CN103765226A - A method and system of determining an inertial sensor orientation offset - Google Patents

Abstract

Inertial sensors are typically mounted at an angular offset relative to a chassis, such as a vehicle chassis or electronic device chassis. This offset can influence the measurements of the angular orientation of said chassis derived from inertial sensors. There is provided a method of determining a sensor orientation offset relative to a chassis by obtaining a first inertial sensor measurement, rotating the chassis approximately 180 DEG, obtaining a second inertial sensor measurement; and then determining the offset from the two inertial sensor measurements.

Description

Determine the method and system of inertial sensor direction skew

Technical field

The present invention relates to determine the method and system of sensor angular deflection.More specifically, the present invention relates to but be not limited to determine that inertial sensor is with respect to the direction on chassis (preferably, the chassis of the vehicles).

Background technology

To quoting of background technology herein, should not be understood to admit that this technology forms Australia or other local common practise.

Inertial sensor is used to the movement of measuring object in many application.For example, such as the vehicles of aircraft and automobile with such as many electronic equipments of smart phone, with inertial sensor, come directions, movement and/or other correlated variables.

Inertial sensor typically comprises take measurement of an angle gyroscope and measure linear acceleration and the accelerometer of speed over time.Conventionally, this sensor is encapsulated in Inertial Measurement Unit (IMU) together.Typical IMU will comprise at least three axis accelerometer, and generally include one or more gyroscopes.IMU also comprises sometimes for 2 axles in the magnetic field of the sensing earth or 3 axle magnetometers (although not being in fact inertial sensor).

Inertial sensor is generally used for determining " attitude " (that is, object or the vehicles are with respect to rotation of infrastructural frame (normally in theory perfectly level ground)) of object or the vehicles.In many application, inertia sensing is crucial accurately.For example, in precision agriculture, need to know that " attitude " of the vehicles is to be changed and to be risen and fallen to compensate the movement of Global Navigation Satellite System (GNSS) antenna by floor level.

In apparatus control is applied as unmanned vehicle, conventionally, the caused skew of inclination that sensor accuracy is high must be enough to make to be arranged on the GNSS antenna on vehicle can produce the positioning error (for example, at least identical with the order of magnitude of GNSS system itself error) that can measure.As a result, compensating for tilt angle is estimated to come in the angle that the sensor measurement of sometimes using the IMU from being arranged on vehicle to produce obtains.

Inertial sensor in IMU is installed into orthogonal configuration and in sensor frame (under normal circumstances, be fixed on the coordinate system of sensor axis) middle generation measurement, and antenna is normally known in vehicle frame (that is, investing the coordinate system of the point of fixity on vehicle) with respect to the position of the vehicles.Therefore,, unless sensor frame is accurately aimed at vehicle frame, otherwise there is fixing angular deflection between sensor frame and vehicle frame.

Typically, want to estimate the attitude of the vehicles and therefore need the angular deflection between vehicle frame and sensor frame.Fabricator, installed in the application of IMU, can determine the angular deflection between sensor frame and vehicle frame by design drawing.Yet, after vehicles manufacture, for example reequiped IMU(, by the third-party vendor of guide device, reequiped) application in, IMU must be installed into makes sensor axis accurately aim at vehicle, or as a part for installation procedure, skew must take measurement of an angle.

Sensor axis is aimed at vehicles axle cause for equipment and can be arranged on the significant restriction that where produced in the vehicles.For example, it can be installed on level land or wall.Yet the position of this aligning may be unsuitable for or be not easy to install this equipment.In addition,, if equipment is not accurately installed perpendicular to the vehicles, can cause measuring error.

If equipment is installed on the surface of not aiming at vehicles axle, the available physical mode skew that takes measurement of an angle.Yet this need to carry out high-acruracy survey such as transit with professional equipment, this is not only consuming time, and is unpractical for many installations, especially the in the situation that of terminal user's erecting equipment.

Goal of the invention

The object of the present invention is to provide a kind of method and system of definite sensor offset, it overcomes or has improved one or more in above-mentioned shortcoming or problem, or available alternative is at least provided.

According to description below, other preferred object of the present invention will become clear.

Summary of the invention

According to an aspect of the present invention, provide a kind of and determined that described method comprises with respect to the method for the sensor orientation skew on chassis:

Obtaining the first inertial sensor measures;

By roughly 180 ° of described chassis rotations;

Obtaining the second inertial sensor measures; And

Utilize described the first inertial sensor to measure and the next described sensor orientation skew of determining with respect to described chassis of described the second inertial sensor measurement.

Preferably, by Inertial Measurement Unit (IMU), carry out described the first inertial sensor measurement and described the second inertial sensor measurement.Preferably, described IMU comprises at least three axis accelerometer.Preferably, described the first inertial sensor measurement and described the second inertial sensor are measured and only gravimetry, are consisted of.

Preferably, the step of definite described sensor orientation skew with respect to described chassis comprises the rotation of estimating between described the first inertial sensor measurement and described the second inertial sensor measurement.The estimation of described rotation is rotation matrix preferably.

Preferably, determine step with respect to the described sensor orientation skew on described chassis comprise determine described sensor with respect to the possible solution of the rotation on described chassis and/or eliminate can not and irrational solution.Alternatively, the step of definite described sensor orientation skew with respect to described chassis can comprise the region of definite feasible solution and selects to separate or estimate the direct the most reasonable sensor orientation skew of determining with respect to described chassis from rotating in this region.

Preferably, any biasing in described IMU all can be ignored or be known.

Preferably, described chassis (preferably, vehicles chassis) turns back to position that first inertial sensor measure to carry out after the second inertial sensor measurement in same position rotation or at swivel base.Described method can comprise the rotation of measuring described chassis between described the first inertial sensor measurement and described the second inertial sensor measurement.Measurement described the first inertial sensor measure measure with described the second inertial sensor between the rotation on described chassis can comprise described in use yaw detector and/or manual measurement and rotating.

Preferably, estimate described the first inertial sensor measure the step of the rotation matrix between measuring with described the second inertial sensor comprise calculate that described the first inertial sensor is measured and described the second inertial sensor measurement between the least-squares estimation of rotation matrix.The least-squares estimation of the rotation matrix between described the first inertial sensor measurement and described the second inertial sensor measurement can be rank defect.

Preferably, determine that described sensor comprises with respect to the possible solution of the rotation on described chassis the feature decomposition of carrying out estimated rotation matrix.

Preferably, eliminate can not and the step of irrational solution comprise eliminate determinant be-1 solution.Preferably, eliminate and can not also comprise that with the step of irrational solution with thick smoothing, eliminating facing upward of described chassis bows and roll.Preferably, eliminate and can not even also comprise with the step of irrational solution the remaining rational solution of selection.Select remaining rational solution to comprise and select the solution corresponding with minimum roll.

Preferably, described chassis is positioned at the cardinal principle flat surfaces for the first inertial sensor is measured and the second inertial sensor is measured.Described cardinal principle flat surfaces can be with completely smooth or ground is angled.Can be in the situation that do not know that described cardinal principle flat surfaces is with respect to the definite sensor orientation skew of angle on complete smooth ground.

According to a further aspect in the invention, provide a kind of being configured to determine the system with respect to the sensor offset on chassis, described system comprises:

Inertial Measurement Unit (IMU); And

Computational resource, itself and described IMU communicate and comprise processor and storer;

Wherein, the described storer of described computational resource is programmed to indicate described processor:

From described IMU, obtaining the first inertial sensor measures;

On described chassis, be rotated roughly and from described IMU, obtained the second inertial sensor measurement after 180 °; And

Utilize described the first inertial sensor to measure and the next described sensor offset of determining with respect to described chassis of described the second inertial sensor measurement.

According to a further aspect in the invention, provide a kind of and determined that described system comprises with respect to the system of the sensor orientation skew on chassis:

Be arranged on the IMU on chassis; And

Computational resource, itself and described IMU communicate and comprise processor and storer; Wherein, described IMU:

Obtaining the first inertial sensor measures; And

On described chassis, be rotated and roughly after 180 °, obtained the second inertial sensor measurement;

Wherein, the processor of described computational resource:

From described IMU, receive described the first inertial sensor measurement and described the second inertial sensor measurement; And

Utilize described the first inertial sensor to measure and the next described sensor offset of determining with respect to described chassis of described the second inertial sensor measurement.

