CN104290755A - Automotive driving state early warning method - Google Patents

Abstract

The invention provides an automotive driving state early warning method. The automotive driving state early warning method includes pre-judging an automotive driving state according to received sensor information; correcting an automotive driving state characteristic value according to the automotive driving state; outputting early warning information according to the corrected automotive driving state characteristic value. According to the automotive driving state early warning method, automotive driving state characteristic values of different automobiles under different states are calculated by analyzing all data collected by a sensor during driving of the automobiles, the automotive driving state characteristic values are used as a criterion for early warning, and the output early warning information prompts drivers to notice potential risks of the mobiles.

Description

A kind of motoring condition method for early warning

Technical field

The present invention relates to field of automobile safety, particularly a kind of motoring condition method for early warning.

Background technology

Real-time Measurement accuracy for automobile sport attitude carries out car chassis Comprehensive Control and the prerequisite evaluated vehicle handling stability and active safety and basis.Have that volume is large, quality heavy, it is loaded down with trivial details to install and use the defects such as inconvenience because traditional employing mechanical rotor gyro forms method that mechanical platform carrys out measured automobiles attitude, make its application on automobile have received considerable restraint.Along with the development of automotive electronic technology, military sensor obtains fast development, be exactly the most typically wherein integrated micro-electro-mechanical subsystem sensor, it has, and volume be little, quality is light, it is fast, highly sensitive to respond, the advantage such as low energy is high, superpower, automobile is used widely.

And integrated micro-electro-mechanical subsystem sensor greatly mainly with gravitational accelerometer as attitude sensor, the pitching of measured automobiles and horizontal surface and roll angle, the angle of such measured automobiles static state is more accurate, and automobile carries attitude its particularity, need the real-time attitude reflecting automobile sport object, due to the sensor that accelerometer is allergen acceleration/accel, this has a larger acceleration/accel when causing automobile in acceleration startup or brake suddenly, be converted into pitch angle angle probably at about 10 degree, directly results in automobile when starting and brake, automobile attitude can not be reflected really.Namely Fig. 1 and Fig. 2 reflect the transient change of automobile pitch angle when starting and brake.

Summary of the invention

The application provides a kind of motoring condition method for early warning, by resolving the data of all collections of sensor in vehicle traveling process, calculate the motoring condition eigenwert of different automobile under different conditions, using the criterion of described motoring condition eigenwert as early warning, export early warning information accordingly to point out the potential danger of chaufeur automobile.

Described motoring condition method for early warning comprises step:

To cut steam vehicle travelling state according to the sensor information anticipation that receives;

According to described motoring condition correction motoring condition eigenwert;

Early warning information is exported according to the motoring condition eigenwert revised.

By upper, by resolving the data of all collections of sensor in vehicle traveling process, calculate the motoring condition eigenwert of different automobile under different conditions, using the criterion of described motoring condition eigenwert as early warning, export early warning information accordingly to point out the potential danger of chaufeur automobile.

Optionally, in steps A, described motoring condition comprises: low-speed running state, rough ride state, frequent uphill/downhill motoring condition or frequent acceleration-deceleration motoring condition.

Optionally, in step B, described motoring condition eigenwert comprises: course angle Ψ, pitch angle γ, roll angle θ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj.

Optionally, when steps A is judged as low-speed running state in advance,

The pitch angle γ in motoring condition eigenwert and roll angle θ is revised in step B.

Optionally, in step B, calculate described pitch angle and roll angle adopts following formula: Ex=a (sin γ-cos θ)+m (Ey+Ez); Ez=a (sin θ-cos γ)+m (Ey+Ex);

In formula, Ex, Ey, Ez represent the magnetic biasing amount of X-axis, Y-axis, Z axis respectively, and a, m represent proportionality coefficient respectively.

By upper, due to low speed driving, the left-hand rotation acceleration/accel measured by accelerometer and right-hand rotation acceleration/accel effectively, need be revised course angle Ψ, pitch angle γ and roll angle θ.Further, in low speed driving process, course angle Ψ is standard value, only needs thus to calculate pitch angle γ and roll angle θ, and due to the low frequency characteristic of magnetic, baud rate is about 10 frames/second, therefore accepts and believe magnetic biasing gauge calculation pitch angle γ and roll angle θ.

Optionally, when steps A is judged as rough ride state in advance,

The course angle Ψ in motoring condition eigenwert, pitch angle γ and roll angle θ is revised in step B.

Optionally, in step B, course angle Ψ, pitch angle γ and the roll angle θ revised in motoring condition eigenwert adopts following formula

Gx1＝gx-Lyx*gy-Lzx*gz；

Gx1＝Gx1-(CCS[0][0]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][0]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][0]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)；

Gy1＝gy-Lxy*gx-Lzy*gz；

Gy1＝Gy1-(CCS[0][1]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][1]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][1]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)；

Gz1＝gz-Lxz*gx-Lyz*gy；

Gz1＝Gz1-(CCS[0][2]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][2]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][2]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)；

In formula, CCS [] [] represents the direction cosine matrix of gyro attitude;

Gx1, Gy1, Gz1 represent the axial angle speed of X-axis, Y-axis, Z axis respectively;

Gx, gy, gz represent the turning rate value of X-axis, Y-axis, Z axis respectively;

Lyx*gy represents to the dynamic effects of X-axis, Lzx*gz, in mechanical erection Y-axis represents that Z axis is to the dynamic effects of X-axis in mechanical erection;

Lxy*gx represents that X-axis is to the dynamic effects of Y-axis in mechanical erection, and Lzy*gz represents that Z axis is to the dynamic effects of Y-axis in mechanical erection;

Lxz*gx represents that X-axis is to the dynamic effects of Z axis in mechanical erection, and Lzy*gz represents that Y-axis is to the dynamic effects of Z axis in mechanical erection;

North_angle is expressed as angle to north orientation, latitude is expressed as latitude, Wie represents earth rotation angular rate, PI represents circular constant.

By upper, when speed is higher, accept and believe gyro attitude matrix algorithm, its baud rate upgrades and reaches 100 frames/second.Further, remove the dynamic effects of three between centers in mechanical erection, improve measurement accuracy.

Optionally, when steps A is judged as frequent uphill/downhill motoring condition or frequent acceleration-deceleration motoring condition in advance,

The course angle in motoring condition eigenwert, pitch angle, roll angle, left-hand rotation acceleration/accel and right-hand rotation acceleration/accel is revised in step B.

Optionally, in step B, revise course angle Ψ, pitch angle γ in motoring condition eigenwert, roll angle θ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj and adopt following formula:

$\begin{array}{cc}T=\left[\begin{array}{ccc}{T}_{11}& T_{12}& T_{13}\\ T_{21}& T_{22}& T_{23}\\ T_{31}& T_{32}& T_{33}\end{array}\right]& \left\{\begin{array}{c}\mathrm{\Ψ}={\mathrm{tg}}^{-1}\left(\frac{T_{12}}{T_{22}}\right)\\ \mathrm{\θ}=\sin \left(T_{32}\right);\\ \mathrm{\γ}={\mathrm{tg}}^{-1}\left(\frac{T_{31}}{T_{T33}}\right)\end{array}\right.\end{array}$

cx*γ＝cx*ax*(0.5f-q2q2-q3q3)+ay*(q1q2-q0q3)+az(q1q3+q0q2)；

cy*θ＝cy*ax*(q1q2-q0q3)+ay*(q1q1-q3q3)+az(q2q3+q0q1)；

cz*Ψ＝cz*ax*(q1q3-q0q2)+az(q1q1+q2q3)；

$q_0=\frac{T22-T21}{T23};q_1=\frac{T33}{T31};q_2=\frac{T22+T23}{T31};q_3=\frac{T22}{T21};$

$\mathrm{axj}=\arctan \left(\frac{\mathrm{ax}}{\sqrt{\mathrm{ay}*\mathrm{ay}+\mathrm{az}*\mathrm{az}}}\right),\mathrm{ayj}=\arctan \left(\frac{\mathrm{ay}}{\sqrt{\mathrm{ax}*\mathrm{ax}+\mathrm{az}*\mathrm{az}}}\right);$

In formula, CCS [] [] represents the direction cosine matrix of gyro attitude;

Cx, cy, cz represent the calibration coefficient of X-axis, Y-axis, Z axis respectively;

Ax, ay, az represent the acceleration/accel of X-axis, Y-axis and Z-direction respectively;

F represents the process variable of computational algorithm.

By upper, when occurring that the whether frequent uphill/downhill of automobile travels or frequent acceleration-deceleration travels, then belong to road conditions the most complicated, therefore the data that sensor gathers all cannot be accepted and believed, and namely revise respectively 5 output valves.In conjunction with above-mentioned magnetic biasing amount computing formula and above-mentioned 5 output valves of gyro attitude matrix correction, provide initial value by magnetic biasing amount computing formula, provide integration by gyro attitude matrix.

Optionally, steps A comprises:

According to the moving velocity over the ground of the automobile received, when the moving velocity over the ground of described automobile is less than the speed of the traveling over the ground threshold value of automobile, be judged as low-speed running state;

When the moving velocity over the ground of described automobile is greater than the speed of the traveling over the ground threshold value of automobile, according to the accekeration in X-axis, Y-axis and Z axis three directions received, calculate the accekeration root of mean square in described three directions, when described accekeration root of mean square is less than accekeration root of mean square threshold value, be judged as rough ride state;

When described accekeration root of mean square is less than accekeration root of mean square threshold value, according to the pitch angle magnitude of angular velocity received and roll angle velocity amplitude, calculate the magnitude of angular velocity root of mean square of pitch angle magnitude of angular velocity and roll angle velocity amplitude, when described magnitude of angular velocity root of mean square is less than magnitude of angular velocity root of mean square threshold value, be judged as rough ride state;

When described magnitude of angular velocity root of mean square is greater than magnitude of angular velocity root of mean square threshold value, be judged as frequent uphill/downhill motoring condition or frequent acceleration-deceleration motoring condition.

By upper, judge the motoring condition of automobile respectively by different acquisition amount, thus revise different motoring condition eigenwert respectively, reach best early warning.

Accompanying drawing explanation

Fig. 1 is the transient change of automobile pitch angle when starting when not adopting the present invention program;

Fig. 2 is the transient change of automobile pitch angle when braking when not adopting the present invention program;

Fig. 3 is the diagram of circuit of motoring condition method for early warning;

Fig. 4 is the transient change of automobile pitch angle when starting after employing the present invention program;

Fig. 5 is the transient change of automobile pitch angle when braking after employing the present invention program.

Detailed description of the invention

Motoring condition method for early warning involved in the present invention, by resolving the data of all collections of sensor in vehicle traveling process, calculate the motoring condition eigenwert of different automobile under different conditions, using the criterion of described motoring condition eigenwert as early warning, export early warning information accordingly to point out the potential danger of chaufeur automobile.Described motoring condition eigenwert comprises: course angle Ψ, pitch angle γ, roll angle θ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj.

In the present embodiment, be arranged at the data that the various kinds of sensors on automobile gathers and comprise: the acceleration/accel of the X-axis that acceleration pick-up gathers, Y-axis and Z-direction: ax, ay and az;

The vehicle that gyroscope gathers is around the turning rate value of X-axis, Y-axis and Z axis axis: gx, gy and gz;

The magnetic biasing amount of the vehicle that magnetoresistive transducer gathers and X-axis, Y-axis and Z axis: Ex, Ey and Ez;

The initial point of above-mentioned system of axes is centre of gravity of vehicle, and X-axis points to north, and Y-axis points to east, and Z axis points to ground along ground vertical line, and X-axis, Y-axis and Z axis form right-handed system.

The diagram of circuit of motoring condition method for early warning as described in Figure 1, specifically comprises:

Step S10: anticipation is cut steam car whether low speed driving, if judged result is no, then enters step S20, otherwise enters step S30.

The moving velocity over the ground collecting automobile when sensor is when being less than 4.8km/h, judges that automobile is as low speed driving, shows that the traveling attitude of automobile is relatively stable, enters step S20.Otherwise three directions add meter root of mean square and whether are greater than setting value, and when the moving velocity over the ground of automobile is higher than 4.8km/h, automobile travels at the higher speeds, enters step S30.In this step, if with the acceleration/accel ax of X-direction for a reference value, then the acceleration/accel ax of X-direction is the moving velocity over the ground of automobile.

Step S20: corrected Calculation pitch angle and roll angle.

Due to speed be less than 4.8km/h time, then think that left-hand rotation acceleration/accel axj measured by accelerometer and right-hand rotation acceleration/accel ayj is effective.Need thus to revise course angle Ψ, pitch angle γ and roll angle θ.

Further, in low speed driving process, course angle Ψ is standard value, only needs thus to calculate pitch angle γ and roll angle θ, and due to the low frequency characteristic of magnetic, baud rate is about 10 frames/second, therefore accepts and believe magnetic biasing gauge calculation pitch angle γ and roll angle θ.

Magnetic biasing amount computing formula is: Ex=a (sin γ-cos θ)+m (Ey+Ez), Ez=a (sin θ-cos γ)+m (Ey+Ex), in formula, a, m represent proportionality coefficient respectively.Calculate pitch angle γ and roll angle θ eventually through above-mentioned two computing formula, the left-hand rotation acceleration/accel axj, the right-hand rotation acceleration/accel ayj and course angle Ψ that detect in combination with sensor export in the lump, jump to step S70.

Step S30: under the non-low-speed running state of automobile, whether the anticipation running car that breaks jolts.

Judge whether jolt road surface Main Basis is X-axis, whether the accelerometer root of mean square in Y-axis and Z axis three directions be greater than threshold value, if be greater than threshold value, then represent that running car jolts, and enters step S50, otherwise enters step S40.

The formula that this step adopts is namely M is greater than threshold value time, represent road bump.

Step S40: corrected Calculation course angle Ψ, pitch angle γ and roll angle θ.

When the speed of a motor vehicle is greater than 4.8km/h, if still adopt magnetic biasing gauge to calculate, then cause its more new data cannot adapt to high-speed case, now need to accept and believe gyro attitude matrix algorithm, its baud rate upgrade reach 100 frames/second.

The direction cosine matrix of gyro attitude is expressed as:

$\begin{array}{cc}T=\left[\begin{array}{ccc}{T}_{11}& T_{12}& T_{13}\\ T_{21}& T_{22}& T_{23}\\ T_{31}& T_{32}& T_{33}\end{array}\right]& \left\{\begin{array}{c}\mathrm{\Ψ}={\mathrm{tg}}^{-1}\left(\frac{T_{12}}{T_{22}}\right)\\ \mathrm{\θ}=\sin \left(T_{32}\right);\\ \mathrm{\γ}={\mathrm{tg}}^{-1}\left(\frac{T_{31}}{T_{T33}}\right)\end{array}\right.\end{array}$

Gyro attitude matrix algorithmic notation is: Gx1=gx-Lyx*gy-Lzx*gz, in formula, Gx1 represents the angular rate of X axis to be solved, gx represents the X turning rate value that sensor detects, Lyx*gy represents that Y-axis is to the dynamic effects of X-axis in mechanical erection, and Lzx*gz represents that Z axis is to the dynamic effects of X-axis in mechanical erection.

Further, also need removal latitude and north component to affect the angular rate of X axis to be solved, calculating formula is expressed as:

Gx1=Gx1-(CCS [0] [0] * cos (north_angle*PI/180.0) * cos (latitude*PI/180.0) * Wie/3600.0+CCS [1] [0] * sin (latitude*PI/180.0) * Wie/3600.0+CCS [2] [0] * sin (north_angle*PI/180.0) * cos (latitude*PI/180.0) * Wie/3600.0), in formula, north_angle is expressed as angle to north orientation, latitude is expressed as latitude, Wie represents earth rotation angular rate, PI represents circular constant.

The angular rate Gx1 of X axis can be obtained by above-mentioned two groups of calculating formulas, by the angular rate Gx1 integration to X axis, namely try to achieve revised pitch angle γ.

Adopt formula Gy1=gy-Lxy*gx-Lzy*gz, in formula, Gy1 represents the angular rate of Y-axis to be solved, gy represents the Y-axis turning rate value that sensor detects, Lxy*gx represents that X-axis is to the dynamic effects of Y-axis in mechanical erection, and Lzy*gz represents that Z axis is to the dynamic effects of Y-axis in mechanical erection.

Removal latitude and north component affect the angular rate of Y-axis to be solved, and calculating formula is expressed as:

Gy1＝Gy1-(CCS[0][1]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][1]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][1]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)。The angular rate Gy1 of Y-axis can be obtained by above-mentioned two groups of calculating formulas, by the angular rate Gy1 integration to Y-axis, namely try to achieve revised roll angle θ.

In like manner, adopt formula Gz1=gz-Lxz*gx-Lyz*gy, in formula, Gz1 represents the angular rate of Z-axis direction to be solved, gz represents the Z axis turning rate value that sensor detects, Lxz*gx represents that X-axis is to the dynamic effects of Z axis in mechanical erection, and Lzy*gz represents that Y-axis is to the dynamic effects of Z axis in mechanical erection.

Removal latitude and north component affect the angular rate of Z-axis direction to be solved, and calculating formula is expressed as:

Gz1＝Gz1-(CCS[0][2]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][2]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][2]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)。The angular rate Gz1 of Z-axis direction can be obtained by above-mentioned two groups of calculating formulas, by the angular rate Gz1 integration to Z-axis direction, namely try to achieve revised course angle Ψ.

By revised for institute pitch angle γ, roll angle θ and course angle Ψ, and the left-hand rotation acceleration/accel axj having sensor to collect, right-hand rotation acceleration/accel ayj export in the lump, jump to step S70.

Step S50: anticipation is cut steam, and frequently whether uphill/downhill traveling or frequent acceleration-deceleration travel car.

Judge frequent uphill/downhill or acceleration-deceleration Main Basis are whether the root of mean square of pitch angle cireular frequency and roll angle speed is greater than setting value, if so, then represent that the frequent uphill/downhill of automobile or acceleration-deceleration run, enter step S60; Otherwise return step S40.

The formula that this step adopts is namely, when N is greater than threshold value N, represent that the whether frequent uphill/downhill of automobile travels or frequent acceleration-deceleration travels.

Step S60: corrected Calculation course angle Ψ, pitch angle γ, roll angle θ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj.

When occurring that the whether frequent uphill/downhill of automobile travels or frequent acceleration-deceleration travels, then belong to road conditions the most complicated, therefore the data that sensor gathers all cannot be accepted and believed, and namely revise respectively 5 output valves.

In the present embodiment, adopt in conjunction with Quaternion algebra, adopt assembled gesture matrix, in conjunction with above-mentioned magnetic biasing amount computing formula and above-mentioned 5 output valves of gyro attitude matrix correction, there is provided initial value by magnetic biasing amount computing formula, provide integration by gyro attitude matrix, its expression formula is:

cx*γ＝cx*ax*(0.5f-q ₂q ₂-q ₃q ₃)+ay*(q ₁q ₂-q ₀q ₃)+az(q ₁q ₃+q ₀q ₂)；

cy*θ＝cy*ax*(q ₁q ₂-q ₀q ₃)+ay*(q ₁q ₁-q ₃q ₃)+az(q ₂q ₃+q ₀q ₁)；

cz*Ψ＝cz*ax*(q ₁q ₃-q ₀q ₂)+az(q ₁q ₁+q ₂q ₃)；

The solving equation group of four elements is expressed as: $q_0=\frac{T22-T21}{T23};q_1=\frac{T33}{T31};q_2=\frac{T22+T23}{T31};$ $q_3=\frac{T22}{T21}.$

In formula, f represents the process variable of computational algorithm, and cx, cy, cz represent the calibration coefficient of X-axis, Y-axis, Z axis respectively.

Above-mentioned calculating formula is adopted to try to achieve pitch angle γ, roll angle θ and course angle Ψ respectively, and acceleration/accel ax, ay and az of X-axis, Y-axis and Z-direction, further, calculate left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj according to above-mentioned three accekerations,

$\mathrm{axj}=\arctan \left(\frac{\mathrm{ax}}{\sqrt{\mathrm{ay}*\mathrm{ay}+\mathrm{az}*\mathrm{az}}}\right),\mathrm{ayj}=\arctan \left(\frac{\mathrm{ay}}{\sqrt{\mathrm{ax}*\mathrm{ax}+\mathrm{az}*\mathrm{az}}}\right).$

Finally, revised for institute pitch angle γ, roll angle θ, course angle Ψ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj are exported in the lump, jumps to step S70.

Step S70: export early warning information according to result of calculation.

This step receives step S20, the pitch angle γ that step S40 or step S60 exports, roll angle θ and course angle Ψ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj, calculate the real-time attitude of automobile according to above-mentioned five data, and be converted to voice messaging output.Further, the standard value of every data under also storing different situations, it is reported to the police by voice messaging that a certain data exceed standard value.Described pitch angle γ, roll angle θ and course angle Ψ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj are calculated automobile attitude and belong to prior art, the present embodiment no longer repeats it.

Further, also can be provided with air mass sensor in driving compartment, to detect the air quality of driving compartment, and pass through voice output.

Be the transient change of the automobile starting pitch angle after the present embodiment revision of option shown in Fig. 4 and Fig. 5, visible, after the present embodiment revision of option, eliminate automobile accelerate to start or suddenly brake time the pitch angle angle change that causes due to acceleration/accel.

The foregoing is only preferred embodiment of the present invention, not in order to limit the present invention, within the spirit and principles in the present invention all in a word, any amendment done, equivalent replacement, improvement etc., all should be included within protection scope of the present invention.

Claims (10)

1. a motoring condition method for early warning, is characterized in that, comprises step:

A, vehicle travelling state of cutting steam according to the sensor information anticipation that receives;

B, according to described motoring condition correction motoring condition eigenwert;

The motoring condition eigenwert that C, foundation are revised exports early warning information.

2. method according to claim 1, is characterized in that, in steps A, described motoring condition comprises: low-speed running state, rough ride state, frequent uphill/downhill motoring condition or frequent acceleration-deceleration motoring condition.

3. method according to claim 1, is characterized in that, in step B, described motoring condition eigenwert comprises: course angle Ψ, pitch angle γ, roll angle θ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj.

4. method according to claim 3, is characterized in that, when steps A is judged as low-speed running state in advance,

The pitch angle γ in motoring condition eigenwert and roll angle θ is revised in step B.

5. method according to claim 4, is characterized in that, in step B, calculates described pitch angle and roll angle adopts following formula: Ex=a (sin γ-cos θ)+m (Ey+Ez); Ez=a (sin θ-cos γ)+m (Ey+Ex);

In formula, Ex, Ey, Ez represent the magnetic biasing amount of X-axis, Y-axis, Z axis respectively, and a, m represent proportionality coefficient respectively.

6. method according to claim 3, is characterized in that, when steps A is judged as rough ride state in advance,

The course angle Ψ in motoring condition eigenwert, pitch angle γ and roll angle θ is revised in step B.

7. method according to claim 6, is characterized in that, in step B, course angle Ψ, pitch angle γ and the roll angle θ revised in motoring condition eigenwert adopts following formula

Gx1＝gx-Lyx*gy-Lzx*gz；

Gx1＝Gx1-(CCS[0][0]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][0]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][0]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)；

Gy1＝gy-Lxy*gx-Lzy*gz；

Gy1＝Gy1-(CCS[0][1]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][1]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][1]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)；

Gz1＝gz-Lxz*gx-Lyz*gy；

Gz1＝Gz1-(CCS[0][2]*cos(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0+CCS[1][2]*sin(latitude*PI/180.0)*Wie/3600.0+CCS[2][2]*sin(north_angle*PI/180.0)*cos(latitude*PI/180.0)*Wie/3600.0)；

In formula, CCS [] [] represents the direction cosine matrix of gyro attitude;

Gx1, Gy1, Gz1 represent the axial angle speed of X-axis, Y-axis, Z axis respectively;

Gx, gy, gz represent the turning rate value of X-axis, Y-axis, Z axis respectively;

Lyx*gy represents to the dynamic effects of X-axis, Lzx*gz, in mechanical erection Y-axis represents that Z axis is to the dynamic effects of X-axis in mechanical erection;

Lxy*gx represents that X-axis is to the dynamic effects of Y-axis in mechanical erection, and Lzy*gz represents that Z axis is to the dynamic effects of Y-axis in mechanical erection;

Lxz*gx represents that X-axis is to the dynamic effects of Z axis in mechanical erection, and Lzy*gz represents that Y-axis is to the dynamic effects of Z axis in mechanical erection;

North_angle is expressed as angle to north orientation, latitude is expressed as latitude, Wie represents earth rotation angular rate, PI represents circular constant.

8. method according to claim 3, is characterized in that, when steps A is judged as frequent uphill/downhill motoring condition or frequent acceleration-deceleration motoring condition in advance,

The course angle in motoring condition eigenwert, pitch angle, roll angle, left-hand rotation acceleration/accel and right-hand rotation acceleration/accel is revised in step B.

9. method according to claim 8, is characterized in that, in step B, revises course angle Ψ, pitch angle γ in motoring condition eigenwert, roll angle θ, left-hand rotation acceleration/accel axj and right-hand rotation acceleration/accel ayj and adopts following formula:

$\begin{array}{cc}T=\left[\begin{array}{ccc}{T}_{11}& T_{12}& T_{13}\\ T_{21}& T_{22}& T_{23}\\ T_{31}& T_{32}& T_{33}\end{array}\right]& \left\{\begin{array}{c}\mathrm{\ψ}={\mathrm{tg}}^{-1}\left(\frac{T_{12}}{T_{22}}\right)\\ \mathrm{\θ}=\sin \left(T_{32}\right)\\ \mathrm{\γ}={\mathrm{tg}}^{-1}\left(\frac{T_{31}}{T_{T33}}\right)\end{array}\right.\end{array};$

cx*γ＝cx*ax*(0.5f-q ₂q ₂-q ₃q ₃)+ay*(q ₁q ₂-q ₀q ₃)+az(q ₁q ₃+q ₀q ₂)；

cy*θ＝cy*ax*(q ₁q ₂-q ₀q ₃)+ay*(q ₁q ₁-q ₃q ₃)+az(q ₂q ₃+q ₀q ₁)；

cz*Ψ＝cz*ax*(q ₁q ₃-q ₀q ₂)+az(q ₁q ₁+q ₂q ₃)；

$q_0=\frac{T22-T21}{T23};q_1=\frac{T33}{T31};q_2=\frac{T22+T23}{T31};q_3=\frac{T22}{T21};$

$\mathrm{axj}=\arctan \left(\frac{\mathrm{ax}}{\sqrt{\mathrm{ay}*\mathrm{ay}+\mathrm{az}*\mathrm{az}}}\right);\mathrm{ayj}=\arctan \left(\frac{\mathrm{ay}}{\sqrt{\mathrm{ax}*\mathrm{ax}+\mathrm{az}*\mathrm{az}}}\right);$

In formula, CCS [] [] represents the direction cosine matrix of gyro attitude;

Cx, cy, cz represent the calibration coefficient of X-axis, Y-axis, Z axis respectively;

Ax, ay, az represent the acceleration/accel of X-axis, Y-axis and Z-direction respectively;

F represents the process variable of computational algorithm.

10. method according to claim 2, is characterized in that, steps A comprises:

According to the moving velocity over the ground of the automobile received, when the moving velocity over the ground of described automobile is less than the speed of the traveling over the ground threshold value of automobile, be judged as low-speed running state;

When the moving velocity over the ground of described automobile is greater than the speed of the traveling over the ground threshold value of automobile, according to the accekeration in X-axis, Y-axis and Z axis three directions received, calculate the accekeration root of mean square in described three directions, when described accekeration root of mean square is less than accekeration root of mean square threshold value, be judged as rough ride state;

When described accekeration root of mean square is less than accekeration root of mean square threshold value, according to the pitch angle magnitude of angular velocity received and roll angle velocity amplitude, calculate the magnitude of angular velocity root of mean square of pitch angle magnitude of angular velocity and roll angle velocity amplitude, when described magnitude of angular velocity root of mean square is less than magnitude of angular velocity root of mean square threshold value, be judged as rough ride state;

When described magnitude of angular velocity root of mean square is greater than magnitude of angular velocity root of mean square threshold value, be judged as frequent uphill/downhill motoring condition or frequent acceleration-deceleration motoring condition.