CN105151047A - Automobile gravity center slip angle measuring method - Google Patents

Abstract

The invention discloses an automobile gravity center slip angle measuring method which comprises the following steps: using a sensor to measure the lateral acceleration ay, the yaw velocity omega r and the longitudinal velocity Vx, gamma and theta of an automobile in the travelling process; using a linear two-freedom model to carry out dynamic parameter estimation so as to obtain the first automobile gravity center slip angle beta L; using a filtering algorithm of a non-linear three-freedom automobile model to estimate the second automobile gravity center slip angle beta N in an integration manner; and using the following formula of beta = (1-zeta) beta L + zeta beta N to calculate out the automobile gravity center slip angle beta. According to the automobile gravity center slip angle measuring method, only the sensor is utilized to measure the three parameters of the lateral acceleration, the yaw velocity and the longitudinal velocity in an automobile running process. Thus, the calculation accuracy is guaranteed while the calculated amount is small.

Description

A kind of automobile side slip angle method of measurement

Technical field

The invention belongs to running car parameter testing technical field, the method for measurement particularly to vehicle centroid sideslip angle in a kind of automobile stability control system.

Background technology

Automobile quantity increases year by year, and the safety problem that vehicle travels causes to be paid close attention to widely.Correlative study shows, the traffic accident that vehicle occurs in the process of running at high speed, and the car accident of about 45% is that vehicle loss of control stability causes.Therefore the entrucking rate of vehicle steadily control system increases year by year, and American-European countries even makes laws and forces vehicle to install electric stabilizing system.

The radical function of vehicle electronic stability system comprises Vehicular yaw control, brakeing during cornereing control etc.Determine the control objectives of automobile stability control system, the running state of current automobile must be obtained accurately, the state parameter of vehicle in machine vehicle traveling process.The main control variable of this system is side slip angle.

The most of electric stabilizing system products sold in the market also using vehicle centroid sideslip angle size as the whether stable basis for estimation of vehicle, but side slip angle cannot directly be measured, existing side slip angle observation procedure mainly contains according to car gage, wheelbase, the preset parameters such as radius of wheel and each wheel wheel speed signal carry out the method observed, and utilize non-contact optical sensor indirectly to obtain the method for data, because this type of optical pickocff is expensive, and be easily interfered in the operating mode that complexity is severe, impact is normal to be used, consider also to be not suitable for applying on a large scale from cost angle.Therefore the electric stabilizing system product of offshore company adopts the method for hard measurement, utilizes the data message of other sensor to estimate side slip angle.Product external in vehicle electronic stability system selling market is in monopoly position, domestic temporary without ripe vehicle electronic stability system product, wherein Domestic Scientific Research mechanism surveys quantifier elimination for side slip angle not yet has ripe Measurement Algorithm to be applied in electric stabilizing system product.

Summary of the invention

Technical matters solved by the invention is to provide a kind of automobile side slip angle method of measurement, under relatively little calculated amount and lower cost, can obtain the result of a measurement of barycenter crab angle accurately.

Technical scheme provided by the invention is:

A kind of automobile side slip angle method of measurement, comprises the following steps:

Step one, the longitudinal acceleration a used in sensor measurement vehicle traveling process _x, lateral acceleration a _y, yaw velocity ω _r, longitudinal velocity V _x, and steering wheel angle δ;

Step 2, adopt linear two-freedom model to carry out Chemical kinetic parameter estimation, obtain automobile first side slip angle β _l;

Step 3, adopt the filtering algorithm of non-linear three degree of freedom auto model to merge mutually to estimate the second side slip angle β of automobile _n;

Step 4, adopt the side slip angle β of following formulae discovery automobile:

Wherein,

for parameter, a ₁and a ₂for lateral acceleration a _ythe first data point and the second data point.

Preferably, in step 2, comprise the following steps:

First side slip angle differential value with lateral acceleration a _y, yaw velocity ω _rwith longitudinal velocity V _xrelation between three meets following formula:

${\stackrel{\·}{\mathrm{\β}}}_L=\frac{a_y}{V_x}-{\mathrm{\ω}}_r$

To the first side slip angle differential value carry out integration, thus obtain the value β of the first side slip angle _l.

Preferably, adopt the method for least square of band forgetting factor to carry out parameter estimation, obtain the first side slip angle differential value value after carry out the value β that integration obtains the first side slip angle _l.

Preferably, and obtain the differential value of the first side slip angle after adopting the method for least square of band forgetting factor to estimate following formula is used to obtain the first real-time side slip angle of vehicle

${\mathrm{\β}}_L=\mathrm{\Σ}(\frac{a_y}{V_x}-{\mathrm{\ω}}_r)*\mathrm{\Δ}t,$

Wherein, Δ t is the integration sampling cycle.

Preferably, in step 3, the equation of motion of three degree of freedom vehicle is:

$\left\{\begin{array}{c}{\stackrel{\·}{\mathrm{\ω}}}_r=\frac{a^2{K}_f+b^2{K}_r}{I_z{V}_x}{\mathrm{\ω}}_r+\frac{{\mathrm{aK}}_f-{\mathrm{bK}}_r}{I_z}{\mathrm{\β}}_N-\frac{{\mathrm{aK}}_r}{I_z}\mathrm{\δ}\\ {\stackrel{\·}{\mathrm{\β}}}_N=(a\frac{{\mathrm{\αK}}_f-{\mathrm{bK}}_r}{{{\mathrm{mV}}_x}^2}-1){\mathrm{\ω}}_r+\frac{K_f+K_r}{{\mathrm{mV}}_x}{\mathrm{\β}}_N-\frac{K_f}{{\mathrm{mV}}_x}\mathrm{\δ}\\ {\stackrel{\·}{V}}_x={\mathrm{\ω}}_r{\mathrm{\β}}_N{V}_x+a_x\end{array}\right.$

$a_y=\frac{{\mathrm{aK}}_f-{\mathrm{bK}}_r}{{\mathrm{mV}}_x}{\mathrm{\ω}}_r+\frac{K_f+K_r}{m}{\mathrm{\β}}_N-\frac{K_f}{m}\mathrm{\δ}$

In formula, K _rand K _fbe respectively trailing wheel cornering stiffness and the front-wheel cornering stiffness of vehicle; A and b is then front axle and the rear axle distance to barycenter respectively; M is vehicle mass; I _zfor around the rotor inertia of z-axis, V _xthe longitudinal direction of car speed of a motor vehicle;

And use volume Kalman filtering algorithm, the estimated valve β of the second side slip angle is gone out through initialization, forecast step and filtering step iterative computation _n.

Preferably, in step one, measure described lateral acceleration a by acceleration pick-up _y, measure described yaw velocity ω by yaw-rate sensor _r, described acceleration pick-up and yaw-rate sensor are arranged on same circuit card.

Preferably, in step one, described longitudinal velocity V _xobtained by GPS module measurement.

Preferably, in step one, by the longitudinal acceleration using the acceleration pick-up measurement be arranged on automobile to obtain automobile, then after integration, obtain described longitudinal velocity V _x.

Preferably, in step one, obtain wheel speed signal by wheel speed sensors measurement, then adopt Kalman filtering mode to estimate the longitudinal velocity V of automobile _x.

The invention has the beneficial effects as follows:

Automobile side slip angle method of measurement provided by the present invention, only needs to use the parameters such as lateral acceleration, yaw velocity and the longitudinal velocity in sensor measurement vehicle traveling process, measures cost lower.The side slip angle computation model that the present invention builds, through simplifying, ensure that the accuracy of calculating while less calculated amount.

Accompanying drawing explanation

Fig. 1 is side slip angle method of measurement diagram of circuit of the present invention.

Fig. 2 is the kinetic model schematic diagram of automobile two degrees of freedom linear model of the present invention.

Fig. 3 is three degree of freedom auto model schematic diagram of the present invention.

Fig. 4 is measurement mechanism johning knot composition of the present invention.

Detailed description of the invention

Below in conjunction with accompanying drawing, the present invention is described in further detail, can implement according to this with reference to specification sheets word to make those skilled in the art.

The invention provides a kind of automobile side slip angle measurement technique, is that the Chemical kinetic parameter estimation based on two degrees of freedom linear model is merged with the filtering algorithm of non-linear three degree of freedom auto model the method estimating vehicle centroid sideslip angle mutually.

Slip angle estimation method of the present invention is based on ESC system conventional sensors used as the design such as the speed measuring module of vehicle, acceleration pick-up (this sensor can the lateral acceleration of measuring vehicle and longitudinal acceleration), yaw-rate sensor, steering wheel angle sensor.

Slip angle estimation method of the present invention is as shown in Figure 1:

Step one S110: first, the lateral acceleration a in measured automobiles implementation process _y, yaw velocity ω _r, longitudinal velocity V _x, and steering wheel angle δ.

Step 2 S120, according to the observed reading of automobile side angle acceleration pick-up, is less than the first acceleration/accel a at lateral acceleration ₁time, adopt linear two-freedom model to carry out Chemical kinetic parameter estimation method, the value of the first acceleration/accel may change because of different vehicles.Preferred as one, a ₁=0.4g, g are acceleration due to gravity, are worth for 9.8m/s ², now automobile first side slip angle β _ldefined formula:

${\mathrm{\β}}_L={\tan }^{-1}\frac{V_y}{V_x},$

Wherein, V _yfor vehicle lateral speed, V _xfor vehicular longitudinal velocity.Due to β _lvery little, can by the first automobile side slip angle β according to trigonometric function relation _lbe reduced to

${\mathrm{\β}}_L={\tan }^{-1}\frac{V_y}{V_x}\≈\frac{V_y}{V_x},$

Vehicle travel process in, vehicular longitudinal velocity V _xat short notice, such as, within 100 milliseconds, can by V _xregard constant as, therefore carrying out differential to the side slip angle in above formula can obtain

${\stackrel{\·}{\mathrm{\β}}}_L=\frac{d}{dt}\left(\frac{V_y}{V_x}\right)\≈\frac{{\stackrel{\·}{V}}_y}{V_x}$

As shown in Figure 2, the model simplification travelled by vehicle is the motion of particle under system of axes, and X-axis and Y-axis are transverse axis and the longitudinal axis of vehicle axis system.Car speed is V at the component of t under transverse axis X _x, the component under longitudinal axis Y is V _y.Vehicle when Turning travel with translation and rotation, after Δ t, velocity magnitude and the direction of vehicle all there occurs change, and X-axis under vehicle axis system and Y direction also there occurs change, and X-axis under vehicle axis system and Y direction also there occurs change, wherein, speed is along the change V of Y-axis _dfor:

V _d＝(V _x+ΔV _x)·sinΔθ+(V _y+ΔV _y)·cosΔθ-V _y

＝V _xsinΔθ+ΔV _xsinΔθ+V _ycosΔθ+ΔV _ycosΔθ-V _y

Wherein, Δ V _xfor automobile longitudinal velocity increment, Δ V _yfor automobile cross velocity increment, Δ θ is that automobile turns over angle.

Because Δ θ is very little at short notice and can ignore its second order trace, therefore there is following simplification:

sinΔθ＝Δθ

ΔV _x＝0

cosΔθ＝1

So, can by V _dbe reduced to

V _d＝V _xΔθ+ΔV _y

By above formula divided by Δ t, the limit when Δ t is tending towards 0 should be got, just obtain the component of vehicle centroid acceleration/accel in Y-axis, be the lateral acceleration in vehicle travel process

$a_y=\underset{t\→0}{\lim }\frac{V_x\mathrm{\Δ}\mathrm{\θ}+{\mathrm{\ΔV}}_y}{\mathrm{\Δ}t}=V_x\·{\mathrm{\ω}}_r+{\stackrel{\·}{V}}_y$

Therefore, can draw

${\stackrel{\·}{V}}_y=a_y-V_x\·{\mathrm{\ω}}_r$

Wherein, ω _rfor yaw velocity.

By what obtain be brought into expression formula in, obtain the differential value of the first side slip angle

${\stackrel{\·}{\mathrm{\β}}}_L=\frac{a_y}{V_x}-{\mathrm{\ω}}_r$

Least Square Method is carried out to above formula, more accurately can obtain the differential value of the first side slip angle

Preferred as one, the present invention adopts a kind of method of least square with forgetting factor to estimate above formula, first above formula is arranged the standard form becoming method of least square parameter estimation:

In formula, y (t) is observed reading output, Y (t)=a _y-ω _rv _x,

Adopt the method for least square of band forgetting factor to estimate the θ (t) in formula 6, algorithm is as follows:

$\widehat{\mathrm{\θ}}\left(t\right)=\widehat{\mathrm{\θ}}(t-1)+K\left(t\right)*\mathrm{\ϵ}\left(t\right|\widehat{\mathrm{\θ}}\left(t-1\right))$

In algorithm, I is identity matrix; ε (t) is prior uncertainty; K (t) is kalman gain; P (t) is covariance matrix; λ is forgetting factor, and the value of λ is less, and the weights of historical data in the new parameter estimation cycle are lower.

By using the method for least square of band forgetting factor to carry out parameter estimation, the differential value of the first side slip angle can be obtained again by using microprocessor pair quadrature, the value of the first real-time side slip angle of vehicle can be obtained, that is:

${\mathrm{\β}}_L=\mathrm{\Σ}(\frac{a_y}{V_x}-{\mathrm{\ω}}_r)*\mathrm{\Δ}t$

Δ t is the sampling period of controller, the general multiple adopting 5ms or 5ms.

From above formula, obtain the lateral acceleration a in vehicle traveling process at use sensor measurement _y, yaw velocity ω _rwith longitudinal velocity V _xafter these three parameters, just directly can obtain the differential value of the first side slip angle again will the first side slip angle β has just been drawn after carrying out integration _l.

Step 3 S130: automobile in the process of moving, when vehicle lateral acceleration is greater than the second acceleration/accel a ₂time, adopt and carry out the second slip angle estimation based on the non-linear car model filtering algorithm of three degree of freedom.The value of the second acceleration/accel may change because of different vehicles.Preferred as one, a ₂=0.45g.

As shown in Figure 3, the equation of motion of vehicle is three degree of freedom auto model:

$\left\{\begin{array}{c}{\stackrel{\·}{\mathrm{\ω}}}_r=\frac{a^2{K}_f+b^2{K}_r}{I_z{V}_x}{\mathrm{\ω}}_r+\frac{{\mathrm{aK}}_f-{\mathrm{bK}}_r}{I_z}{\mathrm{\β}}_N-\frac{{\mathrm{aK}}_r}{I_z}\mathrm{\δ}\\ {\stackrel{\·}{\mathrm{\β}}}_N=(\frac{{\mathrm{aK}}_f-{\mathrm{bK}}_r}{{{\mathrm{mV}}_x}^2}-1){\mathrm{\ω}}_r+\frac{K_f+K_r}{{\mathrm{mV}}_x}{\mathrm{\β}}_N-\frac{K_f}{{\mathrm{mV}}_x}\mathrm{\δ}\\ {\stackrel{\·}{V}}_x={\mathrm{\ω}}_r{\mathrm{\β}}_N{V}_x+a_x\end{array}\right.$

$a_y=\frac{{\mathrm{aK}}_f-{\mathrm{bK}}_r}{{\mathrm{mV}}_x}{\mathrm{\ω}}_r+\frac{K_f+K_r}{m}{\mathrm{\β}}_N-\frac{K_f}{m}\mathrm{\δ}$

In formula, K _rand K _fbe respectively trailing wheel cornering stiffness and the front-wheel cornering stiffness of vehicle; A and b is then front axle and the rear axle distance to barycenter respectively; M is vehicle mass; I _zfor around the rotor inertia of z-axis, ω _rfor yaw velocity, β _nside slip angle, δ front wheel angle (obtain steering wheel angle by steering wheel angle sensor measurement, then steering wheel angle value substitution transmitting ratio formula are tried to achieve front wheel angle.), V _xthe longitudinal direction of car speed of a motor vehicle, a _xand a _ybe respectively automobile longitudinal acceleration and vehicle lateral acceleration, can be recorded by vehicle acceleration sensor.

Adopt nonlinear filtering algorithm can estimate the value of the second side slip angle, usually can adopt volume Kalman filtering algorithm, volume Kalman filtering algorithm.

Preferred as one, the present invention adopts volume Kalman filtering algorithm, estimates the second side slip angle.

Arrange the standard form for root of mean square built-in capacitor G-card Thalmann filter equation structure:

x(t)＝f(x(t),u(t),w(t))

z(t)＝h(x(t),u(t),v(t))

In formula, state vector is x (t)=[ω _r, β, V _x] ^t; Measuring vector is z (t)=[a _y]; Control vector is u (t)=[δ, a _x] ^t; W (t) is process noise; V (t) is measurement noise; It is 0 that w (t) and v (t) meets average, and error covariance matrix is respectively the normal distribution of Q and R.

Volume Kalman filtering algorithm is divided into three parts, is initialization, forecast step and filtering step respectively.According to the equation of state in three degree of freedom auto model, utilize the quantity of state estimated valve in k moment with controlling quantity u _k, rated condition amount predicted value namely step is forecast; Utilize with the controlling quantity u in k+1 moment _k+1predicted value is measured by measurement equation calculated amount and it is right carry out filtering and obtain k+1 moment estimated valve i.e. filtering step.

Initialization: in vehicle travel process, ω _rand V _xinitial value can be obtained by the lateral acceleration measurement module in acceleration pick-up and speed measuring module measurement, and the second side slip angle β _nthen be set to 0, then the state vector initial value of cycle of run k=1 moment auto model is defined as now the error variance initial value of vehicle state estimation is set to P _1|1=diag (10 ^-1010 ^-1010 ^-10)

Forecast:

(1) estimated valve of k moment vehicle state quantity is obtained by k-1 moment estimated result with estimation error variance battle array P _k|k.Owing to using volume Kalman Filter Estimation algorithm, so need the Cubature point generating vehicle running state amount, therefore to P _k|kcarry out Cholesky decomposition, the On Square-Rooting Matrices in k moment estimation error variance battle array can be obtained

(2) according to sphere-radial direction rule, vehicle state quantity estimated valve is generated to the Cubature point of the weights such as a group, namely around vehicle state quantity estimated valve, form one group of well-distributed point with the On Square-Rooting Matrices of estimation error variance battle array.According to two times that the number of the Cubature point of sphere-radial generate rule is state dimension, the state variable number of three degree of freedom auto model is 3, then the Cubature point number of each quantity of state is 6, namely i=1,2 ... 6, X in formula _{i, k|k}i-th Cubature point at k moment quantity of state; ξ _ifor matrix $\sqrt{n}\left[\begin{array}{cccccc}1& 0& 0& -1& 0& 0\\ 0& 1& 0& 0& -1& 0\\ 0& 0& 1& 0& 0& -1\end{array}\right]$ I-th row, n is quantity of state dimension.

(3) utilize the vehicle dynamic equation of discretization each Cubature point to vehicle state quantity to convert, obtain the predicted value of all Cubature points of vehicle state quantity: u in formula _k=[δ a _x] ^t; F is the dynamic equation of the auto model of discrete form.

(4) predicted value of quantity of state is asked according to sphere-radial rule.To all vehicle-states vector Cubature point and newspaper be weighted summation.The weights of each Cubature point are 1/m, m=2n, then Cubature point weights are 1/6.Vehicle-state vector predicted value is

(5), after obtaining the predicted value of vehicle-state vector, CALCULATING PREDICTION error covariance matrix P is also needed _k+1|k: namely $P_{k+1|k}=\frac{1}{m}{\mathrm{\Σ}}_{i=1}^m{X}_{i,k+1|k}^{*}{X}_{i,k+1|k}^{*T}-{\hat{x}}_{k+1|k}{\hat{x}}_{i,k+1|k}^T+Q_k$

In formula: Q _kfor k moment vehicle-state vector prediction error variance matrix.

Filtering:

(1), after the predicted value obtaining auto model state vector and prediction error variance matrix, just need utilization to measure and filtering is carried out to quantity of state predicted value, and then obtain the estimated valve of vehicle state quantity.This process need utilizes the predicted value of measurement equation to vehicle-state vector to convert, and then obtains the predicted value of state estimation amount measurement, and the prediction error variance matrix that calculated amount is measured.Therefore, to quantity of state prediction error conariance battle array P _k+1|kcarry out Cholesky decomposition, obtain the On Square-Rooting Matrices S of quantity of state prediction error variance matrix _k+1|k, namely $S_{k+1|k}=\sqrt{P_{k+1|k}}.$

(2) sphere-radial direction rule is utilized around the predicted value of vehicle state quantity, to generate the point of the weights such as a group according to quantity of state prediction error variance On Square-Rooting Matrices, i.e. Cubature point X _{i, k+1|k}: $X_{i,k+1|k}=S_{k+1|k}{\mathrm{\ξ}}_i+{\hat{x}}_{k+1|k}$

(3) by the measurement equation in dynamic state estimator model, each quantity of state predicted value Cubature point is converted, obtain the Cubature point Z of measurement amount z (t)=[ay] predicted value _{i, k+1|k}: Z _{i, k+1|k}=H (X _{i, k+1|k}, u _k), in formula, H is measurement equation.

(4) summation is weighted to the Cubature point of all measurement amount predicted values, and then obtains measurement amount predicted value.According to sphere-radial direction rule, the weights of each Cubature point are 1/6, then the predicted value of measurement amount ${\hat{Z}}_{k+1|k}=\frac{1}{m}{\mathrm{\Σ}}_{i=1}^m{Z}_{i,k+1|k}$ M=6 in formula;

(5) the prediction error variance matrix P of calculated amount measurement _{zz, k+1|k}:

$P_{zz,k+1|k}=\frac{1}{m}{\mathrm{\Σ}}_{i=1}^m{Z}_{i,k+1|k}{Z}_{i,k+1|k}^T-{\hat{z}}_{k+1|k}{\hat{z}}_{i,k+1|k}^T+P_{k+1}$ R in formula _k+1for k+1 moment error in measurement variance matrix.

(6) the Cross-covariance P between vehicle state quantity predicted value and measurement amount predicted value is calculated _{xz, k+1|k}: $P_{xz,k+1|k}=\frac{1}{m}{\mathrm{\Σ}}_{i=1}^m{X}_{i,k+1|k}{Z}_{i,k+1|k}^T-{\hat{x}}_{k+1|k}{\hat{z}}_{i,k+1|k}^T$

(7) computer card Kalman Filtering gain $W_{k+1}:W_{k+1}=P_{xz,k+1|k}{P}_{zz,k+1|k}^{-1}$

(8) utilize the deviation between the measuring value of k+1 moment measurement amount and measurement amount predicted value, namely newly cease, by Kalman filtering gain, filtering is carried out to vehicle state quantity predicted value, finally obtain k+1 moment quantity of state estimated valve ${\hat{x}}_{k+1|k+1}={\hat{x}}_{k+1|k}+W_{k+1}(z_{k+1}-{\hat{z}}_{k+1|k})$ Z in formula _k+1for the measurement vector in k+1 moment.

(9) auto model quantity of state estimation error variance battle array P is calculated _k+1|k+1:

$P_{k+1|k+1}=P_{k+1|k}-W_{k+1}{P}_{zz,k+1|k}{W}_{k+1}^T$ Obtain k+1 moment quantity of state estimated valve with estimation error variance battle array P _k+1|k+1after, make k=k+1, and time of return upgrades part, carries out next iteration.Iterative algorithm can estimate the estimated valve of current time state vector each time, has the estimated valve β of the second side slip angle in state vector estimated valve _n.

Step 4 S140: when calculating the first side slip angle β _lwith the second side slip angle β _nafter, adopt following formula to obtain the side slip angle β of automobile:

Wherein, for parameter, when vehicle lateral acceleration is less than 0.4g, now β=β _l, auto model adopts and carries out slip angle estimation based on two degrees of freedom linear model Chemical kinetic parameter estimation method; When lateral acceleration is greater than 0.45g, now β=β _n, adopt based on three degree of freedom nonlinear model molded capacity Kalman Filter Estimation side slip angle.When lateral acceleration is between 0.4g and 0.45g, adopt two kinds of methods to carry out estimating then to adopt the mode of Weighted Fusion to obtain the final estimated result of side slip angle, namely as 0.4≤a simultaneously _ywhen≤0.45,

As shown in Figure 4, measurement mechanism of the present invention comprises microcontroller 1, acceleration pick-up 2, yaw-rate sensor 3 and speed measuring module 4.Microcontroller 1, acceleration pick-up 2, yaw-rate sensor 3 and speed measuring module 4 are all welded on double layer printed circuit plate.Microcontroller 1 is connected with speed measuring module 4 with acceleration pick-up 2, yaw-rate sensor 3 respectively, and microcontroller 1 is for gathering the lateral acceleration a of the measurement of acceleration pick-up 2, yaw-rate sensor 3 and speed measuring module 4 _y, yaw velocity ω _rwith longitudinal velocity V _xdata.

Microcontroller 1 adopts 16 micro controller system XC2365A chips.

Acceleration pick-up 2 adopts ADXL203 chip.ADXL203 is complete high precision, low-power consumption, single shaft/twin-axis accelerometer, and provide the voltage through signal condition to export, all functions are all integrated in a single-chip IC.The full scale acceleration analysis scope of these devices is ± 1.7g, both can measure dynamic acceleration, such as, vibrate, and also can measure static acceleration, such as gravity.Adopt ADXL203 chip can measure longitudinal acceleration a accurately _xwith lateral acceleration a _yvalue.

Yaw-rate sensor 3 adopts ADXRS61X chip.ADXRS61X chip adopts the angular velocity sensor of integrated micro-electron machinery system patent technique and BIMOS technique, and inside is integrated with angular rate sensor and signal processing circuit simultaneously.Compared with the sensor of any similar function, ADXRS61X has that size is little, low in energy consumption, shock resistance and the good advantage of vibratility, can accurately measure the yaw velocity ω of automobile _r.

Described speed measuring module 4 is GPS speed measuring module, by receiving and launching gps signal, measures the longitudinal velocity V of automobile _x.

Steering wheel angle information is sent in XC2365A micro controller system by the CAN network of automobile by steering wheel angle sensor.

In another embodiment, speed measuring module 4 adopts wheel speed sensors, the wheel speed signal that microcontroller 1 can be measured according to wheel speed sensors, adopts Kalman filtering or other estimation modes to estimate the longitudinal velocity V of vehicle _x.

In another embodiment, speed measuring module 4 and acceleration pick-up 2 share ADXL203 chip.Because ADXL203 is double-axel acceleration sensor, can the simultaneously longitudinal acceleration of measuring vehicle and lateral acceleration, integration can be carried out by the longitudinal acceleration value that microcontroller 1 couple of ADXL203 measures gained, obtain the longitudinal velocity V of automobile _x.

The lateral acceleration signal of acceleration pick-up 2 measuring vehicle inputs to the analog/digital amount translation interface of microcontroller 1 with the form of analog electric signal.The yaw velocity of yaw-rate sensor 3 measuring vehicle, and analog/digital amount translation interface signal being inputed to microcontroller 1 with the form of analog electric signal.Speed measuring module 4 is responsible for the longitudinal velocity of measuring vehicle, and by real-time vehicle speed by electric signal transmission to microcontroller 1.

Microcontroller 1 is according to the real-time lateral acceleration a of vehicle _y, yaw velocity ω _rand longitudinal vehicle velocity V _x, adopt the formula set to carry out processing and calculating, draw the side slip angle micro component of vehicle side slip angle β can be obtained after carrying out integration again.

Although embodiment of the present invention are open as above, but it is not restricted to listed in specification sheets and embodiment utilization, it can be applied to various applicable the field of the invention completely, for those skilled in the art, can easily realize other amendment, therefore do not deviating under the universal that claim and equivalency range limit, the present invention is not limited to specific details and illustrates here and the legend described.

Claims (9)

1. an automobile side slip angle method of measurement, is characterized in that, comprises the following steps:

Step one, the longitudinal acceleration a used in sensor measurement vehicle traveling process _x, lateral acceleration a _y, yaw velocity ω _r, longitudinal velocity V _x, and steering wheel angle δ;

Step 2, adopt linear two-freedom model to carry out Chemical kinetic parameter estimation, obtain automobile first side slip angle β _l;

Step 3, adopt the filtering algorithm of non-linear three degree of freedom auto model to merge mutually to estimate the second side slip angle β of automobile _n;

Step 4, adopt the side slip angle β of following formulae discovery automobile:

Wherein,

for parameter, a ₁and a ₂for lateral acceleration a _ythe first data point and the second data point.

2. automobile side slip angle method of measurement according to claim 1, is characterized in that, in step 2, comprise the following steps:

First side slip angle differential value with lateral acceleration a _y, yaw velocity ω _rwith longitudinal velocity V _xrelation between three meets following formula:

${\stackrel{\·}{\mathrm{\β}}}_L=\frac{a_y}{V_x}-{\mathrm{\ω}}_r$

To the first side slip angle differential value carry out integration, thus obtain the value β of the first side slip angle _l.

3. automobile side slip angle method of measurement according to claim 2, is characterized in that, adopts the method for least square of band forgetting factor to carry out parameter estimation, obtains the first side slip angle differential value value after carry out the value β that integration obtains the first side slip angle _l.

4. automobile side slip angle method of measurement according to claim 3, is characterized in that,

And after adopting the method for least square of band forgetting factor to estimate, obtain the differential value of the first side slip angle following formula is used to obtain the first real-time side slip angle of vehicle

${\mathrm{\β}}_L=\mathrm{\Σ}(\frac{a_y}{v_x}-{\mathrm{\ω}}_r)*\mathrm{\Δ}t,$

Wherein, Δ t is the integration sampling cycle.

5. automobile side slip angle method of measurement according to claim 1, is characterized in that, in step 3, the equation of motion of three degree of freedom vehicle is:

$\left\{\begin{array}{c}{\stackrel{\·}{\mathrm{\ω}}}_r=\frac{a^2{K}_f+b^2{K}_r}{I_z{V}_x}{\mathrm{\ω}}_r+\frac{{\mathrm{aK}}_f-{\mathrm{bK}}_r}{I_z}{\mathrm{\β}}_N-\frac{{\mathrm{aK}}_r}{I_z}\mathrm{\δ}\\ {\stackrel{\·}{\mathrm{\β}}}_N=(\frac{{\mathrm{\αK}}_f-{\mathrm{bK}}_r}{{{\mathrm{mV}}_x}^2}-1){\mathrm{\ω}}_r+\frac{K_f+K_r}{{\mathrm{mV}}_x}{\mathrm{\β}}_N-\frac{K_f}{{\mathrm{mV}}_x}\mathrm{\δ}\\ {\stackrel{\·}{V}}_x={\mathrm{\ω}}_r{\mathrm{\β}}_N{V}_x+a_x\end{array}\right.$

$a_y=\frac{{\mathrm{aK}}_f-{\mathrm{bK}}_r}{{\mathrm{mV}}_x}{\mathrm{\ω}}_r+\frac{K_f+K_r}{m}{\mathrm{\β}}_N-\frac{K_f}{m}\mathrm{\δ}$

In formula, K _rand K _fbe respectively trailing wheel cornering stiffness and the front-wheel cornering stiffness of vehicle; A and b is then front axle and the rear axle distance to barycenter respectively; M is vehicle mass; I _zfor around the rotor inertia of z-axis, V _xthe longitudinal direction of car speed of a motor vehicle;

And use volume Kalman filtering algorithm, the estimated valve β of the second side slip angle is gone out through initialization, forecast step and filtering step iterative computation _n.

6. automobile side slip angle method of measurement according to claim 1, is characterized in that, in step one, measures described lateral acceleration a by acceleration pick-up _y, measure described yaw velocity ω by yaw-rate sensor _r, described acceleration pick-up and yaw-rate sensor are arranged on same circuit card.

7. automobile side slip angle method of measurement according to claim 1, is characterized in that, in step one, and described longitudinal velocity V _xobtained by GPS module measurement.

8. automobile side slip angle method of measurement according to claim 1, is characterized in that, in step one, by the longitudinal acceleration using the acceleration pick-up measurement be arranged on automobile to obtain automobile, then after integration, obtains described longitudinal velocity V _x.

9. automobile side slip angle method of measurement according to claim 1, is characterized in that, in step one, obtains wheel speed signal by wheel speed sensors measurement, then adopts Kalman filtering mode to estimate the longitudinal velocity V of automobile _x.