Preferably, described computational resource is embedded system.Described computational resource can automatically determine when described chassis has been rotated, or alternatively, described computational resource can provide prompting, and described prompting is suitable for receiving input from user, to confirm when described chassis has been rotated.Described prompting can be figure on display and the rotation of can assisting users determining described chassis.

Described IMU can comprise three axis accelerometer.Described IMU also can comprise one or more angular rate sensors and/or 2 axles or 3 axle magnetometers.Described system preferably also comprises the Global Navigation Satellite System (GNSS) assembly being connected with processor.Can utilize from the output of GNSS assembly and help determine the sensor orientation skew with respect to chassis.GNSS assembly preferably includes gps receiver.

Can determine the sensor orientation skew with respect to chassis according to said method.

According to detailed description below, it is clear that other features and advantages of the present invention will become.

Accompanying drawing explanation

Only, in the mode of example, hereinafter, the preferred embodiment of the present invention is described with reference to the accompanying drawings more fully, wherein:

Fig. 1 is the process flow diagram that the step of the method according to this invention is shown;

Fig. 2 is the process flow diagram of sub-step that the step 130 of the process flow diagram in Fig. 1 is shown.

Embodiment

Present invention relates in general to determine the sensor orientation skew with respect to chassis.Sensor when measuring almost always to install with respect to chassis is angled.Even relatively straight and while flatly installing in chassis when sensor, also likely exist to skew when young.This skew can be expressed as rotation matrix it comprises sensor with respect to the driftage (yaw) on chassis, face upward and bow (pitch) and roll (roll) value, in example embodiment, described chassis is vehicles chassis.

Chassis is considered to be such as the framework of the goods of the vehicles or electronic installation, main body or plane.Although the present invention mainly describes with reference to the vehicles, with reference to ground traffic tools, describe even more specifically, but and mean and be confined to this, the present invention can be applicable to other chassis, comprise (such as) chassis in electronic equipment such as electronics and electromechanical tool, mobile phone, control desk, game console, remote controllers etc.

Although what use when determining sensor orientation skew in a preferred embodiment is rotation matrix, should be appreciated that, can utilize other manifestation mode of rotation, comprise (for example) Eulerian angle (Euler angle), quaternions and shaft angle degree.

Fig. 1 shows the process flow diagram of the step (110 to 130) with general introduction method according to the embodiment of the present invention.By collecting and process from the data that are usually located at the one or more sensors in Inertial Measurement Unit (IMU), obtain the first inertial sensor and measure

in a preferred embodiment, IMU will be a part that comprises the navigational system of computational resource, and described computational resource generally includes processor and storer.At the vehicles when static, system receiving sensor data.

For the static vehicles, acceierometer sensor will be measured as follows:

${\hat{f}}^s=R_n^s{f}^n+b_a+\mathrm{\ϵ}---\left(1\right)$

Wherein,

the certain force measurement in main body frame,

the rotation from navigation (local horizontal) framework to sensor frame, f ⁿ=[0 0-g] ^tthe gravity vector in navigate frame, b _aaccelerometer biasing, on-fixed disturbance when ε is measurement.

Typically, with signal, process process sensor data, to determine the estimation of the certain force of this position.The estimation of certain force comprises carries out signal processing, to consider other factors, all such as (e.g.) removing engine luggine (if engine moves) or other interference.The estimation of treated certain force causes the first inertial sensor to be measured

Then, by chassis (vehicles chassis in this case) Rotate 180 ° (step 110).In a preferred embodiment, once collect enough data at first, system just points out user that chassis is rotated to about 180 °.Some vehicles such as excavator may be able to rotate 180 ° in same point.Yet other vehicles must be driven, thereby face other direction, turn back to this position.In this case, positioning equipment such as GPS may be able to return to same position by assisting users.

Once rotate 180 °, just measured by obtaining the second inertial sensor from sensor collection data

picture the first inertial sensor is measured

the same, with signal, process process sensor data, to determine, cause the second inertial sensor to be measured

the estimation of certain force.

Fig. 2 illustrates in greater detail the step 130 of Fig. 1.First, in step 132, consider first sensor measurement and the second sensor measurement.Consider above-mentioned equation (1), the first measurement and second is measured with the relation of gravity and is:

${\hat{f}}^{s_1}=R_n^{s_1}{f}^n+b_a+{\mathrm{\ϵ}}_1---\left(2\right)$

${\hat{f}}^{{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HDA0000470675720000011.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}=R_n^{s_2}{f}^n+b_a+{\mathrm{\&epsiv;}}_2---\left(3\right)$

be the rotation from navigation (local horizontal) framework to sensor frame, can be broken down into two parts: with

the rotation of from vehicle frame to sensor frame (expectation value), and

from navigating to the rotation of vehicle frame (that is, the attitude of the vehicles).Therefore,

can be expressed as:

$R_n^s=R_v^s{R}_n^v---\left(4\right)$

Equation (4) can be by substitution equation (2) and (3):

${\hat{f}}^{s_1}=R_{v_1}^{s_1}{R}_n^{v_1}{f}^n+b_a+{\mathrm{\&epsiv;}}_1---\left(5\right)$

${\hat{f}}^{{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HDA0000470675720000011.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}=R_{v_2}^{s_2}{R}_n^{v_2}{f}^n+b_a+{\mathrm{\&epsiv;}}_2---\left(6\right)$

When IMU is installed in the fixed position in the vehicles, for , between first sensor measurement and the second sensor measurement, rotation changes.Therefore:

$R_v^s=R_{v_1}^{s_1}=R_{v_2}^{s_2}---\left(7\right)$

Use equation (7), equation (5) can be rewritten as and comprise

${\hat{f}}^{{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HDA0000470675720000011.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}=R_v^s{R}_n^{v_2}{f}^n+b_a+{\mathrm{\&epsiv;}}_2---\left(8\right)$

And vehicle frame is to the rotation of navigate frame owing to measuring for the second time

when measuring for the first time due to the each rotation of measuring in vehicles chassis

and the vehicles carrier frame being further rotated is to the rotation of navigate frame

be identical (suppose to measure at same position, therefore, for different measurements, navigate frame is constant), so equation (8) can be rewritten as:

${\hat{f}}^{{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HDA0000470675720000011.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}=R_v^s{R}_{v_1}^{v_2}{R}_n^{v_1}{f}^n+b_a+{\mathrm{\&epsiv;}}_2---\left(9\right)$

Because known each measured

the rotation of middle vehicle frame is the action of 180 °, so it can be represented as:

$R_{v_1}^{{\mathrm{<figure-callout\; id="2"\; label="v"\; filenames="HDA0000470675720000011.png"\; state="\{\{state\}\}">v</figure-callout>}}_2}=\left[\begin{array}{ccc}-1& 0& 0\\ 0& -1& 0\\ 0& 0& 1\end{array}\right]---\left(10\right)$

Noting, is that the vehicles rather than landform (terrain) are around its z axle rotation, although the two will be consistent on completely smooth ground.

In solve equation (9)

substitution equation (5), supposes ε ₁≈ 0 and ε ₂≈ 0, obtains:

$({\hat{f}}^{s_2}-b_a)=R_v^s{R}_{v_1}^{v_2}{R}_s^v({\hat{f}}^{s_1}-b_a)---\left(11\right)$

For the object of determining skew, can suppose b _a≈ 0, and therefore, the relation between vehicles action is below:

${\hat{f}}^{{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HDA0000470675720000011.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}=R_v^s{R}_{v_1}^{v_2}{R}_s^v{\hat{f}}^{s_1}---\left(12\right)$

$={R\hat{f}}^{s_1}---\left(13\right)$

Wherein,

it represents the rotation of the rotation matrix form between first sensor measurement and the second sensor measurement.R is certain orthogonal matrix (SO (3)) and symmetric matrix, can utilize its attribute to estimate the step 134 of R(Fig. 2), R can be expressed as 3 * 3 matrixes as follows:

$R=\left[\begin{array}{ccc}{R}_{11}& R_{12}& R_{13}\\ R_{12}& R_{22}& R_{23}\\ R_{13}& R_{23}& R_{33}\end{array}\right]---\left(14\right)$

Because R is symmetrical, therefore can understand, only have 6 elements.Yet, in any SO (3) matrix, only have three degree of freedom, therefore, they are not independently entirely.

Due to R=R ^t, therefore can also be expressed as:

${\hat{f}}^{{\mathrm{<figure-callout\; id="2"\; label="s"\; filenames="HDA0000470675720000011.png"\; state="\{\{state\}\}">s</figure-callout>}}_2}=R{\hat{f}}^{s_1}---\left(15\right)$

${\hat{f}}^{s_1}=R{\hat{f}}^{s_2}---\left(16\right)$

Having taken a fancy to is to produce 6 equatioies with 6 unknown numbers, but is order 5 with the least-squares estimation of equation (15) structure, therefore, need to further retrain.

also be that (that is, form is B=P in similar conversion ^-1aP), therefore, reservation matrix trace and final constraint can be confirmed as:

R ₁₁+R ₂₂+R ₃₃=-1 （17）

With equation (15) and (17), the Linear least square estimation of R is configured to:

Ax=b （18）

Wherein,

$A=\left[\begin{array}{cccccc}{f}_x^{s_1}& f_y^{s_1}& f_z^{s_1}& 0& 0& 0\\ 0& f_x^{s_1}& 0& f_y^{s_1}& f_z^{s_1}& 0\\ 0& 0& f_x^{s_1}& 0& f_y^{s_1}& f_z^{s_1}\\ f_x^{s_2}& f_y^{s_2}& f_z^{s_2}& 0& 0& 0\\ 0& f_x^{s_2}& 0& f_y^{s_2}& f_z^{s_2}& 0\\ 0& 0& f_x^{s_2}& 0& f_y^{s_2}& f_z^{s_2}\\ 1& 0& 0& 1& 0& 1\end{array}\right]---\left(19\right)$ $b={\left[\begin{array}{ccccccc}{f}_x^{s_2}& f_y^{s_2}& f_z^{s_2}& f_x^{s_1}& f_y^{s_1}& f_z^{s_1}& -1\end{array}\right]}^T---\left(20\right)$

x=[R ₁₁ R ₁₂ R ₁₃ R ₂₂ R ₂₃ R ₃₃] ^T （21）

Once estimate R, just carry out next step, to determine that sensor orientation is with respect to the feasible solution of the rotation on vehicles chassis any symmetric matrix such as:

$R=R_v^s{R}_{v_1}^{v_2}{R}_s^v---\left(22\right)$

All there is feature decomposition, for the estimation from R is extracted

can utilize this feature decomposition, the form of this feature decomposition is:

A=QΛQ ^T （23）

Wherein, Λ is that the diagonal matrix of eigenwert and Q are the diagonal matrix of the proper vector corresponding with eigenwert.Because feature decomposition is also similarity transformation, so A and the shared identical eigenwert of Λ.

eigenwert can be confirmed as: $\mathrm{eig}\left(R_{v_1}^{v_2}\right)=\left[\begin{array}{ccc}-1& -1& 1\end{array}\right]---\left(24\right)$

Therefore, by

the diagonal matrix that forms of eigenwert be itself, thereby the feature decomposition that causes R is by with coupling

mode provide while arranging

the set of feasible solution.Therefore because the eigenwert in Q is not unique, all can be multiplied by-1 and still keep orthogonality, thereby cause 8 possible solutions.

By elimination process, can not eliminate (step 138) with irrational solution.First, although have ± 1 determinant of Q, certain orthogonal determinant of a matrix must be+1, therefore, half elimination of the solution that can be-1 by determinant.In four remaining solutions, only face upward bow and two uniquenesses of roll (going off course to be fuzzy) to separating corresponding to " multiple twin ".

Reset equation (2):

$R_{s_1}^{v_1}({\hat{f}}^{s_1}-b_a-{\mathrm{\&epsiv;}}_1)=R_{n_1}^{v_1}{f}^n---\left(25\right)$

Can use thick smoothing (coarse leveling) to determine that facing upward of the vehicles (that is) bowed and the estimation of roll.Suppose that biasing and noise are little:

Face upward and bow: $\mathrm{\&theta;}=a\tan 2(f_x^v,\sqrt{{\left(f_y^v\right)}^2,{\left(f_z^v\right)}^2})---\left(26\right)$

Roll: $\mathrm{\&phi;}=a\tan 2(-f_y^v,-f_z^v)---\left(27\right)$

In multiple twin solution one will corresponding to potential solution people another will be corresponding to irrational situation, such as, the vehicles " are suspended from roof ", therefore, have the solution of minimum definitely roll corresponding to physics feasible solution.

As definite solution, then eliminate the alternative of impossible and irrational solution, can first determine the region of feasible solution, then can be from definite potential solution within the scope of this.The alternative method of described method is with rotation, to estimate directly to determine the most reasonably to separate.

Utilize sensor measurement determine exactly two degree of freedom of expected angle skew, that is, and roll and face upward and bow.Can, easily with other sensor (if available) or alternatively by artificial input, determine driftage (around the rotation of Z-axis).For example, can utilize figure demonstration and user that the device with two degree of freedom is shown visually device to be rotated around vertical direction, to input Three Degree Of Freedom.

Advantageously, at sensor biasing b _ain very little or known situation, for example, as at calibration procedure (being at IMU, situation about factory calibrated) being mounted is afterwards the same, the method according to this invention allows the device with IMU to be arranged in chassis with any direction, then, by simple 180 ° of spinning movements, determine the direction of IMU, to proofread and correct the sensor orientation skew with respect to chassis.So do not need the device with IMU and chassis orthogonally or for the installation direction of measuring and manually input is installed.

Because will not have device and the chassis of IMU, install orthogonally, institute thinks that device provides obviously more installation to select.For example, it can be installed on the wall of inclination, or is arranged in arbitrary part of main body (for example, such as, wheel arch) of highway communication instrument.In addition, because determined accurate direction, so any error of having introduced because not making device perfection install orthogonally before having eliminated.

Because needn't be after installing this device of manual measurement with respect to the direction on chassis, so also reduced cost (especially cost of labor) and shut down time.In addition, do not need professional equipment and installation site to be not limited to the position that must be able to measure from outside.

In this manual, such as first and second, the adjective of left and right, top and bottom etc. can be only for an element or action are separated with another element or active region, and not necessarily require or mean this relation or the order of any reality.If context allows, using integral or assembly or step (etc.) by one that is not interpreted as being only limited in integral body, assembly or step, but can be one or more in this integral body, assembly or step etc.

For the those of ordinary skill in correlation technique, the above description to various embodiments of the present invention provides for the purpose of description.It is not intended to exhaustively or limit the invention to single disclosed embodiment.As mentioned above, the technician in the field of above instruction will know numerous alternative form of the present invention and form of distortion.Therefore,, although some alternate embodiments have specifically been discussed, other embodiment will be clearly or relatively easily by those of ordinary skill in the art, to be developed.The present invention is intended to contain of the present invention all alternative forms, modification and the form of distortion of having discussed herein and falls into other embodiment in the spirit and scope of the present invention described above.

In this manual, wording " comprises " or similar wording is intended to represent comprising of nonexcludability, makes to comprise that the method for a column element, system or equipment do not only include these elements, but also can comprise other unlisted element.

Claims (26)

1. determine with respect to a method for the sensor orientation skew on chassis, said method comprising the steps of:

Obtaining the first inertial sensor measures;

By roughly 180 ° of described chassis rotations;

Obtaining the second inertial sensor measures; And

Utilize described the first inertial sensor to measure and the next described sensor orientation skew of determining with respect to described chassis of described the second inertial sensor measurement.

2. method according to claim 1, wherein, described the first inertial sensor measures and described the second inertial sensor measurement is undertaken by Inertial Measurement Unit IMU.

3. method according to claim 2, wherein, described IMU comprises three axis accelerometer.

4. according to claim 1 or method claimed in claim 2, wherein, any biasing in described IMU is all known and is used in the step of definite described sensor orientation skew with respect to described chassis.

5. according to claim 1 or method claimed in claim 2, wherein, any biasing in described IMU is all insignificant.

6. according to method in any one of the preceding claims wherein, wherein, described the first inertial sensor is measured and described the second inertial sensor measurement only consists of gravimetry.

7. according to method in any one of the preceding claims wherein, wherein, the step of definite described sensor orientation skew with respect to described chassis comprises the rotation of estimating between described the first inertial sensor measurement and described the second inertial sensor measurement.

8. method according to claim 7, wherein, rotates and estimates it is rotation matrix.

9. method according to claim 8, wherein, comprises the least-squares estimation of calculating described rotation matrix to the estimation of described rotation matrix.

10. method according to claim 9, wherein, the least-squares estimation of described rotation matrix is rank defect.

11. according to method in any one of the preceding claims wherein, and wherein, the step of definite described sensor orientation skew with respect to described chassis comprises determines that described sensor is with respect to the feasible solution of the rotation on described chassis.

12. methods according to claim 11, wherein, determine that described sensor comprises execution feature decomposition with respect to the step of the feasible solution of the rotation on described chassis.

13. methods according to claim 12, wherein, determine step with respect to the described sensor orientation skew on described chassis comprise elimination can not and irrational solution.

14. methods according to claim 13, wherein, eliminate the solution that can not comprise that with the step of irrational solution elimination determinant is-1.

15. methods according to claim 13, wherein, eliminate and can not comprise that with the step of irrational solution utilizing thick smoothing to eliminate facing upward of described chassis bows and roll.

16. methods according to claim 15, wherein, eliminate and can not also comprise and select the solution corresponding with minimum roll with the step of irrational solution.

17. according to method in any one of the preceding claims wherein, and wherein, described chassis rotates at same position.

18. according to the method described in any one in claim 1 to 16, and wherein, the position that described chassis is measured from described the first inertial sensor is moved and turn back to this position after the described chassis of rotation.

19. according to method in any one of the preceding claims wherein, and wherein, described chassis is vehicles chassis.

20. 1 kinds are configured to determine the system with respect to the sensor offset on chassis, and described system comprises:

Inertial Measurement Unit IMU; And

Computational resource, itself and described IMU communicate and comprise processor and storer;

Wherein, the described storer of described computational resource is programmed to indicate described processor:

From described IMU, obtaining the first inertial sensor measures;

On described chassis, be rotated roughly and from described IMU, obtained the second inertial sensor measurement after 180 °; And

Utilize described the first inertial sensor to measure and the next described sensor offset of determining with respect to described chassis of described the second inertial sensor measurement.

Determine that described system comprises with respect to the system of the sensor orientation skew on chassis for 21. 1 kinds:

Be arranged on the IMU on chassis; And

Computational resource, itself and described IMU communicate and comprise processor and storer; Wherein, described IMU:

Obtaining the first inertial sensor measures; And

On described chassis, be rotated and roughly after 180 °, obtained the second inertial sensor measurement;

And wherein, the described processor of described computational resource:

From described IMU, receive described the first inertial sensor measurement and described the second inertial sensor measurement; And

Utilize described the first inertial sensor to measure and the next described sensor offset of determining with respect to described chassis of described the second inertial sensor measurement.

22. according to the system described in claim 20 or 21, and wherein, described computational resource is embedded system.

23. according to the system described in any one in claim 20 to 22, and wherein, described computational resource automatically determines when described chassis has been rotated.

24. according to the system described in any one in claim 20 to 23, and wherein, described computational resource provides prompting, and described prompting is set to receive input to confirm when described chassis has been rotated from user.

25. systems according to claim 24, wherein, described prompting is the rotation that figure on display and assisting users are determined described chassis.

26. according to the system described in any one in claim 20 to 25, and wherein, described IMU comprises three axis accelerometer.

Images