CN114706110B

CN114706110B - Vehicle satellite dynamic positioning method and system based on vehicle-road cooperation

Info

Publication number: CN114706110B
Application number: CN202210049657.1A
Authority: CN
Inventors: 刘江; 谭思伦; 蔡伯根; 王剑; 陆德彪; 上官伟; 姜维
Original assignee: Beijing Jiaotong University
Current assignee: Beijing Jiaotong University
Priority date: 2022-01-17
Filing date: 2022-01-17
Publication date: 2024-06-11
Anticipated expiration: 2042-01-17
Also published as: CN114706110A

Abstract

The invention provides a satellite dynamic positioning method and system for a vehicle based on vehicle-road cooperation. Comprising the following steps: the method comprises the steps of extracting geographic space information of a target road network, dividing the target road network into a plurality of grid areas, determining modeling ranges of the grid areas, and constructing an area history data set for the grid areas; performing attribute expansion and model training on the observation data samples of each grid area to generate satellite positioning observation quantity weight determining models corresponding to each grid area; and calculating the outline position of the vehicle in the target road network in real time, determining the grid area to which the vehicle belongs, dynamically calculating the observed quantity weight of each visible satellite according to satellite positioning observation data of the vehicle and the cooperative neighboring vehicle and a satellite positioning observation quantity weight determining model of the grid area to which the vehicle belongs, and carrying out positioning calculation on the vehicle in the target road network. The invention uses the vehicle-road cooperative system, and provides favorable support for the update of the fixed weight of the satellite positioning of the vehicle by utilizing the sensing, collecting and processing capacities of the adjacent vehicle set and the road side system in the local road network.

Description

Vehicle satellite dynamic positioning method and system based on vehicle-road cooperation

Technical Field

The invention relates to the technical field of vehicle satellite positioning, in particular to a satellite dynamic positioning method and system for a vehicle based on vehicle-road cooperation.

Background

The intelligent traffic system is to effectively integrate and apply advanced information technology, communication technology, sensing technology, control technology, computer technology and the like to the whole traffic management system, wherein traffic entities such as people, vehicles, roads and the like interact and share information in real time by combining advanced technical means, and a solution is provided for solving traffic efficiency, safety and environmental problems. The gradual maturation and development of the 5G technology and the V2X technology provide more efficient and convenient conditions for information transmission, and further promote the 'collaborative perception, collaborative decision and collaborative control' of vehicles-vehicles and vehicles-roads, so that the collaborative intelligent transportation system has become an important development trend in the future.

The global navigation satellite system (Global Navigation SATELLITE SYSTEM, GNSS) represented by the global positioning system (Global Positioning System, GPS) plays an important role in numerous novel intelligent traffic system applications driven by the vehicle position based on the advantages of high precision, all weather, easy implementation and the like. However, GNSS-based vehicle positioning is also subject to positioning risks caused by various factors, such as being susceptible to signal shielding, multipath interference, electromagnetic interference, spoofing invasion, etc. in a complex urban road environment, which may result in impaired satellite positioning availability, continuity, or degraded satellite positioning performance, so that specific intelligent traffic application requirements cannot be satisfied. Therefore, the combination positioning system is formed by combining different types of auxiliary sensors (such as inertial navigation, wheel speed sensor, vision, laser radar and the like) with satellite navigation by adopting a multi-source information fusion technology, and the problems faced by single satellite positioning can be effectively compensated. Even if the above-mentioned problem is solved by adopting the multisource fusion mode, in the process of implementing positioning calculation by using navigation satellite positioning observation data in different modes and degrees, the quality of observation information from different satellites can be different due to various reasons, therefore, how to implement dynamic weighting for each visible satellite in the actual operation process, so that the positioning calculation logic can effectively adapt to the utilization of the original satellite positioning observation data, and the method is a key subject for ensuring unavoidable positioning calculation performance.

In the traditional satellite positioning application, a plurality of established strategies are generally adopted to simplify the weight calculation to a certain extent, such as distributing the same weight for each visible satellite, or introducing a certain parameter model to substitute parameters such as satellite elevation angle, signal-to-noise ratio and the like to carry out quick estimation on the weight. However, a significant problem with these methods is that the resulting weighting results are difficult to accurately reflect the quality of the observed quality level of the visible satellite signals and establish an effective matching relationship with the observed weight during dynamic operation. In recent years, with the continuous development of novel modes such as the internet of vehicles and a vehicle-road cooperative system, V2X wireless communication technologies represented by special short-range communication (DSRC), 4G/5G and the like become important driving forces for the intelligent and frontier development of a traffic system, researchers at home and abroad widely explore ways for further playing the role value of the intelligent and frontier development, and gradually explore information processing and decision methods based on information interaction cooperation.

The vehicle-vehicle and vehicle-road cooperative interaction provides more sufficient information conditions for dynamic weighting of vehicle satellite positioning, can change the limit of the conventional bicycle for implementing weighting by utilizing limited observation information, further applies the information gathering and information processing capacity of the vehicle, adjacent cooperative vehicles and road side systems to quantitative mapping of the observation quality of the navigation satellite, further enables the implementation of cooperative weighting under the vehicle-road cooperation to be possible, and effectively improves the satellite positioning performance of the vehicle.

Disclosure of Invention

The embodiment of the invention provides a satellite dynamic positioning method of a vehicle based on vehicle-road coordination, which is used for effectively improving satellite positioning performance of the vehicle.

In order to achieve the above purpose, the present invention adopts the following technical scheme.

According to one aspect of the present invention, there is provided a satellite dynamic positioning method for a vehicle based on vehicle-road cooperation, comprising:

s1, extracting geographic space information of a target road network, dividing the target road network into a plurality of grid areas according to the geographic space information, and determining a modeling range of each grid area;

S2, collecting vehicle-mounted satellite positioning observation data of a vehicle in a target road network, and constructing a corresponding region historical data set for each grid region based on the vehicle-mounted satellite positioning observation data of the vehicle;

S3, performing attribute expansion and model training on observation data samples contained in each grid region based on a region history data set of each grid region, and generating a satellite positioning observation quantity weight determining model corresponding to each grid region;

And S4, calculating the outline position of the vehicle in the target road network in real time, determining the grid area to which the vehicle belongs, dynamically calculating the observed quantity weight of each visible satellite according to satellite positioning observation data of the vehicle and the cooperative neighboring vehicle and a satellite positioning observed quantity weight determining model of the grid area to which the vehicle belongs, and carrying out positioning calculation on the vehicle in the target road network based on the observed quantity weight of each visible satellite.

Preferably, the method further comprises:

And S5, the vehicle in the target road network sends satellite positioning observation data of the vehicle to the regional positioning information management unit in real time, and the regional positioning information management unit updates a satellite positioning observation weight model corresponding to the grid region to which the vehicle belongs.

Preferably, the step S1 includes:

S1.1, determining a rough range of a target road network on a map, extracting geographic space information of the target road network to obtain a road network geographic space information set P ⁰{L_max,L_min,B_max,B_min, wherein the road network geographic space information set P ⁰ comprises a longitude upper boundary L _max, a longitude lower boundary L _min, a latitude upper boundary B _max and a latitude lower boundary B _min of the target road network, and the superscript 0 represents the target road network;

S1.2, dividing a target road network into m _L parts and m _B parts according to longitude coordinates and latitude coordinates according to a grid area unit longitude span L _region and a grid area unit latitude span B _region, dividing the target road network into m _L×m_B grid areas, and expanding the geographic space information attribute of the target road network into

S1.3, numbering all grid areas in sequence, determining that the upper longitude boundary of the modeling range of the grid areas is L ⁱ _max, the lower longitude boundary is L ⁱ _min, the upper latitude boundary is B ⁱ _max and the lower latitude boundary is B ⁱ _min, wherein i is the grid area number, and the grid area models are mutually independent.

Preferably, the step S2 includes:

s2.1, after the positioning calculation of a vehicle equipped with a satellite positioning receiver in a target road network is completed at the current moment, extracting satellite positioning observation data, positioning decision information and auxiliary information from actual observation data of the satellite positioning receiver, respectively acquiring a plurality of observation data samples for each grid area by the satellite positioning receiver according to a set acquisition frequency in a period of acquisition time, and forming an original observation data sample by all the observation data samples;

S2.2, determining the regional attribute N _t of each observation data sample based on the positioning decision information and the road network geospatial information of each observation data sample, wherein the specific decision basis is as follows:

wherein l _t and b _t are longitude and latitude in the positioning decision information, floor represents a downward rounding function;

S2.3, determining grid areas corresponding to the observation data samples according to the area attributes N _t and the grid area ranges of the observation data samples, classifying and collecting the observation data samples according to the corresponding grid areas, and obtaining an area history data set of each grid area.

Preferably, the step S3 includes:

S3.1, a satellite elevation threshold value alpha _min, a signal-to-noise ratio threshold value gamma _min and a pseudo-range residual error threshold value zeta _max are set; correcting and compensating a pseudo-range residual value acquired by a satellite receiver to obtain a corrected pseudo-range residual and a pseudo-range residual threshold value xi _max;

S3.2, based on the regional history data set, the availability attribute Y _t,k of the observation data sample which does not meet the satellite elevation angle threshold value alpha _min, the signal-to-noise ratio threshold value gamma _min and the pseudo-range residual error threshold value zeta _max is set to 0, and the formula for judging the availability attribute Y _t,k of the observation data sample X _t,k is as follows:

wherein, the symbol V represents logical judgment OR;

s3.3, extracting observation data samples with the usability attribute Y _t,k of 1 from all the collected observation data samples under each grid area;

extracting a feature quantity from the observation data sample, the feature quantity comprising: elevation angle alpha _t,k, azimuth angle beta _t,k, signal-to-noise ratio gamma _t,k, raw pseudorange χ _t,k, carrier phase delta _t,k, and doppler shift epsilon _t,k;

extracting a residual error including a corrected pseudo-range from the observation data sample Is a target amount of (a).

And performing iterative modeling training on the characteristic quantity and the target quantity of the observed data sample by using the BP neural network, setting a hidden layer of the BP neural network, selecting the minimum error root mean square of the output value and the fitting target value of the BP neural network as a cost function, and stopping iterative modeling training on the characteristic quantity and the target quantity when the maximum iteration number or the cost function reaches a set threshold value to generate an observed quantity performance parameter weighting model corresponding to the grid region.

Preferably, the parameter name of the observed performance parameter weighting model includes: the characteristic variable comprises the following components: elevation angle, azimuth angle, signal-to-noise ratio, raw pseudo-range, carrier phase and Doppler shift, the target variable comprising a pseudo-range residual;

the activation function sigma and the cost function are as follows:

f (x) is an activation function, E is a cost function, P is the total number of samples, j is the number of neurons of an output layer, and d _i(p)、y_i (P) respectively represents an estimated output value of an ith neuron of a current output layer corresponding to a P-th sample and a given target true value.

Preferably, the step S4 includes:

S4.1, each vehicle running in the target road network extracts satellite positioning observation data of the vehicle in real time and receives satellite observation data of a cooperative neighboring vehicle in real time;

S4.2, utilizing satellite positioning observation data of the vehicle, and preliminarily implementing positioning calculation by adopting an equal weight strategy to obtain a rough position of the vehicle, and obtaining a grid region to which the vehicle belongs according to the rough position of the vehicle and a modeling range of the grid region;

determining a vehicle cluster formed by each cooperative adjacent vehicle in the communication range of the vehicle and the vehicle-vehicle, transmitting satellite positioning observation data of the vehicle and the adjacent vehicles to a positioning information management unit of a grid area to which the vehicle belongs, and calculating a pseudo-range residual prediction value of a visible satellite of the jth vehicle in the vehicle cluster at time k by using a satellite positioning observation weight determining model of the grid area by the positioning information management unit See formula (4).

Wherein g _θ (x) represents a satellite positioning observed quantity weight model corresponding to the theta grid region, Respectively representing the elevation angle, azimuth angle and signal-to-noise ratio of the kth satellite observed by the j-number vehicle at the moment t.

The satellite positioning observation data of the vehicle are synthesized, the residual horizontal precision characteristic value RPDOP _k of each visible satellite is calculated, and the larger the residual horizontal precision characteristic value is, the higher the contribution degree in the Wei Xingzhan overall constellation configuration is:

wherein k represents the kth visible satellite, Representing the horizontal accuracy feature value when the first satellite is not in the visible set, PDOP represents the horizontal accuracy feature value when all visible satellite configurations are considered.

For all adjacent vehicles in the vehicle cluster, integrating satellite positioning observation data of all vehicles and a satellite positioning observation quantity weight model of a grid region N _t, and calculating a pseudo-range residual prediction value of a kth visible satellite of the adjacent vehicles at the time t

Pseudo-range residual prediction value of own vehicle combined with each visible satelliteAnd a residual horizontal accuracy feature value RPDOP _k, pseudo-range residual prediction value/>, of all neighbors in the vehicle clusterCalculating the kth visible satellite positioning observed quantity weight w _t,k at the T moment by adopting a T-S fuzzy reasoning method:

wherein h (x) represents a T-S fuzzy reasoning method for weight determination.

Preferably, the positioning calculation of the vehicle in the target road network based on the observed quantity weight of each visible satellite includes:

Carrying out normalization processing on the visible satellite positioning observed quantity weights of the vehicle at the current moment to obtain a fixed weight matrix W _t applied to positioning calculation, wherein the fixed weight matrix W _t is an n multiplied by n diagonal matrix, and the dimension changes along with the change of the current available satellite number n;

Performing pseudo-range observed quantity position calculation on the vehicle by using the weight matrix W _t and adopting a weighted least square method, and performing integrity monitoring on a position calculation result of the vehicle;

specific formulas of the position calculation and the integrity monitoring of the vehicle are shown in formulas (9) to (12)

x＝(H^TWH)^-1HWz (9)

Wherein H is an observation matrix; z is the observation vector; epsilon is an error vector; w is the weight matrix applied; the matrix A and the matrix S are matrices obtained in the resolving process; n is the total number of visible satellites; lambda _W,min is the minimum of chi-squared distributed non-centering parameters that meet the probability of missed detection requirement, and is related to the total number of satellites currently in view.

According to another aspect of the present invention, there is provided a satellite dynamic positioning system for a vehicle based on road cooperation, comprising:

The grid region dividing module is arranged in the region positioning information management unit, extracts the geographic space information of the target road network, divides the target road network into a plurality of grid regions according to the geographic space information, and determines the modeling range of each grid region;

the regional historical data set construction module is arranged in the regional positioning information management unit, and is used for constructing a corresponding regional historical data set for each grid region based on the vehicle-mounted satellite positioning observation data of the vehicle after receiving the vehicle-mounted satellite positioning observation data collected by the vehicle in the target road network;

The satellite positioning observed quantity weight-determining model acquisition module is arranged in the regional positioning information management unit, and is used for carrying out attribute expansion and model training on observed data samples contained in each grid region based on the regional history data set of each grid region, generating a satellite positioning observed quantity weight-determining model corresponding to each grid region and sending the satellite positioning observed quantity weight-determining model to the vehicle positioning calculation module;

The vehicle positioning calculation module is used for setting in the vehicle in the target road network, calculating the outline position of the vehicle in the target road network in real time, determining the grid area to which the vehicle belongs, dynamically calculating the observed quantity weight of each visible satellite according to the satellite positioning observation data of the vehicle and the cooperative adjacent vehicle and the satellite positioning observed quantity weight determining model of the grid area, and carrying out positioning calculation on the vehicle in the target road network based on the observed quantity weight of each visible satellite.

Preferably, the system further comprises:

The satellite positioning observed quantity weight determining model updating module is arranged in the regional positioning information management unit and is used for updating the satellite positioning observed quantity weight determining model corresponding to the grid region to which the vehicle belongs after receiving satellite positioning observed data of the vehicle sent in real time by the vehicle in the target road network.

According to the technical scheme provided by the embodiment of the invention, the satellite dynamic positioning method based on the vehicle-road cooperation provided by the invention can fully utilize the core advantages of the vehicle-road cooperation system mode, and provides favorable support for the fixed weight update of the satellite positioning of the vehicle by utilizing the sensing, collecting and processing capabilities of the adjacent vehicle sets and the road side systems in the local road network, so that the satellite positioning performance is improved in a complex urban road environment, the active tracking and adaptation capability of the satellite positioning to the road space environment where the vehicle is positioned is enhanced, and the satellite dynamic positioning method based on the vehicle-road cooperation system mode has important potential and application value for supporting the application of a plurality of novel cooperative intelligent traffic systems.

Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments 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 other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic diagram of an implementation of a satellite dynamic positioning method for a vehicle based on vehicle-road coordination according to an embodiment of the present invention;

FIG. 2 is a flow chart of a method for dynamically positioning satellites of a vehicle based on vehicle-road coordination according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a grid region division principle based on geospatial information according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a training and predicting process of a region weight model based on a neural network according to an embodiment of the present invention;

FIG. 5 is a graph of time-varying results of weights for each observable satellite in accordance with an embodiment of the present invention;

FIG. 6 is a comparison chart of positioning performance under different weight schemes according to an embodiment of the present invention;

fig. 7 is a block diagram of a satellite dynamic positioning system of a vehicle based on vehicle-road coordination according to an embodiment of the present invention.

Detailed Description

Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals refer to the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the drawings are exemplary only for explaining the present invention and are not to be construed as limiting the present invention.

As used herein, the singular forms "a", "an", "the" and "the" are intended to include the plural forms as well, unless expressly stated otherwise, as understood by those skilled in the art. It will be further understood that the terms "comprises" and/or "comprising," when used in this specification, specify the presence of stated features, integers, steps, operations, elements, and/or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and/or groups thereof. It will be understood that when an element is referred to as being "connected" or "coupled" to another element, it can be directly connected or coupled to the other element or intervening elements may also be present. Further, "connected" or "coupled" as used herein may include wirelessly connected or coupled. The term "and/or" as used herein includes any and all combinations of one or more of the associated listed items.

It will be understood by those skilled in the art that, unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the prior art and will not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

For the purpose of facilitating an understanding of the embodiments of the invention, reference will now be made to the drawings of several specific embodiments illustrated in the drawings and in no way should be taken to limit the embodiments of the invention.

Based on the background, the invention is based on the core advantage of the novel intelligent traffic system vehicle-road cooperative mode, and provides a satellite dynamic positioning method of the vehicle, which can effectively utilize the vehicle-road cooperative interaction information conditions, and provides more solid conditions and foundation for the application of the intelligent traffic system enabled by the satellite positioning of the vehicle.

Example 1

The application scene of the satellite dynamic positioning method of the vehicle based on the vehicle-road coordination provided by the example is as follows: knowing the range of the target road network and being able to extract the geospatial information P ⁰{L_max,L_min,B_max,B_min, wherein the superscript 0 represents the target road network population, including the upper longitude bound L _max, the lower longitude bound L _min, the upper latitude bound B _max, the lower latitude bound B _min; vehicles in the road network are provided with receivers based on a global positioning system (Global Position System, GPS), and can output standard RINEX (RECEIVER INDEPENDENT Exchange Format) protocol and American national marine electronics Association (NMEA-The National Marine Electronics Association) protocol Format data, information interaction can be carried out between the vehicles and the vehicles by utilizing a wireless communication network through an information transmission module, and the vehicle-mounted positioning system and an area positioning information management unit arranged at the road side end in the target road network realize support on hardware and software of the method.

The implementation schematic diagram of the satellite dynamic positioning method of the vehicle based on the vehicle-road cooperation provided by the embodiment of the invention is shown in fig. 1, the specific processing flow is shown in fig. 2, and the implementation schematic diagram comprises the following steps:

s4, calculating the outline position of the vehicle in the target road network in real time, determining the grid area of the vehicle, extracting satellite positioning observation data of the vehicle and a cooperative neighboring vehicle and a satellite positioning observation quantity weight determining model of the grid area, dynamically calculating the observation quantity weight of each visible satellite, and completing the positioning calculation of the vehicle in the target road network based on the observation quantity weight of each visible satellite;

And S5, the vehicle in the target road network sends satellite positioning observation data of the vehicle to an area positioning information management unit at the road side end in the target road network in real time, and updates a satellite positioning observation weight determining model contained in the satellite positioning observation data.

Wherein:

The specific process of step S1 is as follows:

The target road network is a coverage area range of the road side positioning management unit, geographic space information of the target road network is extracted, the target road network is divided into a plurality of grid areas according to the geographic space information, and modeling ranges of the grid areas are determined. Different grid areas covered by the target road network have satellite positioning observation environments with larger difference, so that errors with different degrees are caused by satellite positioning calculation, and the influence degree of regional modeling characterization on satellite signals is needed.

Step S1 further comprises the sub-steps of:

S1.1, for a given target road network using a satellite dynamic positioning method based on a vehicle-road cooperation, determining a rough range of the target road network on a map, extracting geographic space information of the target road network, and obtaining a road network geographic space information set P ⁰{L_max,L_min,B_max,B_min, wherein the geographic space information set P ⁰{L_max,L_min,B_max,B_min comprises a longitude upper boundary L _max, a longitude lower boundary L _min, a latitude upper boundary B _max and a latitude lower boundary B _min of the target road network, and the upper mark 0 represents the target road network.

S1.2, dividing the target road network into m _L parts and m _B parts according to longitude coordinates and latitude coordinates according to the longitude span L _region of the grid area unit and the latitude span B _region of the grid area unit to obtain m _L×m_B grid areas, so that the target road network is divided into a plurality of grid areas.

In this embodiment, consider that dividing the longitude and latitude of the target road network into m _L =5 parts and m _B =5 parts, obtaining 25 grid areas in total, and expanding the geospatial information attribute of the target road network into

S1.3, determining a modeling range of a grid region, numbering all the grid regions in sequence, and determining that the upper longitude limit of the modeling range of the grid region is L ⁱ _max, the lower longitude limit is L ⁱ _min, the upper latitude limit is B ⁱ _max and the lower latitude limit is B ⁱ _min, wherein i is the grid region number, and the grid region models are mutually independent. The schematic diagram of grid region division based on geospatial information provided by the embodiment of the invention, as shown in fig. 3, a plurality of grid regions with geographic space information being Pⁱ{i,Lⁱ _max,Lⁱ _min,Bⁱ _max,Bⁱ _max,P_t ^*}, marked with numbers in grid region i characterize characteristics of road network information of the region where each of the grid regions is located, and the schematic diagram comprises:

(1) With a small number of road segments, such as grid area No. 1;

(2) Having a large number of road segments, such as grid areas No. 5 and No. 11;

(3) There are few road segments, such as a 25 grid area.

The specific process of step S2 is as follows:

The method comprises the steps of collecting vehicle-mounted satellite positioning observation data of a vehicle in a target road network, constructing a corresponding region historical data set for each grid region, and sending the vehicle-mounted satellite positioning observation data to a region positioning information management unit at a road side end after satellite positioning calculation serving the vehicle is completed for all vehicle-mounted satellite positioning receivers, so that accumulation of observation data sample data can be completed.

Step S2 further comprises the sub-steps of:

S2.1, a vehicle with a satellite positioning receiver assembled in a target road network extracts satellite positioning observation data, positioning result information and auxiliary information from actual observation data of the satellite positioning receiver after positioning calculation at the current moment is completed, wherein the actual observation data comprises RINEX (RECEIVER INDEPENDENT Exchange Format independent of the receiver) and NMEA (National Marine Electronics Association, national ocean electronic Association) Format information. The receiver collects the frequency f=1 Hz, namely one second carries on the location to solve, all grid areas will collect several and observe the data sample in a period of collection time, form the primitive observation data sample, wherein:

(1) Satellite positioning observation data comprises UTC time t, visible satellite SAT _t(prn₁,prn₂,...,prn_n), visible satellite elevation angle alpha _t,k, azimuth angle beta _t,k, signal-to-noise ratio gamma _t,k, pseudo range chi _t,k, carrier phase delta _t,k and Doppler frequency shift epsilon _t,k, wherein prn _n is prn number of the nth visible satellite, subscript t is UTC time, k is the kth visible satellite at the current moment, and k epsilon [1, n ];

(2) The positioning decision information comprises a vehicle three-dimensional coordinate (b _t,l_t,h_t) obtained by positioning calculation, a pseudo-range residual xi _t,k and a geometric precision factor set DOP _t, wherein the subscript t is UTC time, and k is the kth visible satellite at the current moment;

(3) Auxiliary information including sampling frequency f, satellite positioning terminal coefficient sigma ^r, satellite positioning terminal ID number r, region attribute N _t, wherein subscript t is UTC time, wherein: the receiver coefficients are a correction factor for the residual results of the position resolution of receivers of different levels.

Wherein L _region、B_region represents a longitude span and a latitude span of a grid area unit, L _max、L_min、B_max、B_min represents a longitude upper boundary, a longitude lower boundary, a latitude upper boundary and a latitude lower boundary of a target road network, m _L、m_B represents the number of parts dividing the target road network into parts according to longitude coordinates and latitude coordinates, L _t and b _t are longitude and latitude in positioning decision information, N _B、N_L is a serial number of the parts of longitude and latitude corresponding to the observed data sample, floor (x) represents a downward rounding function, and N _t is an area attribute of the data sample.

And S2.3, selecting an actual observation data sample of the receiver with UTC= 073510 in the year 1 and the month 15 of 2021, according to the region attribute N _t of the observation data sample, corresponding the observation data sample to a grid region according to the grid region range, and collecting, judging that the region attribute of the observation data sample is 5 by the formula (1), namely, the observation data sample is transmitted to a region historical data set of the grid region No. 5 by a wireless communication network, and expanding the data volume of the observation data sample for subsequent data processing and modeling. And after judging the region attribute of all the acquired observation data samples in sequence, transmitting the region attribute to the corresponding grid region through a wireless communication network, and constructing a region historical data set of each grid region.

Wherein/>Representing corresponding observation data samples for different satellites at different times. Wireless communication networks are representative and include, but are not limited to, 4G/5G public network communications, wi-Fi, DSRC, and the like.

The specific process of step S3 is as follows:

based on the regional history data set, attribute expansion and model training are carried out on the observation data samples contained in the grid region, under the same grid region, the collected observation data samples are obtained under similar observation environments, and similar characteristics can be extracted from a resolving result, so that invalid observation data samples are screened out by setting partial characteristic value thresholds, the observation environments of the grid region are modeled by adopting an intelligent algorithm, and a satellite positioning observation quantity weighting model corresponding to the observation environments of each grid region is generated.

Step S3 further comprises the sub-steps of:

S3.1, expanding availability attribute Y _t,k for all observation data samples of a grid region based on a region history data set, wherein the specific steps comprise:

(1) Setting a satellite elevation threshold value alpha _min =15, a signal-to-noise ratio threshold value gamma _min =24 and a pseudo-range residual threshold value zeta _max =5m;

(2) The satellite positioning terminal coefficients characterize the performance level of the satellite receivers that acquire positioning information, and the satellite positioning terminal ID numbers indicate the identification of satellite receivers of different performance. Under the same observation condition, the satellite receivers of different levels bring pseudo-range residual results of different levels due to the difference of hardware and software, and if the unprocessed results are trained in a model, the influence degree of different satellites on positioning calculation under the current observation environment is difficult to truly represent, so that correction and compensation are necessary to the pseudo-range residual values acquired by the satellite receivers.

The specific implementation steps comprise:

(2-1) acquiring an initial stage, and selecting a receiver coefficient obtained based on information provided by the device parameters in an initial stage of model establishment. The satellite receivers of different models can obtain corresponding precision ranges, the vehicle-mounted satellite positioning terminal precision factor sigma ₀ =20 is extracted from equipment parameters, and the method is calculated by using the following formula:

σ¹＝10

the upper mark 1 represents an ID number of the vehicle-mounted satellite positioning terminal, and the satellite positioning terminal coefficient sigma ¹ represents the calibration precision level of the vehicle-mounted satellite positioning terminal.

And (2-2) based on the sample data of the observation data with a certain scale, obtaining pseudo-range residual distribution with statistical characteristics, and extracting the pseudo-range residual distribution of the same vehicle-mounted satellite positioning terminal ID number. In this embodiment, statistical information is extracted from the original pseudo-range residual values ζ _t,k of all the observation data samples of the satellite receiver model r=1, the confidence level θ=90% is set to obtain the range [ -6.96,7.67] of the corresponding confidence interval, and the satellite positioning terminal coefficient σ ^r of the corresponding model is updated;

σ¹＝max{|-6.96|,|7.67|}＝7.67

(2-3) correcting and compensating the pseudo-range residual error after the absolute value is converted again to obtain a corrected pseudo-range residual error, wherein the corrected pseudo-range residual error is shown in the following formula.

S3.2, for any observation data sample which does not meet the satellite elevation angle threshold value alpha _min, the signal-to-noise ratio threshold value gamma _min and the pseudo-range residual error threshold value xi _max, setting the availability attribute Y _t,k to 0, and judging the availability attribute Y _t,k of the observation data sample X _t,k by combining the following formula (2):

Wherein, the symbol V represents the logical judgment OR.

Taking a specific observation data sample as an example, the following description is given: in the actual acquisition, 8 satellites are observed at a certain moment and positioning calculation is completed, satellite positioning observation data at the moment is extracted, and SAT ₀₇₃₅₁₀ (3,4,9,16,26,27,29,31) can be obtained, wherein,

For satellite number 4, elevation angle a _073510,4 = 55, signal-to-noise ratio γ _073510,4 = 44, raw pseudorange residual xi _0735104 = 3.4, corrected with the receiver coefficients of equation (5),Judging by the formula (2), and obtaining Y _073510,4 =1 through the threshold value;

For satellite number 9, elevation angle a _073510,9 =21, signal-to-noise ratio γ _073510,9 =25, raw pseudorange residual ζ _073510,9 = -55.9, after correction with the receiver coefficients of equation (5), And (3) judging by the formula (2), wherein the pseudo-range residual error does not pass through the threshold value, and obtaining Y _073510,4 =0.

S3.3, performing attribute expansion and model training on the observation data samples contained in the grid region, wherein the specific steps comprise:

(1) Extracting observation data samples with the usability attribute Y _t,k of 1 from all the collected observation data samples under each grid area;

(2) Based on the observation data sample, extracting the feature quantity includes: elevation angle alpha _t,k, azimuth angle beta _t,k, signal-to-noise ratio gamma _t,k, original pseudo-range chi _t,k, carrier phase delta _t,k and Doppler frequency shift epsilon _t,k;

(3) Based on the observation data samples, extracting target quantity comprises correcting pseudo-range residual errors

(4) The schematic diagram of the training and predicting process of the regional weight model based on the neural network is shown in fig. 4.

The method comprises the steps of carrying out observation data sample injection and iteration modeling training on the characteristic quantity and the target quantity, selecting a BP neural network, selecting proper parameters, carrying out training on the characteristic quantity and the target quantity, setting a proper hidden layer, selecting the minimum error root mean square of a network output value and a fitting target value as a cost function, and terminating iteration to generate an observation performance parameter weighting model corresponding to a grid region when the maximum iteration number or the cost function reaches a set threshold value, wherein the parameter names and the set values of the observation performance parameter weighting model are shown in a table 1.

TABLE 1 observance of parameter names and setpoints for a Performance parameter weighting model

The activation function sigma and the cost function are as follows:

Wherein f (x) is an activation function, E is a cost function, P is the total number of samples, j is the number of neurons of an output layer, and d _i(p)、y_i (P) respectively represents the estimated output value of the ith neuron of the current output layer of the corresponding P-th sample and a given target true value.

The specific process of step S4 is:

Each vehicle running in the target road network extracts satellite positioning observation data of the cooperative adjacent vehicles of the vehicle in real time, calculates a sketch position to determine the grid region to which the vehicle belongs, dynamically calculates the weight of each visible satellite positioning observation quantity by combining the satellite positioning observation data of the vehicle, the adjacent vehicles and the satellite positioning observation quantity weight determining model of the grid region to which the vehicle belongs, and completes positioning calculation based on the obtained weight.

Step S4 further comprises the sub-steps of:

S4.1, each vehicle running in the target road network extracts satellite positioning observation data of the vehicle in real time, receives the observation data of the cooperative adjacent vehicles in real time, and takes an actual observation data sample as an example, the satellite positioning observation data of the vehicle observation data and the adjacent vehicle observation data comprise: UTC time t= 073510; the current set of visible satellites SAT ₀₇₃₅₁₀ (3,4,9,16,26,27,29,31); satellite state information, such as elevation angle α _073510,4 =55, azimuth angle β _073510,4 =290, signal-to-noise ratio γ _073510,4 =44 of the satellite of prn=4; original pseudorange χ _073510,4 = 19740068.380; carrier phase δ _073510,4 = 103734739.0612; doppler shift epsilon _073510,4 = 722.173; the auxiliary information comprises acquisition frequency f=1hz, satellite positioning terminal coefficient sigma ¹ =7.67 and satellite positioning terminal ID number r=1.

S4.2, calculating the outline position of the vehicle to determine the grid area of the vehicle, combining the observation data of the vehicle and the adjacent vehicle and the satellite positioning observation quantity weighting model of the grid area, and dynamically calculating the weighting of each visible satellite positioning observation quantity, wherein the specific steps comprise:

(1) And (3) utilizing original satellite positioning observation information of the vehicle, and initially implementing positioning calculation by adopting an equal weight strategy to obtain the outline position of the vehicle, and obtaining the regional attribute N _t =5 of the vehicle according to the grid regional judgment basis.

(2) Determining a vehicle cluster J {0,1,2,3} formed by each cooperative adjacent vehicle in the communication range of the vehicle and the vehicle-vehicle, wherein j=0 represents the vehicle, satellite positioning observation data of the vehicle and the adjacent vehicle are sent to a grid-5 regional positioning information management unit according to the regional attribute N _t =5 of the vehicle, and the pseudo-range residual prediction value of a k-number visible satellite of the J-th vehicle at t time is obtained through regional satellite positioning observation weight model processing and calculationSee formula (4).

At this time, a satellite set which is finally used for calculation is extracted from the observation data obtained by the satellite positioning terminal of the host vehicleThe model prediction is carried out to obtain a predicted pseudo-range residual, namely [0.9763,1.0766,1.1243,3.7337,1.3813], wherein the predicted pseudo-range residual can represent the pseudo-range variance of each satellite to a certain extent in positioning calculation and is an important part of absolute weight.

(3) Calculating the residual horizontal precision characteristic value RPDOP _k of each visible satellite by integrating satellite positioning observation data of the vehicle, wherein the residual horizontal precision characteristic value shows the function of each satellite in the whole satellite configuration, and when the residual horizontal precision characteristic value of a certain satellite is defined as that the satellite is not in a visible set, the horizontal precision characteristic value formed by the residual satellite accounts for the proportion of the horizontal precision characteristic values of all the visible satellites, and the larger the proportion is, the higher the contribution degree of the considered satellite in the whole constellation configuration is:

For all satellites in view of the vehicle at this time in the example RPDOP _k, we obtain RPDOP₄＝1.8092,RPDOP₁₆＝1.6614,RPDOP₂₆＝1.8109,RPDOP₂₇＝2.7432, RPDOP₃₁＝2.0789.

(4) For all adjacent vehicles in the vehicle cluster, namely { J _j }, J is more than or equal to 1, integrating satellite positioning observation data of each vehicle and a grid area N _t satellite positioning observation quantity weight determining model, and calculating pseudo-range residual prediction values of kth visible satellite of the adjacent vehicles at t time in a distributed mannerThe pseudo-range residual error of the adjacent vehicle can represent the observation quality of the whole environment of the area at the current moment.

(5) Pseudo-range residual prediction value of own vehicle combined with each visible satelliteAnd a residual horizontal accuracy feature value RPDOP _k, pseudo-range residual prediction value/>, of all neighbors in the vehicle clusterCalculating the kth visible satellite positioning observed quantity weight w _t,k at the T moment by adopting a T-S fuzzy reasoning method:

Wherein h (x) represents a T-S fuzzy inference method for weight determination.

The T-S fuzzy reasoning method in the model takes a pseudo-range residual predicted value of the vehicle, a precision factor set of the vehicle and a pseudo-range residual predicted value of an adjacent vehicle as inputs, weights as outputs, 27 fuzzy rules are set, weight values are independently obtained for each visible satellite at the same moment, the fuzzy reasoning output value of each visible satellite is w _073510,4＝1.3810 w_073510,16＝1.1357 w_073510,26＝1.0414 w_073510,27＝0.0944 w_073510,31 = 0.6899, and it can be seen that the observation quality of the satellite No. 26 at the moment is lower, and the distribution weight is lowest; the satellite No. 4 has better observation quality and highest distribution weight.

(4) And carrying out normalization processing on the obtained weight information to obtain a fixed weight matrix W _t for position calculation, wherein the fixed weight matrix W _t is an n multiplied by n dimension diagonal array, and the dimension changes along with the change of the number n of currently available satellites.

And S4.3, based on the obtained fixed weight matrix W _t of each visible satellite, the positioning calculation of the vehicle is completed.

The weight of each visible satellite obtained by the vehicle at the moment in the embodiment is normalized, and a fixed weight matrix finally applied to positioning calculation is obtained by:

Corresponding visible satellite sets are In the period t= 073056 ～ 074214, the time-varying result of the weight of each observable satellite in the embodiment is obtained, and the weight distribution situation of the first 100 epochs is shown in fig. 5.

The weight matrix W _t is applied to the processes of positioning integrity monitoring, positioning resolving and the like, so that safer and more reliable guarantee is provided for the application of vehicle positioning information. Positioning calculation selection pseudo-range observed quantity positioning is solved by adopting a weighted least square method, integrity monitoring is carried out by adopting a weighted least square residual method, and specific formulas are shown in formulas (9) to (12)

x＝(H^TWH)^-1HWz (9)

Wherein H is an observation matrix; z is the observation vector; epsilon is an error vector; w is the weight matrix applied; the matrix A and the matrix S are matrices obtained in the resolving process; n is the total number of visible satellites; lambda _W,min is the minimum value of chi-squared distributed non-centering parameters meeting the missing probability requirement, which is only related to the total number of satellites currently in view and can be obtained by looking up a table, taking the moment described in the embodiment as an example, the number of satellites is 5, lambda _W,min＝8.2;H_PL is the level of horizontal protection, and the real-time positioning error protection threshold is calculated according to the warning rate requirement of GNSS application and the actual measurement state, and reflects the performance of the availability of monitoring for the envelope condition of the Horizontal Positioning Error (HPE), wherein HPE is calculated according to the following formula:

wherein x and y refer to the abscissa and the ordinate of the ECEF coordinate system obtained by the positioning and calculating at the current moment, And/>Refers to the true value in the ECEF coordinate system at the current moment.

After the horizontal protection level and the horizontal positioning error of all epochs in the embodiment are calculated, the obtained integrity monitoring and positioning performance is compared with other classical weighting schemes, and the positioning performance advantage of the weighting method provided by the invention can be verified.

Fig. 5 is a weight time-varying result diagram of each observable satellite provided by an embodiment of the present invention, and fig. 6 is a positioning performance comparison diagram under different weight schemes provided by an embodiment of the present invention, where the main comparison schemes include:

(1) An equal weight policy (abbreviated as "equal weight" in the figure);

(2) Elevation weighting parameter model (abbreviated as "elevation weighting" in the figure);

(3) A signal-to-noise ratio weighting parameter model (abbreviated as "signal-to-noise ratio weighting" in the figure);

(4) Elevation signal-to-noise ratio combined weighting model (simply called "parameter combined weighting" in the figure);

(5) The invention provides a satellite dynamic positioning method (called 'cooperative rights' in the figure) of a vehicle based on vehicle-road cooperation;

The specific process of step S5 is:

S5, each vehicle running in the target road network sends satellite positioning observation data of the vehicle to the regional positioning information management unit in real time, namely, each group of data which is subjected to positioning calculation by using a satellite positioning dynamic weight adjustment method based on vehicle-road cooperation can be generated after the positioning calculation is completed, wherein the data has the functions of The observation data sample of the characteristic value carries out increment update on the current area historical data set contained in the grid area through a wireless communication network, and based on the updated historical data set, the step4 is utilized to update the satellite positioning observation quantity weight determining model/>

Example two

The structure diagram of the satellite dynamic positioning system of the vehicle based on the vehicle-road cooperation provided by the embodiment of the invention is shown in fig. 7, and the system comprises the following modules:

the grid region dividing module 71 is configured to be disposed in the region positioning information management unit, extract geospatial information of the target road network, divide the target road network into a plurality of grid regions according to the geospatial information, and determine a modeling range of each grid region;

The regional history data set construction module 72 is configured to be disposed in the regional positioning information management unit, and after receiving the vehicle-mounted satellite positioning observation data collected by the vehicle in the target road network, construct a corresponding regional history data set for each grid region based on the vehicle-mounted satellite positioning observation data of the vehicle;

The satellite positioning observed quantity weight-determining model obtaining module 73 is configured to be disposed in the area positioning information management unit, perform attribute expansion and model training on the observed data samples contained in each grid area based on the area history data set of each grid area, generate a satellite positioning observed quantity weight-determining model corresponding to each grid area, and send the satellite positioning observed quantity weight-determining model to the vehicle positioning resolving module;

The vehicle positioning calculation module 74 is configured to be disposed in a vehicle in the target road network, calculate a rough position of the vehicle in the target road network in real time, determine a grid region to which the vehicle belongs, dynamically calculate an observed quantity weight of each visible satellite according to satellite positioning observation data of the vehicle and a cooperative neighboring vehicle and a satellite positioning observed quantity weight determining model of the grid region, and perform positioning calculation on the vehicle in the target road network based on the observed quantity weights of each visible satellite.

The satellite positioning observed quantity weight determining model updating module 75 is configured to be disposed in the area positioning information management unit, and update the satellite positioning observed quantity weight determining model corresponding to the grid area to which the vehicle belongs after receiving satellite positioning observed data of the vehicle sent in real time by the vehicle in the target road network.

The specific processing procedure of satellite dynamic positioning of vehicles based on vehicle-road cooperation by using the system of the embodiment of the invention is similar to that of the foregoing method embodiment, and will not be repeated here.

In summary, the satellite dynamic positioning method based on the vehicle-road cooperation provided by the invention can fully utilize the core advantages of the vehicle-road cooperation system mode, and provides favorable support for the fixed weight update of the satellite positioning of the vehicle by utilizing the sensing, collecting and processing capabilities of the adjacent vehicle set and the road side system in the local road network, thereby improving the satellite positioning performance in the complex urban road environment, strengthening the active tracking and adaptation capability of the satellite positioning to the road space environment where the vehicle is located, and having important potential and application value for supporting the application of a plurality of novel cooperative intelligent traffic systems.

The invention is applicable to the application of different types of navigation satellites in urban road environments, has universality for different types of road network conditions, different forms of road surrounding buildings and shielding conditions and different road network scales, and has remarkable engineering application value.

Those of ordinary skill in the art will appreciate that: the drawing is a schematic diagram of one embodiment and the modules or flows in the drawing are not necessarily required to practice the invention.

From the above description of embodiments, it will be apparent to those skilled in the art that the present invention may be implemented in software plus a necessary general hardware platform. Based on such understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product, which may be stored in a storage medium, such as a ROM/RAM, a magnetic disk, an optical disk, etc., including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the method described in the embodiments or some parts of the embodiments of the present invention.

In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. In particular, for apparatus or system embodiments, since they are substantially similar to method embodiments, the description is relatively simple, with reference to the description of method embodiments in part. The apparatus embodiments described above are merely illustrative, wherein the elements illustrated as separate elements may or may not be physically separate, and the elements shown as elements may or may not be physical elements, may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.

The present invention is not limited to the above-mentioned embodiments, and any changes or substitutions that can be easily understood by those skilled in the art within the technical scope of the present invention are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. The satellite dynamic positioning method of the vehicle based on the vehicle-road cooperation is characterized by comprising the following steps of:

2. The method of claim 1, wherein the method further comprises:

3. The method according to claim 1 or 2, wherein said step S1 comprises:

S1.2, dividing a target road network into m _L parts and m _B parts according to longitude coordinates and latitude coordinates according to a grid area unit longitude span L _region and a grid area unit latitude span B _regi on, dividing the target road network into m _L×m_B grid areas, and expanding the geographic space information attribute of the target road network into

4. A method according to claim 3, wherein said step S2 comprises:

5. The method according to claim 4, wherein said step S3 comprises:

wherein, the symbol V represents logical judgment OR;

extracting a residual error including a corrected pseudo-range from the observation data sample Is used for the purpose of determining the target amount of (1),

6. The method of claim 5, wherein the parameter names of the observed performance parameter weighting model include: the characteristic variable comprises the following components: elevation angle, azimuth angle, signal-to-noise ratio, raw pseudo-range, carrier phase and Doppler shift, the target variable comprising a pseudo-range residual;

the activation function sigma and the cost function are as follows:

7. The method according to claim 5, wherein said step S4 comprises:

determining a vehicle cluster formed by each cooperative adjacent vehicle in the communication range of the vehicle and the vehicle-vehicle, transmitting satellite positioning observation data of the vehicle and the adjacent vehicles to a positioning information management unit of a grid area to which the vehicle belongs, and calculating a pseudo-range residual prediction value of a visible satellite of the jth vehicle in the vehicle cluster at time k by using a satellite positioning observation weight determining model of the grid area by the positioning information management unit See (4),

Wherein g _θ (x) represents a satellite positioning observed quantity weight model corresponding to the theta grid region, Respectively represents the elevation angle, azimuth angle and signal-to-noise ratio of the kth satellite observed by the j-numbered vehicle at the moment t,

wherein k represents the kth visible satellite, Representing the horizontal accuracy feature value for the time that the first satellite is not in the visible set, PDOP represents the horizontal accuracy feature value when all visible satellite configurations are considered,

wherein h (x) represents a T-S fuzzy reasoning method for weight determination.

8. The method of claim 7, wherein the performing a positioning solution for the vehicle in the target road network based on the observed quantity weight of each visible satellite comprises:

x＝(H^TWH)^-1HWz (9)

9. A satellite dynamic positioning system for a vehicle based on vehicle-road coordination, comprising:

10. The system of claim 9, wherein the system further comprises: