CN115164915A - A multi-source data fusion positioning method applied to driving test vehicles - Google Patents

Abstract

The invention provides a multi-source data fusion positioning method applied to driving test vehicles, belonging to the technical field of intelligent transportation; the technical problem to be solved is: to provide an improvement of the multi-source data fusion positioning method applied to driving test vehicles; Obtain the position information of the test vehicle to be tested and the sensor raw data related to vehicle shaking during the test process, and input the position information and raw sensor data into the fuzzy algorithm model for preprocessing; build the fuzzy algorithm model and train the model: Data fusion of the preprocessed vehicle sensor information, using the basic rules of vehicle shake determination, identify and calculate the time when the vehicle shakes, the driving section, and the initial shaking range; locate the vehicle positioning point to the corresponding digital map road In the driving test vehicle, the distance between the driving test vehicle and the road edge line reference object is automatically identified, and the illegal behavior of the driving test vehicle is determined; the present invention is applied to the violation judgment of the driving test vehicle.

Description

A multi-source data fusion positioning method applied to driving test vehicles

technical field

The invention provides a multi-source data fusion positioning method applied to a driving test vehicle, which belongs to the technical field of intelligent transportation.

Background technique

Driving test vehicle positioning and trajectory monitoring are the prerequisites for driving test behavior analysis. During the driving test vehicle positioning and trajectory monitoring process, test vehicles, especially large vehicles, may shake due to factors such as their service life, body conditions, and driving road conditions. Presenting an unstable driving posture, which leads to deviations in vehicle positioning and affects the accuracy of vehicle trajectory monitoring.

Data fusion is a comprehensive and intelligent information processing technology that combines, processes and analyzes the information collected by sensors. Specifications indicate that it is easy to conduct a more comprehensive and in-depth comprehensive analysis of monitoring information.

Applying China's BeiDou Navigation Satellite System (BDS) to driving test vehicle positioning and combining with map matching technology to achieve real-time trajectory characterization of monitoring vehicles is a reliable method currently used in intelligent driving test systems. The Beidou positioning system fuses the vehicle sensors and vehicle attribute data for information fusion analysis, and combines various related data to correct the vehicle positioning, thereby reducing the positioning error in the vehicle monitoring process and more accurately depicting and displaying the vehicle's driving trajectory. It has very important practical significance for judging the driving attitude of the driving test vehicle and the judgment of violation.

However, in the process of real-time monitoring of driving test vehicles and determination of violations, the positioning technology of Beidou system may be affected by various factors, resulting in different degrees of positioning deviation of vehicles. In order to solve the vehicle positioning error generated under the influence of various factors, reduce the omission of illegal behavior monitoring that may occur during the vehicle shaking process, and realize more accurate trajectory characterization, attitude analysis and violation judgment during the vehicle shaking process, the present invention provides a driving The multi-source data fusion positioning method of the test vehicle. By using Beidou technology combining Kalman filter and real-time differential positioning technology RTK (Real-time kinematic) to achieve centimeter-level positioning of driving test vehicles, and comprehensively considering various factors that affect vehicle shake, the attitude and attitude of driving test vehicles during shaking are analyzed. The details increase the comprehensiveness and accuracy of vehicle driving behavior analysis.

SUMMARY OF THE INVENTION

In order to overcome the deficiencies in the prior art, the technical problem to be solved by the present invention is to provide an improvement of a multi-source data fusion positioning method applied to a driving test vehicle.

In order to solve the above-mentioned technical problems, the technical solution adopted in the present invention is: a multi-source data fusion positioning method applied to a driving test vehicle, comprising the following steps:

S1: Obtain the position information of the test vehicle to be tested during the test process and the sensor raw data related to vehicle shaking, and input the position information and raw sensor data into the fuzzy algorithm model for preprocessing;

S2: construct a fuzzy algorithm model, and train the model: the input of the fuzzy algorithm model is data related to vehicle shaking, and the output is the corrected vehicle positioning information;

The training of the model includes: simulating the degree of shaking of different vehicles under different conditions, collecting sensor data on the vehicle, performing the same-direction gradient analysis on the collected data of a certain frame, and determining the range of the difference between the vehicle's data under different conditions and the normal condition , obtain the corresponding relationship between different degrees of vehicle shaking and vehicle conditions, and input the corresponding relationship into the model in the form of a multi-dimensional array as the basic rule for judging vehicle positioning and shaking;

S3: The information collection node on the vehicle fuses the data of various types of sensor information of the vehicle that has been preprocessed. According to the obtained vehicle data, using the basic rules of vehicle shake judgment, the time, driving section, and shaking of the vehicle are preliminary. Scope identification and calculation;

S4: Use the weight-based map matching algorithm to locate the vehicle positioning point on the corresponding digital map road, put the coordinates of the vehicle marking point on the GIS map to search for matching, and determine whether the monitored vehicle is in the project area and automatically identify it. The distance between the driving test vehicle and the road edge line reference object is used to determine the violation of the driving test vehicle.

In the step S1, the vehicle position information and sensor raw data are obtained through a positioning system, and the positioning system includes a fixed Beidou base station set in the driving test room, a Beidou mobile base station set on the vehicle, and related sensors on the vehicle, wherein the relevant sensors on the vehicle are The vehicle data collected by the sensor includes the data of the vehicle engine state, the vehicle speed, the vehicle acceleration, the vehicle steering angular velocity, and the vehicle distance from the reference object.

In the step 1, the preprocessing of the data through the fuzzy algorithm model includes positioning data processing, vehicle feature data processing and cluster analysis processing.

The positioning data processing steps are as follows:

The pseudo-distance method is used to correct the data information, and the calculation formula is:

σ=t·C=(t _b -t _a )·C;

In the above formula: σ is the pseudo-distance of signal transmission; t _a is the time when the signal satellite sends the signal; t _b is the time when the ground GPS receiver receives the satellite signal; C is the speed of light;

The propagation time is then corrected according to the error between satellite time and ground time:

_ta +V _a =T _a ;

t _b +V _b =T _b ;

In the above formula: V _a is the clock difference of the satellite clock; V _b is the clock difference of the GPS receiver; T _a is the time when the signal satellite sends the signal after correction; T _b is the time when the ground GPS receiver receives the satellite signal after the correction;

The corrected signal propagation time is obtained according to the above formula, and then substituted into the correction formula to obtain:

σ=(t _b −t _a )·C+(V _a −V _b )·C;

In the formula: t _b -t _a is the actual signal propagation time after time error correction; V _a -V _b is the clock error after time error correction;

Perform the impurity removal correction operation again:

σ=(t _b −t _a )·C+(V _a −V _b )·C+I(t)N(t);

In the above formula: I(t) is the delay time caused by ionization refraction when the signal propagates in the atmosphere, N(t) is the delay time caused by convective refraction when the signal propagates in the atmosphere;

According to the above formula, the actual signal propagation distance is obtained by the pseudo-distance method, and the spatial coordinates of the observation point (x', y', z') are substituted to obtain accurate positioning data, and the distance error is corrected:

The processing steps of the vehicle feature data are as follows:

Suppose V = {v ₁ , v ₂ , v ₃ , ..., v _i , ..., v _m } is the set of vehicles, m is the number of vehicles, v _i is the measured distance data of any vehicle, v _i = { p ₁ , p ₂ , p ₃ , ..., p _i , ..., p _s }, where s is the number of data collection points;

The data of the vehicle will contain a variety of feature data at different collection points, _pi is a tuple of all the features of the vehicle at a collection point, defined as _pi = (t, lon, lat, speed, rot, ac, σ, eng , shake), where t, lon, lat, speed, rot, ac, σ, eng, shake represent time, longitude, latitude, vehicle speed, vehicle steering angular velocity, vehicle acceleration, signal propagation distance data, vehicle engine status, vehicle Shake data, where the vehicle steering angular velocity and vehicle acceleration are calculated as:

In the above formula, lon _p , lat _p , t _p , speed _p , rot _p , and ac _p represent the longitude, latitude, timestamp, speed, steering angular velocity and acceleration of the vehicle, respectively.

The steps of the cluster analysis processing are as follows:

Before performing cluster analysis, normalize each vehicle feature data, and fuse the normalized vehicle position feature, vehicle speed feature, vehicle direction feature and vehicle basic information feature into a feature vector ε , and then perform the clustering operation on the feature vector ε, use the Gaussian mixture model algorithm for clustering, and give the vehicle shaking result according to the clustering result.

