CN114541222B - Road network grade pavement flatness detection method based on multi-vehicle crowd funding vibration data - Google Patents
Road network grade pavement flatness detection method based on multi-vehicle crowd funding vibration data Download PDFInfo
- Publication number
- CN114541222B CN114541222B CN202210144895.0A CN202210144895A CN114541222B CN 114541222 B CN114541222 B CN 114541222B CN 202210144895 A CN202210144895 A CN 202210144895A CN 114541222 B CN114541222 B CN 114541222B
- Authority
- CN
- China
- Prior art keywords
- road
- flatness
- vehicle
- vehicles
- vibration data
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 33
- 238000012360 testing method Methods 0.000 claims abstract description 48
- 238000004364 calculation method Methods 0.000 claims abstract description 28
- 238000007781 pre-processing Methods 0.000 claims abstract description 3
- 238000000034 method Methods 0.000 claims description 27
- 230000003595 spectral effect Effects 0.000 claims description 17
- 238000009826 distribution Methods 0.000 claims description 12
- 238000005520 cutting process Methods 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 6
- 239000000725 suspension Substances 0.000 claims description 6
- 230000001133 acceleration Effects 0.000 claims description 4
- 238000005070 sampling Methods 0.000 claims description 4
- 238000012216 screening Methods 0.000 claims description 3
- 230000008901 benefit Effects 0.000 abstract description 3
- 238000005259 measurement Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 230000002159 abnormal effect Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 238000005065 mining Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000006757 chemical reactions by type Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 230000009467 reduction Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000002452 interceptive effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000000691 measurement method Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000000047 product Substances 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000004092 self-diagnosis Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 239000013589 supplement Substances 0.000 description 1
Classifications
-
- E—FIXED CONSTRUCTIONS
- E01—CONSTRUCTION OF ROADS, RAILWAYS, OR BRIDGES
- E01C—CONSTRUCTION OF, OR SURFACES FOR, ROADS, SPORTS GROUNDS, OR THE LIKE; MACHINES OR AUXILIARY TOOLS FOR CONSTRUCTION OR REPAIR
- E01C23/00—Auxiliary devices or arrangements for constructing, repairing, reconditioning, or taking-up road or like surfaces
- E01C23/01—Devices or auxiliary means for setting-out or checking the configuration of new surfacing, e.g. templates, screed or reference line supports; Applications of apparatus for measuring, indicating, or recording the surface configuration of existing surfacing, e.g. profilographs
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Architecture (AREA)
- Civil Engineering (AREA)
- Structural Engineering (AREA)
- Measurement Of Mechanical Vibrations Or Ultrasonic Waves (AREA)
Abstract
The invention relates to a road network grade pavement evenness detection method based on multi-vehicle crowd funding vibration data, which comprises the following steps: obtaining vibration data of a test vehicle and preprocessing the vibration data; taking a road section with the flatness index gradient larger than a preset threshold value in the road network to be detected as a known flatness road section, and acquiring the flatness index of the known flatness road section; acquiring a running track of a test vehicle, acquiring a vehicle number threshold according to the running track of the test vehicle and the information of the known flatness road sections, extracting vehicles passing through the known flatness road sections for times greater than or equal to the vehicle number threshold as calculation vehicles of the iteration, acquiring vibration data of the calculation vehicles and estimating vehicle parameters of the calculation vehicles; calculating a flatness index of the unknown flatness road section based on the estimated vehicle parameters; and repeating the iteration for a plurality of times until the flatness index of all road sections in the road network to be detected is obtained. Compared with the prior art, the invention has the advantages of high accuracy, good stability, low cost and the like.
Description
Technical Field
The invention relates to the field of road surface quality detection, in particular to a road network grade road surface flatness detection method based on multi-vehicle crowd funding vibration data.
Background
Road surface flatness is an important index reflecting road service performance and driving comfort. The traditional flatness detection method based on the laser section can ensure higher measurement precision, but is high in price, limited in coverage range and needs specialized reconstruction of vehicles. With the development of sensing technology, mobile terminals represented by smart phones become an important data acquisition path. However, flatness measurement based on vibration of a bicycle mobile phone is not essentially different from a testing mechanism of a traditional reaction method, complex parameter calibration is still required to be carried out on a special testing vehicle, so that the measurement efficiency is limited, the result discreteness is high, and popularization and application of the method are limited.
The measuring method of the road surface flatness is mainly divided into three types: subjective assessment, section and reaction. The subjective evaluation method adopts an expert scoring mode, has stronger subjectivity and is only used as a road section evaluation reference. The section class method is to measure the elevation change of the road surface under the running track of the vehicle, usually, a vehicle-mounted section instrument, a three-dimensional LiDAR and other precise sensing instruments are deployed on a special detection vehicle, and the international flatness index is calculated according to the road surface section. The section type method has higher measurement precision, but the equipment has high production and application cost and limits the measurement conditions, is mainly suitable for measuring the performance of highways such as expressways, national provinces and main roads, is only used as road evaluation sampling detection in urban roads, and is not suitable for the flatness inspection of large-scale and high-frequency urban roads. The reaction type method is to measure the flatness of the road surface such as a bump accumulator, a BPR meter and the like by measuring the vibration reaction of the vehicle. In addition, obtaining vehicle vibrations through accelerometers, smartphones, and the like is also a reactive class of methods. The reaction type method has low cost and high efficiency, can be used as an important supplement of the section type method, is sensitive to abrupt change values, has high data processing difficulty, is easily influenced by the parameters of the vehicle, the running environment and the running state, and has lower result stability.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a road network grade pavement evenness detection method based on multi-vehicle crowd funding vibration data.
The aim of the invention can be achieved by the following technical scheme:
a road network grade pavement evenness detection method based on multi-vehicle crowd funding vibration data comprises the following steps:
s1: obtaining vibration data of a test vehicle and preprocessing the vibration data;
s2: taking a road section with the flatness index gradient larger than a preset threshold value in the road network to be detected as a known flatness road section, and acquiring the flatness index of the known flatness road section;
s3: acquiring the running track of the test vehicle, acquiring a vehicle number threshold according to the running track of the test vehicle and the known flatness road section information,
s4: extracting vehicles passing through the road sections with known flatness for times greater than or equal to the threshold value of the number of vehicles as calculation vehicles of the iteration, acquiring vibration data of the calculation vehicles and estimating vehicle parameters of the calculation vehicles;
s5: calculating a flatness index of the unknown flatness road section based on the estimated vehicle parameters;
s6: repeating the steps S3-S5 for a plurality of iterations until the flatness index of all road sections in the road network to be detected is obtained.
Preferably, the step S1 specifically includes:
s11: collecting Z-axis vibration data of a test vehicle, setting a reference sampling frequency f, reducing the frequency of the Z-axis vibration data to f when the frequency of the vibration data is higher than f, and increasing the frequency of the Z-axis vibration data to f when the frequency of the vibration data is lower than f;
s12: and calculating the power spectral density of the Z-axis vibration acceleration data of each test vehicle in a calculation period T, cutting a frequency band by an octave distance of delta l distance, and calculating a power spectral density integral value in each octave.
Preferably, the method for obtaining the flatness index of the road section with known flatness in the step S2 is laser detection, vibration detection or level detection.
Preferably, in the step S2, a section with a larger gradient of the flatness index is selected as the section with a known flatness, and the selecting step specifically includes:
when 10% of the total road sections of the road network G to be detected is a non-integer, the total number num (R) of the known flatness road R is rounded upwards;
when the historical flatness data of the road segments in the road network G are known, namely, the road segments IRI with the largest flatness in the road network are selected max With the smallest road section IRI min And by (IRI) max -IRI min ) (num (R) -1) as interval, IRI and IRI are extracted respectively min +N*(IRI max -IRI min ) The closest link of/(num (R) -1) is known as a link, N is a natural integer of 1 to num (R) -2;
when the historical road section flatness data in the road network G is unknown, namely, the road section ACC with the largest average vibration amplitude of the Z axis of the vehicle in the road network is selected max With smallest road section ACC min And by (ACC) max -ACC min ) With (num (R) -1) as interval, IRI and ACC are extracted respectively min +N*(ACC max -IRI min ) The closest link to/(num (R) -1) is known as a link, and N is a natural integer of 1 to num (R) -2.
Preferably, the step S3 specifically includes:
the driving track of the test vehicles is acquired, and the test vehicle set V is classified according to the number of the known road segments: v= { V 1 ,V 2 ,V 3 ,...,V max }，V 1 ,V 2 ,V 3 ,...,V max Respectively passing through 1, 2 and 3 … max vehicles of known road sections, wherein, drawing a frequency distribution histogram of the classified test vehicle set, calculating the number of road sections passing through the known flatness corresponding to the preset vehicle score value q as a vehicle number threshold M q When M q When the number is a non-integer, the number is rounded upwards, and M is more than or equal to 2 q ≤max。
Preferably, the step S4 specifically includes:
extraction ofThe vehicle in (2) is used as a calculating vehicle, Z-axis vibration data of the calculating vehicle is acquired, and vehicle parameters are calculated, wherein +.>In order to make the number of the passing road sections greater than or equal to M q Is a vehicle of (a).
Preferably, the vehicle parameters in the step S4 include a vehicle suspension parameter P, Q and a vehicle model intercept b, and the vehicle parameters are obtained based on least square fitting in the step S4:
wherein IRI is a flatness index, K is the total number of octaves, ω is angular velocity, Δl is the octave distance, S a (ω) is the power spectral density,for power spectral density products within each octaveScore value.
Preferably, the step S5 specifically includes:
s51: calculating based on the estimated vehicle parameters and the power spectral density integral value thereofFlatness index IRI of medium vehicle passing through unknown flatness road section j When a plurality of vehicles pass through the same road section, calculating the estimated flatness standard deviation sigma of the flatness indexes of a plurality of test vehicles passing through the same road section j ；
S52: according to the formula:screening the road sections with the largest number of passing vehicles and the smallest estimated flatness standard deviation>Wherein size (IRI) j ) For the number of vehicles passing by the road section j,
s53: calculation passAverage value of flatness index of a plurality of vehicles of road section as +.>Flatness index of road segment, will ∈ ->The road segments add to the known flatness road segments.
Preferably, after each iteration of step S6, the percentage difference between the parameters P, Q, b of each vehicle after two consecutive iterations is compared:
wherein χ is P 、χ q 、χ b The difference percentages of P, Q, b of the current iteration and the last iteration are respectively, d is the iteration number, if χ P 、χ q 、χ b Any one of the items exceeds a set threshold, the vehicle rejection test vehicle set is subjected to the (d+1) th iteration, and the rejected vehicles are added into the test vehicle set again in the (d+2) th iteration.
Preferably, the speed of the test vehicle is not lower than 20km/h, and the speed variation coefficient of the test vehicle on different road sections within a preset time threshold is not more than 10%.
Compared with the prior art, the invention has the following advantages:
(1) The method can realize the rapid prediction of the flatness of the large-scale road network surface based on crowd-sourced vibration data and track information of the test vehicles, can effectively solve the problems of time and labor consumption, high price and the like of the traditional detection method, improves the updating period of the road surface flatness, timely discovers abnormal jolt and damage of the road, reduces the consumption of manpower and material resources for the complex calibration process of a single vehicle, and has important significance for the high-frequency digital detection of the large-scale road surface performance.
(2) The invention detects based on the data of a plurality of test vehicles, has low cost, and can avoid random errors generated by single-vehicle single-frequency detection and reduce the influence of abnormal data on results by superposing vibration data of different vehicles running on the same road section and mining the stable characteristics of the vibration signals in time-frequency distribution based on the iterative crowd funding detection flow; the road flatness sensing method based on crowd funding data can effectively improve coverage and measurement efficiency, benefit from characteristics of mass data and high frequency, further improve flatness estimation precision and reduce result discreteness.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a schematic diagram of a semi-supervised machine learning model employed in the present invention;
FIG. 3 is a schematic diagram of the number of vehicles traveling on road segments according to the embodiment;
FIG. 4 is a schematic diagram of the number of road segments travelled by a vehicle according to an embodiment;
FIG. 5 is a graph of the vehicle power spectral density score versus IRI for the present embodiment;
FIG. 6 is a graph showing the distribution of estimated flatness parameters for each road segment in the present embodiment;
fig. 7 is a schematic diagram of a semi-supervised machine learning result in the present embodiment.
Detailed Description
The invention will now be described in detail with reference to the drawings and specific examples. Note that the following description of the embodiments is merely an example, and the present invention is not intended to be limited to the applications and uses thereof, and is not intended to be limited to the following embodiments.
Examples
Along with popularization of smart mobile phone equipment, the vibration and position information of the vehicle are collected by utilizing a triaxial acceleration sensor, a gyroscope, a positioning module and the like which are equipped with the smart mobile phone, and the data directly or indirectly reflect the smoothness level of the road surface, so that possibility is provided for large-scale and high-frequency road network level smoothness sensing. Along with popularization of the 'Internet+' travel service mode, massive mass mobile phone crowd funding data with high frequency, wide coverage and low cost are generated in a large number of transportation trips. Although crowd funded data is lower in single point acquisition accuracy than professional equipment, due to the huge volume, stable characteristics in the environment can be reflected. Compared with a flatness measurement method based on a bicycle mobile phone, the pavement flatness sensing method based on crowd funding data can effectively improve coverage and measurement efficiency, and meanwhile, due to the characteristics of mass data and high frequency, flatness estimation accuracy is expected to be further improved, and result discreteness is reduced.
Crowd funding vibration data is a new carrier for traffic analysis and information mining, and improving the quality of the data through the quantity of the data is one of important means for data analysis. During driving, a large amount of sensing data is accumulated in the vehicle, and although the acquisition condition of part of data is unknown to the vehicle parameters, crowd funded data are obviously related in space-time distribution. For example, by superposing vibration data of different vehicles running on the same road section and mining stable characteristics of the vibration signals in time-frequency distribution, random errors generated by single-vehicle single-frequency detection can be avoided, and influence of abnormal data on results is reduced; by comparing the variation trend of vibration data of the same vehicle running on different roads of the road network, the flatness distribution difference between road segments can be analyzed, and the accuracy and the confidence of the model are improved. Therefore, how to extract stable flatness parameters from high-discreteness crowd-sourced vibration data is a key to performing road-network-level road flatness collaborative awareness.
A road network grade road surface flatness detection method based on multi-vehicle crowd funding vibration data mainly estimates suspension parameters of vehicles on a road section through a road based on a small part of known flatness parameters, and when the vehicles with estimated parameters travel to other unknown roads, the flatness of the road section is calculated by a reactive method. In each cycle, selecting a road section with highest calculation confidence level, transferring the road section from an unknown road section set to a known road section level, and repeating the process until the road of the unknown set is an empty set, namely all roads are calculated, so that the road surface flatness of a large-scale road network is rapidly detected, and an interactive semi-supervised learning model of road marking vehicle-vehicle calculation road is constructed.
In the description of the embodiment, the IRI is used for representing the road surface flatness, and other parameters can be used for the road surface flatness parameters in practical application. As shown in fig. 1, the present invention includes the steps of:
s1: vibration data of a test vehicle is obtained and preprocessed, in this embodiment, a set of test vehicles is set as a test vehicle set V, and devices such as a mobile phone, a vehicle sensor, and a peripheral sensor capable of obtaining the vehicle vibration data are mounted on the test vehicle. In this embodiment, when the speed of each individual vehicle is not lower than 20km/h, that is, when the individual vehicle runs at a low speed, the individual vehicle is removed from the vehicle set V, the speed of the road section is relatively stable, the variation coefficient of the speed of the road section is not more than 10% in the time T, the road section of the road network to be detected is a passable lane between adjacent nodes in the road network, different lanes are regarded as different road sections between the same nodes, for example, an urban road can be used as a boundary for cutting the road section, an expressway can be provided with a pile number, a toll gate can be used as a boundary for cutting the road section, and an overhead road can be provided with a ramp position as a boundary for cutting the road section.
The step S1 specifically comprises the following steps:
s11: collecting vibration data of a test vehicle, marking the data as X, Y and Z axis data according to the axial direction of a sensor, extracting Z axis vibration data of the test vehicle, setting a reference sampling frequency f, reducing the frequency of the Z axis vibration data to f by adopting an anti-aliasing frequency reduction method when the frequency of the vibration data is higher than f, and increasing the frequency of the Z axis vibration data to f by adopting a linear interpolation method when the frequency of the vibration data is lower than f. And simultaneously setting a flatness alternating calculation period T.
The vehicle vibration data acquisition can be realized through the modes of mobile phone application acquisition, peripheral vibration sensor acquisition, vehicle-mounted self-diagnosis system reading and the like, the data frequency of the Z axis is not lower than 1Hz, and the data frequency of the Z axis in the embodiment is more than 20Hz. The anti-aliasing frequency reduction adopts a finite long impulse response (FIR) digital low-pass filter, the calculation period T is selected according to the road network scale and is not lower than 30 minutes, and the calculation period is 24 hours in the embodiment.
In order to avoid the influence of the sample size on the detection result, the sample capacity of the known road section can be adjusted in actual operation, and if the road network traffic is smaller and the vehicle vibration data is smaller, the sample capacity of the known road section is increased to 20% or more; when the known samples are limited, the calculation period T is enlarged, so that the data volume requirement is met.
S12: calculating the power spectral density of Z-axis vibration acceleration data of each test vehicle in a calculation period T, cutting a frequency band by an octave distance of delta l distance, wherein the octave length is not more than 10Hz, and calculating a power spectral density integral value in each octave.
Calibrating vehicle parameters requires a small number of road network segment flatness parameters to be known, so S2: taking a road section with the flatness index gradient larger than a preset threshold value in the road network G to be detected as a known flatness road section, wherein the road section is set as R, and the road sections with other unknown flatness are set as U, so as to obtain the flatness index of the road section R with the known flatness.
In this embodiment, S2 selects a road section with a gradient of 10% of the flatness index as a known flatness road section, and obtains the flatness index by adopting modes of laser detection, vibration detection, level gauge detection and the like, where the flatness index includes but is not limited to parameters capable of objectively evaluating smoothness of a road surface, such as an international flatness index, a driving comfort index and the like, and the selecting steps of the known flatness road section specifically include:
when 10% of the total road sections of the road network G to be detected is a non-integer, the total number num (R) of the known flatness road R is rounded upwards;
when the historical flatness data of the road segments in the road network G are known, namely, the road segments IRI with the largest flatness in the road network are selected max With the smallest road section IRI min And by (IRI) max -IRI min ) (num (R) -1) as interval, IRI and IRI are extracted respectively min +N*(IRI max -IRI min ) The closest link of/(num (R) -1) is known as a link, N is a natural integer of 1 to num (R) -2;
when the historical road section flatness data in the road network G is unknown, namely, the road section ACC with the largest average vibration amplitude of the Z axis of the vehicle in the road network is selected max With smallest road section ACC min And by (ACC) max -ACC min ) With (num (R) -1) as interval, IRI and ACC are extracted respectively min +N*(ACC max -IRI min ) The closest link to/(num (R) -1) is known as a link, and N is a natural integer of 1 to num (R) -2.
S3: the method comprises the steps of obtaining the running track of a test vehicle, and obtaining a vehicle number threshold according to the running track of the test vehicle and the known flatness road section information, and specifically comprises the following steps:
the driving track of the test vehicles is acquired, and the test vehicle set V is classified according to the number of the known road segments: v= { V 1 ,V 2 ,V 3 ,...,V max }，V 1 ,V 2 ,V 3 ,...,V max Respectively passing through 1, 2 and 3 … max vehicles of known road sections, wherein, drawing a frequency distribution histogram of the classified test vehicle set, calculating the number of road sections passing through the known flatness corresponding to the preset vehicle score value q as a vehicle number threshold M q When M q When the number is a non-integer, the number is rounded upwards, and M is more than or equal to 2 q ≤max。
In this example, q is 85%.
S4: extraction ofThe vehicle in (2) is used as a calculation vehicle, and the Z-axis vibration data calculator vehicle parameters of the calculation vehicle are acquired,/-degree>In order to make the number of the passing road sections greater than or equal to M q Is a vehicle of (a).
The vehicle parameters include a vehicle suspension parameter P, Q and a vehicle model intercept b, b mainly characterizing the influence of the vehicle engine at a certain steady speed, and the vehicle parameters are obtained based on a least squares fit in step S4:
wherein IRI is a flatness index, K is the total number of octaves, ω is angular velocity, Δl is the octave distance, S a (ω) is the power spectral density,for power spectral density within each octaveIntegral value.
In the steps S1 to S4, the vehicle parameters are estimated based on the road segments with known flatness of the small sample in the first step, the vehicle parameters are estimated based on the road segments with known flatness of the small sample, the vehicle vibration is the most direct and most common expression of the road surface flatness, but the same road surface flatness may generate distinct vibration characteristics due to the influence of the vehicle suspension parameters and the engine performance, so the estimated vehicle parameters are the first step for road network grade road surface flatness detection, the above formula is a linear equation with three parameters of P, Q and b, the model parameters can be fitted by using the least square method, and therefore, at least three known road segments are required to be passed by the vehicle to obtain the IRI and the vibration parameters of the three road segments, and the three parameters of P, Q and b can be calculated. According to the statistical principle, when the number of road sections through which the vehicle passes is larger, namely the sample size of the fitting model is larger, the confidence of the fitting result is higher. Further, the method calculates the number of the roads passing through the known labels corresponding to the number of the 85% of the vehicles, so as to ensure that the confidence of the estimated vehicle parameters is at a higher level and avoid the influence of the parameter estimation errors on the subsequent calculation.
S5: calculating the flatness index of the unknown flatness road section based on the estimated vehicle parameters specifically comprises:
s51: calculating based on the estimated vehicle parameters and the power spectral density integral value thereofFlatness index IRI of medium vehicle passing through unknown flatness road section j When a plurality of vehicles pass through the same road section, calculating the estimated flatness standard deviation sigma of the flatness indexes of a plurality of test vehicles passing through the same road section j ；
S52: according to the formula:screening the road sections with the largest number of passing vehicles and the smallest estimated flatness standard deviation>Wherein size (IRI) j ) For the number of vehicles passing by the road section j,
s53: calculation passThe average value of the flatness index IRI of several vehicles of a road section is used as +.>Flatness index of road segment, will ∈ ->The road segments add to the known flatness road segments R and are removed from U.
In step S5, road segment parameters of the flatness position are calculated based on the estimated vehicle parameters, through S1 to S4, multiple vehicle parameters in the road network are estimated, when the vehicles drive into road segments with unknown flatness parameters, vibration data of the vehicles are collected, IRI of the road segments with the positions can be calculated by using the IRI formula, and the same unknown road segments can be driven by multiple vehicles with known parameters, so that the road segments can be estimated for multiple times. Because of errors in the estimation of the suspension parameters of the vehicle, and the possible difference in the speeds of multiple vehicles passing through the same road section, the estimated IRI is different. To improve confidence level of model prediction, consensus-basedAnd selecting the road with the most collection times and the lowest distribution discreteness in all the estimated road sections, namely the road with the most stable result.
S6: repeating the steps S3-S5 for a plurality of iterations until U is an empty set. After each iteration, only one road segment with the highest confidence of the result is moved from the unknown set to the known set. Therefore, after each iteration, the number of unknown road segments is-1, the known road segment data is +1, the iteration process of the invention is based on the semi-supervised machine learning model, and assuming that the total number of unknown road segments in the road network is num (U), the number of times of iteration of num (U) is needed to calculate the flatness parameters of all road segments, and the iteration process is shown in fig. 2.
In each iteration, the parameters of the vehicle and the flatness of the unknown road segment need to be recalculated. In this embodiment, after each iteration of S6, the percentage difference of the parameter P, Q, b of each vehicle after two consecutive iterations is compared:
wherein χ is P 、χ q 、χ b The difference percentages of P, Q, b of the current iteration and the last iteration are respectively, d is the iteration number, if χ P 、χ q 、χ b Any one of the items exceeds a set threshold, the vehicle rejection test vehicle set is subjected to the (d+1) th iteration, and the rejected vehicles are added into the test vehicle set again in the (d+2) th iteration.
In this embodiment, the implementation process of the scheme of the present invention is as follows:
(1) Vibration data of 500 vehicles in a road network are collected through a vehicle-mounted mobile phone APP and a fixed support, the data collection frequency is 200Hz, 50 road sections are contained in the road network, the number is 1-50, the flatness of the known road sections is shown in a table 1, and all known flatness parameters are international flatness index IRI data collected by a laser detection vehicle.
Table 1 known flatness parameter section
| Numbering device | 4 | 9 | 16 | 28 | 29 |
| IRI | 3.987 | 4.911 | 4.152 | 2.917 | 5.264 |
| Numbering device | 30 | 35 | 39 | 44 | 48 |
| IRI | 5.913 | 6.539 | 9.860 | 2.762 | 3.398 |
(2) Counting the distribution of the vehicles running on each road section and the number of the branches of the road sections on which the vehicles run, as shown in figures 3 and 4, M can be obtained by accumulating the distribution map 85％ For 3, i.e. selecting all vehicles passing through a known road section greater than 3, using vibration data acquired by mobile phone APP, FIG. 5 shows the relationship of the power spectral density integral of three vehicles numbered 3, 280, 400 to the known IRI by multiple linear regressionEquation (1), model parameters P, Q, b are estimated by the least square method.
(3) The road surface flatness IRI of the vehicle passing through the unknown IRI road section is calculated based on the formula (1) using the vibration data of the vehicle of the estimated parameters, as shown in fig. 7. The IRI estimated standard deviation of each road section is calculated,and assigning the IRI of the road section, and iterating to finally obtain the traversal calculation result of all the road sections of the road network, as shown in fig. 6.
(4) Model accuracy evaluation
The average relative error of the road surface flatness calculated according to the model algorithm is 9.71%, the highest road section relative error appears at the number 40 as 46.5%, the absolute error is 1.15m/km, the IRI of the position is considered to be low, and most of the IRI of the trained road sections are distributed at about 4.5m/km, so that the expression is poor at the position with low IRI. As can be seen from fig. 6, the semi-supervised learning model has excellent overall performance, substantially accords with the actual IRI distribution condition of the road surface, most of relative errors can be kept within 10%, and the error fall state occurs at the higher position and the lower position of the IRI.
The above embodiments are merely examples, and do not limit the scope of the present invention. These embodiments may be implemented in various other ways, and various omissions, substitutions, and changes may be made without departing from the scope of the technical idea of the present invention.
Claims (5)
1. A road network grade pavement evenness detection method based on multi-vehicle crowd funding vibration data is characterized by comprising the following steps:
s1: obtaining vibration data of a test vehicle and preprocessing the vibration data;
s2: taking a road section with the flatness index gradient larger than a preset threshold value in the road network to be detected as a known flatness road section, and acquiring the flatness index of the known flatness road section;
s3: acquiring a running track of a test vehicle, and acquiring a vehicle number threshold according to the running track of the test vehicle and the known flatness road section information;
s4: extracting vehicles passing through the road sections with known flatness for times greater than or equal to the threshold value of the number of vehicles as calculation vehicles of the iteration, acquiring vibration data of the calculation vehicles and estimating vehicle parameters of the calculation vehicles;
s5: calculating a flatness index of the unknown flatness road section based on the estimated vehicle parameters;
s6: repeating the steps S3-S5 for a plurality of iterations until the flatness index of all road sections in the road network to be detected is obtained;
the step S1 specifically comprises the following steps:
s11: collecting Z-axis vibration data of a test vehicle, setting a reference sampling frequency f, reducing the frequency of the Z-axis vibration data to f when the frequency of the vibration data is higher than f, and increasing the frequency of the Z-axis vibration data to f when the frequency of the vibration data is lower than f;
s12: calculating the power spectral density of Z-axis vibration acceleration data of each test vehicle in a calculation period T, cutting a frequency band by an octave distance of delta l distance, and calculating a power spectral density integral value in each octave;
the step S3 specifically includes:
the driving track of the test vehicles is acquired, and the test vehicle set V is classified according to the number of the known road segments: v= { V 1 ,V 2 ,V 3 ,...,V max }，V 1 ,V 2 ,V 3 ,...,V max Respectively passing through 1, 2 and 3 … max vehicles of known road sections, wherein, drawing a frequency distribution histogram of the classified test vehicle set, calculating the number of road sections passing through the known flatness corresponding to the preset vehicle score value q as a vehicle number threshold M q When M q When the number is a non-integer, the number is rounded upwards, and M is more than or equal to 2 q ≤max；
The vehicle parameters in the step S4 include vehicle suspension parameters P, Q and vehicle model intercepts b, and the vehicle parameters are obtained based on least square fitting in the step S4:
wherein IRI is a flatness index, K is the total number of octaves, ω is angular velocity, Δl is the octave distance, S a (ω) is the power spectral density,integrating the power spectral density values in each octave;
the step S5 specifically includes:
s51: calculating based on the estimated vehicle parameters and the power spectral density integral value thereofFlatness index IRI of medium vehicle passing through unknown flatness road section j When a plurality of vehicles pass through the same road section, calculating the estimated flatness standard deviation sigma of the flatness indexes of a plurality of test vehicles passing through the same road section j Wherein->In order to make the number of the passing road sections greater than or equal to M q Is a vehicle of (2);
s52: according to the formula:screening the road sections with the largest number of passing vehicles and the smallest estimated flatness standard deviation>Wherein size (IRI) j ) For the number of vehicles passing by the road section j,
s53: calculation passAverage value of flatness index of a plurality of vehicles of road section as +.>Flatness index of road segment, will ∈ ->Adding a road section with known flatness;
the step S4 specifically includes:
and extracting vehicles passing through the road sections with the known flatness and the number of times greater than or equal to the threshold value of the number of vehicles as calculation vehicles of the iteration, acquiring Z-axis vibration data of the calculation vehicles and calculating vehicle parameters of the calculation vehicles.
2. The road network grade road surface flatness detection method based on multi-vehicle crowd funding vibration data according to claim 1, wherein the method for obtaining the flatness index of the road section with known flatness in the step S2 is laser detection, vibration detection or level meter detection.
3. The road network grade road surface flatness detection method based on multi-vehicle crowd funding vibration data according to claim 1, wherein in the step S2, a road section with a larger flatness index gradient of 10% is selected as a road section with a known flatness, and the selecting step specifically comprises:
when 10% of the total road sections of the road network G to be detected is a non-integer, the total number num (R) of the known flatness road R is rounded upwards;
when the historical flatness data of the road segments in the road network G are known, namely, the road segments IRI with the largest flatness in the road network are selected max With the smallest road section IRI min And by (IRI) max -IRI min ) (num (R) -1) as interval, IRI and IRI are extracted respectively min +N*(IRI max -IRI min ) The closest link of/(num (R) -1) is known as a link, N is a natural integer of 1 to num (R) -2;
when the historical road section flatness data in the road network G is unknown, namely, the road section ACC with the largest average vibration amplitude of the Z axis of the vehicle in the road network is selected max With smallest road section ACC min And by (ACC) max -ACC min )/(num(R)-1)For interval, respectively extract IRI and ACC min +N*(ACC max -IRI min ) The closest link to/(num (R) -1) is known as a link, and N is a natural integer of 1 to num (R) -2.
4. The road network grade road surface flatness detection method based on multi-vehicle crowd funding vibration data according to claim 1, wherein after each iteration of step S6, the difference percentages of the parameters P, Q, b of each vehicle after two consecutive iteration processes are compared:
wherein χ is P 、χ q 、χ b The difference percentages of P, Q, b of the current iteration and the last iteration are respectively, d is the iteration number, if χ P 、χ q 、χ b Any one of the items exceeds a set threshold, the vehicle rejection test vehicle set is subjected to the (d+1) th iteration, and the rejected vehicles are added into the test vehicle set again in the (d+2) th iteration.
5. The road network grade road surface flatness detection method based on multi-vehicle crowd funded vibration data according to claim 1, wherein the speed of the test vehicle is not lower than 20km/h, and the speed variation coefficient of the test vehicle on different road sections within a preset time threshold is not more than 10%.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210144895.0A CN114541222B (en) | 2022-02-17 | 2022-02-17 | Road network grade pavement flatness detection method based on multi-vehicle crowd funding vibration data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210144895.0A CN114541222B (en) | 2022-02-17 | 2022-02-17 | Road network grade pavement flatness detection method based on multi-vehicle crowd funding vibration data |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114541222A CN114541222A (en) | 2022-05-27 |
| CN114541222B true CN114541222B (en) | 2024-01-26 |
Family
ID=81675291
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210144895.0A Active CN114541222B (en) | 2022-02-17 | 2022-02-17 | Road network grade pavement flatness detection method based on multi-vehicle crowd funding vibration data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114541222B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104120644A (en) * | 2013-04-26 | 2014-10-29 | 同济大学 | Gravitational acceleration sensor based road surface flatness detection method |
| JP2015028456A (en) * | 2013-07-31 | 2015-02-12 | 国立大学法人 東京大学 | Road surface evaluation device and method |
| CN108839657A (en) * | 2018-06-05 | 2018-11-20 | 吉林大学 | A method of online recognition road roughness information is responded based on automobile vibration |
| CN111275363A (en) * | 2020-03-27 | 2020-06-12 | 山东智路信息科技有限公司 | Road network flatness evaluation system and method based on vehicle vibration data |
| CN113434954A (en) * | 2021-06-15 | 2021-09-24 | 同济大学 | Calibration method of vibrating type pavement flatness test vehicle |
-
2022
- 2022-02-17 CN CN202210144895.0A patent/CN114541222B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104120644A (en) * | 2013-04-26 | 2014-10-29 | 同济大学 | Gravitational acceleration sensor based road surface flatness detection method |
| JP2015028456A (en) * | 2013-07-31 | 2015-02-12 | 国立大学法人 東京大学 | Road surface evaluation device and method |
| CN108839657A (en) * | 2018-06-05 | 2018-11-20 | 吉林大学 | A method of online recognition road roughness information is responded based on automobile vibration |
| CN111275363A (en) * | 2020-03-27 | 2020-06-12 | 山东智路信息科技有限公司 | Road network flatness evaluation system and method based on vehicle vibration data |
| CN113434954A (en) * | 2021-06-15 | 2021-09-24 | 同济大学 | Calibration method of vibrating type pavement flatness test vehicle |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114541222A (en) | 2022-05-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Alessandroni et al. | Smartroadsense: Collaborative road surface condition monitoring | |
| CN109410586B (en) | Traffic state detection method based on multi-metadata fusion | |
| CN104164829B (en) | Detection method of road-surface evenness and intelligent information of road surface real-time monitoring system based on mobile terminal | |
| CN107330217B (en) | Mesoscopic oil consumption prediction method based on RBFNN | |
| CN107207009B (en) | Pavement state judgment method | |
| CN104792937B (en) | Bridge head bump detection evaluation method based on vehicle-mounted gravitational acceleration sensor | |
| CN105005694B (en) | A kind of bridge fatigue life frequency-domain analysis method based on dynamic weighing system | |
| Hanson et al. | Evaluation of low-cost consumer-level mobile phone technology for measuring international roughness index (IRI) values | |
| CN101275900A (en) | Method for recognizing road surface types based on vehicle wheel vibration | |
| CN109829252B (en) | Influence line identification-based bridge condition rapid rating method | |
| CN104233935B (en) | A kind of pavement quality grade discrimination method based on profile of road information | |
| CN111127906B (en) | Intelligent road surface management system and method based on Internet of things | |
| Song et al. | Characteristics of low-speed vehicle-specific power distributions on urban restricted-access roadways in beijing | |
| CN109886724B (en) | Robust resident travel track identification method | |
| CN110852542B (en) | Road flatness calculation method and system | |
| CN102721397A (en) | Method for extracting road surface characteristic parameters based on modern time series of vertical dynamic load | |
| CN105588599B (en) | The adaptive corrective method of vehicle-mounted mobile measuring system vibration displacement error | |
| CN114264597B (en) | Low-cost road adhesion coefficient determination method and system | |
| CN115544763A (en) | Road surface flatness prediction method, system and medium | |
| CN112100721B (en) | Bridge structure modal identification method based on mobile crowd sensing and deep learning | |
| CN114541222B (en) | Road network grade pavement flatness detection method based on multi-vehicle crowd funding vibration data | |
| CN105066959A (en) | Pavement vertical section elevation information acquisition method | |
| Yunusov et al. | Signal processing and conditioning tools and methods for road profile assessment | |
| CN115964905A (en) | Method and medium for constructing emission test working condition of plateau mountain road | |
| Zhang et al. | Feasibility study on pavement rutting evaluation method based on smartphone |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |