloud computing platform, and the like, the server may be an independent server, or a server cluster, and the like, and these are not listed here one by one.

The preset high-precision road network may be a lane road network compiled in advance. For example, in combination with the above analysis, the road network is usually a road-level road network, and in this embodiment, in order to implement lane-level map matching, the road-level road network may be compiled in advance to obtain a lane-level road network (i.e., a preset high-precision road network).

Illustratively, as shown in fig. 2, the preset high-precision road network includes 4 road segments, each road segment includes a lane group, and accordingly, the lane groups corresponding to the 4 road segments are a, B, C, and D as shown in fig. 2. The lane group includes a plurality of lanes, and the lane group a as shown in fig. 2 includes A1, A2, A3, A4,4 lane groups; the lane group B comprises 4 lane groups B1, B2, B3 and B4; the lane group C comprises C1, C2, C3, C4 and 4 lane groups; the lane group D comprises D1, D2, D3, D4,4 lane groups.

It should be understood that fig. 2 is only used for exemplary illustration, and the preset high-precision road network may include the number of road segments, the number of lane groups, and the number of lanes in the lane groups, and is not to be construed as a limitation on the content of the preset high-precision road network.

Similarly, the number of observation points in the observation sequence is not limited in this embodiment. For example, 15 observation points, 1, 2, up to 15 as shown in fig. 2, may be included in the observation sequence.

S102: and acquiring the switching type of two adjacent observation points in the observation sequence. The switching type is used for representing whether two adjacent observation points are observation points of the same lane group.

Illustratively, the observation sequence includes a plurality of observation points, and two adjacent observation points may be referred to as two adjacent observation points.

As shown in fig. 2, observation point 1 and observation point 2 are two adjacent observation points, and observation point 1 and observation point 2 can be referred to as two adjacent observation points, and so on, and observation point 14 and observation point 15 can be referred to as two adjacent observation points.

Accordingly, the switching type may be understood that two adjacent observation points are two observation points of the same lane group, or that two adjacent observation points are not two observation points of the same lane group.

For example, observation points 2 and 3 are two adjacent observation points, observation point 2 is an observation point of lane A1, lane A1 is a lane in lane group a, observation point 3 is an observation point of lane A1, and lane A1 is a lane in lane group a, and therefore observation point 2 and observation point 3 are both observation points in lane group a, the switching type of observation point 2 and observation point 3 can be determined as: and the switching type representation observation point 2 and the observation point 3 are observation points of the same lane group.

For another example, observation point 5 and observation point 6 are two adjacent observation points, observation point 5 is an observation point of lane A1, lane A1 is a lane in lane group a, observation point 6 is an observation point of lane B2, and lane B2 is a lane in lane group B, and therefore observation point 5 and observation point 6 are observation points in different lane groups, the switching type between observation point 5 and observation point 6 can be determined as: the switching type representation observation point 5 and the switching type representation observation point 6 are not observation points of the same lane group, namely the switching type representation observation point 5 and the switching type representation observation point 6 are observation points of different lane groups.

The following example implementation can be employed with respect to acquiring observation sequences:

in one example, the matching device may be connected to the trajectory data acquisition device and receive the image transmitted by the trajectory data acquisition device. The track data acquisition device can be a vehicle and other devices which can be in communication connection with the vehicle.

In another example, the matching device may provide a tool to load trajectory data through which a user may transmit a sequence of observations to the matching device.

The tool for loading the track data may be an interface for connecting with the external device, such as an interface for connecting with other storage devices, and the track data transmitted by the external device is acquired through the interface; the tool for loading the trajectory data may also be a display device, for example, the matching device may input an interface with a function of loading the trajectory data on the display device, and a user may import the observation sequence into the matching device through the interface, and the matching device obtains the imported observation sequence.

S103: and calculating to obtain the transition probability between two adjacent observation points by a calculation method corresponding to the sampling and switching type, and determining a target lane-level path matched with the observation sequence according to the transition probability and the obtained emission probability. Wherein, the calculation methods corresponding to different switching types are different.

In combination with the above analysis, the switching type may represent that two adjacent observation points are observation points of the same lane group, and may also represent that two adjacent observation points are not observation points of the same lane group.

In the related art, map matching is road-level map matching, and the same calculation method is used when calculating the transition probability between two different adjacent observation points, but in the present embodiment, the switching type between two adjacent observation points is determined by combining a preset high-precision network, so that the transition probability is calculated by using different calculation methods for different switching types, thereby implementing lane-level map matching.

In this embodiment, the calculation methods corresponding to different switching types are not limited, for example, the calculation methods corresponding to different switching types may be determined based on features of the different switching types, or the calculation methods capable of implementing lane-level map matching corresponding to different switching types may be determined based on a test manner, and so on, which are not listed here one by one.

After the transition probability is obtained through calculation, the target lane-level path can be determined and obtained by combining the transition probability and the obtained emission probability.

The embodiment does not limit the way of determining the target lane-level path by specifically combining the transition probability and the acquired transmission probability. Illustratively, in conjunction with the above example, a viterbi probability may be calculated in conjunction with the transition probability and the acquired transmission probability to determine the target lane-level path based on the viterbi probability.

Similarly, the method for obtaining the emission probability is not limited in this embodiment, and the emission probability may be obtained by calculating according to the method described in the above embodiment.

Based on the above analysis, the present disclosure provides a lane-level map matching method, including: the method comprises the steps of obtaining an observation sequence of a vehicle running on a preset high-precision road network, wherein the preset high-precision road network comprises a plurality of road sections, each road section comprises a lane group, each lane group comprises a plurality of lanes, obtaining a switching type of two adjacent observation points in the observation sequence, wherein the switching type is used for representing whether the two adjacent observation points are observation points of the same lane group, sampling a calculation method corresponding to the switching type, calculating to obtain a transition probability between the two adjacent observation points, and determining a target lane-level path matched with the observation sequence according to the transition probability and the obtained emission probability, wherein the calculation methods corresponding to different switching types are different.

That is, by the lane-level map matching method provided by the embodiments of the present disclosure, lane-level map matching may be implemented to relatively accurately determine, based on the travel track of the vehicle, the road on which the vehicle travels and the lane of the road.

Correspondingly, in the application of vehicle navigation, the lane-level map matching method of the embodiment can be adopted to provide lane-level guidance, lane-level road conditions, lane-level route planning and lane-level accident mining capability for a user, and improve the navigation experience of the user.

To facilitate the reader's understanding more deeply, the principles of the present disclosure that improve transition probabilities, a lane-level map matching method of an embodiment of the present disclosure is now set forth in detail in conjunction with fig. 3.

Fig. 3 is a schematic diagram according to a second embodiment of the present disclosure, and as shown in fig. 3, the lane-level map matching method of the embodiment of the present disclosure includes:

s301: and acquiring an observation sequence of the vehicle running in a preset high-precision road network.

It should be understood that, in order to avoid tedious statements, the present embodiment will not be described again with respect to the same technical features as those in the above embodiments.

For example, regarding the implementation principle of S301, reference may be made to S101, and details are not described herein.

S302: and acquiring the switching type of two adjacent observation points in the observation sequence.

The switching type is used for representing whether two adjacent observation points are observation points of the same lane group.

For example, with respect to the implementation principle of S302, reference may be made to S102, which is not described herein again.

S303: and if the switching type represents that the two adjacent observation points are observation points of the same lane group, acquiring the distance between the two adjacent observation points, the distance between the two adjacent observation points and a penalty coefficient between the two adjacent observation points.

Wherein the penalty coefficient is determined based on the traffic relation between the lanes where the vehicles are located at two adjacent observation points.

As can be known from the foregoing embodiments, in the related art, the transition probability may be calculated based on the distance and the distance, but in the present embodiment, technical features of penalty coefficients are introduced to calculate the transition probability by combining the penalty coefficients.

The manner in which the distance and the distance are acquired in this embodiment is not limited in this embodiment. If the distance can be calculated based on the coordinates of two adjacent observation points, the distance can be determined based on the travel length of the vehicle from one observation point to the other observation point of the two adjacent observation points.

The traffic relation between lanes may be understood as a relation whether traffic between lanes is possible, such as whether a vehicle may travel from one lane to another.

Accordingly, the traffic relationship between the lanes may include two types, one type may be that traffic is allowed between the lanes, such as that vehicles may travel from one lane to another lane, and the other type may be that traffic is not allowed between the lanes, such as that vehicles may not travel from one lane to another lane.

In the present embodiment, the penalty coefficient is determined by combining the traffic relationship between the lanes where the vehicle is located at two adjacent observation points, so that the penalty coefficient is strongly associated with the traffic relationship between the lanes.

Illustratively, in combination with the above example, observation point 2 and observation point 3 are observation points of the same lane group, and the penalty coefficient between observation point 2 and observation point 3 may be determined based on the traffic relationship of the vehicle between the lanes in which observation point 2 and observation point 3 are located.

Accordingly, observation point 2 and observation point 3 are both observation points in lane A1, and therefore, a penalty coefficient between observation point 2 and observation point 3 can be determined based on the traffic relationship of the vehicle in lane A1.

As another example, observation point 5 and observation point 6 are not observation points of the same lane group, a penalty coefficient between observation point 5 and observation point 6 may be determined based on a traffic relationship of the vehicle between lanes in which observation point 5 and observation point 6 are located, respectively.

Accordingly, observation point 5 is an observation point of lane A1, and observation point 6 is an observation point of lane B2, so that the penalty coefficient between observation point 5 and observation point 6 can be determined from the traffic relationship between lane A1 and lane B2.

S304: and calculating to obtain the transition probability according to one or more of the distance, the distance and the penalty coefficient.

For example, in some embodiments, the transition probability may be calculated based on distance and distance. In other embodiments, the transition probability may also be calculated according to the distance, and the penalty coefficient.

In this embodiment, the penalty coefficient is determined by combining the traffic relationship between the lanes where the vehicles are located at two adjacent observation points, so that the penalty coefficient can be strongly associated with the traffic relationship between the lanes, the transition probability is strongly associated with the traffic relationship between the lanes, and the transition probability and whether the vehicles can pass between the lanes in a real scene are improved, so that the effectiveness and the reliability of the transition probability are improved.

In some embodiments, if the traffic relationship characterizes that the vehicle can transit traffic between lanes where two adjacent observation points are located, the penalty factor is 1.

If the passing relationship represents that the vehicle cannot transfer to pass between the lanes where the two adjacent observation points are located, the penalty coefficient is 0.

It should be understood that the penalty factor of 1 in this embodiment means that the penalty factor is 1 or close to 1 (i.e. there may be some reasonable error). A penalty factor of 0 means that the penalty factor is 0 or close to 0 (i.e. there may be some reasonable error).

For example, if the vehicle can travel from the lane where one of the two adjacent observation points is located to the lane where the other of the two adjacent observation points is located (i.e., if the traffic relationship is characterized, the vehicle can transit between the lanes where the two adjacent observation points are located), the penalty coefficient of the two adjacent observation points is close to 1, so that the transition probability calculated based on the penalty coefficient can be relatively determined based on the distance and the distance.

If the vehicle can not travel from the lane where one of the two adjacent observation points is located to the lane where the other of the two adjacent observation points is located (namely if the traffic relation is characterized, the vehicle can not transfer traffic between the lanes where the two adjacent observation points are located), the penalty coefficients of the two adjacent observation points are close to 0, so that the transition probability calculated based on the penalty coefficients is close to 0.

For example, as shown in fig. 2, observation point 4 is an observation point of lane A1, observation point 5 is an observation point of lane A2, and a lane line between observation point 4 and observation point 5 is a solid line, that is, in an actual road scene, a vehicle cannot travel from lane A1 to lane A2, and if a penalty coefficient is not introduced, a calculated transition probability between observation point 4 and observation point 5 is relatively large, a target lane-level path may be determined as a path including lane A2, but in an actual road scene, a travel path of a vehicle cannot travel from lane A1 to lane A2, and therefore, if a penalty coefficient is not introduced, the accuracy of the determined target lane-level path is relatively low.

In the embodiment, by introducing the penalty coefficient, because the penalty coefficient is determined based on the traffic relation, when the traffic relation indicates that the vehicle cannot drive from the lane A1 to the lane A2, the penalty coefficient is close to 0, the calculated transition probability is also close to 0, and the target lane path does not include the lane A2 at a high probability, so that the target lane path has high effectiveness and reliability.

Therefore, if the vehicle cannot drive to the lane where the other observation point of the two adjacent observation points is located from the lane where the one observation point of the two adjacent observation points is located, the transition probability close to 0 is obtained through calculation, the target lane-level path where the vehicle cannot normally pass can be prevented from being determined based on the transition probability, and therefore the effectiveness and the reliability of lane-level map matching are improved.

For example, traffic relationships (which may also be referred to as road network constraints) may be determined during the compilation of the network. For example, when compiling a road network, lane lines in the road network may be obtained, and the lane lines may be represented by a dotted line and a solid line, and in general, the dotted line is used to represent that two lanes can pass through each other, and the solid line is used to represent that two lanes cannot pass through each other.

Therefore, the traffic relationship between the two lanes corresponding to the lane line indicated by the broken line can be determined as: the vehicle cannot transfer between the two lanes, i.e. the vehicle cannot travel from one of the two lanes to the other of the two lanes; determining a traffic relationship between two lanes corresponding to a lane line indicated by a solid line as: the vehicle may make a transition between the two lanes, i.e. the vehicle may travel from one of the two lanes to the other of the two lanes.

In other embodiments, the transition probability may also be calculated in conjunction with the switching frequency.

Illustratively, S303 may be replaced by: and if the switching type represents that the two adjacent observation points are observation points of the same lane group, acquiring the distance between the two adjacent observation points, the penalty coefficient between the two adjacent observation points and the switching frequency of the vehicle in the same lane group. Wherein the penalty coefficient is determined based on the traffic relation of the vehicle between the lanes where the two adjacent observation points are located.

Accordingly, S304 may be replaced with: and calculating to obtain the transition probability according to one or more of the distance, the distance and the penalty coefficient and the switching frequency.

For example, in some embodiments, the transition probability may be calculated based on distance, range, and switching frequency. In other embodiments, the transition probability may also be calculated according to the distance, the penalty coefficient, and the switching frequency.

The switching frequency may be understood as a frequency of switching of the vehicle within the same lane group, such as a frequency of switching of the vehicle between different lanes within the same lane group.

In the embodiment, the transition probability is calculated by combining the switching frequency, which is equivalent to calculating the transition probability by considering characteristics such as stability and safety of vehicle running, so that the effectiveness and reliability of the transition probability are further improved.

In some embodiments, calculating the transition probability according to the switching frequency, the distance, and the penalty coefficient may include the following steps:

the first step is as follows: and adjusting the penalty coefficient according to the switching frequency to obtain the adjusted penalty coefficient. Wherein, the larger the switching frequency is, the closer the penalty factor is to 0.

In contrast, in consideration of the safety and stability of vehicle driving, the vehicle is not switched in the same lane group too frequently, and therefore, in the embodiment, the penalty factor is adjusted by combining the switching frequency, so that when the switching frequency is higher, the penalty factor is adjusted to be a value closer to 0, so that the penalty factor is associated with the safety and stability of the vehicle, and the accuracy and reliability of the adjusted penalty factor are improved.

The second step is as follows: and calculating to obtain the transition probability according to the distance, the distance and the adjusted penalty coefficient.

In this embodiment, since the adjusted penalty factor is associated with the safety and stability of the vehicle, and has higher accuracy and reliability, the transition probability calculated by combining the adjusted penalty factor has higher accuracy and reliability.

S305: and if the switching type represents that the two adjacent observation points are not observation points of the same lane group, acquiring the distance between the two adjacent observation points and the distance between the two adjacent observation points.

Similarly, regarding the implementation principle of S305, reference may be made to the partial implementation principle in S303, for example, the implementation principle of obtaining the distance and the route in S303, which is not described herein again.

S306: and calculating the transition probability according to the distance and the distance.

For the implementation principle of S306, refer to the implementation principle of 3) the calculation result in the main flow of the hidden markov model in the above embodiment, which is not described herein again.

In conjunction with S302-306, the principle of calculating the transition probability in this embodiment can be understood as follows:

and judging whether the two adjacent observation points are observation points of the same lane group, if so, acquiring the distance between the two adjacent observation points, the distance between the two adjacent observation points and a penalty coefficient between the two adjacent observation points, and calculating the transition probability between the two adjacent observation points based on the distance, the distance and the penalty coefficient.

Otherwise, if not (namely, the two adjacent observation points are not the observation points of the same lane group), the distance between the two adjacent observation points and the distance between the two adjacent observation points are obtained, and the transition probability between the two adjacent observation points is calculated based on the distance and the distance.

As can be seen by combining the explanation of the main flow of the hidden markov model in the above embodiment, in the related art, the transition probability is directly obtained by distance and distance calculation, and in the present embodiment, the penalty coefficient is combined, and even on the basis of the penalty coefficient, the transition probability can be calculated by combining the switching frequency, and the penalty coefficient is strongly related to the traffic relation, and the switching frequency is strongly related to the characteristics of the driving safety, reliability, and the like of the vehicle.

S307: and determining a target lane level path matched with the observation sequence according to the transition probability and the acquired emission probability.

The above-described embodiment explains the lane-level map matching method of the present embodiment from the principle of improving the transition probability. As shown in fig. 4, in contrast to the principle that the transition probability is calculated based on the distance and the distance in the road-level map matching, in the lane-level map matching of the present embodiment, the calculation of the transition probability may be implemented in various ways, such as:

one way is to calculate the transition probability based on distance and distance; the other mode is that the transition probability is calculated based on the distance, the distance and the penalty coefficient; yet another way is to calculate the transition probability based on the switching frequency, distance, and penalty factor.

In other embodiments, the present disclosure also provides a lane-level map matching method implemented based on the principle of improving the transmission probability. For the reader's understanding, the lane-level map matching method of the present disclosure is now set forth below in connection with FIG. 5 from the principle of improving the probability of transmission.

Wherein, fig. 5 is a schematic diagram according to a third embodiment of the present disclosure, and as shown in fig. 5, the lane-level map matching method of the embodiment of the present disclosure includes:

s501: and acquiring an observation sequence of the vehicle running in a preset high-precision road network.

Similarly, in order to avoid tedious statements, the technical features of the present embodiment that are the same as those of the above embodiments are not repeated.

For example, with respect to the implementation principle of S501, reference may be made to S101, which is not described herein again.

S502: and acquiring a distance factor and an angle factor between the observation sequence and each lane, and acquiring a preset attribute factor of each lane.

Wherein the attribute factor of each lane is used for representing the appropriateness of the vehicle to run on the lane.

For example, the suitability of the vehicle to travel on different lanes may be different, such as for an emergency lane, where the vehicle may not travel on the emergency lane easily, and it may be determined that the suitability of the vehicle to travel on the emergency lane is low.

Similarly, the manner of obtaining the distance factor and the angle factor in the present embodiment is not limited, and the distance factor and the angle factor may be obtained by the manner in the related art.

S503: and determining the emission probability between the observation sequence and each lane according to the acquired distance factor, angle factor and attribute factor.

In combination with the above analysis, it can be known that each lane has an attribute factor, and the attribute factor of each lane is used for representing the appropriate degree of the vehicle running on the lane, that is, the attribute factor is determined based on the running characteristics of the vehicle in the actual scene of the vehicle.

S504: and determining a target lane level path matched with the observation sequence according to the emission probabilities and the acquired transition probabilities.

For example, in some embodiments, the transition probability may be obtained by sampling a manner of calculating the transition probability in the related art, and in other embodiments, the transition probability may also be obtained by using the manner described in the above first embodiment or second embodiment, which is not limited in this embodiment.

Accordingly, regarding the manner of determining the target lane-level path matching the observation sequence by combining the emission probability and the transition probability, this embodiment is not limited, and the calculation result (e.g. 3 in the main flow of the hidden markov model) may also be implemented by using the manner in the related art.

The above-described embodiment explains the lane-level map matching method of the present embodiment from the principle of improving the emission probability. As shown in fig. 4, in contrast to the principle that the emission probability is calculated based on the distance factor and the angle factor in the road-level map matching, in the lane-level map matching of the present embodiment, the emission probability may be calculated by combining the distance factor, the angle factor, and the attribute factor.

It should be noted that, in some embodiments, the first embodiment, the second embodiment, and the third embodiment may be independent embodiments as described above, in other embodiments, some of the embodiments may also be combined to obtain a new embodiment, some of the technical features in the three embodiments may also be combined to obtain a new embodiment, some of the technical features may also be added or deleted on the basis of any embodiment to obtain a new embodiment, and the like, and the present embodiment is not limited.

For example, a new embodiment may be obtained by combining the first embodiment and the third embodiment, or a new embodiment may be obtained by combining the second embodiment and the third embodiment.

In conjunction with the above analysis, the first and second embodiments are lane-level map matching implemented for the improvement of the transition probability, and the third embodiment is a lane-level map matching method implemented for the improvement of the transmission probability.

Therefore, if the first embodiment and the third embodiment are combined, or the second embodiment and the third embodiment are combined, the lane-level matching method can be understood as a lane-level matching method in which both the emission probability and the transition probability are improved.

In combination with the above analysis, it can be seen that, in 3) the calculation result of the main flow of the hidden markov model can be implemented by calculating the viterbi probability, and accordingly, in some embodiments, the present disclosure further provides a lane-level map matching method implemented based on the principle of improving the viterbi probability.

For the reader's understanding, the lane-level map matching method of the present disclosure is now set forth below in connection with FIG. 6 from the principles of improving Viterbi probability. Fig. 6 is a schematic diagram according to a fourth embodiment of the present disclosure, and as shown in fig. 6, the lane-level map matching method according to the embodiment of the present disclosure includes:

s601: and acquiring an observation sequence of the vehicle running in a preset high-precision road network.

For example, with respect to the implementation principle of S601, reference may be made to S101, which is not described herein again.

S602: and calculating to obtain the emission probability and the transition probability corresponding to the observation sequence.

The transmission probability is a lane-level transmission probability, and the transition probability is a lane-level transition probability.

By way of example, a lane-level emission probability may be understood as the probability that an observation sequence corresponds to a nearby lane. The lane-level transition probability is understood to be the probability that a vehicle travels from the lane where one of the two adjacent observation points is located to the lane where the other of the two adjacent observation points is located.

The principle of calculating the transmission probability and the transition probability is not limited in this embodiment. For example, when the emission probability is calculated, the emission probability may be calculated by combining the distance factor and the angle factor, or may be calculated by combining the distance factor, the angle factor, and the attribute factor. For a specific implementation principle, reference may be made to the above embodiments, which are not described herein again.

For example, when the transition probability is calculated, the transition probability can be calculated by combining the distance and the distance, and can also be calculated by combining the distance, the distance and the penalty coefficient. For a specific implementation principle, reference may be made to the above embodiments, which are not described herein again.

S603: and calculating to obtain the Viterbi probability corresponding to each candidate lane level path matched with the obtained observation sequence according to the transition probability and the emission probability.

The present embodiment does not limit the manner of obtaining each candidate lane-level path, and does not limit the manner of calculating the viterbi probability corresponding to each candidate lane-level path.

For example, 1) of the main flow such as the hidden markov model may be adopted to find candidate route points, obtain the candidate route points, and determine each candidate lane-level path based on the candidate route points.

Correspondingly, the viterbi probability corresponding to each candidate lane level path can be calculated by adopting the calculation result of 3) in the main flow of the hidden markov model.

S605: and sequentially rejecting the Viterbi probabilities meeting preset conditions from the Viterbi probabilities.

Wherein the preset conditions include: the current Viterbi probability is the maximum Viterbi probability, the transition probability of the current Viterbi probability is calculated to be larger than a preset first threshold, and the emission probability of the current Viterbi probability is calculated to be smaller than a preset second threshold.

Similarly, regarding the preset first threshold and the preset second threshold, this embodiment is not limited, and may be determined based on the needs, the history, and the tests.

This step can be understood as rejecting the viterbi probability if the transition probability calculated to obtain the maximum viterbi probability is greater but the transmission probability calculated to obtain the viterbi probability is less.

For example, for a maximum viterbi probability in the viterbi probabilities, it is determined whether a transition probability used for calculating the maximum viterbi probability is greater than a preset first threshold and whether a transmission probability used for calculating the maximum viterbi probability is smaller than a preset second threshold, if the transition probability used for calculating the maximum viterbi probability is greater than the preset first threshold and the transmission probability used for calculating the maximum viterbi probability is smaller than the preset second threshold, the maximum viterbi probability is rejected from the viterbi probabilities, and for each rejected viterbi probability, it is determined whether the rejection processing needs to be performed on the maximum viterbi probability in the rejected viterbi probabilities, so that the rejection is not repeated one by one.

It should be understood that the present embodiment is applicable to, among preset conditions: the judgment sequence is not limited, if the transition probability of the calculated current Viterbi probability is greater than the preset first threshold, and the emission probability of the calculated current Viterbi probability is less than the preset second threshold, after the current maximum Viterbi probability is determined, whether the transition probability used for calculating the calculated maximum Viterbi probability is greater than the preset first threshold is judged, if yes, whether the emission probability used for calculating the maximum Viterbi probability is less than the preset second threshold is continuously judged, and if yes, the maximum Viterbi probability is rejected.

For another example, after determining the current maximum viterbi probability, it may be determined whether the transmission probability used for calculating the maximum viterbi probability is smaller than a preset second threshold, if so, it is continuously determined whether the transition probability used for calculating the maximum viterbi probability is larger than a preset first threshold, and if so, the maximum viterbi probability is rejected.

For another example, after the current maximum viterbi probability is determined, whether the transition probability used for calculating the maximum viterbi probability is greater than a preset first threshold and whether the transmission probability used for calculating the maximum viterbi probability is less than a preset second threshold may be respectively determined, and if both the determination results are yes, the maximum viterbi probability is rejected.

In combination with the above analysis, the emission probability can be understood as the probability that the observation sequence corresponds to the nearby lane, and the transition probability can be understood as the probability that the vehicle travels from the lane where one of the two adjacent observation points is located to the lane where the other of the two adjacent observation points is located. Therefore, if the transition probability at which the maximum viterbi probability is calculated is large but the transmission probability is small, it is highly likely that the maximum viterbi probability is calculated due to an error or the like.

Therefore, in this embodiment, the maximum viterbi probabilities calculated based on the larger transition probabilities and the smaller transmission probabilities are sequentially removed from the viterbi probabilities, so that the maximum viterbi probabilities with higher error probability can be removed, and the accuracy and reliability of each viterbi probability in the viterbi probabilities satisfying the preset condition are removed.

S605: and selecting the maximum Viterbi probability from the Viterbi probabilities which meet the preset conditions, and determining the candidate lane level path corresponding to the selected maximum Viterbi probability as a target lane level path.

Because each Viterbi in the Viterbi probability which meets the preset condition is removed has higher accuracy and reliability, when the maximum Viterbi probability is selected from the Viterbi probabilities which meet the preset condition and the target lane level path is determined based on the maximum Viterbi probability, the accuracy and the reliability of the target lane level path can be improved, namely the accuracy and the reliability of the lane level map matching can be improved.

The above embodiment explains the lane-level map matching method of the present embodiment from the principle of improving the viterbi probability. As shown in fig. 4, with respect to the principle of determining a road-level path based on the maximum viterbi probability in the road-level map matching, in the lane-level map matching according to this embodiment, the viterbi probabilities satisfying the preset condition may be removed to obtain the removed viterbi probabilities, and a target lane-level path is determined based on the maximum viterbi probability in the removed viterbi probabilities.

Similarly, the fourth embodiment may be an independent embodiment, or may be combined with at least one of the first, second, and third embodiments described above to obtain a new embodiment.

For example, the fourth embodiment can be combined with the first embodiment to obtain a new embodiment, and this embodiment is an embodiment combining the principle of improving the transition probability and the principle of improving the viterbi probability.

As another example, the fourth embodiment can be combined with the third embodiment to obtain a new embodiment, and this embodiment is an embodiment combining the principle of improving the transmission probability and the principle of improving the viterbi probability.

As another example, the fourth embodiment can be combined with the first and third embodiments to obtain a new embodiment, and this embodiment is an embodiment combining the principle of improving the transition probability, the principle of improving the transmission probability, and the principle of improving the viterbi probability.

It should be understood that the above examples are only for illustrative purposes, and possible combinations between the fourth embodiment and the first to third embodiments are not to be construed as limitations on the combinations.

As can be seen from the above analysis, the lane-level map matching method implemented in combination with the preset high-precision map can be understood as the four embodiments of the first to fourth embodiments. In other embodiments, a lane-level map matching method may be implemented by combining a preset high-precision road network and a preset general road network.

For the convenience of the reader to understand, the lane-level map matching method implemented by combining the preset high-precision road network and the preset general road network is now explained with reference to fig. 7 as follows. Fig. 7 is a schematic diagram according to a fifth embodiment of the present disclosure, and as shown in fig. 7, the lane-level map matching method according to the embodiment of the present disclosure includes:

s701: and acquiring an observation sequence of the vehicle running in a preset high-precision road network.

Similarly, in order to avoid the tedious statements, the technical features of the present embodiment that are the same as those of the above embodiments are not repeated herein.

For example, with respect to the implementation principle of S701, reference may be made to S101, which is not described herein again.

In some embodiments, the observation sequence is high-precision trajectory data, and in other embodiments, the observation sequence may also be non-high-precision trajectory data (which may also be referred to as "fine trajectory data").

Illustratively, the coverage rate of the lane-level map matching is solved by mining high-quality non-high-precision track data to solve the problem that the coverage rate of the lane-level map matching is relatively small.

The high quality can be determined based on the driving behavior of the user, and if the non-high-precision track data of the user with relatively stable driving behavior and large safety factor can be selected as the observation sequence.

S702: and acquiring the switching type of two adjacent observation points in the observation sequence.

For example, with respect to the implementation principle of S702, reference may be made to S101, which is not described herein again.

S703: and calculating to obtain the transition probability between two adjacent observation points by a calculation method corresponding to the sampling and switching type, and determining a target lane-level path matched with the observation sequence according to the transition probability and the obtained emission probability.

Wherein, the calculation methods corresponding to different switching types are different.

For example, regarding the implementation principle of S703, reference may be made to S103, which is not described herein again.

S704: and matching according to a preset common road network corresponding to the preset high-precision road network to obtain a road-level matching result corresponding to the observation sequence.

The preset general road network comprises a plurality of road sections.

For example, for convenience of distinction and description, we may refer to a precompiled road-level road network as a pre-planned road network and a pre-compiled roadway road network as a pre-planned road network. The preset high-precision road network and the preset general road network have a corresponding relationship, and the corresponding relationship can be understood as that the preset high-precision road network and the preset general road network are different expression modes for the same road network, the preset high-precision road network is the lane-level expression for the same road network, and the preset general road network is the road-level expression for the same road network.

As shown in fig. 2, the predetermined general road network includes a plurality of road segments, which are X, Y, and Z shown in fig. 2. The road in the real scene represented by the road segment X, the road in the real scene represented by the lane group a, and a part of the roads in the real scene represented by the lane group B are the same roads in the real scene, and so on, which are not listed one by one here.

In some embodiments, the correspondence between the road segments of the preset general road network and the lane groups of the preset high-precision road network may be determined in the road network compiling process.

In this embodiment, for the road-level matching result corresponding to the observation sequence obtained by matching, reference may be made to the principle of applying the hidden markov model to the map matching field.

S705: and determining a target matching result matched with the observation sequence according to the road level matching result and the target lane level path.

In this embodiment, a final matching result with the observation sequence is determined by combining matching results of two dimensions, one dimension is a preset high-precision road network dimension, the matching result is a target lane-level path, the other dimension is a preset general road network dimension, the matching result is a road-level matching result, and the final matching result is a target matching result.

It should be noted that, although the lane-level matching result has higher accuracy and can determine the lane-level path corresponding to the observation sequence, the technical solution of determining the final matching result with the observation sequence by combining the two-dimensional matching result of this embodiment can avoid the disadvantages of lane-level matching error, drift of the observation point in the observation sequence, road missing in the road network, and the like as much as possible, and further improve the validity and reliability of map matching.

In some embodiments, S705 may include the steps of:

the first step is as follows: compatible attributes are determined for each observation point in the observation sequence.

The second step is as follows: and if the compatible attribute of each observation point represents the lane of the observation point on the target lane-level path and is positioned on the road section of the observation point on the road-level matching result, determining the target lane-level path as the target matching result.

The third step: and if the compatible attribute of any observation point represents the lane of the observation point on the target lane level path and no road section of the observation point on the road level matching result is located, replacing part of the path corresponding to any observation point in the target lane level path with the road level matching result of any observation point to obtain the target matching result.

It should be understood that there is no necessarily sequential logic between the second step and the third step, and the processing manner of the compatible attributes of different contents is divided into two steps for convenience of description and understanding.

For example, as can be seen from fig. 8 (fig. 8 is a schematic diagram of a lane-level map matching method according to an embodiment of the present disclosure), the implementation principle of the present embodiment can be understood as follows:

and performing road-level matching based on a preset general road network to obtain a road-level matching result of the observation sequence.

And performing lane-level matching based on a preset high-precision road network to obtain a target lane-level path of the observation sequence.

And judging whether the road is compatible with the target lane, if so, adopting a target lane level path, and if not, adopting a road level matching result.

For example, for each observation point, whether the lane of the observation point on the target lane-level path is located on the road segment of the observation point on the road-level matching result is judged, if yes, the target lane-level path is not deviated from the road segment, and the target lane-level path is determined to be the target matching result.

Otherwise, if the target matching result is not the target matching result (i.e. the observation point is in the lane of the target lane-level path, and there is no road segment located at the observation point in the road-level matching result), it is indicated that a part of the lanes corresponding to the observation point in the target lane-level path deviate from the road segment, and the lane corresponding to the observation point in the target lane-level path is replaced by the road segment corresponding to the observation point in the road-level matching result, so as to obtain the target matching result.

Illustratively, as shown in fig. 9, the observation sequence includes 1 to 15 observation points.

The road-level matching result determined based on the preset general road network represents the road sections corresponding to the observation points in the observation sequence, as shown in fig. 9: the section corresponding to observation points 1 to 6 is section X, the section corresponding to observation points 7 to 9 is section Y, and the section corresponding to observation points 10 to 15 is section Z.

A target lane-level path determined based on a preset high-precision road network represents lanes corresponding to observation points in an observation sequence, as shown in fig. 9: the lane from observation point 2 to observation point 5 is A1, the lane from observation point 6 to observation point 8 is B2, the lane from observation point 9 to observation point 11 is lane C3, and the lane from observation point 12 to observation point 14 is D3.

Accordingly, in combination with the above compatibility attributes, the lane of each observation point on the target lane-level route shown in fig. 9 is located on the road segment of the road-level matching result of the observation point, and therefore, the target lane-level route can be determined as the target matching result.

In the embodiment, the lane-level matching and the road-level matching are combined to determine the target matching result of the observation sequence, so that the diversity and richness of the matching can be improved, and the accuracy, reliability and effectiveness of the target matching result are improved.

Similarly, the present embodiment may be combined with at least one of the other embodiments to obtain a new embodiment, and the present embodiment is not limited.

The lane-level map matching method described in any of the above embodiments may be applied to a service system as shown in fig. 10. Wherein, fig. 10 is a schematic view of a service system of the lane-level map matching method according to the present disclosure, as shown in fig. 10, the service system includes:

and the trace receiving proxy server (nginx-its-ugc) is used for receiving the trace data.

And generating a service-high-precision positioning service (hps) layer by the user for accessing the high-precision track data.

And the user generates a content-service (ugc-server) layer, which is used for accessing the track data, unpacking and analyzing the accessed track data according to a transmission protocol, and differentially acquiring the coordinates, the time stamps, the speed and the like of track points in the track data.

The user generated content service layer may include three modules, such as a promptuous module for accessing promptuous track data, a high-precision module for accessing high-precision track data, and a text module for accessing text track data.

Illustratively, in conjunction with the above analysis, in some embodiments, the user-generated content service layer may also access generic trace data, such as high-quality non-high-precision trace data.

And the track data distribution layer (Gps-transfer) is used for acquiring the high-precision track data and the general-precision track data and sending the high-precision track data and the general-precision track data to the servers of the corresponding areas.

And the high-precision map matching server (maps-match-server (mms) -hps) is used for matching the high-precision track data transmitted by the track data distribution layer with a pre-compiled road network and outputting a possible lane-level sequence of the high-precision track data.

Illustratively, the input of the high-precision map matching service is the observation sequence in the above embodiment, and the high-precision map matching service may execute the lane-level map matching method in any of the above embodiments to obtain a lane-level sequence, where the lane-level sequence is a target lane-level path.

And the high-precision map matching database (map-match-data-base (mmdb) -hps) is used for post-processing the lane-level sequence transmitted by the high-precision map matching server to obtain a post-processed road sequence. Post-processing may include merging, pre-inspection repair replacement, etc., among others.

A producer module for writing the post-processed road sequence transmitted by the high-precision map-matching database to a distributed publish-subscribe message system (kafka), such as a producer-consumer model (producer-consumer) in the publish-subscribe message system.

And the remote procedure call (rpc) module is used for transmitting the post-processed road sequence transmitted by the high-precision map matching database to other applications.

And the map matching service (map-match-server) is used for matching the high-precision track data transmitted by the track data distribution layer to obtain a road-level sequence.

Illustratively, the input of the map matching service is the observation sequence in the above embodiment, and the map matching service performs road-level matching on the observation sequence to obtain a road-level matching result (e.g., a road-level sequence).

A map matching database (mmdb) for post-processing the road-level sequence. Similarly, post-processing may include merging, pre-inspection revision replacement, and the like.

It should be understood that fig. 10 is for exemplary purposes only, and is not to be construed as limiting the contents of a service system that may be used to implement the lane-level map matching service system of the present disclosure. For example, in other embodiments, the components in FIG. 10 may be reduced or new components may be added or subtracted from FIG. 10.

For example, in combination with the above analysis, if the target matching result corresponding to the observation sequence is determined by combining the road-level matching and the lane-level matching, a merging module may be added to the service system shown in fig. 10 to implement the principle of determining the target matching result matching the observation sequence according to the road-level matching result and the target lane-level path as described in the above embodiment.

Fig. 11 is a schematic diagram according to a sixth embodiment of the present disclosure, and as shown in fig. 11, a lane-level map matching apparatus 1100 of the embodiment of the present disclosure includes:

the first obtaining unit 1101 is configured to obtain an observation sequence of a vehicle traveling in a preset high-precision road network, where the preset high-precision road network includes a plurality of road segments, each road segment includes a lane group, and the lane group includes a plurality of lanes.

A second obtaining unit 1102, configured to obtain a switching type of two adjacent observation points in the observation sequence, where the switching type is used to represent whether the two adjacent observation points are observation points of the same lane group.

A calculating unit 1103, configured to sample a calculating method corresponding to the switching type, and calculate a transition probability between the two adjacent observation points.

A first determining unit 1104, configured to determine, according to the transition probability and the obtained emission probability, a target lane-level path matching the observation sequence, where the calculation methods corresponding to different switching types are different.

Fig. 12 is a schematic diagram according to a seventh embodiment of the present disclosure, and as shown in fig. 12, a lane-level map matching apparatus 1200 of the embodiment of the present disclosure includes:

the first obtaining unit 1201 is configured to obtain an observation sequence of a vehicle traveling in a preset high-precision road network, where the preset high-precision road network includes a plurality of road segments, each road segment includes a lane group, and the lane group includes a plurality of lanes.

A second obtaining unit 1202, configured to obtain a switching type of two adjacent observation points in the observation sequence, where the switching type is used to represent whether the two adjacent observation points are observation points of a same lane group.

A calculating unit 1203, configured to sample a calculating method corresponding to the switching type, and calculate a transition probability between the two adjacent observation points.

As can be seen in fig. 12, in some embodiments, the calculating unit 1203 includes:

a first obtaining subunit 12031, configured to, if the switching type indicates that the two adjacent observation points are observation points of the same lane group, obtain a distance between the two adjacent observation points, and a penalty coefficient between the two adjacent observation points, where the penalty coefficient is determined based on a traffic relationship between lanes where the two adjacent observation points are located by the vehicle.

A first calculating subunit 12032, configured to calculate the transition probability according to one or more of the distance, and the penalty coefficient.

In some embodiments, if the traffic relationship indicates that the vehicle can transit traffic between lanes where the two adjacent observation points are located, the penalty factor is 1.

And if the passing relationship represents that the vehicle cannot transfer the passing between the lanes where the two adjacent observation points are located, the penalty coefficient is 0.

As can be seen from fig. 12, in other embodiments, if the switching type represents that the two adjacent observation points are observation points of the same lane group, the calculating unit 1203 further includes:

a second acquiring subunit 12033, configured to acquire a switching frequency of the vehicle in the same lane group.

And the first calculating subunit 12032 is configured to calculate the transition probability according to the switching frequency and one or more of the distance, and the penalty coefficient.

In some embodiments, the first computing subunit 12032, includes:

and the adjusting module is used for adjusting the penalty coefficient according to the switching frequency to obtain the adjusted penalty coefficient, wherein the larger the switching frequency is, the larger the penalty coefficient is, the more 0 the penalty coefficient is.

And the calculation module is used for calculating and obtaining the transition probability according to the distance, the distance and the adjusted penalty coefficient.

As can be seen in fig. 12, in other embodiments, the calculating unit 1203 further includes:

a third obtaining subunit 12034, configured to obtain a distance between the two adjacent observation points and a distance between the two adjacent observation points if the switching type indicates that the two adjacent observation points are not observation points of the same lane group.

And the second calculating subunit 12035 is configured to calculate the transition probability according to the distance and the distance.

A first determining unit 1204, configured to determine, according to the transition probability and the obtained emission probability, a target lane-level path matching the observation sequence, where the calculation methods corresponding to different handover types are different.

In some embodiments, as can be seen in fig. 12, the first determining unit 1204 includes:

and the third calculating subunit 12041 is configured to calculate, according to the transition probability and the obtained emission probability, a viterbi probability corresponding to each candidate lane-level path matched with the obtained observation sequence.

A rejecting subunit 12042, configured to sequentially reject viterbi probabilities that meet preset conditions from the viterbi probabilities, where the preset conditions include: the current Viterbi probability is the maximum Viterbi probability, the transition probability of the current Viterbi probability is calculated to be larger than a preset first threshold, and the emission probability of the current Viterbi probability is calculated to be smaller than a preset second threshold.

A selecting subunit 12043, configured to select a maximum viterbi probability from the viterbi probabilities that satisfy the preset condition are rejected.

A first determining subunit 12044, configured to determine, as the target lane-level path, the candidate lane-level path corresponding to the selected maximum viterbi probability.

In some embodiments, as can be seen in fig. 12, the lane-level map matching apparatus 1200 further includes:

a third obtaining unit 1205, configured to obtain a distance factor and an angle factor between the observation sequence and each lane, and obtain a preset attribute factor of each lane, where the attribute factor of each lane is used to represent a suitable degree of the vehicle traveling on the lane.

A second determining unit 1206, configured to determine, according to the obtained distance factor, angle factor, and attribute factor, a transmission probability between the observation sequence and each lane, where the obtained transmission probability includes the transmission probability between the observation sequence and each lane.

a matching unit 1207, configured to obtain a road-level matching result corresponding to the observation sequence by matching according to a preset general road network corresponding to the preset high-precision road network, where the preset general road network includes a plurality of road segments.

A third determining unit 1208, configured to determine a target matching result matching the observation sequence according to the road-level matching result and the target lane-level path.

In some embodiments, as can be seen in fig. 12, the third determining unit 1208 includes:

a second determining subunit 12081, configured to determine a compatibility attribute for each observation point in the observation sequence.

A third determining subunit 12082, configured to determine, if the compatible attribute of each observation point represents that the observation point is on the lane of the target lane-level route and is located on the road segment of the road-level matching result of the observation point, the target lane-level route as the target matching result.

A replacing subunit 12083, configured to replace, if the compatible attribute of the arbitrary observation point represents the lane of the observation point on the target lane-level path, and there is no road segment of the observation point on the road-level matching result, a part of the path corresponding to the arbitrary observation point in the target lane-level path with the road-level matching result of the arbitrary observation point, so as to obtain the target matching result.

Fig. 13 is a schematic diagram according to an eighth embodiment of the present disclosure, and as shown in fig. 13, an electronic device 1300 in the present disclosure may include: a processor 1301 and a memory 1302.

A memory 1302 for storing programs; the Memory 1302 may include a volatile Memory (RAM), such as a Static Random Access Memory (SRAM), a Double Data Rate Synchronous Dynamic Random Access Memory (DDR SDRAM), and the like; the memory may also comprise a non-volatile memory, such as a flash memory. The memory 1302 is used to store computer programs (e.g., applications, functional modules, etc. that implement the above-described methods), computer instructions, etc., which may be stored in partitions in the one or more memories 1302. And the above-described computer programs, computer instructions, data, etc., may be called by the processor 1301.

The computer programs, computer instructions, etc. described above may be stored in one or more memories 1302 in a partitioned manner. And the above-mentioned computer program, computer data, and the like can be called by the processor 1301.

A processor 1301 for executing the computer program stored in the memory 1302 to implement the steps of the methods according to the embodiments described above.

Reference may be made in particular to the description relating to the preceding method embodiment.

The processor 1301 and the memory 1302 may be separate structures or may be integrated structures that are integrated together. When the processor 1301 and the memory 1302 are separate structures, the memory 1302 and the processor 1301 may be coupled through a bus 1303.

The electronic device of this embodiment may execute the technical solution in the method, and the specific implementation process and the technical principle are the same, which are not described herein again.

In the technical scheme of the disclosure, the collection, storage, use, processing, transmission, provision, disclosure and other processing of the personal information of the related user are all in accordance with the regulations of related laws and regulations and do not violate the good customs of the public order.

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

According to an embodiment of the present disclosure, the present disclosure also provides a computer program product comprising: a computer program, stored in a readable storage medium, from which at least one processor of the electronic device can read the computer program, and the execution of the computer program by the at least one processor causes the electronic device to perform the solutions provided by any of the above embodiments.

FIG. 14 shows a schematic block diagram of an example electronic device 1400 that can 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 devices, such as personal digital assistants, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the disclosure described and/or claimed herein.

As shown in fig. 14, the device 1400 includes a computing unit 1401 that can perform various appropriate actions and processes in accordance with a computer program stored in a Read Only Memory (ROM) 1402 or a computer program loaded from a storage unit 1408 into a Random Access Memory (RAM) 1403. In the RAM 1403, various programs and data required for the operation of the device 1400 can also be stored. The calculation unit 1401, the ROM 1402, and the RAM 1403 are connected to each other via a bus 1404. An input/output (I/O) interface 1405 is also connected to bus 1404.

Various components in device 1400 connect to I/O interface 1405, including: an input unit 1406 such as a keyboard, a mouse, or the like; an output unit 1407 such as various types of displays, speakers, and the like; a storage unit 1408 such as a magnetic disk, optical disk, or the like; and a communication unit 1409 such as a network card, a modem, a wireless communication transceiver, and the like. The communication unit 1409 allows the device 1400 to exchange information/data with other devices via a computer network such as the internet and/or various telecommunication networks.

The computing unit 1401 may be a variety of general purpose and/or special purpose processing components having processing and computing capabilities. Some examples of the computing unit 1401 include, but are not limited to, a Central Processing Unit (CPU), a Graphics Processing Unit (GPU), various dedicated Artificial Intelligence (AI) computing chips, various computing units running machine learning model algorithms, a Digital Signal Processor (DSP), and any suitable processor, controller, microcontroller, and the like. The calculation unit 1401 performs the respective methods and processes described above, such as the lane-level map matching method. For example, in some embodiments, the lane-level map matching method may be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 1408. In some embodiments, part or all of the computer program may be loaded and/or installed onto device 1400 via ROM 1402 and/or communication unit 1409. When the computer program is loaded into the RAM 1403 and executed by the computing unit 1401, one or more steps of the lane-level map matching method described above may be performed. Alternatively, in other embodiments, the computing unit 1401 may be configured to perform the lane-level map matching method by any other suitable means (e.g. by means of firmware).

Various implementations of the systems and techniques described here above may be implemented in digital electronic circuitry, integrated circuitry, field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), system on a chip (SOCs), complex Programmable Logic Devices (CPLDs), computer hardware, firmware, software, and/or combinations thereof. These various embodiments may include: implemented in one or more computer programs that are executable and/or interpretable on a programmable system including at least one programmable processor, which may be special or general purpose, receiving data and instructions from, and transmitting data and instructions to, a storage system, at least one input device, and at least one output device.

Program code for implementing the methods of the present disclosure may be written in any combination of one or more programming languages. These program codes 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 codes, when executed by the processor or controller, cause the functions/operations specified in the flowchart and/or block diagram to be performed. 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. A 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 (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a compact disc 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., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to a user; and a keyboard and a pointing device (e.g., a mouse or a 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 can 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, speech, or tactile input.

The systems and techniques described here can be implemented in a computing system that includes a back-end 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 back-end, 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 Networks (LANs), wide Area Networks (WANs), and the Internet.

The computer system may include clients and servers. A client and server are generally 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 can be a cloud Server, also called a cloud computing Server or a cloud host, and is a host product in a cloud computing service system, so as to solve the defects of high management difficulty and weak service expansibility in the traditional physical host and VPS service ("Virtual Private Server", or simply "VPS"). The server may also be a server of a distributed system, or a server incorporating a blockchain.

It should be understood that various forms of the flows shown above may be used, with steps reordered, added, or deleted. For example, the steps described in the present disclosure may be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in the present disclosure can be achieved, and the present disclosure is not limited herein.

The above detailed description should not be construed as limiting the scope of the disclosure. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may be made in accordance with design requirements and other factors. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present disclosure should be included in the scope of protection of the present disclosure.

Claims

1. A lane-level map matching method, comprising:

sampling a calculation method corresponding to the switching type, calculating to obtain a transition probability between the two adjacent observation points, and determining a target lane-level path matched with the observation sequence according to the transition probability and the obtained emission probability, wherein the calculation methods corresponding to different switching types are different;

the calculating method of the sampling corresponding to the switching type calculates the transition probability between the two adjacent observation points, and includes:

if the switching type represents that the two adjacent observation points are observation points of the same lane group, acquiring a distance between the two adjacent observation points, a distance between the two adjacent observation points and a penalty coefficient between the two adjacent observation points, wherein the penalty coefficient is determined based on a traffic relation of the vehicle between lanes where the two adjacent observation points are located; and acquiring the switching frequency of the vehicle in the same lane group;

adjusting the penalty coefficient according to the switching frequency to obtain an adjusted penalty coefficient, wherein the larger the switching frequency is, the closer the penalty coefficient is to 0;

and calculating to obtain the transition probability according to the distance, the distance and the adjusted penalty coefficient.

2. The method according to claim 1, wherein the penalty factor is 1 if the traffic relation indicates that the vehicle can transit traffic between the lanes where the two adjacent observation points are located;

and if the traffic relation indicates that the vehicle cannot transit between the lanes where the two adjacent observation points are located, the penalty coefficient is 0.

3. The method of claim 1, wherein the sampling a calculation method corresponding to the switching type to calculate a transition probability between the two adjacent observation points comprises:

if the switching type represents that the two adjacent observation points are not observation points of the same lane group, acquiring the distance between the two adjacent observation points and the distance between the two adjacent observation points;

and calculating to obtain the transition probability according to the distance and the distance.

4. The method of claim 1, further comprising:

acquiring a distance factor and an angle factor between the observation sequence and each lane, and acquiring a preset attribute factor of each lane, wherein the attribute factor of each lane is used for representing the appropriate degree of the vehicle running on the lane;

and determining the emission probability between the observation sequence and each lane according to the acquired distance factor, angle factor and attribute factor, wherein the acquired emission probability comprises the emission probability between the observation sequence and each lane.

5. The method of claim 1, wherein determining a target lane-level path matching the observation sequence according to the transition probability and the obtained transmission probability comprises:

calculating to obtain the Viterbi probability corresponding to each candidate lane level path matched with the obtained observation sequence according to the transition probability and the obtained emission probability;

sequentially eliminating Viterbi probabilities meeting preset conditions from each Viterbi probability, wherein the preset conditions comprise: the current Viterbi probability is the maximum Viterbi probability, the transition probability of the current Viterbi probability obtained by calculation is greater than a preset first threshold, and the emission probability of the current Viterbi probability obtained by calculation is smaller than a preset second threshold;

and selecting the maximum Viterbi probability from the Viterbi probabilities which meet the preset conditions, and determining the candidate lane level path corresponding to the selected maximum Viterbi probability as the target lane level path.

6. The method according to any one of claims 1-5, further comprising:

matching according to a preset general road network corresponding to the preset high-precision road network to obtain a road-level matching result corresponding to the observation sequence, wherein the preset general road network comprises a plurality of road sections;

and determining a target matching result matched with the observation sequence according to the road level matching result and the target lane level path.

7. The method of claim 6, wherein determining a target match matching the observation sequence based on the road-level match and the target lane-level path comprises:

determining a compatibility attribute for each observation point in the observation sequence;

if the compatible attribute of each observation point represents the lane of the observation point on the target lane-level path and is located on the road section of the observation point on the road-level matching result, determining the target lane-level path as the target matching result;

and if the compatible attribute of any observation point represents the lane of the observation point on the target lane-level path and no road section of the observation point on the road-level matching result is located, replacing part of the path corresponding to any observation point in the target lane-level path with the road-level matching result of any observation point to obtain the target matching result.

8. A lane-level map matching apparatus, characterized by comprising:

the first determining unit is used for determining a target lane level path matched with the observation sequence according to the transition probability and the acquired emission probability, wherein the calculation methods corresponding to different switching types are different;

the calculation unit includes:

a first obtaining subunit, configured to, if the switching type indicates that the two adjacent observation points are observation points of the same lane group, obtain a distance between the two adjacent observation points, and a penalty coefficient between the two adjacent observation points, where the penalty coefficient is determined based on a traffic relationship between lanes where the two adjacent observation points are located by the vehicle;

the first calculating subunit is used for calculating the transition probability according to one or more of the distance, the distance and the penalty coefficient;

if the switching type represents that the two adjacent observation points are observation points of the same lane group, the calculation unit further includes:

the second acquisition subunit is used for acquiring the switching frequency of the vehicle in the same lane group;

the first calculating subunit is configured to calculate the transition probability according to the switching frequency and one or more of the distance, and the penalty coefficient;

the first computing subunit includes:

the adjusting module is used for adjusting the penalty coefficient according to the switching frequency to obtain an adjusted penalty coefficient, wherein the larger the switching frequency is, the closer the penalty coefficient is to 0;

9. The apparatus of claim 8, wherein the penalty factor is 1 if the traffic relationship indicates that the vehicle can transit traffic between the lanes in which the two adjacent observation points are located;

10. The apparatus of claim 8, wherein the computing unit comprises:

the third obtaining subunit is configured to obtain a distance between the two adjacent observation points and a distance between the two adjacent observation points if the switching type represents that the two adjacent observation points are not observation points of the same lane group;

and the second calculating subunit is used for calculating the transition probability according to the distance and the distance.

11. The apparatus of claim 8, further comprising:

a third obtaining unit, configured to obtain a distance factor and an angle factor between the observation sequence and each lane, and obtain a preset attribute factor of each lane, where the attribute factor of each lane is used to represent a suitable degree of driving of the vehicle on the lane;

and a second determining unit, configured to determine, according to the obtained distance factor, angle factor, and attribute factor, a transmission probability between the observation sequence and each lane, where the obtained transmission probability includes the transmission probability between the observation sequence and each lane.

12. The apparatus of claim 8, wherein the first determining unit comprises:

the third calculation subunit is used for calculating to obtain the Viterbi probability corresponding to each candidate lane level path matched with the obtained observation sequence according to the transition probability and the obtained emission probability;

and the rejection subunit is used for sequentially rejecting the Viterbi probabilities meeting preset conditions from the Viterbi probabilities, wherein the preset conditions comprise: the current Viterbi probability is the maximum Viterbi probability, the transition probability of the current Viterbi probability obtained by calculation is greater than a preset first threshold, and the emission probability of the current Viterbi probability obtained by calculation is smaller than a preset second threshold;

a selecting subunit, configured to select a maximum viterbi probability from the viterbi probabilities that have been rejected to satisfy the preset condition;

and the first determining subunit is used for determining the candidate lane-level path corresponding to the selected maximum Viterbi probability as the target lane-level path.

13. The apparatus of any one of claims 8-12, further comprising:

the matching unit is used for matching according to a preset general road network corresponding to the preset high-precision road network to obtain a road-level matching result corresponding to the observation sequence, wherein the preset general road network comprises a plurality of road sections;

and the third determining unit is used for determining a target matching result matched with the observation sequence according to the road level matching result and the target lane level path.

14. The apparatus of claim 13, wherein the third determining unit comprises:

a second determining subunit, configured to determine a compatibility attribute of each observation point in the observation sequence;

a third determining subunit, configured to determine, if the compatible attribute of each observation point represents that the observation point is on the lane of the target lane-level route, and the observation point is located on the road segment of the road-level matching result, the target lane-level route as the target matching result;

and the replacing subunit is configured to replace, if the compatible attribute of the arbitrary observation point represents that the observation point is on the lane of the target lane-level path, and there is no road segment where the observation point is on the road-level matching result, a part of the path corresponding to the arbitrary observation point in the target lane-level path with the road-level matching result of the arbitrary observation point, so as to obtain the target matching result.

15. An electronic device, comprising:

at least one processor; and

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 1-7.

16. A non-transitory computer readable storage medium having stored thereon computer instructions for causing the computer to perform the method of any one of claims 1-7.