CN114065497A - Viaduct section identification method based on multi-sensor data fusion - Google Patents

Viaduct section identification method based on multi-sensor data fusion Download PDF

Info

Publication number
CN114065497A
CN114065497A CN202111326382.3A CN202111326382A CN114065497A CN 114065497 A CN114065497 A CN 114065497A CN 202111326382 A CN202111326382 A CN 202111326382A CN 114065497 A CN114065497 A CN 114065497A
Authority
CN
China
Prior art keywords
vehicle
viaduct
classified
threshold value
current position
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111326382.3A
Other languages
Chinese (zh)
Inventor
史秀纺
甘帅
张文安
刘浩淼
沈林强
付明磊
杨旭升
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Zhejiang University of Technology ZJUT
Original Assignee
Zhejiang University of Technology ZJUT
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Zhejiang University of Technology ZJUT filed Critical Zhejiang University of Technology ZJUT
Priority to CN202111326382.3A priority Critical patent/CN114065497A/en
Publication of CN114065497A publication Critical patent/CN114065497A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
    • G06F2119/02—Reliability analysis or reliability optimisation; Failure analysis, e.g. worst case scenario performance, failure mode and effects analysis [FMEA]

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Physics (AREA)
  • Computational Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • General Engineering & Computer Science (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Optimization (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Position Fixing By Use Of Radio Waves (AREA)
  • Software Systems (AREA)
  • Computer Hardware Design (AREA)
  • Operations Research (AREA)
  • Probability & Statistics with Applications (AREA)
  • Geometry (AREA)
  • Algebra (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Evolutionary Computation (AREA)
  • Navigation (AREA)
  • Evolutionary Biology (AREA)

Abstract

A viaduct section identification method based on multi-sensor data fusion comprises the following steps: (1) fixing the smart phone in a running vehicle; (2) collecting data generated by a current position sensor, wherein the data comprises linear acceleration information and GNSS information of a current vehicle; (3) calculating the variance of the number of the received satellites at the current moment relative to the number of the previous N positions to serve as a characteristic for distinguishing whether the vehicle is on the viaduct or not; (4) the linear acceleration received in the step (2) is compared with a threshold value ls1And a threshold value ls2Performing comparison and classification; the speed in the GNSS information received in the step (2) is compared with a threshold upsilonsPerforming comparison and classification; the variance obtained in the step (3) is compared with a threshold value dsPerforming comparison and classification; (5) and inputting the classified data into the constructed decision tree model for recognition, and outputting the current position of the vehicle.When the invention is used, the vehicle can be effectively distinguished from being positioned on the viaduct or under the viaduct.

Description

Viaduct section identification method based on multi-sensor data fusion
Technical Field
The invention relates to the field of navigation positioning, in particular to a viaduct section identification method based on multi-sensor data fusion.
Background
Global Navigation Satellite Systems (GNSS) play an important role in the fields of vehicle navigation, intelligent traffic systems, and surveying and mapping. Through the GNSS, the navigation system can provide a planned driving path and real-time navigation for a driver, and great convenience is brought to people.
However, the electromagnetic waves used by the navigation satellite are transmitted over long distances and are affected by troposphere, ionosphere, and high buildings and viaducts in urban areas, so that the electromagnetic waves may generate errors of several tens of meters or even several hundreds of meters. Since the distribution of the navigation satellite in the elevation direction is limited, the vertical accuracy of the navigation satellite is more error than the horizontal accuracy, and particularly, the vertical accuracy of the navigation satellite is further limited in such an area under the viaduct where the overhead shielding is severe.
With the rapid development of cities, a large number of viaducts are built in order to relieve traffic pressure of citizens during traveling. However, when the GNSS navigation is used to navigate through the viaduct area, sometimes it is impossible to distinguish whether the vehicle is located on the viaduct or under the viaduct, and certain threats may be brought to the trip and safety of the driver.
Disclosure of Invention
In order to solve the problems in the prior art, the invention provides a viaduct section identification method based on multi-sensor data fusion.
Based on the fact that the road conditions of the viaduct and the ground road are different, the vibration and the speed of the vehicle and the stability of the number of received satellites are different, and the three characteristics are analyzed to determine whether the vehicle runs on the viaduct or under the viaduct.
The invention uses a mobile phone acceleration sensor and a mobile phone GNSS receiver; because the road conditions of the elevated road and the ground road are different, the speed, the vehicle vibration and the number of receivable satellites of the vehicle when the vehicle runs on the two road surfaces are different, and the vehicle is predicted whether to run on the elevated road or not based on the three characteristics; the method mainly comprises the steps of collecting data of a mobile phone sensor placed in a vehicle in real time, then extracting features, and inputting into a decision tree model for calculation so as to predict whether the vehicle runs on the viaduct or not.
The invention is realized by the following technical scheme: a viaduct section identification method based on multi-sensor data fusion comprises the following steps:
(1) the smart phone is fixed in a running vehicle, so that the smart phone is ensured not to slide along with the bumping of the vehicle;
(2) collecting data generated by a mobile phone sensor at the current moment, wherein the data comprises the linear acceleration information and GNSS information of a current vehicle; the GNSS information comprises vehicle speed information and currently received satellite number information;
(3) calculating the variance of the number of the received satellites at the current moment relative to the number of the previous N positions to serve as a characteristic for distinguishing whether the vehicle is on the viaduct or not;
(4) comparing and classifying the linear acceleration received in the step (2) with a threshold ls1 and a threshold ls 2; comparing and classifying the speed in the GNSS received in the step (2) with a threshold value vs; comparing and classifying the variance obtained in the step (3) with a threshold value ds;
(5) and inputting the classified data into the constructed decision tree model for recognition, and outputting the current position of the vehicle.
Further, the step (3) of calculating the variance of the number of received satellites at the current time with respect to the previous N positions is as follows:
(3.1) the sequence of the number of satellites initially received to the number of satellites received at the current position obtained by the sensor is [ o ]1，o2，o3，…，ot]；
(3.2) if the current position t is less than N, calculating the current position variance:
Figure BDA0003347318310000031
wherein the average of the number of satellites received from the initial position to the current position
Figure BDA0003347318310000032
(3.3) if the current position t is more than or equal to N, calculating the current position variance:
Figure BDA0003347318310000033
where the average of the current position to the first N positions
Figure BDA0003347318310000034
Further, the step (4) of comparing the sensor data with the corresponding threshold value for classification is as follows:
(4.1) acquiring the three data obtained by different brands of smart phones in the vehicle in different overhead areas and under different traffic flows in advance; analyzing and comparing the same data, and selecting a numerical value capable of effectively distinguishing the vehicles running on the elevated roads and the ground roads as a threshold value;
(4.2) Linear acceleration two thresholds l are selecteds1And ls2Of which ls1＞ls2The specific value is not limited herein; the linear acceleration a obtained currently is subjected to the two threshold valuesiIs classified intos1≥ai≥ls2They are classified into one class; if ai＞ls1or ai＜ls2They are classified as another class;
(4.3) speed selection of a threshold value vsThe specific value is not limited herein; by said threshold value for the current speed viIs classified, if vi＞vsIt is classified asOne type; if v isi≤vsThey are classified into one class;
(4.4) selecting a threshold d for the variance of the number of satellites receivedsThe specific value is not limited herein; variance delta of the number of current receiving satellites through the threshold valuetIs classified intot＞dsThey are classified into one class; if deltat≤dsThey are classified into one group.
Further, the step (5) inputs the classified data into the constructed decision tree model for recognition, and the step of constructing the decision tree model is as follows: and (4) collecting the three data obtained by different brands of smart phones in the vehicle under the conditions of different elevated areas and different vehicle flows in advance according to the step (4.1) in the step (4), classifying the collected data by using the decision tree algorithm after the other steps in the step (4), and then training by using the decision tree algorithm to obtain the viaduct identification decision tree model.
Compared with the prior art, the viaduct section identification method based on multi-sensor data fusion has the following advantages that:
1. in the prior art, the vertical position of the vehicle is detected directly through elevation data obtained by a GNSS, and whether the vehicle is positioned on the viaduct cannot be accurately determined due to low vertical precision of the GNSS. Because the road conditions of the viaduct and the ground road are different, the vibration and the running speed of the vehicle have different characteristics, and the quantity of the received satellites is different due to the shielding of the viaduct. The invention uses the relevant characteristics to evaluate the vertical position of the vehicle, and effectively avoids the error brought by the vertical precision of the GNSS.
2. In the prior art, equipment such as a barometer and the like is used for calculating the real-time height of a vehicle so as to determine whether the vehicle is positioned on a viaduct, only part of high-end mobile phones are provided with the barometer at present, and the barometer is relatively expensive in price. In the case of a vehicle window opening, the barometer reading can be subject to significant errors as the vehicle speed changes. According to the invention, the acceleration sensor and the GNSS receiving sensor which are assembled on the smart phone are used instead of the barometer, so that whether the vehicle is positioned on the viaduct can be detected in a low-cost manner.
Drawings
Fig. 1 is a flowchart of a method for identifying overpass sections based on multi-sensor data fusion according to the present invention.
Fig. 2 is a state diagram of the actual position of the vehicle traveling on the overpass.
FIG. 3 is a state diagram of vehicle position predicted using the present invention.
Fig. 4 is a diagram showing a state of a real position of a vehicle traveling on a road under an overpass.
FIG. 5 is a state diagram of vehicle position predicted using the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
referring to fig. 1 to 5, a viaduct bridge identification method based on multi-sensor data fusion includes the following steps:
(1) the smart phone is fixed in a running vehicle, so that the smart phone is ensured not to slide along with the bumping of the vehicle;
(2) collecting data generated by a current position sensor, wherein the data comprises linear acceleration information and GNSS information of a current vehicle; the GNSS information comprises vehicle speed information and currently received satellite number information;
(3) calculating the variance of the number of the received satellites at the current moment relative to the number of the previous N positions to serve as a characteristic for distinguishing whether the vehicle is on the viaduct or not;
in the step (3), the step of calculating the variance of the number of the received satellites at the current time with respect to the previous N positions is as follows:
(3.1) the sequence of the number of satellites initially received to the number of satellites received at the current position obtained by the sensor is [ o ]1，o2，o3，…，ot]；
(3.2) if the current position t is less than N, calculating the current position variance:
Figure BDA0003347318310000061
wherein the average of the number of satellites received from the initial position to the current position
Figure BDA0003347318310000062
(3.3) if the current position t is more than or equal to N, calculating the current position variance:
Figure BDA0003347318310000063
where the average of the current position to the first N positions
Figure BDA0003347318310000064
(4) The linear acceleration received in the step (2) is compared with a threshold value ls1And a threshold value ls2Performing comparison and classification; the speed in the GNSS received in the step (2) is compared with a threshold value vsPerforming comparison and classification; the variance obtained in the step (3) is compared with a threshold value dsPerforming comparison and classification;
in the step (4), the step of comparing the sensor data with the corresponding threshold value to classify the sensor data is as follows:
(4.1) acquiring the three data obtained by different brands of smart phones in the vehicle in different overhead areas and under different traffic flows in advance; analyzing and comparing the same data, and selecting a numerical value capable of effectively distinguishing the vehicles running on the elevated roads and the ground roads as a threshold value;
(4.2) Linear acceleration two thresholds l are selecteds1And ls2Of which ls1＞ls2The specific value is not limited herein; the linear acceleration a obtained currently is subjected to the two threshold valuesiIs classified intos1≥ai≥ls2They are classified into one class; if ai＞ls1or ai＜ls2They are classified as another class;
(4.3) speed selectionA threshold value vsThe specific value is not limited herein; by said threshold value for the current speed viIs classified, if vi＞vsThey are classified into one class; if v isi≤vsThey are classified into one class;
(4.4) selecting a threshold d for the variance of the number of satellites receivedsThe specific value is not limited herein; variance delta of the number of current receiving satellites through the threshold valuetIs classified intot＞dsThey are classified into one class; if deltat≤dsThey are classified into one class;
(5) and inputting the classified data into the constructed decision tree model for recognition, and outputting the current position of the vehicle.
The decision tree model in the step (5) is constructed by the following steps:
and (4) according to the three data acquired by the smart phones of different brands in the vehicle under the conditions of different elevated areas and different vehicle flows collected in advance in the step (4.1), classifying the collected data by using the steps (4.1) to (4.4), and then training by using a decision tree algorithm to obtain an elevated bridge recognition decision tree model.
Examples
Part of the data collected using the river overhead in Hangzhou city was verified using the present invention, and the usage data set included the linear acceleration, velocity and number of satellites received data required by the present invention.
Fig. 2 is a schematic view of a vehicle traveling on an overpass, where 1 denotes the vehicle traveling on the overpass, and 0 denotes the vehicle traveling on a ground road. FIG. 3 shows a prediction of vehicle position using the present invention with an accuracy of 88.46%.
Fig. 4 is a schematic view of a road on which a vehicle is traveling under an overhead bridge, and fig. 5 is a schematic view of a prediction of the vehicle position using the present invention, calculated with 90% accuracy.

Claims (4)

1. A viaduct section identification method based on multi-sensor data fusion is characterized by comprising the following steps:
(1) the smart phone is fixed in a running vehicle, so that the smart phone is ensured not to slide along with the bumping of the vehicle;
(2) collecting data generated by a current position sensor, wherein the data comprises linear acceleration information and GNSS information of a current vehicle; the GNSS information comprises vehicle speed information and currently received satellite number information;
(3) calculating the variance of the number of the received satellites at the current moment relative to the number of the previous N positions to serve as a characteristic for distinguishing whether the vehicle is on the viaduct or not;
(4) the linear acceleration received in the step (2) is compared with a threshold value ls1And a threshold value ls2Performing comparison and classification; the speed in the GNSS received in the step (2) is compared with a threshold value vsPerforming comparison and classification; the variance obtained in the step (3) is compared with a threshold value dsPerforming comparison and classification;
(5) and inputting the classified data into the constructed decision tree model for recognition, and outputting the current position of the vehicle.
2. The viaduct section identification method based on multi-sensor data fusion as claimed in claim 1, wherein in the step (3), detailed steps are as follows:
(3.1) the sequence of the number of satellites initially received to the number of satellites received at the current position obtained by the sensor is [ o ]1，o2，o3，…，ot]；
(3.2) if the current position t is less than N, calculating the current position variance:
Figure FDA0003347318300000011
wherein the average of the number of satellites received from the initial position to the current position
Figure FDA0003347318300000021
(3.3) if the current position t is more than or equal to N, calculating the current position variance:
Figure FDA0003347318300000022
where the average of the current position to the first N positions
Figure FDA0003347318300000023
3. The viaduct section identification method based on multi-sensor data fusion as claimed in claim 1, wherein in the step (4), the detailed steps are as follows:
(4.1) acquiring the three data obtained by different brands of smart phones in the vehicle in different overhead areas and under different traffic flows in advance; analyzing and comparing the same data, and selecting a numerical value capable of effectively distinguishing the vehicles running on the elevated roads and the ground roads as a threshold value;
(4.2) Linear acceleration two thresholds l are selecteds1And ls2Of which ls1＞ls2The specific value is not limited herein; the linear acceleration a obtained currently is subjected to the two threshold valuesiIs classified intos1≥ai≥ls2They are classified into one class; if ai＞ls1 or ai＜ls2They are classified as another class;
(4.3) speed selection of a threshold value vsThe specific value is not limited herein; by said threshold value for the current speed viIs classified, if vi＞vsThey are classified into one class; if v isi≤vsThey are classified into one class;
(4.4) selecting a threshold d for the variance of the number of satellites receivedsThe specific value is not limited herein; variance delta of the number of current receiving satellites through the threshold valuetIs classified intot＞dsThey are classified into one class; if deltat≤dsTo make it fall intoAre a class.
4. The viaduct identification method based on multi-sensor data fusion according to claim 3, characterized in that the three data obtained by different brands of smart phones in the vehicle under different viaduct areas and different vehicle flows are collected in advance according to the step (4.1), and the collected data are classified by the steps (4.2) to (4.4) and then trained by using a decision tree algorithm to obtain a viaduct identification decision tree model.
CN202111326382.3A 2021-11-10 2021-11-10 Viaduct section identification method based on multi-sensor data fusion Pending CN114065497A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111326382.3A CN114065497A (en) 2021-11-10 2021-11-10 Viaduct section identification method based on multi-sensor data fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111326382.3A CN114065497A (en) 2021-11-10 2021-11-10 Viaduct section identification method based on multi-sensor data fusion

Publications (1)

Publication Number Publication Date
CN114065497A true CN114065497A (en) 2022-02-18

Family

ID=80274586

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111326382.3A Pending CN114065497A (en) 2021-11-10 2021-11-10 Viaduct section identification method based on multi-sensor data fusion

Country Status (1)

Country Link
CN (1) CN114065497A (en)

Similar Documents

Publication Publication Date Title
CN107103775B (en) Road quality detection method based on crowd-sourcing calculation
CN102208013B (en) Landscape coupling reference data generation system and position measuring system
CN102997926B (en) A kind of method for obtaining navigation data
JP2012520491A (en) How to update a digital map with altitude information
CN111275969B (en) Vehicle track filling method based on intelligent identification of road environment
EP2633270B1 (en) Method for detecting grade separated crossings and underpasses
CN101958043A (en) Highway section calculation element hourage and method thereof
CN104408915A (en) Traffic state parameter estimation method and system
CN104395944A (en) Carriageway recognition
CN104900057A (en) City expressway main and auxiliary road floating vehicle map matching method
Byon et al. Travel time collection and traffic monitoring via GPS technologies
CN110018503B (en) Vehicle positioning method and positioning system
Rogers Creating and evaluating highly accurate maps with probe vehicles
CN109979194B (en) Heavy vehicle ramp crawling behavior identification method based on Markov chain
CN114511080A (en) Model construction method and device and abnormal track point real-time detection method
JP7091620B2 (en) Computer program, road judgment method, road judgment device and in-vehicle device
US11238291B2 (en) Method, apparatus, and computer program product for determining if probe data points have been map-matched
Li et al. Driving performances assessment based on speed variation using dedicated route truck GPS data
Vasantha Kumar et al. Urban arterial travel time estimation using buses as probes
Gim et al. Landmark attribute analysis for a high-precision landmark-based local positioning system
CN114065497A (en) Viaduct section identification method based on multi-sensor data fusion
WO2018139258A1 (en) Information processing device and method, program, and information processing system
Pribe et al. Learning to associate observed driver behavior with traffic controls
CN105741554A (en) Traffic flow congestion determining method and traffic flow congestion determining device based on mobile phone motion sensor data
CN112053571A (en) Expressway vehicle trajectory tracking method and system

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination