CN205239485U - Heavy road train parameter estimation system - Google Patents

Abstract

This paper discloses a heavy road train parameter estimation system. Aim at is according to heavy road train environment perception apparatus information and sensor information real -time online estimate road grade, whole car quality and barycenter position. This method considers fully that the vehicle angle of pitch to slope estimated accuracy's influence, calculates according to environment perception apparatus information and estimates that road grade subtracts and calculate the tractor angle of pitch promptly for actual road grade according to air spring height sensor, the quality estimates fully to consider slope resistance and aerodynamic drag's influence, regards as the unknown quantity to estimate aerodynamic drag, improves estimated accuracy, estimate the barycenter position according to automatically controlled air braking system pressure sensor information.

Description

A kind of Heavy Duty Truck parameter estimation system

Technical field

The utility model is designed for vehicle, especially the parameter estimation system of Heavy Duty Truck.

Background technology

Along with the development of electronic technology, the integrated control of automobile electric control system improves automotive electronics, intelligent level greatly, in automobile active safety andMore and more important effect is being brought into play in driver comfort aspect. Electric-control system judges travel condition of vehicle by vehicle parameter information and then controls certainlyPlan. Therefore, accurately vehicle parameter information is particularly important for electric-control system. Although sensor can provide vehicle and environmental information to electric-control system,As the speed of a motor vehicle, wheel speed, acceleration and brake pressure etc.; But road grade, complete vehicle quality and centroid position etc. are difficult to pass through sensor measurement. Setting upOn the basis of auto model, utilize control algolithm estimated sensor to become gradually research focus by measurement parameter.

It is strongly non-linear that Heavy Duty Truck has motion, the feature such as between tractor and trailer brake coordination is poor. In addition, Heavy Duty Truck existsIn transportation loading mass change greatly, driver when braking in the impercipient situation of loading condition, easily there is braking deficiency or brakedDegree, reduces braking safety and comfortableness; In braking procedure, centroid position changes, and axle load shifts, and affects braking force distribution control effect.

Document " SimultaneousMassandTime-VaryingGradeEstimationforHeavy-D utyVehicles ", " ExperimentsforOnlineEstimationofHeavyVehicle’sMassandTime-VaryingRoadGrade”,“RecursiveLeastSquareswithForgettingforOnlineEstimationofVehicleMassandRoadGrade:T heoryandExperiments " employing is minimum respectively in propositionSquare law and state observer carry out road grade and heavy vehicle complete vehicle quality estimate to reach with less instrument estimate vehicle mass simultaneously, the resistance of travellingThe object of power and road grade.

Document " On-boardpayloadidentificationforcommercialvehicles " has designed a kind of commercial car matter based on airsuspension systemHeart height On-line Estimation device. Electronics load monitoring system is estimated commercial car load-carrying by air suspension left and right sides air pressure difference, utilizes band external sourceInput automatic returning model least square method is estimated height of center of mass. Due to road excitation, air spring intraluminal pressure, in disorderly change procedure, utilizesThe force value that pressure sensor obtains is inaccurate.

Document " ParameterIdentificationofaVehicleforAutomaticPlatooningC ontrol " is by between load and Suspension DeformationRelation is estimated four axle truck load and centroid positions. Pressure sensor between tire and ground is measured axle load, and line decoder is measured truck front twoAxle spring deflection, by obtaining truck centroid position to analysis of experimental data. But the use of sensor has increased estimated cost greatly.

Existing heavy vehicle parameter identification method one class is based on model, and a class is based on sensor. Method based on model is because model and algorithm answersPolygamy real-time is poor; Sensor-based method is due to the expensive practicality that greatly reduces of sensor.

Utility model content

The utility model object is to provide a kind of vehicle parameter identification system for Heavy Duty Truck, meets precision and requirement of real-time simultaneously. VapourThe car team that car team car is made up of automobile or tractor and trailer, is mainly divided into full trailer train, semi-trailer train, double trailer train and lengthFour kinds of goods truck combinations. Research object is herein semi-trailer train.

For this reason, the utility model provides a kind of Heavy Duty Truck parameter identification method, comprises the following steps:

1) Longitudinal Dynamic Model while setting up vehicle drive system kinetic model, braking;

2) set up parameter identification equation by described vehicle dynamic model, described parameter identification equation with air spring height, brake pressure,Heavy vehicle positional informations etc. are as input, and road grade, heavy vehicle quality and centroid position are as unknown quantity;

3) in the time of normal vehicle operation, obtain car status information, as the speed of a motor vehicle, longitudinal acceleration, brake pressure;

4) calculate the tractor angle of pitch according to the air spring elevation information of automatically controlled air spring system;

5) according to the current value of slope of navigation system positional information calculation;

6) calculating value of slope deducts the tractor angle of pitch and is the real road gradient;

7) estimate complete vehicle quality;

8), in the time that braking deceleration is greater than lowest threshold, centroid position is estimated.

Wherein, in step 8) in, change the impact on barycenter position estimation accuracy for reducing braking deceleration, be greater than minimum threshold at braking decelerationWhen value, at interval of 0.5m/s^-2Estimate one time centroid position.

According to a kind of preferred embodiment of the present utility model, the air drag of heavy vehicle is estimated as unknown quantity, thereby is improved parameter EstimationPrecision.

According to a kind of preferred embodiment of the present utility model, in heavy vehicle running, estimate in real time road actual grade.

According to a kind of preferred embodiment of the present utility model, the quality of heavy vehicle is estimated to carry out in accelerator.

According to a kind of preferred embodiment of the present utility model, the centroid position of heavy vehicle is estimated to carry out in braking procedure.

According to a kind of preferred embodiment of the present utility model, heavy vehicle assembling electronic control air suspension system.

According to a kind of preferred embodiment of the present utility model, heavy vehicle assembling electronic control pneumatic brake system.

According to a kind of preferred embodiment of the present utility model, heavy vehicle assembling global positioning system.

According to a kind of preferred embodiment of the present utility model, the height that the air spring elevation information of heavy vehicle is given by electronic control air suspension system passesSensor records.

According to a kind of preferred embodiment of the present utility model, the brake pressure of heavy vehicle is recorded by the pressure sensor of electronic control pneumatic brake system.

According to a kind of preferred embodiment of the present utility model, the run location information of heavy vehicle is provided by navigation system.

According to a kind of preferred embodiment of the present utility model, in accelerator, estimate complete vehicle quality, defeated using rough estimate quality as centroid position moduleEnter amount, centroid position estimation module is estimated complete vehicle quality and centroid position.

The utility model provides a kind of heavy vehicle parameter identification system, comprising: environment sensing module, car status information acquisition module, roadGradient estimation module, quality estimation module and centroid position estimation module, its formation is utilized above-mentioned heavy vehicle method for parameter estimation of the present utility modelEstimate road grade, quality and centroid position.

Advantage of the present utility model is: 1) utilize electronic control air suspension system air spring elevation information to estimate the vehicle angle of pitch; 2) road gradeAlgorithm for estimating takes into full account the impact of the vehicle angle of pitch; 2) while estimating complete vehicle quality in accelerator, air drag is estimated as unknown quantity, carriedHigh estimated accuracy; 3) utilize electronic control pneumatic brake system pressure sensor, in braking procedure, estimate centroid position; 4) based on information fusion technologyVehicle parameter method of estimation improve computational speed, realize in real time estimate.

Brief description of the drawings

Certain preferred embodiments of the present utility model is described below with reference to accompanying drawings, in the accompanying drawings:

Fig. 1 is vehicle parameter estimating system structure chart.

Fig. 2 is air spring suspension system schematic.

Fig. 3 is road grade estimation module structure chart.

Fig. 4 is centroid position estimation module structure chart.

Detailed description of the invention

Research object of the present utility model is Heavy Duty Truck, especially equips the vehicle of electronic control pneumatic brake system and electronic control air suspension system.Record longitudinal acceleration by acceleration transducer; Record spring height by height sensor; Record brake pressure by pressure sensor. VehicleStarting stage, acceleration hour, fluctuate larger, thereby method of estimation is preferably carried out quality estimation in the time that acceleration is greater than lowest threshold. Due to heavyType vehicle mass change in running is little, therefore complete vehicle quality is considered as to constant; And braking deceleration variation causes centroid position to change, thereforeThis method of estimation is when preferably braking deceleration is greater than lowest threshold again, at interval of 0.5m/s^-2Estimate centroid position every.

Consult Fig. 1, environmental perception device, for obtaining vehicle location; Vehicle-state sensor, for obtaining air spring height, brake pressure;Computing module, comprises road grade estimation module, quality estimation module and centroid position estimation module three part compositions; Between modules, moduleAnd carry out transfer of data by bus between vehicle.

According to a kind of preferred embodiment of the present utility model, parameter estimation algorithm has taken into full account the impact that the vehicle angle of pitch is estimated road grade.Due to the unknown of vehicle centroid position, when gps signal receiver installation site does not overlap with centroid position, the road obtaining according to the positional information of GPSIt is larger that the road gradient is affected by the vehicle angle of pitch. Therefore, the utility model is estimated the vehicle angle of pitch according to air spring height change, is believed by GPS positionThe road grade estimated value that breath obtains deducts the vehicle angle of pitch and is road actual grade.

Consult Fig. 2, air spring height sensor exports elevation information number to vehicle angle of pitch computing module, and vehicle angle of pitch computing module will be bowedElevation information exports road slope calculation module to: wherein a, b, c point air spring coordinate are known, the unknown of d point coordinates,a(0,0,0)b(0,-B,Z_rr-Z_rl)c(L,-B,Z_rr-Z_fr). Wherein L antero posterior axis air spring spacing, the distance between left and right that B is air spring, angle of pitch noteFor θ, angle of heel is designated asDirection shown in figure is θ,Positive direction. Can obtain following formula by geometrical relationship:

Can calculate the angle of pitch by formula (1) and (2) and be designated as θ, angle of heel is designated as

$\mathrm{\θ}=\arctan \left(\frac{Z_{rr}-Z_{fr}}{L}\right)$

Consult Fig. 3, heavy vehicle gps receiver is accepted gps satellite signal, and vehicle location signal is passed to road slope calculation module,Road slope calculation module exports road grade signal to rapid prototyping control module, and rapid prototyping control module will produce according to corresponding control strategyReal time control command, car status information is passed in real time rapid prototyping control module by onboard sensor. Horizontal level calculation element obtains road slopeDegree estimated valueReal road value of slope (plus-minus symbol is determined by gps receiver and barycenter relative position)

$i=\hat{i}\±\mathrm{\θ}$

According to a kind of preferred embodiment of the present utility model, heavy vehicle quality is carried out in accelerator, in the time that acceleration is greater than lowest threshold,Quality algorithm for estimating is started working. Quality estimation module is according to onboard sensor information estimator complete vehicle quality. First, set up heavy vehicle LongitudinalLearn model;

T_i＝T_tqi_gi₀η_T＝F_tr_d

Wherein, be driving moment, be engine output torque, be transmission ratio, be final driver ratio, be driving force, for tire rollsMoving radius.

$F_t=F_f+F_w+F_i+F_j=Gf+\frac{C_{D}A}{21.15}{u}_a^2+Gi+\mathrm{\δ}m\frac{du}{dt}$

Wherein, be resistance to rolling, be air drag, be grade resistance, be acceleration resistance, be complete vehicle weight, be coefficient of rolling resistance, be airResistance coefficient, is front face area, is the speed of a motor vehicle, is the gradient, is complete vehicle quality, is correction coefficient of rotating mass.

According to a kind of preferred embodiment of the present utility model, road grade resistance can obtain by the road grade of estimating, air drag and car load matterAmount, as unknown quantity, adopts least square method to estimate, ensures real-time and the estimated accuracy of algorithm. Least square method can be described with following formula

And then can obtain

Introduce variable

P(t)＝[Φ^T(t)Φ(t)]^-1

Unknown quantity can be expressed as

$\widehat{\mathrm{\θ}}\left(t\right)=P\left(t\right){\mathrm{\Φ}}^T\left(t\right)Y\left(t\right)$

Wherein,

$Y\left(t\right)=\left[\begin{array}{c}y\left(1\right)\\ .\\ .\\ .\\ y\left(t\right)\end{array}\right],\mathrm{\Σ}\left(t\right)=\left[\begin{array}{c}\mathrm{\ϵ}\left(1\right)\\ .\\ .\\ .\\ \mathrm{\ϵ}\left(t\right)\end{array}\right]$

Step estimated value is

Wherein,

Evaluated error is

When the evaluated error of adjacent two stepsTime, calculating stops, delivery air resistance and complete vehicle quality information.

According to a kind of embodiment of optimizing of the present utility model, in moderating process, estimate centroid position, in the time that being greater than lowest threshold, calculates braking decelerationMethod is started working. Consider that braking deceleration variation causes centroid position to estimate to change, the utility model is at interval of 0.5m/s^-2Again to centroid positionEstimate once.

Consult Fig. 4, according to a kind of embodiment of optimizing of the present utility model, centroid position estimates to use the two tracklesses based on three grades of information fusion technologies againKalman filter, one of them Kalman filter is proofreaied and correct and is estimated complete vehicle quality, the heart position of another wave filter estimation. Three grades of information with and beTurnkey is drawn together signal detection level, state/parameter Estimation level and Performance Evaluation level. The advantage of this optimization embodiment is: quality estimated accuracy meets to be wantedWhile asking, can close this wave filter, reduce parameter uncertainty on the one hand, avoid on the other hand causing centroid position estimation because model parameter changesPrecision reduces.

Vehicle sensors signal, comprises each wheel brake pressure, the speed of a motor vehicle, longitudinal acceleration, brake pressure etc., is input to signal detection level (Sensor information preprocessing part). Signal detection level is made pretreatment to sensor signal, and cancellation cusp and high-frequency noise are state/parameter EstimationLevel is ready. Performance Evaluation and be that the performance of information fusion system is carried out to Real-Time Monitoring, as the covariance monitoring of Kalman filter and onThe fault detect of two-stage working condition.

Claims (1)

1. a Heavy Duty Truck parameter estimation system, comprising:

Environmental perception device, for obtaining vehicle location;

Vehicle-state sensor, for obtaining air spring height, brake pressure;

Computing module, comprises road grade estimation module, quality estimation module and centroid position estimation module three part compositions, to road grade, heavy vapourCar train weight and centroid position are estimated.