In the step S3, the preprocessed vehicle sensor information is fused, wherein the vehicle sensor information is collected through a positioning AT330 measurement antenna, a directional AT330 measurement antenna, a control switch sensor and a camera device arranged on the vehicle, and the collection is performed. The information includes vehicle switch control information, vehicle engine status information, vehicle real-time positioning and orientation information, and vehicle driving information. The steps of data fusion are:

Firstly, the spatiotemporal reference of multi-source data is unified, data redundancy items are removed, spatiotemporal labels are added to the multi-source data, and the specification is converted into the final vehicle monitoring data in a unified format, and finally the fusion data is transmitted to the control and monitoring system terminal.

The steps of the weight-based map matching algorithm in step S4 are as follows:

Taking the positioning point (t, lon, lat, speed, rot, ac, σ, eng, shake) as the position of the vehicle on the current road, different weights are given to different influencing factors. During the driving process of the vehicle, the steering angular velocity of the vehicle is used as The important factor that controls the direction of the vehicle has a higher weight than other influencing factors, and the vehicle speed has the same weight as the vehicle acceleration.

Compared with the prior art, the present invention has the following beneficial effects: the present invention mainly aims at the real-time monitoring and violation judgment scenarios of driving test vehicles, and realizes centimeter-level positioning of moving vehicles through BDS technology. Focusing on the shaking of the driving test vehicle during driving, the influence of vehicle shaking on BDS vehicle positioning is explored through the relationship between vehicle-related attributes and vehicle shaking. Using the method of data clustering and grouping to locate the stage where the vehicle shakes, it can achieve a more comprehensive and detailed description and characterization of the driving process of the vehicle.

The invention utilizes the data fusion technology to realize the centralized and unified processing of the data collected by the multi-sensors of the vehicle, and obtains the time-space unified data of the vehicle driving test process data, including the vehicle switch control information, the vehicle engine state information, the vehicle real-time positioning and orientation information, and the Other driving information of the vehicle, etc. The centralized and unified processing of vehicle collection information ensures the unity of sensor information, and also reduces the difficulty of data processing in the terminal control system.

The invention combines the vehicle's own attribute data, engine state data, road condition information, etc., to perform positioning correction and trajectory range definition on the vehicle's positioning trajectory, and explores the influence of different degrees of vehicle jitter on the violation of the body line. The implementation of more frequent detection and analysis has enhanced the accuracy of vehicle violation determination, and has practical guiding significance for the further advancement of intelligent driving test evaluation.

Description of drawings

The present invention will be further described below in conjunction with the accompanying drawings:

Fig. 1 is the system structure schematic diagram of the present invention;

FIG. 2 is a flow chart of the method of the present invention.

Detailed ways

As shown in FIG. 1 to FIG. 2 , in order to solve the positioning error caused by various factors in the monitoring process of driving test vehicles, the present invention provides a multi-source data fusion positioning method applied to driving test vehicles. Based on the data, the positioning error caused by vehicle shaking has been reasonably corrected, which is of great guiding significance for reducing the omission of illegal behaviors that may occur in the process of vehicle shaking, and achieving more accurate trajectory characterization, attitude analysis and violation judgment during the vehicle shaking process. . The centimeter-level positioning of the driving test vehicle is realized by using the Beidou technology combined with Kalman filtering and real-time differential positioning technology, and the posture and details of the driving test vehicle shaking process are analyzed by considering various factors that affect the vehicle shake. The specific technical solutions are as follows:

Step 1: Set up Beidou fixed base stations in the driving test room, and set up Beidou mobile base stations and related sensors for test vehicles and test vehicles. The Beidou fixed base station is used as a reference point, and the position can be adjusted according to actual needs; the Beidou mobile base station is set on the vehicle to be tested, and the real-time positioning of the monitored vehicle is realized through the RTK-based navigation and positioning system, as shown in Figure 1. Vehicle sensing data is collected through a positioning AT330 measurement antenna, a directional AT330 measurement antenna, control switch sensors and camera equipment.

Step 2: Survey and map the whole picture of the test site and draw a GIS electronic map of the site. The main information of the electronic map is the area coordinates and area number of each item in subjects two and three.

Step 3: Carry out a simulation experiment, simulate the shaking of the vehicle to different degrees, and collect its sensor data. Specifically: first collect the vehicle data when the vehicle is in a stationary state, and then collect the rotational speed of the vehicle when the vehicle shakes and the rotational speed when the shake is maximum at the same position, and classify according to the degree of shaking to complete the calibration of the vehicle shaking.

然后计算

方向范围内数据点之间距离差的最大值，其中μ为可调整的自定义参数，并将结果与车辆平稳行驶时采集数据的极差值进行对比，确定不同情况与正常情况数据的差额范围。同时定义和标识车辆相关属性，将车辆不同程度抖动与车辆发动机状态、车辆速度、车辆加速度、车辆转向角速度等与车辆抖动相关的属性采集数据进行对应，得到其对应关系，并将结果以多维数组的形式输入系统，作为后续系统分析车辆定位和抖动行为的基本规则。Perform the same-direction range analysis on the collected data of a certain frame, that is, first obtain the mean value of the direction data

then calculate

The maximum value of the distance difference between data points in the direction range, where μ is an adjustable custom parameter, and the result is compared with the extreme difference value of the data collected when the vehicle is running smoothly to determine the difference range between the data in different situations and the normal situation . At the same time, define and identify vehicle-related attributes, correlate different degrees of vehicle jitter with vehicle engine state, vehicle speed, vehicle acceleration, vehicle steering angular velocity and other attribute collection data related to vehicle jitter, obtain their corresponding relationships, and put the results in a multi-dimensional array The form of the input system is used as the basic rule for the subsequent system to analyze the vehicle positioning and shaking behavior.

When judging the line pressing of the vehicle, it can be judged whether the vehicle violates the regulations according to the division of the body, such as the front and the rear, according to the position of the body where the line is pressed, combined with its shaking situation.

According to the simulation experiment, the basic framework of the fuzzy algorithm model is obtained. The data of vehicle-related attributes is used as the input, and the judged vehicle shaking result is used as the output. In the fuzzy algorithm model, RTK data processing, feature processing, and cluster analysis are used to analyze the input data. to be processed.

Step 4: After obtaining the original sensor data, use RTK-based navigation and positioning technology to obtain its positioning data, and then process the data. The vehicle positioning data is clustered and grouped, and the data is grouped and clustered according to the direction data information in the data message. Include the following steps:

1. RTK data processing, the signal collected by the data acquisition equipment has certain errors, and it is easily interfered by many factors such as time difference, atmospheric refraction, etc. in the process of signal transmission. The data information is corrected by the pseudo-range method, and the time difference or other information errors in the receiving process are calculated according to the distance between the receiver and the signal satellite, so that the final positioning measurement result is more accurate. The operation process of the pseudo distance method:

σ=t·C=(t _b −t _a )·C (1);

In the formula, σ is the pseudo-distance of signal transmission; t _a is the time for the signal satellite to send the signal; t _b is the time required for the ground GPS receiver to receive the satellite signal. Pseudorange is equal to the time difference of signal propagation times the speed of light C.

The propagation time is then corrected according to the error between satellite time and ground time:

_ta +V _a =T _a (2);

t _b +V _b =T _b (3);

According to the above formula, the corrected signal propagation time can be obtained, and then substituted into formula (1) to obtain:

σ=(t _b −t _a )·C+(V _a −V _b )·C (4);

In the formula: V _a -V _b is the clock error after time error correction, t _b -t _a is the actual signal propagation time after time error correction, but the signal measured during this process will still be affected by atmospheric ions and other factors interference, so it is necessary to perform the impurity removal correction operation again:

σ=(t _b −t _a )·C+(V _a −V _b )·C+I(t)N(t) (5);

In the formula, I(t) and N(t) are the delay times caused by ionization refraction and convective refraction respectively when the signal propagates in the atmosphere:

According to the above formula, the actual signal propagation distance is obtained by the pseudo-distance method, and the spatial coordinates of the observation point (x', y', z') are substituted to obtain accurate positioning data, and the distance error is corrected:

Through the above calculation process, the positioning information and distance measurement data with small error and high precision are obtained.

2. Feature processing, assuming that V={v ₁ , v ₂ , v ₃ , ..., v _i , ..., v _m } is the set of vehicles, m is the number of vehicles, and v _i is the measured distance data of any vehicle . v _i ={p ₁ , p ₂ , p ₃ ,..., _pi ,...,ps }, where _s is the number of data collection points. The data of the vehicle will contain a variety of feature data at different collection points, _pi is a tuple of all the features of the vehicle at a collection point, including time, longitude, latitude, vehicle speed, vehicle steering angular velocity, vehicle acceleration, distance data, vehicle Engine status, vehicle shake data. Defined as p _i = (t, lon, lat, speed, rot, ac, σ, eng, shake), and the original vehicle data lacks the vehicle steering angular velocity and vehicle acceleration, it is necessary to calculate the characteristics, and further conduct the vehicle shaking. analyze. Therefore, the angular velocity and acceleration are calculated using formula (7)-formula (9).

Among them, lon _p , lat _p , t _p , speed _p , rot _p , and ac _p represent the longitude, latitude, time stamp, speed, steering angular velocity and acceleration of the vehicle, respectively, and the time stamp information involved needs to be numerically converted for calculate.

Due to the large differences in the values of the collected data, in order to eliminate the dimensional influence between the data indicators, data normalization processing is required. The min-max normalization is used so that the results can fall in the [0,1] interval.

Among them, max and min represent the maximum and minimum values in a set of vehicle data, respectively.

3. Cluster analysis, the data after normalization is processed by using the vehicle position feature, vehicle speed feature, vehicle direction feature and vehicle basic information feature extracted from the encoder, and the extracted features are fused to form feature vector ε, and then perform a clustering operation on the feature vector ε. The Gaussian mixture model (GMM) algorithm is used for clustering. The Gaussian mixture model refers to the linear combination of multiple Gaussian distribution functions. There are N categories, and each category corresponds to a data point. At this time, any given data can be perfectly fitted. Through the clustering results, the vehicle shaking results are given.

Step 5: Set up information collection nodes to process all kinds of sensor information of vehicles that have been preprocessed, so that each sensor data is concentrated on a data mainboard, which is convenient for unified reception and processing with the control system. Including vehicle switch control information, vehicle engine status information, vehicle real-time positioning and orientation information, and other vehicle driving information. To fuse the above data, first unify the spatiotemporal reference of multi-source data, remove redundant data items, add spatio-temporal labels to the multi-source data, standardize and convert it into final vehicle monitoring data in a unified format, and finally transmit the fused data to the control and monitoring system. Monitoring system terminal.

Step 6: Vehicle shake identification and scope definition. According to the obtained vehicle data, using the basic rules for vehicle shake determination obtained from the vehicle shake simulation experiment performed in advance, the time, driving section, and approximate range of the vehicle shake are identified and calculated.

Step 7: The map matching technology is to correct the position information received by the vehicle positioning information system to eliminate the influence of various errors, so that the vehicle positioning point is closer to the real map positioning point, and the positioning point is located to the corresponding location. In the road of digital map, it provides necessary support for vehicle monitoring and analysis.

The specific algorithm used is a weight-based map matching algorithm, and the positioning point (t, lon, lat, speed, rot, ac, σ, eng, shake) is used as the position of the vehicle on the current road, and the vehicle in the vehicle is comprehensively considered. speed, vehicle steering angular velocity and other factors. Since different influence factors have different effects on the matching map, they are given different weights. In the process of vehicle driving, the vehicle steering angular velocity (rot), as an important factor to control the direction of the vehicle, should be more influential than other weights. The vehicle speed (speed) weight is similar to the vehicle acceleration (ac) weight, and the same weight is assigned to it.

W= _Wrot + _Wspe + _Wac (11);

Among them, W _rot , W _spe , and W _ac are the weights of vehicle steering angular velocity, vehicle speed, and vehicle acceleration, respectively. The total weight W determines the map segments matched by the positioning data.

Step 8: Vehicle violation determination and driving behavior analysis. By placing the coordinates of the vehicle marking points in the GIS map to search and match, it is possible to determine whether the monitored vehicle is in the project area, and at the same time, it can also automatically identify the distance between the vehicle and the road edge line and other reference objects, so as to realize the pressure line between the vehicle wheel and the body. Or the determination of violations such as the wrong vehicle driving route. The control and monitoring system can observe and analyze the monitored vehicle in real time, visually display and record the driving dynamics of the vehicle, and provide data basis for the analysis of driving behavior of driving test candidates.

The invention adopts the Beidou positioning system independently developed by my country to realize the precise positioning of the driving test vehicle, and uses the electronic map matching technology to realize the vehicle driving trajectory description, and realizes a more intuitive and intelligent whole-process monitoring of the driving test vehicle. The internal relationship between factors such as vehicle service life, body condition, driving road condition and vehicle positioning is explored, and the vehicle positioning is integrated with the analysis of the vehicle shaking process, which further enhances the accuracy of the monitored vehicle line pressure detection during the driving test process. . The information collection node of the present invention can process different sensor signals, including vehicle switch control information, vehicle engine status information, vehicle real-time positioning and orientation information, and other vehicle driving information. The centralized and unified processing of vehicle information ensures the unity of sensor information and reduces the difficulty of data processing in the terminal control system.

Regarding the specific structure of the present invention, it should be noted that the connection relationship between the various component modules adopted in the present invention is determined and achievable. Technical effect, and based on the premise of not relying on the execution of the corresponding software program, to solve the technical problem proposed by the present invention, the model of the components, modules, specific components, and the connection method between the components, modules, and components appearing in the present invention, and the above-mentioned technical features bring about The conventional methods of use and predictable technical effects, unless specifically stated, belong to the disclosed content of patents, journal papers, technical manuals, technical dictionaries, and textbooks that can be obtained by those skilled in the art before the application date, or belong to the field. Existing technologies such as conventional technology and common knowledge need not be repeated, so that the technical solution provided in this case is clear, complete and achievable, and the corresponding physical product can be reproduced or obtained according to the technical means.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: The technical solutions described in the foregoing embodiments can still be modified, or some or all of the technical features thereof can be equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the technical solutions of the embodiments of the present invention. scope.

Claims (8)

1. a multi-source data fusion positioning method applied to a driving test vehicle, is characterized in that: comprise the steps: S1: Obtain the position information of the test vehicle to be tested during the test process and the sensor raw data related to vehicle shaking, and input the position information and raw sensor data into the fuzzy algorithm model for preprocessing; S2: construct a fuzzy algorithm model, and train the model: the input of the fuzzy algorithm model is data related to vehicle shaking, and the output is the corrected vehicle positioning information; The training of the model includes: simulating the degree of shaking of different vehicles under different conditions, collecting sensor data on the vehicle, performing the same-direction gradient analysis on the collected data of a certain frame, and determining the range of the difference between the vehicle's data under different conditions and the normal condition , obtain the corresponding relationship between different degrees of vehicle shaking and vehicle conditions, and input the corresponding relationship into the model in the form of a multi-dimensional array as the basic rule for judging vehicle positioning and shaking; S3: The information collection node on the vehicle fuses the data of various types of sensor information of the vehicle that has been preprocessed. According to the obtained vehicle data, using the basic rules of vehicle shake judgment, the time, driving section, and shaking of the vehicle are preliminary. Scope identification and calculation; S4: Use the weight-based map matching algorithm to locate the vehicle positioning point on the corresponding digital map road, put the coordinates of the vehicle marking point on the GIS map to search for matching, and determine whether the monitored vehicle is in the project area and automatically identify it. The distance between the driving test vehicle and the road edge line reference object is used to determine the violation of the driving test vehicle. 2. A multi-source data fusion positioning method applied to a driving test vehicle according to claim 1, characterized in that: in the step S1, vehicle position information and sensor raw data are obtained through a positioning system, and the positioning system comprises: The fixed Beidou base station set in the driving test room, the Beidou mobile base station set on the vehicle, and the related sensors on the vehicle, where the vehicle data collected by the relevant sensors on the vehicle include the vehicle engine state, vehicle speed, vehicle acceleration, vehicle steering angular velocity, and vehicle distance. data of the reference object. 3. a kind of multi-source data fusion positioning method applied to driving test vehicle according to claim 2, is characterized in that: in described step 1, preprocessing data by fuzzy algorithm model comprises positioning data processing, vehicle characteristic data processing and cluster analysis processing. 4. a kind of multi-source data fusion positioning method applied to driving test vehicle according to claim 3, is characterized in that: described positioning data processing step is as follows: The pseudo-distance method is used to correct the data information, and the calculation formula is: σ=t·C=(t _b -t _a )·C; In the above formula: σ is the pseudo-distance of signal transmission; t _a is the time when the signal satellite sends the signal; t _b is the time when the ground GPS receiver receives the satellite signal; C is the speed of light; The propagation time is then corrected according to the error between satellite time and ground time: _ta +V _a =T _a ; t _b +V _b =T _b ; In the above formula: V _a is the clock difference of the satellite clock; V _b is the clock difference of the GPS receiver; T _a is the time when the signal satellite sends the signal after correction; T _b is the time when the ground GPS receiver receives the satellite signal after the correction; The corrected signal propagation time is obtained according to the above formula, and then substituted into the correction formula to obtain: σ=(t _b −t _a )·C+(V _a −V _b )·C; In the formula: t _b -t _a is the actual signal propagation time after time error correction; V _a -V _b is the clock error after time error correction; perform impurity removal correction operation again: σ=(t _b −t _a )·C+(V _a −V _b )·C+I(t)N(t); In the above formula: I(t) is the delay time caused by ionization refraction when the signal propagates in the atmosphere, N(t) is the delay time caused by convective refraction when the signal propagates in the atmosphere; According to the above formula, the actual signal propagation distance is obtained by the pseudo-distance method, and the spatial coordinates of the observation point (x', y', z') are substituted to obtain accurate positioning data, and the distance error is corrected:

5. a kind of multi-source data fusion positioning method applied to driving test vehicle according to claim 4, is characterized in that: described vehicle characteristic data processing step is as follows: Suppose V = {v ₁ , v ₂ , v ₃ , ..., v _i , ..., v _m } is the set of vehicles, m is the number of vehicles, v _i is the measured distance data of any vehicle, v _i = { p ₁ , p ₂ , p ₃ , ..., p _i , ..., p _s }, where s is the number of data collection points; The data of the vehicle will contain a variety of feature data at different collection points, _pi is a tuple of all the features of the vehicle at a collection point, defined as _pi = (t, lon, lat, speed, rot, ac, σ, eng , shake), where t, lon, lat, speed, rot, ac, σ, eng, shake represent time, longitude, latitude, vehicle speed, vehicle steering angular velocity, vehicle acceleration, signal propagation distance data, vehicle engine status, vehicle Shake data, where the vehicle steering angular velocity and vehicle acceleration are calculated as:

In the above formula, lon _p , lat _p , t _p , speed _p , rot _p , and ac _p represent the longitude, latitude, timestamp, speed, steering angular velocity and acceleration of the vehicle, respectively. 6. a kind of multi-source data fusion positioning method applied to driving test vehicle according to claim 5, is characterized in that: the step of described cluster analysis processing is as follows: Before performing cluster analysis, normalize each vehicle feature data, and fuse the normalized vehicle position feature, vehicle speed feature, vehicle direction feature and vehicle basic information feature into a feature vector ε , and then perform the clustering operation on the feature vector ε, use the Gaussian mixture model algorithm for clustering, and give the vehicle shaking result according to the clustering result. 7. A multi-source data fusion positioning method applied to a driving test vehicle according to claim 2, wherein in the step S3, the preprocessed information of each sensor of the vehicle is fused, wherein the information of each sensor of the vehicle is fused. Collected by a positioning AT330 measuring antenna, a directional AT330 measuring antenna, control switch sensor and camera device set on the vehicle, the collected information includes vehicle switch control information, vehicle engine status information, vehicle real-time positioning and orientation information and vehicle The steps of driving information and data fusion are: Firstly, the spatiotemporal reference of multi-source data is unified, data redundancy items are removed, spatiotemporal labels are added to the multi-source data, and the specification is converted into the final vehicle monitoring data in a unified format, and finally the fusion data is transmitted to the control and monitoring system terminal. 8. a kind of multi-source data fusion positioning method applied to driving test vehicle according to claim 2, is characterized in that: in described step S4, the map matching algorithm steps based on weight are as follows: Taking the positioning point (t, lon, lat, speed, rot, ac, σ, eng, shake) as the position of the vehicle on the current road, different weights are given to different influencing factors. During the driving process of the vehicle, the steering angular velocity of the vehicle is used as The important factor that controls the direction of the vehicle has a higher weight than other influencing factors, and the vehicle speed has the same weight as the vehicle acceleration.

Images