1. A vehicle positioning method, characterized by comprising:

acquiring vehicle running data sent by a plurality of road side devices in the running process of a vehicle, wherein the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be positioned;

inputting the vehicle driving data into a pre-trained fingerprint library positioning model for identification, and determining the position information of the vehicle, wherein the fingerprint library positioning model is obtained by training based on fingerprint library information corresponding to each road side device;

and sending the determined at least one position information of the vehicle to a corresponding server every preset time length, so that the server determines the state of the vehicle according to the at least one position information of the vehicle.

2. The method of claim 1, wherein the vehicle driving data includes an arrival angle and a roadside device identifier sent by at least two adjacent roadside devices, and the inputting the vehicle driving data into a pre-trained fingerprint database location model for recognition to determine the location information of the vehicle includes:

and inputting the arrival angles sent by the at least two adjacent roadside devices and the corresponding roadside device identifications into a pre-trained fingerprint database positioning model for identification, and determining the position information of the vehicle.

3. The method according to claim 1, wherein the sending the determined at least one position information of the vehicle to a corresponding server every preset time period so that the server determines the state of the vehicle according to the at least one position information of the vehicle comprises:

and sending the determined at least one position information of the vehicle to a corresponding server every preset time length, so that the server determines the driving direction and the driving speed of the vehicle and the driving state of the surrounding vehicle according to the at least one position information of the vehicle.

4. The method according to any one of claims 1 to 3, wherein, before the obtaining of vehicle driving data sent by a plurality of road side devices during driving of the vehicle, the road side devices in a pre-constructed grid area corresponding to an area to be located further comprise:

acquiring absolute position information of each piece of road side equipment and a corresponding road side equipment identifier;

and constructing a grid region containing each road side device based on a preset grid region construction rule, the absolute position of each road side device and the corresponding road side device identification.

5. The method according to claim 4, wherein after the step of constructing the grid region containing the road side devices based on the preset grid region construction rules, the absolute position of each road side device and the corresponding road side device identification, the method further comprises the steps of:

for each roadside device in the grid area, constructing a polar coordinate according to the coverage area of the roadside device and the absolute position of the roadside device;

dividing the grid region in the polar coordinate based on a preset specification to obtain a plurality of sub-grid regions;

determining an arrival angle corresponding to each sub-grid area and sub-polar coordinates corresponding to a coverage area of the road side equipment in the driving process of the training vehicle in the grid area;

determining fingerprint database information of the road side equipment according to the arrival angle corresponding to each sub-grid region, the sub-polar coordinates corresponding to the coverage region of the road side equipment and the corresponding road side equipment identifier;

and inputting the fingerprint library information corresponding to each road side device into a training model for training to obtain a fingerprint library positioning model.

6. The method according to claim 5, wherein after the inputting the fingerprint library information corresponding to each road side device into the training model for training to obtain the fingerprint library positioning model, the method further comprises:

inputting the test fingerprint library information sets corresponding to the devices of each road into the fingerprint library positioning model for identification to obtain a test position information set of the vehicle;

determining the accuracy of the fingerprint database positioning model according to the testing position information in the testing position information set of the vehicle and the corresponding actual position information;

and if the accuracy of the fingerprint database positioning model is not greater than the preset accuracy, re-executing the steps of establishing polar coordinates for each roadside device in the grid area according to the coverage area of the roadside device and the absolute position of the roadside device, and the subsequent steps.

7. A vehicle positioning device, comprising:

the system comprises an acquisition module, a positioning module and a positioning module, wherein the acquisition module is used for acquiring vehicle running data sent by a plurality of road side devices in the running process of a vehicle, and the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be positioned;

the processing module is used for inputting the vehicle driving data into a pre-trained fingerprint library positioning model for identification and determining the position information of the vehicle, wherein the fingerprint library positioning model is obtained by training based on the fingerprint library information corresponding to each road side device;

the processing module is further configured to send the determined at least one position information of the vehicle to a corresponding server every preset time period, so that the server determines the state of the vehicle according to the at least one position information of the vehicle.

8. A vehicle positioning apparatus, comprising: at least one processor and memory;

the memory stores computer-executable instructions;

the at least one processor executing the computer-executable instructions stored by the memory causes the at least one processor to perform the vehicle localization method of any of claims 1-6.

9. A computer-readable storage medium having computer-executable instructions stored thereon which, when executed by a processor, implement the vehicle localization method according to any one of claims 1 to 6.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, carries out the vehicle positioning method according to any one of claims 1 to 6.

Background

Currently, vehicle positioning can be mainly realized by means of an outdoor positioning technology based on a wireless mobile communication network, and the outdoor positioning technology based on the wireless mobile communication network can also comprise a positioning technology based on a ranging model and a positioning technology based on a position fingerprint.

In the prior art, for the location technology based on location fingerprints, the location technology can be mainly implemented by vehicle-road cooperative equipment, and the specific implementation process can include two stages: the first stage is a training stage, signal characteristic parameters of the positions of all reference nodes in a required positioning area can be collected mainly through an RSU (road side Unit) to form a unified fingerprint characteristic library, the second stage is a vehicle positioning stage, a vehicle-mounted device trains a fingerprint library positioning model in advance through the unified fingerprint characteristic library collected by all RUSs, and then newly collected vehicle driving data can be identified through the fingerprint library positioning model to further determine the position information of a vehicle.

However, different from the cellular deployment mode of the traditional base station, the deployment mode of the RSU is often deployed along the road, the deployment position is not regular, the accuracy of the formed unified fingerprint feature library data is reduced, the accuracy of the training result of the fingerprint library positioning model is influenced, and the accuracy of vehicle positioning is further influenced.

Disclosure of Invention

The embodiment of the invention provides a vehicle positioning method, a vehicle positioning device and vehicle positioning equipment, which are used for improving the accuracy of vehicle positioning.

In a first aspect, an embodiment of the present invention provides a vehicle positioning method, including:

acquiring vehicle running data sent by a plurality of road side devices in the running process of a vehicle, wherein the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be positioned;

inputting the vehicle driving data into a pre-trained fingerprint library positioning model for identification, and determining the position information of the vehicle, wherein the fingerprint library positioning model is obtained by training based on fingerprint library information corresponding to each road side device;

and sending the determined at least one position information of the vehicle to a corresponding server every preset time length, so that the server determines the state of the vehicle according to the at least one position information of the vehicle.

Optionally, the step of inputting the vehicle driving data into a pre-trained fingerprint database location model for identification to determine the location information of the vehicle includes:

and inputting the arrival angles sent by the at least two adjacent roadside devices and the corresponding roadside device identifications into a pre-trained fingerprint database positioning model for identification, and determining the position information of the vehicle.

Optionally, the sending the determined at least one position information of the vehicle to a corresponding server every preset time period, so that the server determines the state of the vehicle according to the at least one position information of the vehicle, includes:

and sending the determined at least one position information of the vehicle to a corresponding server every preset time length, so that the server determines the driving direction and the driving speed of the vehicle and the driving state of the surrounding vehicle according to the at least one position information of the vehicle.

Optionally, before the acquiring vehicle driving data sent by a plurality of roadside devices in the driving process of the vehicle, where the roadside device is a roadside device in a pre-constructed grid region corresponding to the region to be located, the method further includes:

acquiring absolute position information of each piece of road side equipment and a corresponding road side equipment identifier;

and constructing a grid region containing each road side device based on a preset grid region construction rule, the absolute position of each road side device and the corresponding road side device identification.

Optionally, after the constructing the grid region including each roadside device based on the preset grid region construction rule, the absolute position of each roadside device, and the corresponding roadside device identifier, the method further includes:

for each roadside device in the grid area, constructing a polar coordinate according to the coverage area of the roadside device and the absolute position of the roadside device;

dividing the grid region in the polar coordinate based on a preset specification to obtain a plurality of sub-grid regions;

determining an arrival angle corresponding to each sub-grid area and sub-polar coordinates corresponding to a coverage area of the road side equipment in the driving process of the training vehicle in the grid area;

determining fingerprint database information of the road side equipment according to the arrival angle corresponding to each sub-grid region, the sub-polar coordinates corresponding to the coverage region of the road side equipment and the corresponding road side equipment identifier;

and inputting the fingerprint library information corresponding to each road side device into a training model for training to obtain a fingerprint library positioning model.

Optionally, after inputting the fingerprint library information corresponding to each road side device into a training model for training to obtain a fingerprint library positioning model, the method further includes:

inputting the test fingerprint library information sets corresponding to the devices of each road into the fingerprint library positioning model for identification to obtain a test position information set of the vehicle;

determining the accuracy of the fingerprint database positioning model according to the testing position information in the testing position information set of the vehicle and the corresponding actual position information;

and if the accuracy of the fingerprint database positioning model is not greater than the preset accuracy, re-executing the steps of establishing polar coordinates for each roadside device in the grid area according to the coverage area of the roadside device and the absolute position of the roadside device, and the subsequent steps.

In a second aspect, an embodiment of the present invention provides a vehicle positioning apparatus, including:

the system comprises an acquisition module, a positioning module and a positioning module, wherein the acquisition module is used for acquiring vehicle running data sent by a plurality of road side devices in the running process of a vehicle, and the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be positioned;

the processing module is used for inputting the vehicle driving data into a pre-trained fingerprint library positioning model for identification and determining the position information of the vehicle, wherein the fingerprint library positioning model is obtained by training based on the fingerprint library information corresponding to each road side device;

the processing module is further configured to send the determined at least one position information of the vehicle to a corresponding server every preset time period, so that the server determines the state of the vehicle according to the at least one position information of the vehicle.

In a third aspect, an embodiment of the present invention provides a vehicle positioning apparatus, including: at least one processor and memory;

the memory stores computer-executable instructions;

the at least one processor executing the computer-executable instructions stored by the memory causes the at least one processor to perform the vehicle localization method of any of the first aspects.

In a fourth aspect, the embodiment of the present invention provides a computer-readable storage medium, in which a computer executes instructions, and when a processor executes the computer to execute the instructions, the vehicle positioning method according to any one of the first aspect is implemented.

In a fifth aspect, embodiments of the present invention provide a computer program product comprising a computer program that, when executed by a processor, implements a vehicle localization method as described above in the first aspect and in various possible designs of the first aspect.

The embodiment of the invention provides a vehicle positioning method, a device and equipment, after the scheme is adopted, vehicle driving data sent by a plurality of road side devices in the driving process of a vehicle can be obtained firstly, wherein the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be positioned, then the vehicle driving data can be input into a pre-trained fingerprint library positioning model for identification, the position information of the vehicle is determined, the fingerprint library positioning model is obtained by training the fingerprint library information corresponding to each road side device, then at least one position information of the determined vehicle can be sent to a corresponding server every preset time so that the server can determine the state of the vehicle according to the at least one position information of the vehicle, the area where the road side devices are located is divided into regular grid areas in advance, and then the vehicle driving data in the regular grid area driving process of the vehicle are obtained, and the position information of the vehicle is determined by a fingerprint database positioning model obtained by training based on the fingerprint database information corresponding to each road side device, so that the accuracy of the training result of the fingerprint database positioning model is improved, and the accuracy of vehicle positioning is further improved.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

Fig. 1 is a schematic architecture diagram of an application system of a vehicle positioning method according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart of a vehicle positioning method according to an embodiment of the present invention;

fig. 3 is an application schematic diagram of a vehicle-road cooperation scenario provided in the embodiment of the present invention;

fig. 4 is an application schematic diagram of a line-of-sight multi-path signal of a vehicle-mounted terminal according to an embodiment of the present invention;

FIG. 5 is a schematic flow chart of a vehicle positioning method according to another embodiment of the present invention;

FIG. 6 is a schematic structural diagram of a vehicle positioning device according to an embodiment of the present invention;

fig. 7 is a schematic hardware structure diagram of a vehicle positioning apparatus according to an embodiment of the present invention.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of including other sequential examples in addition to those illustrated or described. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

In the prior art, for the location technology based on location fingerprints, the location technology can be mainly implemented by vehicle-road cooperative equipment, and the specific implementation process can include two stages: the first stage is a training stage, signal characteristic parameters of the positions of all reference nodes in a required positioning area can be mainly acquired through an RSU (remote subscriber Unit) to form a unified fingerprint characteristic library, the second stage is a vehicle positioning stage, a vehicle-mounted device can pre-train a fingerprint library positioning model through the unified fingerprint characteristic library acquired by all RUSs, and then newly acquired vehicle driving data can be identified through the fingerprint library positioning model to further determine the position information of a vehicle. However, different from the cellular deployment mode of the traditional base station, the deployment mode of the RSU is often deployed along the road, and the deployment position is not regular, that is, a regular grid region cannot be formed, so that the signal strength received by the vehicle in the driving process is also continuously changed, the acquired vehicle driving data is incomplete, the accuracy of the formed unified fingerprint database data is reduced, the accuracy of the training result of the fingerprint database positioning model is influenced, and the accuracy of the vehicle positioning is further influenced.

Based on the problems, the method divides the area where the road side equipment is located into the regular grid area in advance, then obtains the vehicle running data of the vehicle in the running process of the regular grid area, and determines the position information of the vehicle through the fingerprint library positioning model obtained through training based on the fingerprint library information corresponding to each road side equipment, so that the technical effects of solving the problem that the intensity of the received signals is large in change in the running process of the vehicle due to the fact that no rule exists in the arrangement of the road side equipment, improving the accuracy of the fingerprint library positioning model and further improving the accuracy of vehicle positioning are achieved.

Fig. 1 is a schematic architecture diagram of an application system of a vehicle positioning method according to an embodiment of the present invention, as shown in fig. 1, a vehicle 101, an on-board terminal 102 deployed on the vehicle 101, a roadside device 103 located in a grid area constructed in advance, and a server 104 for transmitting, during the travel of the vehicle 101 in the grid area constructed in advance, the in-vehicle terminal 102 disposed on the vehicle 101 may receive the vehicle travel data transmitted by the roadside apparatus 103, wherein, the vehicle terminal 102 can receive the vehicle running data transmitted by a plurality of road side devices 103, the vehicle driving data can be input into a pre-trained fingerprint database positioning model for identification, the position information of the vehicle is determined, and the determined at least one position information of the vehicle is sent to the corresponding server 104 every preset time period, so that the server 104 determines the state of the vehicle according to the at least one position information of the vehicle.

The technical solution of the present invention will be described in detail below with specific examples. The following several specific embodiments may be combined with each other, and details of the same or similar concepts or processes may not be repeated in some embodiments.

Fig. 2 is a schematic flowchart of a vehicle positioning method according to an embodiment of the present invention, where the method of this embodiment may be executed by the vehicle terminal 102. As shown in fig. 2, the method of this embodiment may include:

s201: the method comprises the steps of obtaining vehicle running data sent by a plurality of road side devices in the running process of a vehicle, wherein the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be positioned.

In this embodiment, the vehicle may be positioned in real time during the driving process, the position information of the vehicle is determined, and then further analysis processing may be performed according to the position information of the vehicle, so as to finally realize functions of navigation positioning, unmanned driving, and the like.

Furthermore, each region to be positioned can be pre-divided, a plurality of road side devices are deployed in each region to be positioned, and then the region to be positioned is divided according to the actual deployment condition of the road side devices and the grid region division rule, so that a grid region matched with the road side devices is obtained. When the vehicle runs through the area to be positioned, the vehicle-mounted terminal on the vehicle can acquire vehicle running data sent by the multiple road side devices in the running process of the vehicle.

S202: and inputting the vehicle driving data into a pre-trained fingerprint library positioning model for identification, and determining the position information of the vehicle, wherein the fingerprint library positioning model is obtained by training based on the fingerprint library information corresponding to each road side device.

In this embodiment, after the vehicle driving data is acquired, the vehicle driving data may be directly input into a pre-trained fingerprint database location model for identification, so as to determine the location information of the vehicle. The vehicle driving data may be a plurality of pieces, that is, the vehicle driving data transmitted by the plurality of road side devices and received by the vehicle-mounted terminal at the same time may be the plurality of pieces of vehicle driving data.

Further, the vehicle driving data may include the arrival angle and the roadside device identifier sent by at least two adjacent roadside devices, and then the vehicle driving data is input into a pre-trained fingerprint database positioning model for identification, and the determining of the position information of the vehicle may specifically include:

and inputting the arrival angles sent by at least two adjacent roadside devices and the corresponding roadside device identifications into a pre-trained fingerprint database positioning model for recognition, and determining the position information of the vehicle.

Specifically, because the coverage area of the roadside equipment is limited, the deployment mode is not a honeycomb mode, and the roadside equipment does not regularly cover along the road, when the vehicle driving data is acquired, the vehicle driving data of the current roadside equipment and the adjacent roadside equipment can be set to be at least acquired, more data can be acquired, the current position of the vehicle can be accurately positioned, and the requirement that the roadside equipment is continuously switched due to the continuous switching of the regions under the condition of high-speed driving can be met.

In addition, in the range covered by the equipment without the road side, the fingerprint positioning can be seamlessly switched to other outdoor positioning or GNSS, so that the effect of fusing positioning is achieved.

S203: and sending the determined at least one position information of the vehicle to a corresponding server every preset time length so that the server determines the state of the vehicle according to the at least one position information of the vehicle.

In this embodiment, after at least one position information of the vehicle is determined, the determined at least one position information of the vehicle may be sent to the corresponding remote server every preset time period, and the server may further determine the state of the vehicle according to the obtained position information of the vehicle.

Furthermore, the server can determine the driving direction, the driving speed, the driving state of the surrounding vehicle and other states according to at least one piece of position information of the vehicle, and further can realize functions such as navigation positioning and unmanned driving according to the driving direction, the driving speed, the driving state of the surrounding vehicle and the like.

After the scheme is adopted, vehicle running data sent by a plurality of road side devices in the running process of the vehicle can be obtained firstly, wherein the road side devices are road side devices in a pre-constructed grid area corresponding to the area to be positioned, then the vehicle running data can be input into a pre-trained fingerprint library positioning model for identification, and the position information of the vehicle is determined, wherein the fingerprint library positioning model is obtained by training based on the fingerprint library information corresponding to each road side device, then at least one position information of the determined vehicle can be sent to a corresponding server at intervals of preset time, so that the server determines the state of the vehicle according to the at least one position information of the vehicle, the area where the road side devices are located is divided into regular grid areas in advance, then the vehicle running data of the vehicle in the regular grid areas are obtained, and the vehicle running data are determined through the fingerprint library positioning model obtained by training based on the fingerprint library information corresponding to each road side device The method for determining the position information of the vehicle improves the accuracy of the training result of the positioning model of the fingerprint database, and further improves the accuracy of vehicle positioning.

Based on the method of fig. 2, the present specification also provides some specific embodiments of the method, which are described below.

In addition, in another embodiment, before obtaining vehicle driving data sent by a plurality of road side devices during the driving process of a vehicle, where the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be located, the method may further include:

and acquiring absolute position information of each road side device and a corresponding road side device identifier.

And constructing a grid region containing each road side device based on a preset grid region construction rule, the absolute position of each road side device and the corresponding road side device identification.

In this embodiment, based on the deployment mode and the broadcast characteristics of the roadside apparatus, the relative positions of the vehicle and the roadside apparatus may be recorded by adopting a grid method of polar coordinates. The grid region construction rule may adopt an existing grid region construction rule, and then the grid region including each roadside device may be created according to the existing grid region construction rule and the absolute position of each roadside device. Further, each roadside device may be represented in the grid area by roadside device identification.

In addition, after the grid region including each roadside device is constructed based on a preset grid region construction rule, the absolute position of each roadside device, and the corresponding roadside device identifier, the method may further include:

for each roadside device in the grid area, polar coordinates are constructed from the coverage area of the roadside device and the absolute position of the roadside device.

And dividing the grid region in the polar coordinate based on a preset specification to obtain a plurality of sub-grid regions.

And determining an arrival angle corresponding to each sub-grid area and sub-polar coordinates corresponding to the coverage area of the road side equipment in the driving process of the training vehicle in the grid area.

And determining fingerprint database information of the road side equipment according to the arrival angle corresponding to each sub-grid region, the sub-polar coordinates corresponding to the coverage region of the road side equipment and the corresponding road side equipment identifier.

And inputting the fingerprint library information corresponding to each road side device into a training model for training to obtain a fingerprint library positioning model.

In this embodiment, the process of constructing the fingerprint database location model through the fingerprint database information of each roadside device may be divided into the processes of grid area construction, data cleaning, fingerprint database construction, fingerprint database location model construction, and the like.

Specifically, the process of constructing the grid region may include:

based on the deployment mode and the broadcasting characteristics of the roadside devices, the relative positions of the vehicle and the roadside devices can be recorded by adopting a grid method of polar coordinates. The radiation range of each roadside device can be determined by taking each roadside device as an origin and taking a preset distance as a distance radius. Since the position parameters of the roadside devices are known, the vehicle positioning estimation can be obtained through the position parameters of the corresponding roadside devices, and the position estimation under the condition of fast region switching can also be carried out by using the device identifiers among different roadside devices. And in the driving process of the vehicle, the vehicle-mounted terminal can simultaneously receive and identify a plurality of road side equipment signals. In addition, in order to meet the switching requirement of vehicle driving, at least information of two adjacent road side devices needs to be acquired. The coverage of the roadside equipment is an irregular sector area with an effective radius within about 500 m. Therefore, coverage edges of different roadside devices are partially overlapped, and comprehensive weighted evaluation is often required to be performed on fingerprint libraries of different roadside devices at a point of coverage edge measurement so as to improve prediction accuracy. For example, the characteristic value of the fingerprint library information may be expressed as < RSUID (roadside device identification), AOA (angle of arrival), R (radius corresponding to the roadside device), RSRP1 (reference signal received power), RSRP2 (reference signal received power) … >.

The process of cleansing data may include: data cleansing is critical to the accuracy of fingerprint localization, and cleansing criteria for data may include: the fingerprint database has definite characteristic information, the missing information is as little as possible, and no obvious discrete characteristics exist, namely, the data received by the vehicle-mounted terminals can be cleaned according to the standard, and the fingerprint database information corresponding to each vehicle-mounted terminal can be constructed according to the cleaned data.

The process of fingerprint library construction may include:

the coverage area of each roadside device can be divided into grids according to the specification of Xm by taking the roadside device as an origin and taking the east direction of a map as a radius position of a polar coordinate 0 degree, the whole area is uniformly divided into a plurality of grids according to the area of a relative square area, then each grid is numbered, correspondingly, the central point of each grid is converted into a unique AOAgrid and a polar coordinate consisting of the radius length Rgrid of an RSU, each grid comprises a plurality of pieces of < RSUID, R, AOA and RSRP … > feature information, and through the process, each roadside device correspondingly has respective fingerprint library information.

The process of constructing the fingerprint library positioning model can comprise the following steps:

the training model can be trained through the fingerprint library information corresponding to each roadside device to obtain a fingerprint library positioning model, specifically, the model can be trained through a WKNN algorithm, a fingerprint library is formed through direct training of a position-signal strength library, and the actual position of the target to be positioned is directly predicted and output through the model.

And (3) generating a fingerprint library: the method mainly comprises the steps of establishing a plurality of fingerprint libraries corresponding to complete RSRP and AOA with RSU as an origin and polar coordinate radius. In fact, in a practical environment, the RSRP fingerprint of a particular location may be constantly affected by changes in the surrounding environment. In fact, in order to maintain the positioning accuracy, the grid fingerprint database information needs to be continuously maintained and updated to protect the robustness and accuracy.

Taking the grid central point as a grid coordinate point, the coordinate mark of the fingerprint library information can be expressed as:

wherein the signature of the ith training point can be recorded as

R_i＝{l_i,P_i}；

In the signature of each training point,/_i＝[l_i,1,l_i,2,l_i,3...l_i,k]^TRepresenting the position vectors, P, of the current RSU and the neighboring RSUs acquired by the terminal_i＝[p_i,1,p_i,2,p_i,3...p_i,k]^TRSRP components at each RSU region for training points. Since the training points may coincide at the same position, P_iThe average value of the predetermined time is set. Considering that the initial position of the terminal is random, and the cell information is lost due to various fading, switching and other reasons, after the data is cleaned, the signature set of the fingerprint database can be expressed as

In combination with the aforementioned establishment of the polar fingerprint database, the established joint signature set is:

wherein the content of the first and second substances,to obtain the angle of the grid AOA, r_i,gridIs the grid polar radius. The invention selects a weighted K-nearThe neighbor algorithm predicts the test data, and belongs to an evolution algorithm generated by increasing the component weight of the KNN in consideration of the reliability and the priority of the signature of the KNN algorithm. The signature of the test point is R^E＝{r^E,P^EAnd f, taking k nearest neighbor points, and comparing the similarity distance of the KNN as:

D(R^E,R_i ^E)＝||d(P^E)-d(P_I,GRID)||²；

in addition, considering that the RSRP weights of the cells are different, and the predicted result weights are also different in different RSU sub-regions (the predicted weight of the sub-grid position corresponding to the current main service RSU is the highest), in the WKNN algorithm, a weight may be added in the process of establishing the fingerprint database, and the weighting coefficient may be set to w ═ f (P is the highest weight) in the WKNN algorithm₁,P₂...P_c)。

Meanwhile, in order to reduce numerical value influence in position estimation, the distance weighting coefficient is normalized, the neighbor distance of N degrees of freedom is selected, and the test estimation distance of WKNN is calculated as follows:

s.t.3≤N，0≤gridID≤max(gridID)。

in addition, when AOA measurement is performed in a vehicle-road cooperation scene, in the current vehicle-road system scene based on the RSU, the RSU communicates with the vehicle-mounted OBU through broadcast information, and functions such as MDT are difficult to realize. In consideration of the deployment characteristics of RSUs, necessary characteristic values such as AOAs and RSRPs can be selected to construct novel fingerprint library information, and necessary fingerprint library parameters can be obtained based on single-time sampling. In the present embodiment, in order to improve the positioning accuracy, the vehicle AOA may be measured by using the minimum variance matrix beam method and the receiving array antenna of the vehicle itself, including but not limited to this method. Fig. 3 is a schematic application diagram of a vehicle-road cooperation scenario provided in an embodiment of the present invention, and as shown in fig. 3, an RSU broadcasts at a fixed frequency, assuming that a vehicle is in a scenario of a uniform array receiver with multiple antennas, where the number of antennas is N.

Fig. 4 is a schematic diagram illustrating an application of line-of-sight multi-path signals of a vehicle-mounted terminal according to an embodiment of the present invention, and as shown in fig. 4, a least square estimation of a channel response of a kth OFDM subcarrier and an nth antenna element may be represented as

Wherein epsilon_p，κ_p，Respectively representing the complex channel gain, channel delay and arrival angle on the p-th path.Is the array response of the nth antenna element.The time delay interval from the nth array element of the vehicle speed to the reference array element is considered. f is the subcarrier frequency, c is the electromagnetic wave propagation speed, and v is the road traveling speed, and can be obtained from the on-board OBU. w is additive white gaussian noise.

Further, consider that the spacing d of the duplex antennas is typically designed to be a half wavelength and much less than the propagation path length. Meanwhile, when the adjacent subcarrier frequency spacing is larger than the coherence bandwidth, the subcarrier attenuations are mutually independent, epsilon_pCan be considered as an independent attenuation of the k sub-carrier. The vehicle speed can be approximately regarded as constant within this short time interval and is much smaller than the electromagnetic wave transmission rate. The channel estimation at the receiving end can be simplified as follows:

the CFR estimation matrix at the receiving end is,

Y＝[y₁,y₂....y_K]；

wherein, the k column of the matrix isSampling at a single time may establish a frequency domain covariance matrix of

According to the information theory criterion, the frequency domain covariance matrix is decomposed into a signal source matrix and a noise matrix, and the following can be obtained:

X＝Z_s∑_sZ_s ^*+Z_n∑_nZ_n ^*；

wherein, sigma_s＝diag{λ₁,λ₂...λ_sAnd the matrix is an X signal source matrix and is a collection of larger eigenvalues. Sigma_n＝diag{λ_s+1,λ_s+2...λ_NAnd the smaller eigenvalue collection is used as the noise matrix. With s as the matrix bundle window length, the scalar ζ is used to define the matrix bundle as Z₁-ζZ₂，Z₁For signal source matrix 1 to s-1 columns, Z₂Signal source matrices 2 to s columns.

Solving generalized eigenvalues of the matrix beam in the null space as follows:

(Z₁Z₂ ^-1-ζI)x＝0；

by the formulaFor the corresponding eigenvalue component, an estimate of AOA can be obtained

In addition, in another embodiment, after inputting the fingerprint library information corresponding to each road side device into the training model for training to obtain the fingerprint library positioning model, the method may further include:

and inputting the test fingerprint library information set corresponding to each path of equipment into a fingerprint library positioning model for identification to obtain a test position information set of the vehicle.

And determining the accuracy of the fingerprint database positioning model according to the testing position information in the testing position information set of the vehicle and the corresponding actual position information.

And if the accuracy of the fingerprint database positioning model is not greater than the preset accuracy, re-executing the steps of constructing polar coordinates for each road side device in the grid area according to the coverage area of the road side device and the absolute position of the road side device, and the subsequent steps.

In this embodiment, the established fingerprint library positioning model may be used to predict the information set of the test fingerprint library, and then the predicted result is compared with the actual terminal location information, so that a quantized result of the accuracy of the prediction of the model may be obtained. And according to the position result of the terminal position prediction, the training model is closed-loop corrected, and the training parameters of the model are adjusted, so that the prediction capability of the model is enhanced. Because the revised model and the predicted result are in a closed-loop feedback relationship, the correction of the model can be effectively completed in the closed loop, and the accuracy of the fingerprint database positioning model is improved.

In addition, fig. 5 is a schematic flow chart of a vehicle positioning method according to another embodiment of the present invention, as shown in fig. 5, since the deployment of the roadside devices is along the road, the division of the main cell and the neighboring cells is relatively random compared to the conventional cellular architecture, which makes it difficult to divide the standard fingerprint grid. Meanwhile, in the conventional vehicle-road cooperative system, the RSU bears less calculation function, and is more responsible for broadcasting and communicating road conditions in the controlled area, and the vehicle-mounted terminal and the line control device can continuously judge and control the vehicle according to the collected information. This is also different compared to the data processing and signaling interaction functions of the eNB/gNB. Based on the above problem, a corresponding sub-fingerprint library information may be set for each roadside device individually. Specifically, each RSU is used as the origin of the fingerprint positioning sub-region, and the RSU coverage area is used as the radius to perform independent signature training, thereby obtaining each sub-fingerprint library. And when the vehicle enters the region under the jurisdiction of the RSU, the RSU broadcasts the sub-fingerprint library information to the vehicle. And training the vehicle in the sub-fingerprint database, and finally comparing the obtained relative positioning result with the inherent coordinates of the RSU, so as to obtain the current position. Meanwhile, the training results of the vehicle in the plurality of sub-fingerprint libraries can be subjected to combined processing and correction, and the positioning longitude is improved. Because the single RSU is not large in area, the number of signaling bytes occupied by the sub-fingerprint database information is small, and the calculation function is mainly processed by the vehicle end, the transmission delay of signaling interaction is greatly reduced, accurate error correction can be made according to the OBU parameters, and the positioning accuracy and the real-time performance are guaranteed. Since the coverage area of the RSU is very limited and the deployment is not cellular. And in the driving process of the train coupler, at least the sub-fingerprint libraries of the current RSU and the RSU of the adjacent region are obtained. On one hand, in order to acquire more signatures, the current position is accurately positioned. On the other hand, the requirement of continuous switching of the sub-fingerprint database caused by continuous switching of the cells under high-speed driving can be met. In the coverage range of the equipment without the road side, fingerprint positioning can be seamlessly switched to other outdoor positioning or GNSS, so that the effect of fusion positioning is achieved.

In addition, because a fingerprint library utilizing new signature elements is used, the originally required uploaded GNSS longitude and latitude determination fingerprint library is replaced by a road side equipment ID (RSU ID) and an angle of arrival (AOA). And fingerprint positioning of the independent cell is carried out under the condition of no adjacent cell information, and a polar coordinate fingerprint database is constructed. Based on the cooperative scene of the vehicle and the road, the obtained test data is cleaned and trained by adopting a machine learning method, and meanwhile, the test error is reduced based on a Kalman filtering mode, so that the reliability of the data is increased.

The method for issuing the fingerprint database by the roadside device takes the vehicle as a positioning main body. The vehicle utilizes the fingerprint database and self test parameters, calculates the relative position of the vehicle and the road side equipment by adopting a machine learning method, and then positions the vehicle in real time according to the absolute position of the road side equipment. The positioning calculation is placed at the terminal, so that the signaling interaction is greatly reduced, the transmission delay is reduced, and the positioning calculation efficiency is improved. And proposes separate roadside device coverage areas as independent fingerprint repositories and independent grid areas. The vehicle can obtain at least the fingerprint library information of the coverage area of the current road side equipment and the fingerprint library information of the adjacent equipment at the same time, the problem of communication switching of the vehicle in high-speed advancing is solved, and meanwhile, the positioning precision can be improved.

For the method for fingerprint positioning of the independent cell under the condition of missing the adjacent cell information, the method adopts a novel fingerprint database signature, utilizes devices such as a uniform array receiver of a vehicle and the like, estimates the arrival angle by using methods such as a minimum variance matrix beam and the like, and combines other parameters such as RSRP and the like to form a polar coordinate fingerprint database. Therefore, the problem that the road side equipment cell cannot perform fingerprint positioning due to insufficient channel parameters is solved.

Based on the same idea, an embodiment of the present specification further provides a device corresponding to the foregoing method, and fig. 6 is a schematic structural diagram of a vehicle positioning device provided in an embodiment of the present invention, as shown in fig. 6, the vehicle positioning device may include:

the obtaining module 601 is configured to obtain vehicle driving data sent by a plurality of road side devices in a driving process of a vehicle, where the road side devices are road side devices in a pre-constructed grid area corresponding to an area to be located.

The processing module 602 is configured to input the vehicle driving data into a pre-trained fingerprint library positioning model for identification, and determine location information of the vehicle, where the fingerprint library positioning model is obtained by training based on fingerprint library information corresponding to each roadside device.

In this embodiment, the vehicle driving data includes an arrival angle and a roadside device identifier sent by at least two adjacent roadside devices, and the processing module 602 is further configured to:

and inputting the arrival angles sent by the at least two adjacent roadside devices and the corresponding roadside device identifications into a pre-trained fingerprint database positioning model for identification, and determining the position information of the vehicle.

The processing module 602 is further configured to send the determined at least one position information of the vehicle to a corresponding server every preset time period, so that the server determines the state of the vehicle according to the at least one position information of the vehicle.

In this embodiment, the processing module 602 is further configured to:

and sending the determined at least one position information of the vehicle to a corresponding server every preset time length, so that the server determines the driving direction and the driving speed of the vehicle and the driving state of the surrounding vehicle according to the at least one position information of the vehicle.

Furthermore, in another embodiment, the processing module 602 is further configured to:

and acquiring absolute position information of each road side device and a corresponding road side device identifier.

And constructing a grid region containing each road side device based on a preset grid region construction rule, the absolute position of each road side device and the corresponding road side device identification.

Furthermore, in another embodiment, the processing module 602 is further configured to:

and constructing polar coordinates for each roadside device in the grid area according to the coverage area of the roadside device and the absolute position of the roadside device.

And dividing the grid region in the polar coordinate based on a preset specification to obtain a plurality of sub-grid regions.

And determining an arrival angle corresponding to each sub-grid area and sub-polar coordinates corresponding to the coverage area of the road side equipment in the driving process of the training vehicle in the grid area.

And determining fingerprint database information of the road side equipment according to the arrival angle corresponding to each sub-grid region, the sub-polar coordinates corresponding to the coverage region of the road side equipment and the corresponding road side equipment identifier.

And inputting the fingerprint library information corresponding to each road side device into a training model for training to obtain a fingerprint library positioning model.

Furthermore, in another embodiment, the processing module 602 is further configured to:

and inputting the test fingerprint library information set corresponding to each path of equipment into the fingerprint library positioning model for identification to obtain a test position information set of the vehicle.

And determining the accuracy of the fingerprint database positioning model according to the testing position information in the testing position information set of the vehicle and the corresponding actual position information.

And if the accuracy of the fingerprint database positioning model is not greater than the preset accuracy, re-executing the steps of establishing polar coordinates for each roadside device in the grid area according to the coverage area of the roadside device and the absolute position of the roadside device, and the subsequent steps.

The apparatus provided in the embodiment of the present invention may implement the method in the embodiment shown in fig. 2, and the implementation principle and the technical effect are similar, which are not described herein again.

Fig. 7 is a schematic diagram of a hardware structure of a vehicle positioning apparatus according to an embodiment of the present invention, and as shown in fig. 7, an apparatus 700 according to the embodiment includes: at least one processor 701 and a memory 702. The processor 701 and the memory 702 are connected by a bus 703.

In a specific implementation process, the at least one processor 701 executes the computer-executable instructions stored in the memory 702, so that the at least one processor 701 executes the method in the above-described method embodiment.

For a specific implementation process of the processor 701, reference may be made to the above method embodiments, which implement principles and technical effects similar to each other, and details of this embodiment are not described herein again.

In the embodiment shown in fig. 7, it should be understood that the Processor may be a Central Processing Unit (CPU), other general purpose processors, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of a method disclosed in connection with the present invention may be embodied directly in a hardware processor, or in a combination of the hardware and software modules within the processor.

The memory may comprise high speed RAM memory and may also include non-volatile storage NVM, such as at least one disk memory.

The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, an Extended ISA (EISA) bus, or the like. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of illustration, the buses in the figures of the present application are not limited to only one bus or one type of bus.

The embodiment of the invention also provides a computer-readable storage medium, wherein a computer executing instruction is stored in the computer-readable storage medium, and when a processor executes the computer executing instruction, the vehicle positioning method of the embodiment of the method is realized.

Embodiments of the present invention further provide a computer program product, which includes a computer program, and when the computer program is executed by a processor, the vehicle positioning method as described above is implemented.

The computer-readable storage medium may be implemented by any type of volatile or non-volatile memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic or optical disk. Readable storage media can be any available media that can be accessed by a general purpose or special purpose computer.

An exemplary readable storage medium is coupled to the processor such the processor can read information from, and write information to, the readable storage medium. Of course, the readable storage medium may also be an integral part of the processor. The processor and the readable storage medium may reside in an Application Specific Integrated Circuits (ASIC). Of course, the processor and the readable storage medium may also reside as discrete components in the apparatus.

Those of ordinary skill in the art will understand that: all or a portion of the steps of implementing the above-described method embodiments may be performed by hardware associated with program instructions. The program may be stored in a computer-readable storage medium. When executed, the program performs steps comprising the method embodiments described above; and the aforementioned storage medium includes: various media that can store program codes, such as ROM, RAM, magnetic or optical disks.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.