CN106295505A - State estimating system during pavement usage - Google Patents

Abstract

The present invention discloses the state estimating system during a kind of pavement usage, solves polynary, the synthesis measuring problem of pavement state during use.State estimating system during pavement usage of the present invention, first road pavement type is measured, and then measures the flatness on road surface, finally whether there is breakage on detection road surface；Described system includes: detection car；Road surface kind determination unit, for measuring the type on road surface, output road surface category identification result；Measuring flatness of road surface unit, for measuring the flatness on road surface；Road surface breakage positioning unit, is used for detecting whether road surface has breakage, when road surface breakage being detected, gathers the image information of damaged road surface, and obtains the positional information of described damaged road surface.The present invention, the accuracy of pavement state measurement data is high, it is possible to reacted pavement state accurately.

Description

State estimating system during pavement usage

Technical field

The present invention relates to the state estimating system during a kind of pavement usage.

Background technology

Along with the fast development of road haulage, the maintenance of highway and management are the most important, wherein, pavement destruction information Obtain and just become extremely important.

At present, in the collection of pavement state information, it is common that carry out adopting of information of road surface by information of road surface collecting vehicle Collection, is wherein provided with video camera on information acquisition vehicle, by information acquisition vehicle traveling on road surface, by information of road surface by taking the photograph The mode of picture stores, and after system-wide section collection completes, will storage data transmission to control centre, by control centre according to Shooting content, judges the breakage in road surface by artificial or graphical analysis.

In existing this information of road surface gatherer process, need all for road surface information, i.e. damaged road surface and non-damaged road surface All collecting, information of road surface collection capacity is big；Meanwhile, also have no idea to determine the particular geographic location destroying road surface, and, by Unknown in the geographical position of damaged road surface, it is impossible to notify timely and effectively to the row on road surface in time, accurately by damaged road surface Sail vehicle, cause vehicle, through this damaged road surface, the vehicle accidents such as traffic accident easily occur.

Summary of the invention

For the problems referred to above, the present invention provides a kind of collection pavement state data by diversification, determines pavement state Pavement usage during state estimating system.

State estimating system during pavement usage of the present invention, first road pavement type is measured, and then measures road Whether the flatness in face, finally exist breakage on detection road surface；Described system includes:

Detection car；

Road surface kind determination unit, for measuring the type on road surface, output road surface category identification result；

Measuring flatness of road surface unit, for measuring the flatness on road surface；

Road surface breakage positioning unit, is used for detecting whether road surface has breakage, when road surface breakage being detected, gathers damaged road surface Image information, and obtain the positional information of described damaged road surface；

Wherein, road surface kind determination unit, including detection data acquisition module, the first pretreatment module of microprocessor, First computing module of Cloud Server；

Described detection data acquisition module, including be arranged on detection car on forward laser light radar, acceleration transducer, after Put laser radar and ccd image sensor；

First pretreatment module of microprocessor, including sensing data acquisition module, described preposition swashs for obtaining Optical radar, acceleration transducer, rearmounted laser radar and ccd image sensor；

Identification module, acceleration identification module and picture recognition module, described laser identification module, for obtaining respectively The laser data that forward laser light radar, rearmounted laser radar gather, generates road surface category identification result according to laser data；Described Acceleration identification module, for obtaining the acceleration information that acceleration transducer obtains, generates road according to described acceleration information Face category identification result；Picture recognition module, for obtaining the acceleration information that image acquisition device obtains, according to described picture number According to generating road surface category identification result；And by the first computing module of described road surface recognition result output to Cloud Server；

First computing module of described Cloud Server, merges module, the generation of road surface kind including road surface category identification result Unit, wherein said road surface category identification combining unit, the road surface kind result described in obtaining according to ballot method is entered Row merging treatment, obtains recognition result after the road surface category identification result merging of corresponding each sensor；Wherein, the Ma Er of establishment The energy equation of section's husband's random field models is:

$\begin{array}{c}E(x,y,u,w,v)=\mathrm{\ρ}\underset{i}{\mathrm{\Σ}}\sqrt{\left|x_i^2-x_{i-1}^2\right|}+\underset{i}{\mathrm{\Σ}}{r}_i\sqrt{\left|x_i^2-y_i^2\right|}+\\ \underset{i}{\mathrm{\Σ}}k{1}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^1\right)}^2\right|}+\underset{i}{\mathrm{\Σ}}k{2}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^2\right)}^2\right|}+\underset{i}{\mathrm{\Σ}}k{3}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^3\right)}^2\right|}+\mathrm{...}+\underset{i}{\mathrm{\Σ}}{\mathrm{kn}}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^n\right)}^2\right|}\end{array}$

Wherein, i is the sequence number in identified section, and i is integer and i >=2, x_iFor the forward laser light that i-th section is to be optimized The road surface category identification result of radar, x_i-1It is the final road surface category identification result in the i-th-1 section, y_iBefore i-th section Put the road surface category identification result after the merging of laser radar, u_i-1It it is the road surface after the merging of the i-th-1 section imageing sensor Category identification result, w_i-1It is the road surface category identification result after the merging of the i-th-1 section acceleration transducer, v_i-1Be i-th- Road surface category identification result after the merging of 1 rearmounted laser radar in section, ρ is x_iWith x_i-1Between link potential energy, r_iFor x_i With y_iBetween link potential energy, k1_i-1For x_iWith u_i-1Between link potential energy, k2_i-1For x_iWith w_i-1Between link potential energy, k3_i-1For x_iWith v_i-1Between link potential energy；

By road surface category identification result x of forward laser light radar to be optimized for described i-th section_iCorresponding numerical value divides Do not bring described energy equation into calculate, will enable energy equation E (x, y, u, w, the x that value v) is minimum_iCorresponding road surface Kind is as described final road surface category identification result；

Measuring flatness of road surface unit, including the flatness detector being arranged on detection car, the second of microprocessor is pre- Processing module, the second computing module of Cloud Server；

Described roughness measurement instrument includes two axle acceleration sensors, laser range sensor, mileage sensor for countering, top Spiral shell instrument, wherein said two axle acceleration sensors are being perpendicular to for measuring described surface evenness detector when t The first acceleration a_measured on described surface evenness detector direction_y(t) and be parallel to described surface evenness The second acceleration a_measured on detector direction_x(t)；Described laser range sensor is used for measuring described evenness of road surface Degree detector distance h (t) when t and between road surface；Described mileage sensor for countering is used for measuring described road surface puts down The whole degree detector horizontal displacement s when t_x(t)；Described gyroscope is used for measuring described surface evenness detector Anglec of rotation θ (t) when t and between level road；

Second pretreatment module of described microprocessor, for obtaining the first described acceleration a_measured_y(t)、 Second acceleration a_measured_x(t), distance h (t) measured between described surface evenness detector and road surface, horizontal position Shifting amount s_x(t), anglec of rotation θ (t)；And the first acceleration a_measured described in obtaining_y(t), the second acceleration a_ measured_xT (), anglec of rotation θ (t) and gravity acceleration g export the second computing module to Cloud Server；

Described second computing module, according to when t, described the first acceleration a_measuredy (t), second Acceleration a_measuredx (t), anglec of rotation θ (t) and gravity acceleration g, when calculating described surface evenness detector t Acceleration of vibration ay (t) of in the vertical direction during quarter；According to described acceleration of vibration a_yT (), calculates described surface evenness The displacement s of in the vertical direction vibration during detector t_y(t)；According to described surface evenness detector and described road Distance h (t) between face and the displacement s of described surface evenness detector in the vertical direction vibration_yT (), obtains in institute State horizontal displacement s_xThe surface evenness w on described road surface in the range of (t)_y(t)；

Road surface breakage positioning unit, including the 3rd pretreatment module, the cloud of video camera, GPS locator and microprocessor 3rd computing module of server；

Described video camera obtains pavement image data, the geography position, road surface that GPS locator record current image frame is corresponding Confidence ceases, and exports view data, the 3rd pretreatment module of GPS location data to microprocessor；

Described 3rd pretreatment unit is used for carrying out described view data pretreatment, and by pretreated view data And the shooting time of view data and GPS location information export the 3rd computing module to Cloud Server；

Described 3rd computing module of described Cloud Server, for data separate image recognition side, pretreated road surface Method road pavement road surface data are identified obtaining the road surface data comprising breakage, it is thus achieved that comprise road surface data and this figure of breakage As shooting time and the GPS of data position information；

Wherein, described microprocessor is arranged on detection car, and described microprocessor and described Cloud Server are by wireless Network carries out data transmission.

Further, the 3rd pretreatment unit of described microprocessor, the preprocess method of road pavement image includes:

The view data obtained is carried out color analysis and gray processing processes, coloured image is converted to gray level image；

View data is made discrete Daubechies4 wavelet transformation, obtains the wavelet conversion coefficient of view data, The high frequency filter coefficient [G0 G1 G2 G3] of Daubechies4 small echo and low-frequency filter coefficient [H0 H1 H2 G3] are respectively For:

$G_0=\frac{(1+\sqrt{3})}{4\sqrt{2}},G_1=\frac{(3+\sqrt{3})}{4\sqrt{2}},G_2=\frac{(3-\sqrt{3})}{4\sqrt{2}},G_3=\frac{(1-\sqrt{3})}{4\sqrt{2}}$

$H_0=\frac{(1-\sqrt{3})}{4\sqrt{2}},H_1=\frac{(-3+\sqrt{3})}{4\sqrt{2}},H_2=\frac{(3+\sqrt{3})}{4\sqrt{2}},H_3=\frac{(-1-\sqrt{3})}{4\sqrt{2}};$

The window using n × n is smooth mobile along view data, calculates being correlated with between view data wavelet conversion coefficient Property, obtain the view data after shrinking weights, wherein the computing formula of dependency R is:

$R=\frac{\mathrm{\Σ}XY-\frac{\mathrm{\Σ}X\·\mathrm{\Σ}Y}{n}}{\sqrt{{\mathrm{\ΣX}}^2-\frac{{\left(\mathrm{\Σ}X\right)}^2}{n}\]\·\[{\mathrm{\ΣY}}^2-\frac{{\left(\mathrm{\Σ}Y\right)}^2}{n}\]}}$

In above formula, X, Y are the combination of any two n in image × n window subimage, ∑_X, ∑_YIt is respectively subgraph gray value Summation, ∑_XYFor suing for peace after two subgraph correspondence position gray value phase products of X, Y；

View data after shrinking weights is carried out soft-threshold process, and the computing formula of soft-threshold λ is:

$\mathrm{\λ}=\mathrm{\σ}\sqrt{2ln\left(N\right)}$

In above formula, N × N is to shrink the size of view data after weights, and σ is noise variance；

View data is carried out discrete Daubechies4 inverse wavelet transform, obtains the view data after denoising；

View data after denoising is strengthened.

Further, the concrete steps of laser radar identification road surface types include:

Gather coordinate data and the reflected energy data of the sampled point of laser radar output；

The road surface outline data of Cartesian form, the number of coordinates of wherein said sampled point is generated according to described coordinate data Change into Cartesian form according to by polar form, utilize following formula to realize:

x_i=r_isin(θ_i), z_i=H-r_icos(θ_i)

Wherein, i is the sequence number of the sampled point of laser radar, r_iFor the polar data of the sampled point of laser radar, θ_iIt is sharp Light light beam and laser radar are perpendicular to the angle in direction, ground, and H is the laser radar absolute setting height(from bottom) relative to ground grading, The center of circle of described rectangular coordinate system is positioned at the laser emission point of laser radar, x_iFor horizontal relative to laser emission point of sampled point Coordinate, z_iFor sampled point relative to the longitudinal coordinate of laser emission point；

Described road surface outline data is processed, generation transverse axis be spatial frequency, the longitudinal axis be the road of spatial frequency component Facial contour spatial frequency data；

In units of the S of horizontal displacement interval, described road surface outline data is carried out segmentation, generates n road surface outline data Sample；

Use Lomb algorithm respectively described n road surface profile data sample to be processed, generate and comprise n transverse axis for sky Between frequency, the longitudinal axis be the road surface profile spatial frequency data of sub-road surface profile spatial frequency data of spatial frequency component；

Described road surface profile characteristic and described reflected energy data are merged process, generates grader input number According to；

By the input data input of described grader to trained grader, thus obtain road surface classification results.

Further, the method generating road surface profile characteristic specifically includes:

According to initial frequency set in advance, step pitch frequency and termination frequency, calculate every sub-road wheel exterior feature spatial frequency In data, from the beginning of described initial frequency, to described termination frequency, spatial frequency component corresponding in every section of step pitch frequency Sum, thus generate n sub-road surface profile spatial frequency features data, described sub-road surface profile spatial frequency features packet contains The sum of m spatial frequency component；

Described n sub-road surface profile spatial frequency features data are combined, generate described first via facial contour feature Matrix Y1:

Wherein, the every string in described first via facial contour eigenmatrix Y1 corresponds to a described sub-road surface profile space Frequency characterization data, each element in described first via facial contour eigenmatrix Y1 divides corresponding to a described spatial frequency The sum of amount；

Described reflected energy data is specially the first reflected energy eigenmatrix Y2:

Wherein, the every string in described road surface profile eigenmatrix Y2 corresponds to a road surface profile data sample, described Each element in first reflected energy eigenmatrix Y2 is corresponding to the reflected energy numerical value of the sampled point of laser radar.

Further, road surface structare layer stealth Defect inspection unit, described road surface structare layer stealth Defect inspection are also included Unit includes being arranged on the GPR on detection car and excitation road surface sound-producing device, and the 4th pretreatment mould of microprocessor Block, the 4th computing module of Cloud Server；

Wherein, excitation road surface sound-producing device taps road surface continuously so that produce the sound wave of road vibration because tapping road surface Along road surface structare layer by surface layer travel downward, road GPR launches electromagnetic wave to road surface simultaneously, is sent out by road GPR Penetrate the time synchronized of signal and generating means percussion road surface, excitation road surface；

4th pretreatment module of microprocessor, for obtaining the electromagnetic wave propagation time in road surface structare layer, according to The electromagnetic wave propagation time in road surface structare layer calculates the thickness of every layer in road surface structare layer, and exports this thickness to cloud clothes 4th computing module of business device, thickness equations is as follows:

$h_1=\frac{c_0{t}_1}{2\sqrt{{\mathrm{\ξ}}_1}};h_2=\frac{c_0(t_2-t_1)}{2\sqrt{{\mathrm{\ξ}}_2}};......;h_n=\frac{c_0(t_n-t_{n-1})}{2\sqrt{{\mathrm{\ξ}}_n}},$

Wherein, h₁, h₂..., h_nRepresent the thickness of pavement structure every layer, c₀The electromagnetic wave representing road GPR exists The speed propagated under vacuum state, t₁, t₂..., t_nRepresent the electromagnetic wave of road GPR and arrive in road surface structare layer every layer The propagation time of structure, ξ₁, ξ₂..., ξ_nRepresent the dielectric constant of every layer in road surface structare layer；

4th pretreatment module of microprocessor, for obtaining the sound wave the encouraging road surface sound-producing device biography on every layer of road surface Between sowing time, according to the sound wave of the excitation road surface sound-producing device propagation time T on every layer of road surface_p1, T_p2..., T_pn, and road surface knot The thickness h of every layer in structure layer₁, h₂..., h_n, it is thus achieved that sound wave is the spread speed of every layer in road surface structare layer, and exports this and wear Broadcasting speed to the 4th computing module of Cloud Server, it is as follows that spread speed calculates formula:

$v_1=\frac{h_1}{T_{P1}};v_2=\frac{h_2}{T_{P2}-T_{P1}};......v_n=\frac{h_n}{T_{Pn}-T_{Pn-1}};$

4th computing module of Cloud Server, for according to SVEL V_iWith modulus of resilience E_iRelation formula, calculate Modulus of resilience E of every layer in road surface structare layer_i, analyze the decay of the strength of materials in road surface structare layer: wherein, V_iRepresent sound wave to exist Spread speed v of i-th layer in road surface structare layer_i, ρ_iRepresent the density of the i-th layer material in road surface structare layer.

Further, the density p of i-th layer in road surface structare layer_iRecord by the section detected is carried out coring test.

Further, also include that thermal source affects road surface determination unit, provide road pavement to provide contactless for road pavement Thermal source, monitors the variations in temperature on described road surface, determines the state on described road surface according to described temperature variation data；Supply including thermal source To device, device for detecting temperature, the 5th pretreatment module of microprocessor, the 5th processing module of Cloud Server, wherein

Thermal source feedway, provides non-contact heat source for road pavement, makes the temperature on described road surface change, thermal source Feedway is active visible ray thermal light source or infrared laser source；

Temperature-detecting device, for monitoring the variations in temperature on described road surface, it is thus achieved that temperature variation data, and by described temperature Delta data exports to microprocessor；

5th pretreatment module of microprocessor, for obtaining temperature variation curve, root according to described temperature variation data The state on described road surface is determined according to the slope of described temperature variation curve and slope variation；Slope when described temperature variation curve Constant, and when slope is more than or equal to first threshold, determine that described pavement state is drying regime；When described temperature variation curve Slope is constant, and when slope is less than described first threshold and is more than or equal to Second Threshold, determines that described pavement state is hydrops shape State；When the slope of described temperature variation curve is not 0, and when at a time becoming big suddenly, determine that described pavement state is long-pending Water state；When the slope of described temperature variation curve is occurred jumping characteristic change in certain a period of time by zero, determine described road surface State is icing condition；5th pretreatment module of microprocessor, is additionally operable to by described pavement temperature delta data and by really Fixed pavement state result exports the 5th computing module to cloud processor；

5th computing module of cloud processor, for the described temperature variation data that will obtain and the different road surfaces shape preset State is mated by corresponding temperature history data base, determines the state on described road surface according to matching result；

If the pavement state that the pavement state that the 5th computing module of cloud processor obtains obtains with microprocessor is identical, then Do not process；If it is different, then with the pavement state of microprocessor output as final result, and update historical data base.

Further, the pavement skid resistance determination unit for testing pavement skid resistance condition, the 6th of microprocessor are also included Pretreatment module, the 6th computing module of Cloud Server,

Wherein, described pavement skid resistance condition determination unit include polylith same size, be placed in detection vehicle tyre and road surface Between pressure sensitive film, the area of pressure sensitive film is more than or equal to the face of tire and road surface interface to be tested Long-pending；Detection car makes tire press on pressure sensitive film with its deadweight, and standing is until the contacting of pressure sensitive film and tire Pressure distribution state during pressurized is demonstrated completely on face；

6th pretreatment module of microprocessor, for being scanned pressure sensitive film, obtains every piece of pressure sensitive The pressure values of each test point on film, determines that every piece of pressure sensitive film is 0～0.2MPa with pressure values on tire contact plane Area M, calculate the area of P=M/ pressure sensitive film and tire contact plane；

Relatively P value and pre-stored threshold values, the pavement skid resistance condition on evaluation road surface:

If P value is more than or equal to pre-stored threshold values, then the antiskid excellent performance on road surface；

If P value is less than pre-stored threshold values, then the antiskid poor-performing on road surface.

Further, according to formula a_y(t)=a_measured_x(t)·sinθ+a_measured_yT () cos θ-g, asks Go out described surface evenness detector acceleration of vibration a of in the vertical direction when t_y(t)；

To described acceleration of vibration a_yT () carries out secondary dual-integration computing, obtain described surface evenness detector t The displacement s of in the vertical direction vibration during the moment_y(t)；

According to formula w_y(t)=s_yT ()-h (t), obtains described surface evenness detector at described horizontal displacement s_xModel Enclose the surface evenness w on interior described road surface_y(t)。

Further, described thermal source feedway is active visible ray thermal light source or infrared laser source；Described temperature is examined Survey device, for infrared thermopile detector or non-refrigerate infrared focal plane array seeker.

Beneficial effect

State estimating system during pavement usage of the present invention and prior art possess following beneficial effect:

1, by obtaining the road surface types recognition result of multiple sensors, and the road surface types identification to each sensor is tied Fruit merges process, and using forward laser light radar as master reference, other sensors, as from sensor, recycle Ma Erke Road surface types after the merging of multiple sensors is known by husband's random field (Markov Radom Field, MRF) model energy equation Other result is optimized process, improves the accuracy rate that road surface identifies.

2, by utilizing two axle acceleration sensors measurement surface evenness detectors being perpendicular to surface evenness detection The first acceleration on instrument direction and at the second acceleration being parallel on described surface evenness detector direction, utilizes laser Distance measuring sensor measures the distance between surface evenness detector and road surface to be tested, utilizes mileage to count sensor measurement road The horizontal displacement of surface evenness detector, utilizes gyroscope to measure the rotation between surface evenness detector and level road Angle and utilize processor to process above-mentioned first acceleration, the second acceleration, distance, accurately measures road to be tested The surface evenness in face, has calibrated the error brought due to test vehicle chassis rotary motion, improves surface evenness detection The certainty of measurement of instrument.

3, utilize the collection of the speed controlling road pavement image of the pavement detection car collected, and gathered road by acquisition module Face image recognition goes out in the pavement image collected containing damaged pavement image；Further, to above-mentioned collect all Information processes such that it is able to calculate the position containing damaged pavement image by the described information collected.Can In the case of reducing cost, road pavement breakage positions neatly, and setting accuracy is higher.

Detailed description of the invention

The present invention will be further described below.

Embodiment 1

State estimating system during the present embodiment pavement usage, first road pavement type is measured, and then measures Whether the flatness on road surface, finally exist breakage on detection road surface；Described system includes:

Detection car；

Road surface kind determination unit, for measuring the type on road surface, output road surface category identification result；

Measuring flatness of road surface unit, for measuring the flatness on road surface；

Road surface breakage positioning unit, is used for detecting whether road surface has breakage, when road surface breakage being detected, gathers damaged road surface Image information, and obtain the positional information of described damaged road surface；

Wherein, road surface kind determination unit, including detection data acquisition module, the first pretreatment module of microprocessor, First computing module of Cloud Server；

Described detection data acquisition module, including be arranged on detection car on forward laser light radar, acceleration transducer, after Put laser radar and ccd image sensor；

First pretreatment module of microprocessor, including sensing data acquisition module, described preposition swashs for obtaining Optical radar, acceleration transducer, rearmounted laser radar and ccd image sensor；

Identification module, acceleration identification module and picture recognition module, described laser identification module, for obtaining respectively The laser data that forward laser light radar, rearmounted laser radar gather, generates road surface category identification result according to laser data；Described Acceleration identification module, for obtaining the acceleration information that acceleration transducer obtains, generates road according to described acceleration information Face category identification result；Picture recognition module, for obtaining the acceleration information that image acquisition device obtains, according to described picture number According to generating road surface category identification result；And by the first computing module of described road surface recognition result output to Cloud Server；

First computing module of described Cloud Server, merges module, the generation of road surface kind including road surface category identification result Unit, wherein said road surface category identification combining unit, the road surface kind result described in obtaining according to ballot method is entered Row merging treatment, obtains recognition result after the road surface category identification result merging of corresponding each sensor；Wherein, the Ma Er of establishment The energy equation of section's husband's random field models is:

$\begin{array}{c}E(x,y,u,w,v)=\mathrm{\ρ}\underset{i}{\mathrm{\Σ}}\sqrt{\left|x_i^2-x_{i-1}^2\right|}+\underset{i}{\mathrm{\Σ}}{r}_i\sqrt{\left|x_i^2-y_i^2\right|}+\\ \underset{i}{\mathrm{\Σ}}k{1}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^1\right)}^2\right|}+\underset{i}{\mathrm{\Σ}}k{2}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^2\right)}^2\right|}+\underset{i}{\mathrm{\Σ}}k{3}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^3\right)}^2\right|}+\mathrm{...}+\underset{i}{\mathrm{\Σ}}{\mathrm{kn}}_{i-1}\sqrt{\left|x_i^2-{\left(U_{i-1}^n\right)}^2\right|}\end{array}$

Wherein, i is the sequence number in identified section, and i is integer and i >=2, x_iFor the forward laser light that i-th section is to be optimized The road surface category identification result of radar, x_i-1It is the final road surface category identification result in the i-th-1 section, y_iBefore i-th section Put the road surface category identification result after the merging of laser radar, u_i-1It it is the road surface after the merging of the i-th-1 section imageing sensor Category identification result, w_i-1It is the road surface category identification result after the merging of the i-th-1 section acceleration transducer, v_i-1Be i-th- Road surface category identification result after the merging of 1 rearmounted laser radar in section, ρ is x_iWith x_i-1Between link potential energy, r_iFor x_i With y_iBetween link potential energy, k1_i-1For x_iWith u_i-1Between link potential energy, k2_i-1For x_iWith w_i-1Between link potential energy, k3_i-1For x_iWith v_i-1Between link potential energy；

By road surface category identification result x of forward laser light radar to be optimized for described i-th section_iCorresponding numerical value divides Do not bring described energy equation into calculate, will enable energy equation E (x, y, u, w, the x that value v) is minimum_iCorresponding road surface Kind is as described final road surface category identification result；

Measuring flatness of road surface unit, including the flatness detector being arranged on detection car, the second of microprocessor is pre- Processing module, the second computing module of Cloud Server；

Described roughness measurement instrument includes two axle acceleration sensors, laser range sensor, mileage sensor for countering, top Spiral shell instrument, wherein said two axle acceleration sensors are being perpendicular to for measuring described surface evenness detector when t The first acceleration a_measured on described surface evenness detector direction_y(t) and be parallel to described surface evenness The second acceleration a_measured on detector direction_x(t)；Described laser range sensor is used for measuring described evenness of road surface Degree detector distance h (t) when t and between road surface；Described mileage sensor for countering is used for measuring described road surface puts down The whole degree detector horizontal displacement s when t_x(t)；Described gyroscope is used for measuring described surface evenness detector Anglec of rotation θ (t) when t and between level road；

Second pretreatment module of described microprocessor, for obtaining the first described acceleration a_measured_y(t)、 Second acceleration a_measured_x(t), distance h (t) measured between described surface evenness detector and road surface, horizontal position Shifting amount s_x(t), anglec of rotation θ (t)；And the first acceleration a_measured described in obtaining_y(t), the second acceleration a_ measured_xT (), anglec of rotation θ (t) and gravity acceleration g export the second computing module to Cloud Server；

Described second computing module, according to when t, described the first acceleration a_measuredy (t), second Acceleration a_measuredx (t), anglec of rotation θ (t) and gravity acceleration g, when calculating described surface evenness detector t Acceleration of vibration ay (t) of in the vertical direction during quarter；According to described acceleration of vibration a_yT (), calculates described surface evenness The displacement s of in the vertical direction vibration during detector t_y(t)；According to described surface evenness detector and described road Distance h (t) between face and the displacement s of described surface evenness detector in the vertical direction vibration_yT (), obtains in institute State horizontal displacement s_xThe surface evenness w on described road surface in the range of (t)_y(t)；

Road surface breakage positioning unit, including the 3rd pretreatment module, the cloud of video camera, GPS locator and microprocessor 3rd computing module of server；

Described video camera obtains pavement image data, the geography position, road surface that GPS locator record current image frame is corresponding Confidence ceases, and exports view data, the 3rd pretreatment module of GPS location data to microprocessor；

Described 3rd pretreatment unit is used for carrying out described view data pretreatment, and by pretreated view data And the shooting time of view data and GPS location information export the 3rd computing module to Cloud Server；

Described 3rd computing module of described Cloud Server, for data separate image recognition side, pretreated road surface Method road pavement road surface data are identified obtaining the road surface data comprising breakage, it is thus achieved that comprise road surface data and this figure of breakage As shooting time and the GPS of data position information；

Wherein, described microprocessor is arranged on detection car, and described microprocessor and described Cloud Server are by wireless Network carries out data transmission.

According to formula a in the present embodiment_y(t)=a_measured_x(t)·sinθ+a_measured_y(t) cos θ-g, Obtain described surface evenness detector acceleration of vibration a of in the vertical direction when t_y(t)；

To described acceleration of vibration a_yT () carries out secondary dual-integration computing, obtain described surface evenness detector t The displacement s of in the vertical direction vibration during the moment_y(t)；

According to formula w_y(t)=s_yT ()-h (t), obtains described surface evenness detector at described horizontal displacement s_xModel Enclose the surface evenness w on interior described road surface_y(t)。

In this enforcement, by obtaining the road surface types recognition result of multiple sensors, and the road surface class to each sensor Type recognition result merges process, and using forward laser light radar as master reference, other sensors are as from sensor more sharp With the road after markov random file (Markov Radom Field, MRF) the model energy equation merging to multiple sensors Face type identification result is optimized process, improves the accuracy rate that road surface identifies.

It is being perpendicular to surface evenness detector by utilizing two axle acceleration sensors to measure surface evenness detector The first acceleration on direction and at the second acceleration being parallel on described surface evenness detector direction, utilizes Laser Measuring Distance between sensor measurement surface evenness detector and road surface to be tested, utilizes mileage to count sensor measurement road surface The horizontal displacement of flatness detector, utilizes gyroscope to measure the anglec of rotation between surface evenness detector and level road Spend and utilize processor that above-mentioned first acceleration, the second acceleration, distance are processed, accurately measuring road surface to be tested Surface evenness, calibrated due to the test error brought of vehicle chassis rotary motion, improve surface evenness detector Certainty of measurement.

Utilize the collection of the speed controlling road pavement image of the pavement detection car collected, and gathered road surface by acquisition module Image recognition goes out in the pavement image collected containing damaged pavement image；Further, to the above-mentioned all letters collected Breath processes such that it is able to calculate the position containing damaged pavement image by the described information collected.Can be In the case of reduction cost, road pavement breakage positions neatly, and setting accuracy is higher.

Embodiment 2

State estimating system during the present embodiment pavement usage, on the basis of embodiment 1, also includes pavement structure The stealthy Defect inspection unit of layer, described road surface structare layer stealth Defect inspection unit includes the GPR being arranged on detection car With excitation road surface sound-producing device, and the 4th pretreatment module of microprocessor, the 4th computing module of Cloud Server；

Wherein, excitation road surface sound-producing device taps road surface continuously so that produce the sound wave of road vibration because tapping road surface Along road surface structare layer by surface layer travel downward, road GPR launches electromagnetic wave to road surface simultaneously, is sent out by road GPR Penetrate the time synchronized of signal and generating means percussion road surface, excitation road surface；

4th pretreatment module of microprocessor, for obtaining the electromagnetic wave propagation time in road surface structare layer, according to The electromagnetic wave propagation time in road surface structare layer calculates the thickness of every layer in road surface structare layer, and exports this thickness to cloud clothes 4th computing module of business device, thickness equations is as follows:

$h_1=\frac{c_0{t}_1}{2\sqrt{{\mathrm{\ξ}}_1}};h_2=\frac{c_0(t_2-t_1)}{2\sqrt{{\mathrm{\ξ}}_2}};......;h_n=\frac{c_0(t_n-t_{n-1})}{2\sqrt{{\mathrm{\ξ}}_n}},$

Wherein, h₁, h₂..., h_nRepresent the thickness of pavement structure every layer, c₀The electromagnetic wave representing road GPR exists The speed propagated under vacuum state, t₁, t₂..., t_nRepresent the electromagnetic wave of road GPR and arrive in road surface structare layer every layer The propagation time of structure, ξ₁, ξ₂..., ξ_nRepresent the dielectric constant of every layer in road surface structare layer；

4th pretreatment module of microprocessor, for obtaining the sound wave the encouraging road surface sound-producing device biography on every layer of road surface Between sowing time, according to the sound wave of the excitation road surface sound-producing device propagation time T on every layer of road surface_p1, T_p2..., T_pn, and road surface knot The thickness h of every layer in structure layer₁, h₂..., h_n, it is thus achieved that sound wave is the spread speed of every layer in road surface structare layer, and exports this and wear Broadcasting speed to the 4th computing module of Cloud Server, it is as follows that spread speed calculates formula:

$v_1=\frac{h_1}{T_{P1}};v_2=\frac{h_2}{T_{P2}-T_{P1}};......v_n=\frac{h_n}{T_{Pn}-T_{Pn-1}};$

4th computing module of Cloud Server, for according to SVEL V_iWith modulus of resilience E_iRelation formula, calculate Modulus of resilience E of every layer in road surface structare layer_i, analyze the decay of the strength of materials in road surface structare layer: wherein, V_iRepresent sound wave to exist Spread speed v of i-th layer in road surface structare layer_i, ρ_iRepresent the density of the i-th layer material in road surface structare layer.In road surface structare layer The density p of i layer_iRecord by the section detected is carried out coring test.

The present embodiment can detect pavement disease initial situation in real time, and follows the tracks of its development, analyzes road surface Disease development trend, has preferable real-time, and detection efficiency is high；And the method pavement disease detect in application can Save highway maintenance cost, it is ensured that traffic safety, extend highway service life.

Embodiment 3

State estimating system during the present embodiment pavement usage, on the basis of embodiment 2, also includes that thermal source affects Road surface determination unit, provides road pavement to provide non-contact heat source for road pavement, monitors the variations in temperature on described road surface, according to Described temperature variation data determines the state on described road surface；Including thermal source feedway, device for detecting temperature, the of microprocessor Five pretreatment module, the 5th processing module of Cloud Server, wherein said thermal source feedway is active visible ray thermal light source Or infrared laser source；Described temperature-detecting device, for infrared thermopile detector or non-refrigerate infrared focal plane array seeker.

Thermal source feedway, provides non-contact heat source for road pavement, makes the temperature on described road surface change, thermal source Feedway is active visible ray thermal light source or infrared laser source；

Temperature-detecting device, for monitoring the variations in temperature on described road surface, it is thus achieved that temperature variation data, and by described temperature Delta data exports to microprocessor；

5th pretreatment module of microprocessor, for obtaining temperature variation curve, root according to described temperature variation data The state on described road surface is determined according to the slope of described temperature variation curve and slope variation；Slope when described temperature variation curve Constant, and when slope is more than or equal to first threshold, determine that described pavement state is drying regime；When described temperature variation curve Slope is constant, and when slope is less than described first threshold and is more than or equal to Second Threshold, determines that described pavement state is hydrops shape State；When the slope of described temperature variation curve is not 0, and when at a time becoming big suddenly, determine that described pavement state is long-pending Water state；When the slope of described temperature variation curve is occurred jumping characteristic change in certain a period of time by zero, determine described road surface State is icing condition；5th pretreatment module of microprocessor, is additionally operable to by described pavement temperature delta data and by really Fixed pavement state result exports the 5th computing module to cloud processor；

5th computing module of cloud processor, for the described temperature variation data that will obtain and the different road surfaces shape preset State is mated by corresponding temperature history data base, determines the state on described road surface according to matching result；

If the pavement state that the pavement state that the 5th computing module of cloud processor obtains obtains with microprocessor is identical, then Do not process；If it is different, then with the pavement state of microprocessor output as final result, and update historical data base.

The present embodiment, it is thus achieved that temperature variation data, and then determine pavement state according to the trend of variations in temperature.Due to This supply thermal source and temperature monitoring are contactless, so avoiding the damage to pavement of road, and, this variations in temperature number According to being directly to obtain at road surface, react pavement state more accurately, improve the accuracy of measurement.

Embodiment 4

State estimating system during the present embodiment pavement usage, on the basis of embodiment 3, also includes for testing The pavement skid resistance determination unit of pavement skid resistance condition, the 6th pretreatment module of microprocessor, the 6th computing mould of Cloud Server Block,

Wherein, described pavement skid resistance condition determination unit include polylith same size, be placed in detection vehicle tyre and road surface Between pressure sensitive film, the area of pressure sensitive film is more than or equal to the face of tire and road surface interface to be tested Long-pending；Detection car makes tire press on pressure sensitive film with its deadweight, and standing is until the contacting of pressure sensitive film and tire Pressure distribution state during pressurized is demonstrated completely on face；

6th pretreatment module of microprocessor, for being scanned pressure sensitive film, obtains every piece of pressure sensitive The pressure values of each test point on film, determines that every piece of pressure sensitive film is 0～0.2MPa with pressure values on tire contact plane Area M, calculate the area of P=M/ pressure sensitive film and tire contact plane；

Relatively P value and pre-stored threshold values, the pavement skid resistance condition on evaluation road surface:

If P value is more than or equal to pre-stored threshold values, then the antiskid excellent performance on road surface；

If P value is less than pre-stored threshold values, then the antiskid poor-performing on road surface.

The present embodiment, uses pressure sensitive film to measure the pressure distribution of tire and road surface, and then evaluates skid resistance of pavement Can, artifical influence factor is little, and evaluation precision is high, overcomes existing pavement skid resistance evaluation methodology anthropic factor impact big, and result is inclined The problem that difference is big.Implement step simple, easy to use, mainly have employed the pressure sensitive film that price is relatively low, cost is relatively low, easily In universal.Be not limited to existing from testing friction coefficient and construction depth the two index to consider pavement skid resistance condition, but The angle road pavement antiskid performance from tire with road surface pressure distribution characteristic of novelty is evaluated, and breaches art technology The thinking model that personnel are usual.

In the various embodiments described above, the concrete steps of laser radar identification road surface types include:

Gather coordinate data and the reflected energy data of the sampled point of laser radar output；

The road surface outline data of Cartesian form, the number of coordinates of wherein said sampled point is generated according to described coordinate data Change into Cartesian form according to by polar form, utilize following formula to realize:

x_i=r_isin(θ_i), z_i=H-r_icos(θ_i)

Wherein, i is the sequence number of the sampled point of laser radar, r_iFor the polar data of the sampled point of laser radar, θ_iIt is sharp Light light beam and laser radar are perpendicular to the angle in direction, ground, and H is the laser radar absolute setting height(from bottom) relative to ground grading, The center of circle of described rectangular coordinate system is positioned at the laser emission point of laser radar, x_iFor horizontal relative to laser emission point of sampled point Coordinate, z_iFor sampled point relative to the longitudinal coordinate of laser emission point；

Described road surface outline data is processed, generation transverse axis be spatial frequency, the longitudinal axis be the road of spatial frequency component Facial contour spatial frequency data；

In units of the S of horizontal displacement interval, described road surface outline data is carried out segmentation, generates n road surface outline data Sample；

Use Lomb algorithm respectively described n road surface profile data sample to be processed, generate and comprise n transverse axis for sky Between frequency, the longitudinal axis be the road surface profile spatial frequency data of sub-road surface profile spatial frequency data of spatial frequency component；

Described road surface profile characteristic and described reflected energy data are merged process, generates grader input number According to；

By the input data input of described grader to trained grader, thus obtain road surface classification results.

The method generating road surface profile characteristic specifically includes:

According to initial frequency set in advance, step pitch frequency and termination frequency, calculate every sub-road wheel exterior feature spatial frequency In data, from the beginning of described initial frequency, to described termination frequency, spatial frequency component corresponding in every section of step pitch frequency Sum, thus generate n sub-road surface profile spatial frequency features data, described sub-road surface profile spatial frequency features packet contains The sum of m spatial frequency component；

Described n sub-road surface profile spatial frequency features data are combined, generate described first via facial contour feature Matrix Y1:

Wherein, the every string in described first via facial contour eigenmatrix Y1 corresponds to a described sub-road surface profile space Frequency characterization data, each element in described first via facial contour eigenmatrix Y1 divides corresponding to a described spatial frequency The sum of amount；

Described reflected energy data is specially the first reflected energy eigenmatrix Y2:

Wherein, the every string in described road surface profile eigenmatrix Y2 corresponds to a road surface profile data sample, described Each element in first reflected energy eigenmatrix Y2 is corresponding to the reflected energy numerical value of the sampled point of laser radar.

By laser radar coordinate acquisition data and reflected energy data, and carry out coordinate data processing generation road wheel Wide characteristic, utilizes the acceleration information of output of acceleration sensor to the coordinate data of sampled point or road surface outline data It is corrected, based on the road surface profile characteristic after correction and reflected energy data, it becomes possible to road pavement type is identified, Get rid of laser radar measurement error produced by up-down vibration in vehicle travel process simultaneously, thus improve road surface and know Other scope and accuracy rate.

In the various embodiments described above, the 3rd pretreatment unit of described microprocessor, the preprocess method of road pavement image Including:

The view data obtained is carried out color analysis and gray processing processes, coloured image is converted to gray level image；

View data is made discrete Daubechies4 wavelet transformation, obtains the wavelet conversion coefficient of view data, The high frequency filter coefficient [G0 G1 G2 G3] of Daubechies4 small echo and low-frequency filter coefficient [H0 H1 H2 G3] are respectively For:

$G_0=\frac{(1+\sqrt{3})}{4\sqrt{2}},G_1=\frac{(3+\sqrt{3})}{4\sqrt{2}},G_2=\frac{(3-\sqrt{3})}{4\sqrt{2}},G_3=\frac{(1-\sqrt{3})}{4\sqrt{2}}$

$H_0=\frac{(1-\sqrt{3})}{4\sqrt{2}},H_1=\frac{(-3+\sqrt{3})}{4\sqrt{2}},H_2=\frac{(3+\sqrt{3})}{4\sqrt{2}},H_3=\frac{(-1-\sqrt{3})}{4\sqrt{2}};$

The window using n × n is smooth mobile along view data, calculates being correlated with between view data wavelet conversion coefficient Property, obtain the view data after shrinking weights, wherein the computing formula of dependency R is:

$R=\frac{\mathrm{\Σ}XY-\frac{\mathrm{\Σ}X\·\mathrm{\Σ}Y}{n}}{\sqrt{{\mathrm{\ΣX}}^2-\frac{{\left(\mathrm{\Σ}X\right)}^2}{n}\]\·\[{\mathrm{\ΣY}}^2-\frac{{\left(\mathrm{\Σ}Y\right)}^2}{n}\]}}$

In above formula, X, Y are the combination of any two n in image × n window subimage, ∑_X, ∑_YIt is respectively subgraph gray value Summation, ∑_XYFor suing for peace after two subgraph correspondence position gray value phase products of X, Y；

View data after shrinking weights is carried out soft-threshold process, and the computing formula of soft-threshold λ is:

$\mathrm{\λ}=\mathrm{\σ}\sqrt{2ln\left(N\right)}$

In above formula, N × N is to shrink the size of view data after weights, and σ is noise variance；

View data is carried out discrete Daubechies4 inverse wavelet transform, obtains the view data after denoising；

View data after denoising is strengthened.

To the present invention it should be appreciated that embodiment described above, to the purpose of the present invention, technical scheme and useful effect Fruit carried out further details of explanation, these are only embodiments of the invention, be not intended to limit the present invention, every Within the spiritual principles of the present invention, done any modification, equivalent substitution and improvement etc., should be included in the protection of the present invention Within the scope of, protection scope of the present invention should be as the criterion with the protection domain that claim is defined.

Claims (10)

1. the state estimating system during a pavement usage, it is characterised in that first road pavement type is measured, then Measure the flatness on road surface, finally whether there is breakage on detection road surface；Described system includes:

Detection car；

Road surface kind determination unit, for measuring the type on road surface, output road surface category identification result；

Measuring flatness of road surface unit, for measuring the flatness on road surface；

Road surface breakage positioning unit, is used for detecting whether road surface has breakage, when road surface breakage being detected, gathers the figure of damaged road surface As information, and obtain the positional information of described damaged road surface；

Wherein, road surface kind determination unit, including detection data acquisition module, the first pretreatment module of microprocessor, cloud clothes First computing module of business device；

Described detection data acquisition module, including be arranged on detection car on forward laser light radar, acceleration transducer, rearmounted swash Optical radar and ccd image sensor；

First pretreatment module of microprocessor, including sensing data acquisition module, for obtaining described forward laser light thunder Reach, acceleration transducer, rearmounted laser radar and ccd image sensor；

Identification module, acceleration identification module and picture recognition module, described laser identification module, preposition for obtaining respectively The laser data that laser radar, rearmounted laser radar gather, generates road surface category identification result according to laser data；Described acceleration Degree identification module, for obtaining the acceleration information that acceleration transducer obtains, generates road surface kind according to described acceleration information Class recognition result；Picture recognition module, for obtaining the acceleration information that image acquisition device obtains, raw according to described view data Become road surface category identification result；And by the first computing module of described road surface recognition result output to Cloud Server；

First computing module of described Cloud Server, merges module, road surface kind signal generating unit including road surface category identification result, Wherein said road surface category identification combining unit, the road surface kind result described in obtaining according to ballot method merges Process, obtain recognition result after the road surface category identification result merging of corresponding each sensor；Wherein, the Markov of establishment with The energy equation of airport model is:

Wherein, i is the sequence number in identified section, and i is integer and i >=2, and xi is the forward laser light radar that i-th section is to be optimized Road surface category identification result, x_i-1It is the final road surface category identification result in the i-th-1 section, y_iSwash for i-th section is preposition Road surface category identification result after the merging of optical radar, u_i-1It it is the road surface kind after the merging of the i-th-1 section imageing sensor Recognition result, w_i-1It is the road surface category identification result after the merging of the i-th-1 section acceleration transducer, v_i-1It is the i-th-1 Road surface category identification result after the merging of the rearmounted laser radar in section, ρ is x_iWith x_i-1Between link potential energy, r_iFor x_iWith y_i Between link potential energy, k1_i-1For x_iWith u_i-1Between link potential energy, k2_i-1For x_iWith w_i-1Between link potential energy, k3_i-1For x_iWith v_i-1Between link potential energy；

By road surface category identification result x of forward laser light radar to be optimized for described i-th section_iCorresponding numerical value carries respectively Enter described energy equation to calculate, will enable energy equation E (x, y, u, w, the x that value v) is minimum_iCorresponding road surface kind As described final road surface category identification result；

Measuring flatness of road surface unit, including the flatness detector being arranged on detection car, the second pretreatment of microprocessor Module, the second computing module of Cloud Server；

Described roughness measurement instrument includes two axle acceleration sensors, laser range sensor, mileage sensor for countering, gyro Instrument, wherein said two axle acceleration sensors for measuring described surface evenness detector being perpendicular to when t State the first acceleration a_measured on surface evenness detector direction_y(t) and be parallel to described surface evenness inspection Survey the second acceleration a_measured on instrument direction_x(t)；Described laser range sensor is used for measuring described surface evenness Detector distance h (t) when t and between road surface；Described mileage sensor for countering is used for measuring described evenness of road surface The degree detector horizontal displacement s when t_x(t)；Described gyroscope is used for measuring described surface evenness detector and exists Anglec of rotation θ (t) during t and between level road；

Second pretreatment module of described microprocessor, for obtaining the first described acceleration a_measured_y(t), second add Speed a_measured_x(t), distance h (t) measured between described surface evenness detector and road surface, horizontal displacement s_x (t), anglec of rotation θ (t)；And the first acceleration a_measured described in obtaining_y(t), the second acceleration a_ measured_xT (), anglec of rotation θ (t) and gravity acceleration g export the second computing module to Cloud Server；

Described second computing module, according to when t, described the first acceleration a_measuredy (t), second accelerates Degree a_measuredx (t), anglec of rotation θ (t) and gravity acceleration g, when calculating described surface evenness detector t Acceleration of vibration ay (t) of in the vertical direction；According to described acceleration of vibration a_yT (), calculates the detection of described surface evenness The displacement s of in the vertical direction vibration during instrument t_y(t)；According to described surface evenness detector and described road surface it Between distance h (t) and described surface evenness detector in the vertical direction vibration displacement s_yT (), obtains at described water Flat displacement s_xThe surface evenness w on described road surface in the range of (t)_y(t)；

Road surface breakage positioning unit, including the 3rd pretreatment module, the cloud service of video camera, GPS locator and microprocessor 3rd computing module of device；

Described video camera obtains pavement image data, geographical position, the road surface letter that GPS locator record current image frame is corresponding Breath, and export view data, the 3rd pretreatment module of GPS location data to microprocessor；

Described 3rd pretreatment unit for described view data being carried out pretreatment, and by pretreated view data and The shooting time of view data and GPS location information export the 3rd computing module to Cloud Server；

Described 3rd computing module of described Cloud Server, for pretreated road surface data separate image-recognizing method pair Road surface, road surface data are identified obtaining the road surface data comprising breakage, it is thus achieved that comprise road surface data and this picture number of breakage According to shooting time and GPS position information；

Wherein, described microprocessor is arranged on detection car, and described microprocessor and described Cloud Server pass through wireless network Carry out data transmission.

State estimating system during pavement usage the most according to claim 1, it is characterised in that described micro-process 3rd pretreatment unit of device, the preprocess method of road pavement image includes:

The view data obtained is carried out color analysis and gray processing processes, coloured image is converted to gray level image；

View data is made discrete Daubechies4 wavelet transformation, obtains the wavelet conversion coefficient of view data, The high frequency filter coefficient [G0 G1 G2 G3] of Daubechies4 small echo and low-frequency filter coefficient [H0 H1 H2 G3] are respectively For:

The window using n × n is smooth mobile along view data, calculates the dependency between view data wavelet conversion coefficient, Obtaining the view data after shrinking weights, wherein the computing formula of dependency R is:

In above formula, X, Y are the combination of any two n in image × n window subimage,It is respectively subgraph gray value total With,For suing for peace after two subgraph correspondence position gray value phase products of X, Y；

View data after shrinking weights is carried out soft-threshold process, and the computing formula of soft-threshold λ is:

In above formula, N × N is to shrink the size of view data after weights, and σ is noise variance；

View data is carried out discrete Daubechies4 inverse wavelet transform, obtains the view data after denoising；

View data after denoising is strengthened.

State estimating system during pavement usage the most according to claim 1, it is characterised in that laser radar identification The concrete steps of road surface types include:

Gather coordinate data and the reflected energy data of the sampled point of laser radar output；

According to described coordinate data generate Cartesian form road surface outline data, the coordinate data of wherein said sampled point by Polar form changes into Cartesian form, utilizes following formula to realize:

Wherein, i is the sequence number of the sampled point of laser radar, r_iFor the polar data of the sampled point of laser radar, θ_iFor laser light Bundle and laser radar are perpendicular to the angle in direction, ground, and H is the laser radar absolute setting height(from bottom) relative to ground grading, described The center of circle of rectangular coordinate system is positioned at the laser emission point of laser radar, x_iSit relative to the horizontal of laser emission point for sampled point Mark, z_iFor sampled point relative to the longitudinal coordinate of laser emission point；

Described road surface outline data is processed, generation transverse axis be spatial frequency, the longitudinal axis be the road wheel of spatial frequency component Wide spatial frequency data；

In units of the S of horizontal displacement interval, described road surface outline data is carried out segmentation, generates n road surface profile data sample；

Using Lomb algorithm to process described n road surface profile data sample respectively, it is space frequency that generation comprises n transverse axis Rate, the longitudinal axis are the road surface profile spatial frequency data of the sub-road surface profile spatial frequency data of spatial frequency component；

Described road surface profile characteristic and described reflected energy data are merged process, generates grader input data；

By the input data input of described grader to trained grader, thus obtain road surface classification results.

State estimating system during pavement usage the most according to claim 3, it is characterised in that generate road surface profile The method of characteristic specifically includes:

According to initial frequency set in advance, step pitch frequency and termination frequency, calculate every sub-road wheel exterior feature spatial frequency data In, from the beginning of described initial frequency, to described termination frequency, spatial frequency component corresponding in every section of step pitch frequency With, thus generating n sub-road surface profile spatial frequency features data, described sub-road surface profile spatial frequency features packet contains m The sum of individual spatial frequency component；

Described n sub-road surface profile spatial frequency features data are combined, generate described first via facial contour eigenmatrix Y1:

Wherein, the every string in described first via facial contour eigenmatrix Y1 corresponds to a described sub-road surface profile spatial frequency Characteristic, each element in described first via facial contour eigenmatrix Y1 is corresponding to a described spatial frequency component With；

Described reflected energy data is specially the first reflected energy eigenmatrix Y2:

Wherein, the every string in described road surface profile eigenmatrix Y2 correspond to a road surface profile data sample, described first Each element in reflected energy eigenmatrix Y2 is corresponding to the reflected energy numerical value of the sampled point of laser radar.

State estimating system during pavement usage the most according to claim 1, it is characterised in that also include that road surface is tied Structure layer stealth Defect inspection unit, described road surface structare layer stealth Defect inspection unit includes the spy Rhizoma Anemones flaccidae being arranged on detection car Reach and encourage road surface sound-producing device, and the 4th pretreatment module of microprocessor, the 4th computing module of Cloud Server；

Wherein, excitation road surface sound-producing device taps road surface continuously so that produce the sound wave of road vibration along road because tapping road surface Face structure sheaf is by surface layer travel downward, and road GPR launches electromagnetic wave to road surface simultaneously, and road GPR is launched letter Number tap the time synchronized on road surface with excitation road surface generating means；

4th pretreatment module of microprocessor, for obtaining the electromagnetic wave propagation time in road surface structare layer, according to electromagnetism The ripple propagation time in road surface structare layer calculates the thickness of every layer in road surface structare layer, and exports this thickness to Cloud Server The 4th computing module, thickness equations is as follows:

Wherein, h₁, h₂..., h_nRepresent the thickness of pavement structure every layer, c₀Represent the electromagnetic wave of road GPR at vacuum shape The speed propagated under state, t₁, t₂..., t_nThe electromagnetic wave representing road GPR arrives every Rotating fields in road surface structare layer Propagation time, ξ₁, ξ₂..., ξ_nRepresent the dielectric constant of every layer in road surface structare layer；

4th pretreatment module of microprocessor, for obtaining the sound wave of excitation road surface sound-producing device when the propagation on every layer of road surface Between, according to the sound wave of the excitation road surface sound-producing device propagation time T on every layer of road surface_p1, T_p2..., T_pn, and road surface structare layer In the thickness h of every layer₁, h₂..., h_n, it is thus achieved that sound wave is the spread speed of every layer in road surface structare layer, and exports this and wear and broadcast speed Spending the 4th computing module to Cloud Server, it is as follows that spread speed calculates formula:

4th computing module of Cloud Server, for according to SVEL V_iWith modulus of resilience E_iRelation formula, calculate road surface Modulus of resilience E of every layer in structure sheaf_I,Analyze the decay of the strength of materials in road surface structare layer: wherein, V_iRepresent sound wave on road surface Spread speed v of i-th layer in structure sheaf_i, ρ_iRepresent the density of the i-th layer material in road surface structare layer.

State estimating system during pavement usage the most according to claim 1, it is characterised in that in road surface structare layer The density p of i-th layer_iRecord by the section detected is carried out coring test.

State estimating system during pavement usage the most according to claim 1, it is characterised in that also include thermal source shadow Ring road surface determination unit, provide road pavement to provide non-contact heat source for road pavement, monitor the variations in temperature on described road surface, root The state on described road surface is determined according to described temperature variation data；Including thermal source feedway, device for detecting temperature, microprocessor 5th pretreatment module, the 5th processing module of Cloud Server, wherein

Thermal source feedway, provides non-contact heat source for road pavement, makes the temperature on described road surface change, and thermal source supplies Device is active visible ray thermal light source or infrared laser source；

Temperature-detecting device, for monitoring the variations in temperature on described road surface, it is thus achieved that temperature variation data, and by described variations in temperature Data export to microprocessor；

5th pretreatment module of microprocessor, for obtaining temperature variation curve according to described temperature variation data, according to institute State the slope of temperature variation curve and slope variation determines the state on described road surface；When described temperature variation curve slope not Become, and when slope is more than or equal to first threshold, determine that described pavement state is drying regime；Oblique when described temperature variation curve Rate is constant, and when slope is less than described first threshold and is more than or equal to Second Threshold, determines that described pavement state is hydrops state； When the slope of described temperature variation curve is not 0, and when at a time becoming big suddenly, determine that described pavement state is hydrops shape State；When the slope of described temperature variation curve is occurred jumping characteristic change in certain a period of time by zero, determine described pavement state For icing condition；5th pretreatment module of microprocessor, is additionally operable to described pavement temperature delta data and will determine Pavement state result exports the 5th computing module to cloud processor；

5th computing module of cloud processor, for the described temperature variation data that will obtain and the different pavement states pair preset The temperature history data base answered is mated, and determines the state on described road surface according to matching result；

If the pavement state that the pavement state that the 5th computing module of cloud processor obtains obtains with microprocessor is identical, do not do Process；If it is different, then with the pavement state of microprocessor output as final result, and update historical data base.

State estimating system during pavement usage the most according to claim 1, it is characterised in that also include for surveying The pavement skid resistance determination unit of examination pavement skid resistance condition, the 6th pretreatment module of microprocessor, the 6th computing of Cloud Server Module,

Wherein, described pavement skid resistance condition determination unit include polylith same size, be placed in detection vehicle tyre and road surface between Pressure sensitive film, the area of pressure sensitive film is more than or equal to the area of tire and road surface interface to be tested；Inspection Measuring car makes tire press on pressure sensitive film with its deadweight, stands until pressure sensitive film is complete with on the contact surface of tire Entirely demonstrate pressure distribution state during pressurized；

6th pretreatment module of microprocessor, for being scanned pressure sensitive film, obtains every piece of pressure sensitive film The pressure values of upper each test point, determines every piece of pressure sensitive film and the face that pressure values on tire contact plane is 0～0.2MPa Long-pending M, calculates the area of P=M/ pressure sensitive film and tire contact plane；

Relatively P value and pre-stored threshold values, the pavement skid resistance condition on evaluation road surface:

If P value is more than or equal to pre-stored threshold values, then the antiskid excellent performance on road surface；

If P value is less than pre-stored threshold values, then the antiskid poor-performing on road surface.

9. require the state estimating system during the pavement usage described in 1 according to power, it is characterised in that according to formula a_y(t)= a_measured_x(t)·sinθ+a_measured_yT () cos θ-g, obtains described surface evenness detector in t Time in the vertical direction acceleration of vibration a_y(t)；

To described acceleration of vibration a_yT () carries out secondary dual-integration computing, obtain described surface evenness detector t Time in the vertical direction vibration displacement s_y(t)；

According to formula w_y(t)=s_yT ()-h (t), obtains described surface evenness detector at described horizontal displacement s_xIn the range of The surface evenness w on described road surface_y(t)。

State estimating system during pavement usage the most according to claim 7, it is characterised in that described thermal source supplies It is active visible ray thermal light source or infrared laser source to device；Described temperature-detecting device, for infrared thermopile detector or Non-refrigerate infrared focal plane array seeker.