RU2706444C1 - Method for local navigation of mobile object - Google Patents

Abstract

FIELD: aviation.

SUBSTANCE: invention relates to navigation and is intended for determination of coordinates of a mobile object on a runway (taxiway, motorway, etc.) with code marks installed on it, as well as coordinates and dimensions of damages and obstacles on the runway. It can be used both for autonomous operation and for operation in complex with other navigation systems. Disclosed method comprises successive photo exposure of the earth's surface by means of a camera mounted on the mobile object, selecting on each pair of two consecutive images a group of at least three code marks, determining coordinates of label centres in the associated coordinate system, reading from each mark its code and from it coordinates of mark in horizontal coordinate system are determined, linear and angular coordinates of movable object, as well as coordinates and dimensions of damages and obstacles on runway are calculated.

EFFECT: technical result is higher accuracy of measuring coordinates of location on entire runway (taxiway).

1 cl, 2 dwg

Description

The invention relates to navigation and is intended to determine the coordinates of a moving object on the runway (taxiway, highway, etc.). It can be used both for stand-alone operation, and for working in conjunction with other navigation systems.

A known method for determining the orientation of a moving object by the coordinates of reference sources (RI), including determining signals corresponding to the position of the image of the RI in the plane of the photodetector, determining the angles of bearings of each RI from the received signals taking into account the focal length of the lens and determining the coordinates of the RI according to the angles-bearings taking into account the distance between the RI.

This method is implemented in a device containing reference sources on a moving object, and on a stationary one - two photodetector optical-location blocks and a block for determining the coordinates of the RI and the orientation of the moving object, described in the patent for the invention [GB 2002986 A publ. 02/28/1979.]

The disadvantage of the described analogue is the use of reference sources with a known location on a moving object, as well as determining only the orientation of a moving object at an unknown location.

Known closest to the claimed invention is a method for determining the location and angular orientation of a moving object relative to the runway, based on the reception of radiation from three ground laser beacons mounted on board a moving object and system (light contrasts), described in the patent for the invention [ RU 2347240, publ. 02/20/2009] The coordinates of the images of laser beacons on a photosensitive matrix are determined, and then the coordinates and angular position of the moving object relative to the runway are calculated.

The disadvantage of this method is the low accuracy of determining coordinates at significant distances from the beacons, due to the quadratic dependence of the errors on the distance to the beacons.

The technical result when using the invention is to increase the accuracy of measuring location coordinates on the entire runway (taxiway), this is achieved by using a series of code marks with known coordinates installed along the runway so that at least three marks are in the camera’s field of view at any time Therefore, the measurement process is carried out at their best mutual arrangement.

The specified technical result of the invention is achieved by the fact that in the method of navigating a moving object, based on photo exposure of the earth's surface by the camera, processing their digitized image and calculating the coordinates of the location of the moving object, sequential photo exposure of the earth's surface with code marks installed on it is carried out using a camera mounted on a moving object, select on each image a group of at least three code labels, determine the coordinates you their centers in the related coordinate system, is read from each tag and its code thereon tags define the coordinates in the horizontal coordinate system is calculated linear and angular coordinates of the mobile object and the position and dimensions of the obstacles on the runway.

The invention consists in the following:

consecutive photo exposure of the earth's surface with code marks installed on it by means of a camera mounted on a moving object;

selection on each image of a group of at least three code labels and determining the coordinates of the centers of the code labels in the associated coordinate system;

reading from each label its code and determining from it the coordinates of the label in a horizontal coordinate system;

calculating the linear and angular coordinates of a moving object, as well as the coordinates and dimensions of obstacles on the runway.

The invention is illustrated in FIG. 1 and FIG. 2.

In FIG. 1 shows examples of nine-digit binary code codes 000 000 001 and 000 000 010.

Figure 2 shows the optical measurement scheme by means of a camera, where are indicated:

OXYZ - coordinate system associated with a moving object (OX - longitudinal axis, OY - normal axis, OZ - transverse axis - GOST 20058-80);

O'X'Y'Z 'is the measuring coordinate system associated with the photomatrix (the O'X axis is the optical axis of the digital camera, O'Y' and O'Z 'are the symmetry axes of the photomatrix);

O''X''Y''Z '' - the normal coordinate system associated with the runway (O''X '' - the axis of symmetry of the runway, O''Y '' - the vertical axis, O'Z '- forms the right system coordinates);

F is the optical center of the photo lens;

FM - photomatrix;

M ₁ M ₂ , M ₃ - centers of navigation marks;

R ₁ , R ₂ , R ₃ - image labels; G is the center of the funnel.

The problem of determining the coordinates of a moving object is solved by using the necessary number of stationary optical navigation marks along the runway, the coordinates of which are known on the earth’s surface, and the position of the moving object relative to the marks is determined by optical-electronic means. For reliable recognition of marks, they are numbered by means of a code, and their coordinates in the O''X''Y''Z '' system are determined by the coded numbers. A technology known as a barcode can be used that successfully works when marking products. We use a multi-line nine-digit code tag (3 × 3) with a capacity of 2 ⁹ = 512, shown in FIG. one.

The moving object is equipped with a monocular gyrostabilized camera connected to a computer, which, using images from the photomatrix, solves the problem of determining the coordinates of the points of the runway landscape being photographed. All navigation marks are horizontal.

In the image field of the runway or taxiway, a group of navigation marks (M1, M2, M3 ...) is selected, center points in the image field of the marks are selected and the coordinates of these points are determined on the photomatrix y _n , z _n , where index n = 1,2,3 ... - tag code number, respectively. Moreover, a correspondence is established between the coordinates of these points in the normal O``X''Y''Z '' system and the code numbers.

The images of the central mark on the photomatrix, in which 0 or 1 are encoded, is a figure close to a circle with a diameter of tens of pixels, and depending on the recorded code, this image can be in the form of a ring or a circle. The coordinates of the center of this image are determined in the presence of photomatrix noise superimposed on the actual image obtained by the lens under various lighting conditions and from different angles. Therefore, to solve this problem with an error of not more than 0.1 pixels, the integrated brightness of the entire image of the circle is taken into account

The center of the brightness function of the label image can be determined in the same way as the center of mass of a volumetric body is calculated [Maths manual for students of technical schools. Bronstein I.N., Semendyaev K.A.- M .: Nauka, 1981. - P. 332].

where r is the pixel size;

- интегральная яркость изображения области W_n;

- integrated brightness of the image area W _n ;

W _n - the inner region of the image label, satisfying the condition

R (i, j) ≥ R _nop ;

V _n - the outer region of the image label, satisfying the condition

R (i, j) <R _nop , while the outer boundary of the region V _n is a circle of radius equal to the diameter of the inner boundary of the region V _n ;

R (i, j) is the brightness of the pixels depending on the row number i and column j;

- пороговое значение функции яркости;

- threshold value of the brightness function;

I _vn is the number of pixels in the inner region W of the tag image;

I _extern - the number of pixels in the outer region V of the label image,

единичные орты связанной системы координат OXYZ;

unit vectors of the associated coordinate system OXYZ;

единичные орты измерительной системы координат O'X'Y'Z';

single unit vectors of the measuring coordinate system O'X'Y'Z ';

,

,

единичные орты нормальной системы координат O''X''Y''Z''; F - фокусное расстояние фотообъектива, м.

,

,

unit vectors of the normal coordinate system O``X''Y''Z ''; F is the focal length of the photo lens, m

и

,

,

связаны между собой известным соотношениемSingle unit vectors

and

,

,

are interconnected by a known relation

where b _ij are the elements of the matrix of guide cosines (c), i-1 ... 3, j-1 ... 3.

where the angles θ _υ , θ _ψ , θ _γ are the orientations of the cardan suspension relative to the connected system

связаны между собой соотношениемSingle unit vectors

related by the ratio

where α _ij are the elements of the matrix of guide cosines (A), i-1 ... 3, j-1 ... 3.

ψ is the yaw angle, υ is the pitch angle, γ is the angle of heel.

The transformation of the coordinates of an arbitrary vector from a coupled coordinate system OXYZ to the normal coordinate system O''X''Y''Z '' is carried out by means of a well-known relation that takes into account only the mutual angular position of the coordinate systems. The relationship between the old and new coordinates of the vector for the case when both bases are orthonormal is defined by the following formula:

Then the coordinates of the measuring and normal systems are connected by the following relation

и центра воронки

в нормальной системе координат, ее полуоси α, β и глубину h.It is necessary to determine the coordinates of the moving object

and the center of the funnel

in the normal coordinate system, its semiaxes α, β and depth h.

The vector describing the spatial position of the lens in the O''X''Y''Z '' system (point F) relative to the nth beacon is denoted by FM _n '':

,

,

- координаты вектора

,

- искомые координаты вектора FO'' (положение объектива в системе O''X''Y''Z'').Where

,

,

- vector coordinates

,

- the desired coordinates of the vector FO '' (the position of the lens in the O``X''Y''Z '' system).

на оси измерительной системы координат O'X'Y'Z' для чего используем соотношениеProject a vector

on the axis of the measuring coordinate system O'X'Y'Z 'for which we use the relation

в системе координат O'X'Y'Z' выражается следующим образомVector

in the coordinate system O'X'Y'Z 'is expressed as follows

и

лежат на одной прямой, поэтому должны удовлетворять условию коллинеарностиVectors

and

lie on one straight line, therefore they must satisfy the collinearity condition

In scalar form, this condition is rewritten as follows

Since the obtained equations are linearly dependent, which is easily verified, discarding the first of them and transforming the other two, we obtain

If there are six unknown two equations obtained, it is not enough to solve the problem of determining the location of a moving object, so we use three spaced code labels (n = 1,2,3 ...) that are not located on one straight line.

при этом величины

, F известны.Thus, the resulting equations bind six unknown variables

while the quantities

, F are known.

Let the gyro-stabilized platform frames be rotated at angles θ _ψ = 0, θ _υ = -π / 2, θ _γ = 0, then the matrix

L/2, где L - ширина ВПП, k - номер меткиPlace the marks around the runway as shown in FIG. 2 as follows

L / 2, where L is the runway width, k is the mark number

With this arrangement, each triple of marks forms an equilateral triangle. Given the accepted location for three consecutive marks and the camera, equations (6) will take the following form

γ, υ, ψ из любой точки над ВПП, при этом решение такой системы осуществляется методом последовательных приближений, который обеспечивает очень быструю сходимость.By solving the system of equations (8), the coordinates of the moving object are determined

γ, υ, ψ from any point above the runway, and the solution of such a system is carried out by the method of successive approximations, which ensures very fast convergence.

Determining the coordinates of a point feature on a runway

осуществляется по координатам его изображения на фотоматрице

Учитывая линейный характер преобразования изображения посредством фотообъектива связь координат точечных объектов на ВПП и их изображений на фотоматрице описывается следующим образом.Determining the coordinates of a point object (the center of the funnel, mine, foreign objects, etc.) on the runway

carried out by the coordinates of its image on the photomatrix

Given the linear nature of image conversion by means of a photo lens, the relationship between the coordinates of point objects on the runway and their images on the photomatrix is described as follows.

Unknown parameters c ₁₁ , c ₁₂ , c _xo , c ₂₁ , c ₂₂ , c _{z0 are} determined by substituting the known coordinates of three navigation marks into these ratios

Considering three even and three odd equations of this system, we write two systems of three equations

We replace the second equation of system (10) with the difference of the second equation and the first, and replace the third with the difference of the second and third, we obtain

Their solution has the form

We replace the second equation of system (11) with the difference of the second equation and the first, and the third we replace with the difference of the second and third, we obtain

Their solution has the form

точечного объекта с координатами его изображения на фотоматрице

Solutions of systems (10) and (11) are used to calculate the coordinates

point object with the coordinates of its image on the photomatrix

Funnel area calculation

This problem is solved taking into account the scaling of the image, which is described by system (9). The number of pixels in the funnel image is calculated, and then the area of the funnel is calculated

where k is the number of the obstacle (funnel).

The image conversion matrix is written as follows

- определитель матрицы С.Where

is the determinant of matrix C.

то эта матрица описывает поворот изображения объекта, а множитель

- масштабный коэффициент преобразования линейных размеров объекта. Поэтому площадь воронки вычисляется следующим образом.Since the determinant of the matrix

then this matrix describes the rotation of the image of the object, and the factor

- scale conversion coefficient of the linear dimensions of the object. Therefore, the area of the funnel is calculated as follows.

- линейный размер пиксела фотоматрицы.where N is the number of pixels in the image of the funnel,

- linear pixel size of the photomatrix.

Claims (1)

A method of local navigation of a moving object, based on photo exposure of the earth's surface by a camera, processing of a digitized image and calculating the coordinates of the location of a moving object, characterized in that the earth is sequentially exposed to the earth with code marks installed on it by a camera mounted on a moving object, and it is highlighted on each image a group of at least three code marks, determine the coordinates of their centers in the associated coordinate system nat, read its code from each mark and determine the coordinates of the mark in the horizontal coordinate system from it, calculate the linear and angular coordinates of the moving object, as well as the coordinates and dimensions of obstacles on the runway.

Images