CN102706365A - Calibration method for three-beam laser velocimeter on basis of navigation system - Google Patents

Abstract

A calibration method for a three-beam laser velocimeter on basis of a navigation system comprises the following seven steps: 1, installing an inertial unit, a differential GPS and the laser velocimeter on a carrier, and binding initial position parameters onto a navigation computer; 2, preheating a strap down inertial unit, and collecting the output data of a gyroscope and an accelerometer; 3, preprocessing the collected output data; 4, initiating the differential GPS and the laser velocimeter, allowing the system to switch into an SINS/DGPS mode from an alignment mode, and allowing the carrier to begin moving; 5, collecting the output of the velocimeter and the output of an SINS/DGPS combination navigation system when the carrier is in different motion states at different times; 6, respectively identifying the error of three installation angles and the scale factor of the laser velocimeter by a least square method; and 7, evaluating and analyzing the calibration precision. According to the invention, the accurate calibration of the three-beam laser velocimeter is realized through utilizing the characteristics of the strap down inertial unit and the combination navigation system that the output speed precision is high. Therefore, the calibration method has a practical value in the inertial navigation technology field.

Description

A kind of three wave beam laser velocimeter scaling methods based on navigational system

Technical field:

The present invention relates to a kind of three wave beam laser velocimeter scaling methods, belong to inertial navigation technology/integrated navigation technical field based on navigational system.

Background technology:

Laser velocimeter is as a kind of speed pickup, has advantage autonomous fully, that precision is high, the wide ranges that tests the speed, dynamic property reach non-cpntact measurement well.Independent laser velocimeter does not possess navigation locating function, but can have complementary advantages with the inertial navigation system combination, can realize complete autonomous, high precision navigator fix.

Inertial navigation system and three wave beam laser velocimeters are contained in the diverse location of carrier respectively in the actual use of integrated navigation system; Need to demarcate the knotmeter established angle; The laser velocimeter constant multiplier that records in the laboratory simultaneously can change in actual use, needs to demarcate the constant multiplier error.The laser velocimeter that is used for the navigator fix field at present in the open source literature does not have unified scaling method; This paper has proposed the scaling method of a kind of three wave beam laser velocimeter established angles and constant multiplier error, has solved the underlying issue of laser velocimeter and inertial navigation combined system navigator fix in the engineering reality.

Summary of the invention:

1, purpose: the purpose of this invention is to provide a kind of three wave beam laser velocimeter scaling methods based on navigational system, it has overcome the deficiency of prior art, has solved the problem that need demarcate established angle and constant multiplier error when laser velocimeter installs on the carrier.

2, technical scheme: a kind of three wave beam laser velocimeter scaling methods of the present invention based on navigational system, these method concrete steps are following:

Step 1, SINS, differential GPS and laser velocimeter are installed on the carrier navigational computer of bookbinding initial position parameters (comprising initial longitude, latitude and height) to inertial navigation.

Step 2, inertial navigation preheating, the output data of gathering gyro and accelerometer then.

Step 3, the gyro and the accelerometer data that collect are handled,, adopted second order leveling and orientation estimation algorithm to accomplish the coarse alignment of system, tentatively definite attitude of carrier angle according to SINS error Propagation Property and classic control theory.The coarse alignment time is 1 minute.Utilized the Kalman Filter Technology fine alignment behind the coarse alignment 5 minutes.

Step 4, startup differential GPS and laser velocimeter, inertial navigation system switches to SINS/DGPS (SINS/differential GPS) integrated navigation pattern (the combinational algorithm block diagram is seen Fig. 1) by alignment pattern, carrier setting in motion after switching is accomplished.

Three different outputs are constantly adopted in step 5, the output of taking knotmeter under the carrier different motion state and the output of SINS/DGPS integrated navigation system at least.

Step 6, utilize the least square method respectively established angle and the constant multiplier error of three wave beams of identification laser velocimeter.

Step 7, stated accuracy evaluation analysis.

Step 1-7 is divided into three phases, and step 1-3 is the preparatory stage, and step 4-6 is a laser velocimeter calibration phase (the scaling method block diagram is seen Fig. 2), and step 7 is the calibration result evaluation phase.

Wherein, the established angle and the constant multiplier error of the least squares identification three wave beam laser velocimeters described in the step 6, concrete implementation procedure is explained as follows:

With certain wave beam in the three wave beam laser velocimeters is established angle and the constant multiplier error calibrating method that example is introduced it, and the established angle of two other wave beam is identical therewith with the constant multiplier error calibrating method.

Definition V _LBe laser velocimeter output, Being respectively SINS/DGPS is three of x, y, the z output on axially at carrier, and δ K is a laser velocimeter constant multiplier error, and α, β, γ are respectively laser velocimeter and carrier is x, y, three axial angles of z.Record n point (n>=3) single beam knotmeter output V _L(1), V _L(2) ... V _L(n) and n point (n>=3) SINS/DGPS output

By $V_L=(1+\mathrm{\δ\; K})(V_x^b\mathrm{Cos}\mathrm{\α}+V_y^b\mathrm{Cos}\mathrm{\β}+V_z^b\mathrm{Cos}\mathrm{\γ})$ Can get

$\left[\begin{array}{c}{V}_L\left(1\right)\\ V_L\left(2\right)\\ \·\\ \·\\ \·\\ V_L\left(n\right)\end{array}\right]=\left[\begin{array}{ccc}{V}_x^b\left(1\right)& V_y^b\left(1\right)& V_z^b\left(1\right)\\ V_x^b\left(2\right)& V_y^b\left(2\right)& V_z^b\left(2\right)\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ V_x^b\left(n\right)& V_y^b\left(n\right)& V_z^b\left(n\right)\end{array}\right]\left[\begin{array}{c}\cos \mathrm{\α}\\ \cos \mathrm{\β}\\ \cos \mathrm{\γ}\end{array}\right][1+\mathrm{\δK}]$

Order $Z=\left[\begin{array}{c}{V}_L\left(1\right)\\ V_L\left(2\right)\\ \·\\ \·\\ \·\\ V_L\left(n\right)\end{array}\right],$ $H=\left[\begin{array}{ccc}{V}_x^b\left(1\right)& V_y^b\left(1\right)& V_z^b\left(1\right)\\ V_x^b\left(2\right)& V_y^b\left(2\right)& V_z^b\left(2\right)\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ V_x^b\left(n\right)& V_y^b\left(n\right)& V_z^b\left(n\right)\end{array}\right],$ $X=\left[\begin{array}{c}\mathrm{Cos}\mathrm{\α}\\ \mathrm{Cos}\mathrm{\β}\\ \mathrm{Cos}\mathrm{\γ}\end{array}\right][1+\mathrm{\δ\; K}]$

By the least squares estimation formula?

can solve for X, least squares estimation?

By cos ²α+cos ²β+cos ²γ=1

$\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2}=\sqrt{{(1+\mathrm{\δ\; K})}^2({\mathrm{Cos}}^2\mathrm{\α}+{\mathrm{Cos}}^2\mathrm{\β}+{\mathrm{Cos}}^2\mathrm{\γ})}=1+\mathrm{\δ\; K},$ Therefore $\mathrm{\δ\; K}=\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2}-1$

$\mathrm{\α}=\arccos (\hat{X}\left(1\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

$\mathrm{\β}=\arccos (\hat{X}\left(2\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

$\mathrm{\γ}=\arccos (\hat{X}\left(3\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

Wherein, the stated accuracy evaluation analysis described in the step 7, concrete implementation procedure explanation are as follows:

With calibration result δ K, α, β, γ substitution

${\hat{V}}_L=(1+\mathrm{\δ\; K})(V_x^b\mathrm{Cos}\mathrm{\α}+V_y^b\mathrm{Cos}\mathrm{\β}+V_z^b\mathrm{Cos}\mathrm{\γ}),$

Calculate through standards difference computing formula With knotmeter real output value V _L(1), V _L(2) ... V _L(n) dispersion degree between is judged the degree of accuracy of calibration result thus.

$\mathrm{\σ}=\sqrt{({\left[\widehat{V_L\left(1\right)-V_L\left(1\right)}\right]}^2+{\left[\widehat{V_L\left(2\right)-V_L\left(2\right)}\right]}^2+\·\·\·+{\left[\widehat{V_L\left(n\right)-V_L\left(n\right)}\right]}^2)/(n-1)}.$

3, advantage and effect: a kind of three wave beam laser velocimeter scaling methods of the present invention based on navigational system; The advantage of this method is to utilize inertial navigation and the high characteristics of differential GPS integrated navigation system output speed precision that three wave beam laser velocimeters are realized accurately demarcating; Solved the problem that three wave beam laser velocimeter established angles and scale error are difficult to test, for the raising of inertial navigation system/laser velocimeter integrated navigation precision provides the foundation.

Description of drawings

Fig. 1 is a SINS/DGPS combinational algorithm block diagram;

Fig. 2 is a laser velocimeter scaling method block diagram;

Fig. 3 is the process flow diagram of the present invention's three wave beam laser velocimeter scaling methods.

Symbol description is following among the figure:

DGPS: differential GPS

V ^t: the speed under the Department of Geography of integrated navigation system output

V ^b: the speed under the carrier system of integrated navigation system output

V _L: the speed of laser velocimeter output

α, β, γ: laser velocimeter wave beam and carrier are x, y, three axial angles of z

δ K: laser velocimeter constant multiplier error

SINS/DGPS: SINS/differential GPS

Embodiment:

See Fig. 3, a kind of three wave beam laser velocimeter scaling methods of the present invention based on navigational system, these method concrete steps are following:

Step 1, SINS, differential GPS and laser velocimeter are installed on the carrier navigational computer of bookbinding initial position parameters (comprising initial longitude, latitude and height) to inertial navigation.

Step 2, inertial navigation preheating, the output data of gathering gyro and accelerometer then.

Step 3, the gyro and the accelerometer data that collect are handled,, adopted second order leveling and orientation estimation algorithm to accomplish the coarse alignment of system, tentatively definite attitude of carrier angle according to SINS error Propagation Property and classic control theory.The coarse alignment time is 1 minute.Utilized the Kalman Filter Technology fine alignment behind the coarse alignment 5 minutes.

Step 4, startup differential GPS and laser velocimeter, inertial navigation system switches to SINS/DGPS integrated navigation pattern (the combinational algorithm block diagram is seen Fig. 1) by alignment pattern, carrier setting in motion after switching is accomplished.

Three different outputs are constantly adopted in step 5, the output of taking knotmeter under the carrier different motion state and the output of SINS/DGPS integrated navigation system at least.

Step 6, utilize the least square method respectively established angle and the constant multiplier error of three wave beams of identification laser velocimeter.

Step 7, stated accuracy evaluation analysis.

Can step 1-7 be divided into three phases, step 1-3 is the preparatory stage, and step 4-6 is a laser velocimeter calibration phase (the scaling method block diagram is seen Fig. 2), and step 7 is the calibration result evaluation phase.

Wherein, the established angle and the constant multiplier error of the least squares identification three wave beam laser velocimeters described in the step 6, concrete implementation procedure is explained as follows:

With certain wave beam in the three wave beam laser velocimeters is established angle and the constant multiplier error calibrating method that example is introduced it, and the established angle of two other wave beam is identical therewith with the constant multiplier error calibrating method.

Definition V _LBe laser velocimeter output,

Being respectively SINS/DGPS is three of x, y, the z output on axially at carrier, and δ K is a laser velocimeter constant multiplier error, and α, β, γ are respectively laser velocimeter and carrier is x, y, three axial angles of z.

Record n point (n>=3) single beam knotmeter output V _L(1), V _L(2) ... V _L(n) and n point (n>=3) SINS/DGPS output

By $V_L=(1+\mathrm{\δ\; K})(V_x^b\mathrm{Cos}\mathrm{\α}+V_y^b\mathrm{Cos}\mathrm{\β}+V_z^b\mathrm{Cos}\mathrm{\γ})$ Can get

$\left[\begin{array}{c}{V}_L\left(1\right)\\ V_L\left(2\right)\\ \·\\ \·\\ \·\\ V_L\left(n\right)\end{array}\right]=\left[\begin{array}{ccc}{V}_x^b\left(1\right)& V_y^b\left(1\right)& V_z^b\left(1\right)\\ V_x^b\left(2\right)& V_y^b\left(2\right)& V_z^b\left(2\right)\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ V_x^b\left(n\right)& V_y^b\left(n\right)& V_z^b\left(n\right)\end{array}\right]\left[\begin{array}{c}\cos \mathrm{\α}\\ \cos \mathrm{\β}\\ \cos \mathrm{\γ}\end{array}\right][1+\mathrm{\δK}]$

Order $Z=\left[\begin{array}{c}{V}_L\left(1\right)\\ V_L\left(2\right)\\ \·\\ \·\\ \·\\ V_L\left(n\right)\end{array}\right],$ $H=\left[\begin{array}{ccc}{V}_x^b\left(1\right)& V_y^b\left(1\right)& V_z^b\left(1\right)\\ V_x^b\left(2\right)& V_y^b\left(2\right)& V_z^b\left(2\right)\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ V_x^b\left(n\right)& V_y^b\left(n\right)& V_z^b\left(n\right)\end{array}\right],$ $X=\left[\begin{array}{c}\mathrm{Cos}\mathrm{\α}\\ \mathrm{Cos}\mathrm{\β}\\ \mathrm{Cos}\mathrm{\γ}\end{array}\right][1+\mathrm{\δ\; K}]$

By the least squares estimation formula?

can solve for X, least squares estimation?

By cos ²α+cos ²β+cos ²γ=1

$\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2}=\sqrt{{(1+\mathrm{\δ\; K})}^2({\mathrm{Cos}}^2\mathrm{\α}+{\mathrm{Cos}}^2\mathrm{\β}+{\mathrm{Cos}}^2\mathrm{\γ})}=1+\mathrm{\δ\; K},$ Therefore $\mathrm{\δ\; K}=\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2}-1$

$\mathrm{\α}=\arccos (\hat{X}\left(1\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

$\mathrm{\β}=\arccos (\hat{X}\left(2\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

$\mathrm{\γ}=\arccos (\hat{X}\left(3\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

Wherein, the stated accuracy evaluation analysis described in the step 7, concrete implementation procedure explanation are as follows:

With calibration result δ K, α, β, γ substitution

${\hat{V}}_L=(1+\mathrm{\δ\; K})(V_x^b\mathrm{Cos}\mathrm{\α}+V_y^b\mathrm{Cos}\mathrm{\β}+V_z^b\mathrm{Cos}\mathrm{\γ}),$

Calculate the velocity amplitude that knotmeter should responsive arrive through standard deviation computing formula (following formula)

With knotmeter real output value V _L(1), V _L(2) ... V _L(n) dispersion degree between is judged the degree of accuracy of calibration result thus.

$\mathrm{\σ}=\sqrt{({\left[\widehat{V_L\left(1\right)-V_L\left(1\right)}\right]}^2+{\left[\widehat{V_L\left(2\right)-V_L\left(2\right)}\right]}^2+\·\·\·+{\left[\widehat{V_L\left(n\right)-V_L\left(n\right)}\right]}^2)/(n-1)}.$ 。

Claims (3)

1. three wave beam laser velocimeter scaling methods based on navigational system, it is characterized in that: these method concrete steps are following:

Step 1, SINS, differential GPS and laser velocimeter are installed on the carrier, the bookbinding initial position parameters is the navigational computer of initial longitude, latitude and height to inertial navigation;

Step 2, inertial navigation preheating, the output data of gathering gyro and accelerometer then;

Step 3, the gyro and the accelerometer data that collect are handled,, adopted second order leveling and orientation estimation algorithm to accomplish the coarse alignment of system, tentatively definite attitude of carrier angle according to SINS error Propagation Property and classic control theory; The coarse alignment time is 1 minute, utilizes the Kalman Filter Technology fine alignment behind the coarse alignment 5 minutes;

Step 4, startup differential GPS and laser velocimeter, it is SINS/differential GPS integrated navigation pattern that inertial navigation system switches to SINS/DGPS by alignment pattern, carrier setting in motion after switching is accomplished;

Three different outputs are constantly adopted in step 5, the output of taking knotmeter under the carrier different motion state and the output of SINS/DGPS integrated navigation system at least;

Step 6, utilize the least square method respectively established angle and the constant multiplier error of three wave beams of identification laser velocimeter;

Step 7, stated accuracy evaluation analysis.

2. a kind of three wave beam laser velocimeter scaling methods according to claim 1 based on navigational system; It is characterized in that: the established angle and the constant multiplier error of the least squares identification three wave beam laser velocimeters described in the step 6, concrete implementation procedure is explained as follows:

With certain wave beam in the three wave beam laser velocimeters is established angle and the constant multiplier error calibrating method that example is introduced it, and the established angle of two other wave beam is identical therewith with the constant multiplier error calibrating method;

Definition V _LBe laser velocimeter output,

Being respectively SINS/DGPS is three of x, y, the z output on axially at carrier, and δ K is a laser velocimeter constant multiplier error, and α, β, γ are respectively laser velocimeter and carrier is x, y, three axial angles of z; Record n point (n>=3) single beam knotmeter output V _L(1), V _L(2) ... V _L(n) and n point (n>=3) SINS/DGPS output

By $V_L=(1+\mathrm{\δ\; K})(V_x^b\mathrm{Cos}\mathrm{\α}+V_y^b\mathrm{Cos}\mathrm{\β}+V_z^b\mathrm{Cos}\mathrm{\γ})$

$\left[\begin{array}{c}{V}_L\left(1\right)\\ V_L\left(2\right)\\ \·\\ \·\\ \·\\ V_L\left(n\right)\end{array}\right]=\left[\begin{array}{ccc}{V}_x^b\left(1\right)& V_y^b\left(1\right)& V_z^b\left(1\right)\\ V_x^b\left(2\right)& V_y^b\left(2\right)& V_z^b\left(2\right)\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ V_x^b\left(n\right)& V_y^b\left(n\right)& V_z^b\left(n\right)\end{array}\right]\left[\begin{array}{c}\cos \mathrm{\α}\\ \cos \mathrm{\β}\\ \cos \mathrm{\γ}\end{array}\right][1+\mathrm{\δK}]$

Order $Z=\left[\begin{array}{c}{V}_L\left(1\right)\\ V_L\left(2\right)\\ \·\\ \·\\ \·\\ V_L\left(n\right)\end{array}\right],$ $H=\left[\begin{array}{ccc}{V}_x^b\left(1\right)& V_y^b\left(1\right)& V_z^b\left(1\right)\\ V_x^b\left(2\right)& V_y^b\left(2\right)& V_z^b\left(2\right)\\ \·& \·& \·\\ \·& \·& \·\\ \·& \·& \·\\ V_x^b\left(n\right)& V_y^b\left(n\right)& V_z^b\left(n\right)\end{array}\right],$ $X=\left[\begin{array}{c}\mathrm{Cos}\mathrm{\α}\\ \mathrm{Cos}\mathrm{\β}\\ \mathrm{Cos}\mathrm{\γ}\end{array}\right][1+\mathrm{\δ\; K}]$

By the least squares estimation formula

that the solution of the least squares estimate of X

By cos ²α+cos ²β+cos ²γ=1

$\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2}=\sqrt{{(1+\mathrm{\δ\; K})}^2({\mathrm{Cos}}^2\mathrm{\α}+{\mathrm{Cos}}^2\mathrm{\β}+{\mathrm{Cos}}^2\mathrm{\γ})}=1+\mathrm{\δ\; K},$ Therefore $\mathrm{\δ\; K}=\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2}-1$

$\mathrm{\α}=\arccos (\hat{X}\left(1\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

$\mathrm{\β}=\arccos (\hat{X}\left(2\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

$\mathrm{\γ}=\arccos (\hat{X}\left(3\right)/\sqrt{\hat{X}{\left(1\right)}^2+\hat{X}{\left(2\right)}^2+\hat{X}{\left(3\right)}^2})$

3. a kind of three wave beam laser velocimeter scaling methods based on navigational system according to claim 1 is characterized in that: the stated accuracy evaluation analysis described in the step 7, and concrete implementation procedure is explained as follows:

With calibration result δ K, α, β, γ substitution

Calculate through standards difference computing formula

With knotmeter real output value V _L(1), V _L(2) ... V _L(n) dispersion degree between is judged the degree of accuracy of calibration result thus;

$\mathrm{\σ}=\sqrt{({\left[\widehat{V_L\left(1\right)-V_L\left(1\right)}\right]}^2+{\left[\widehat{V_L\left(2\right)-V_L\left(2\right)}\right]}^2+\·\·\·+{\left[\widehat{V_L\left(n\right)-V_L\left(n\right)}\right]}^2)/(n-1)}.$

Images