RU2722599C1 - Method for correcting strapdown inertial navigation system of unmanned aerial vehicle of short range using intelligent system of geospatial information - Google Patents

Abstract

FIELD: aviation.SUBSTANCE: invention relates to an unmanned aerial vehicle (UAV) positioning method in off-line state. For this purpose, current coordinates of information-measuring devices are continuously determined by low-precision strapdown inertial navigation system (SINS) of UAV. Periodical correction of current position based on signals of satellite navigation (SSN) is performed. Validation of SSN data is performed. In automatic mode, regions or objects of observation are fixed based on information of optoelectronic system. Corrections for current coordinates of the latter are generated. At the pre-start preparation stage, the period of forming the data areas of the strap-down inertial system is calculated. In flight, processing, geo-spatial referencing and storing images on UAV are stored.EFFECT: higher accuracy of drone positioning in autonomous mode.1 cl, 1 dwg

Description

The invention relates to the field of navigation of mobile technical systems and can be used in autonomous navigation systems for unmanned aerial vehicles (UAVs) of short range, which require the determination with high accuracy of the absolute coordinates of the location in the conditions of complete (partial) absence of signals from satellite navigation systems.

The UAV flight and navigation system is considered as an integrated navigation system (SPS), which includes a strapdown inertial navigation system (SINS) that operates continuously when measuring current acceleration (accelerometers), horizontal and vertical angles (gyroscopes), and a satellite receiver of a satellite signal from GLONASS global navigation system groupings, periodically adjusting current navigation information from SINS

Satellite navigation systems (GPS, GLONASS, GALILEO, BEIDOU) provide the necessary accuracy of determining the location on the Earth's surface, due to the influence of various factors, they also have disadvantages:

- the occurrence of an error caused by "multipath", which is caused by repeated reflections of the signal from the satellite from surrounding objects and surfaces before it enters the receiver.

- interference from closely located powerful sources of radio emissions: locators, television and radio transmitting stations, etc.

- poor quality or complete absence of a satellite signal caused by the unsuccessful geometric relative position of the UAV and navigation satellites.

- blocking associated with the use of special electronic warfare systems or intentional substitution of the navigation field (GPS spoofing).

As additional meters for correcting SINS on board an UAV, meters based on various physical principles for determining motion parameters can be used.

Known navigation system (NS) based on the visual navigation of UAVs. There are three main approaches to solving the problem of determining the parameters of the movement of an object:

1. Determining the parameters of movement and orientation by tracking the movement in the frame of images of characteristic (special) points of the area, with information prepared in advance. After identifying the points in each image, an accurate determination of the motion parameters allows you to determine the current location of the mobile technical system. The most difficult problem remains the detection of singular points and their correlation with a set of characteristic features.

2. The “ego motion” approach allows you to evaluate the relative movement of the camera in relation to its previous position. The data source is video frames. Having estimated the motion parameters and initial conditions, it is possible by integration to calculate the trajectory of the camera. An important advantage of this method (in contrast to the previous one) is the absence of the need to identify special points on the ground. The problem side is the growth of errors in the integration process. Thus, there is an analogue of the “drift” of the position estimation, and the accuracy of the estimation deteriorates over time (or with the growth of the distance traveled).

The two approaches outlined are combined by the need for sequential image analysis.

3. The third approach allows you to compensate for the effect of "drift", using an additional algorithm for direct estimation of location. The digital terrain data (DTM) is used, which allows to discretely determine the coordinates of the location through the matrix of heights above sea level.

The methods used make it possible to complex the CRM data and the species data obtained from the object’s camera to assess the position and orientation of the object, bypassing the relief reconstruction stage from the video data [1].

A known system for autonomous landing of an unmanned aerial vehicle (UAV) on a moving vessel (RU No. 110070, 2011), containing a television homing system associated with the UAV motion control system, and ship landing equipment, which includes a gripping device mounted on remote end of the rotary crane-beam, and light beacons, characterized in that the structure of the ship's landing equipment includes at least three light beacons, the first of which is mounted on an arm, the rack of which is located outside the UAV stopping distance and is carried along the side the vessel forward relative to the crane beam, on which the second and third light beams are installed, the second and third light beacons form the base of an equilateral triangle, the apex of which is the projection of the first light beacon on a vertical plane passing through the second and third light beacons when installing the crane beams perpendicular to the side of the ship, and in the geometric center of this triangle is the aiming point at which the UAV is guided by a television homing system that contains a television camera, a field of view control unit, the output of which is connected to the control input of the television camera, a video signal processing unit, a direction determination unit for the aiming point, a range determination unit , a unit for determining the pitching parameters, a memory unit and a block for correcting the direction to the aiming point, while the output of the ranging unit and the output of the unit for determining the direction to the aiming point, on which the current value of the angle of rotation of the line of sight of the aiming point is formed, are connected to the corresponding inputs of the direction correction unit to the aiming point and the block for determining the pitching parameters, the outputs of which according to the signals of the current values of the amplitude and the period of the pitching and the output of the block determining the direction to the aiming point, on which the current value of the angle and sighting of the aiming point, connected to the corresponding inputs of the direction correction unit to the aiming point, the inputs of which are signals from the current values of the yaw angle and pitch of the UAV connected to the corresponding outputs of the navigation device, which is part of the UAV's motion control system, which also contains a block for calculating the angle of attack and the slip module for determining the speed module and the block for generating UAV position control signals, the outputs of which through the block of steering drives are connected to the corresponding inputs of the UAV rudder block, while the inputs of the UAV block for generating signals for controlling the position of the UAV receive the correct values for the rotation angles and tilt of the line of sight aiming, connected to the corresponding outputs of the direction correction block to the aiming point, the inputs to which the values of the transmission coefficients of the control loops along the yaw, pitch and roll angles and the values of the transmission coefficients of the control loops yaw, pitch and roll speeds are connected to the corresponding outputs of the memory unit, inputs to which the current yaw, pitch and roll angles and current values of the yaw, pitch and roll angular speeds of the UAV are connected to the corresponding outputs of the navigation device, and the inputs, which receive the current values of the angle of attack and slip are connected to the outputs of the block for calculating the angle of attack and slip, the inputs of the block for calculating the speed module are connected to the outputs of the navigation device, which form the current values of the longitudinal, vertical and lateral components of the speed of the UAV, and its output is connected to the corresponding inputs of the direction correction unit for the aiming point and the unit for calculating the angle of attack and slip, whose inputs are connected to the outputs of the navigation device by the signals of the vertical and lateral components of the UAV speed, the output of the television camera, which forms the current value of the angle of view of the television camera in horizontal ntal plane, connected to the corresponding inputs of the direction determining unit to the aiming point and the range determination unit, the output on which the current value of the angle of view of the television camera in the vertical plane is formed is connected to the corresponding input of the direction determining unit to the aiming point, and the outputs of the television camera to of which the synchronization signal is generated in rows and the amplitude value of the interrogated resolution element is connected to the corresponding inputs of the video processing unit, the outputs of the video processing unit, on which the current values of the vertical and horizontal coordinates of the first, second and third beacons are formed, are connected to the corresponding inputs of the direction determination unit the aiming point, in addition, the output of the video signal processing unit, on which the current value of the vertical coordinate of the first beacon is generated, is connected to the corresponding input of the field of view control unit, and the outputs, on which current values of the horizontal coordinates of the second and third beacons are generated, connected to the corresponding inputs of the range-determining unit and the field of view control unit, the outputs of the memory unit, on which the number of horizontal and vertical resolution elements are formed between the center of the frame and each of the beacons in the case where the center of the triangle formed by the first, second and third light beacons coincides with the center of the frame of the television camera, connected to the corresponding inputs of the unit for determining the direction to the aiming point, the outputs on which the values of the mean square deviation of the noise amplitude and the probability of false alarm are formed are connected to the corresponding the inputs of the video signal processing unit, the outputs of the memory block, on which the values of the number of resolution elements are generated in the row and column of the television camera, are connected to the corresponding inputs of the video signal processing unit and the field of view control unit, output, to which generates the value of the transmission coefficient of the control field of view, connected to the corresponding input of the control unit of the field of view, and the outputs on which the value of the distance along the crane beam between the second and third light beacons and the value of the number of resolution elements in the line of the television camera are connected the inputs of the range determination unit [2].

The well-known UAV is designed to perform a narrow specific task - landing it on a ship on which special beacons are equipped - landmarks to ensure auto tracking to the touchdown point, and braking devices.

The disadvantage of this auto tracking system is that in order to obtain the necessary positioning accuracy, special beacons and other devices are needed, which limits the radius of its application.

Known unmanned aerial vehicle for monitoring extended objects (RU No. 137016, 2014) including a glider, a power plant, an automatic and remote control system for the flight of the aircraft and the operation of its systems, an on-board diagnostic system for the condition of extended objects, is equipped with a universal gyro-stabilized platform with the possibility of simultaneous placing on it at least one sensor of the device selected from the group of photo or video, as well as a thermal imager and gas analyzer, and the glider includes a removable fuselage and wing [3].

The disadvantages of this system are the need for continuous emission of electromagnetic energy throughout the flight, therefore, the insecurity from possible organized interference, as well as the absence of signs of local identification of extended objects, and as a result, the unsatisfactory accuracy of UAV guidance, determined only by the GLONASS / GPS system data.

The closest in technical essence to the claimed invention is a control system for an unmanned aerial vehicle (RU155323, 2015). comprising an automatic and remote flight control system of an aircraft, including a satellite navigation system of an unmanned aerial vehicle, a remote control signal receiver, a correction signal control unit, and an autopilot for controlling the aerodynamic surface of an unmanned aerial vehicle, and an optoelectronic system consisting of a gyro-stabilized platforms with image sensors located on it, operating in the visible and infrared ranges of radiation associated with the information transmitter, a system for automatic recognition and auto tracking of observation objects, including a block of a reference image, a block for recognizing a reference, an image preparation block, a decision block, an auto tracking and coordinate correction unit, a screen information shaper, wherein the output of the reference image block is connected to the first input of the recognition unit by the standard, the second input of which it is connected to the output of the image preparation unit, the output of the recognition unit by reference is connected to the input of the decision unit, the first output of which is connected to the input of the auto tracking unit and coordinate correction, and the second output is connected to the first input of the image preparation unit, the second input of which is connected to the optical output -electronic system, to the first input of the screen information generator and to the second input of the auto tracking and coordinate correction unit, the first output of which is connected to the second input of the screen information generator, the output of which is connected to the input of the information transmitter, the second output of the auto tracking unit and coordinate correction is connected to the input block correction of control signals [4].

The disadvantages of the prototype are:

- Mandatory advance preparation of relevant information in the form of images of protected areas or landmarks;

- the lack of a solution in cases of substitution or failure of the satellite signal receiver to the zone of likely organized interference of the radio link and, as a result, the UAV not falling into the area of the reference images of the protected area or landmarks, due to growing errors in determining the coordinates;

- the length of time spent on comparing the images is not taken into account in the case of a large number of embedded reference images of the protected area or landmarks.

The aim of the present invention is to provide a method for correcting low-current SINS of an autonomous UAV, short-range and short-range using an intelligent geospatial information system (ISGI), which allows correcting SINS in the absence of prepared reference images in advance with the complete or partial absence of signals from the SNA, taking into account the length of time spent comparing reference images with current ones.

The required technical result is achieved by the fact that with the correction method of the strapdown inertial navigation system of a short-range unmanned aerial vehicle using an intelligent geospatial information system, which consists in automatically detecting an area or object of observation, obtaining a reliable result of automatic recognition, auto tracking of the object and forming correction coordinate coefficients due to to obtain a mismatch with respect to the current coordinates of the recognized image, an intelligent image processing process is additionally presented, consisting in calculating the period of formation of correction areas for creating a reference image when flying to a pivot point and justifying the height of the return flight using the calculated coefficients characterizing the sparseness of the common points of the underlying surface and the actual share of commonality reference image from the current.

The invention is illustrated by a functional diagram, where in FIG. 1 are shown:

1 - automatic and remote flight control system;

2 - an intelligent system of geospatial information;

3 - optoelectronic system (ECO);

4 - information transmitter;

5 - block correction of control signals;

6 - block determining reliable coordinates;

7 - receiver of remote control signals;

8 - autopilot;

9 - bar altimeter;

10 - magnetometer;

11 - inertial navigation system;

12 - satellite navigation system;

13 - multiplexer;

14 - chronograph;

15 - block auto tracking and coordinate correction;

16 - block preparation of information;

17 - decision block;

18 - recognition unit by reference;

19 is a block creating a reference image;

20 - block storage of reference images.

The invention works as follows: in the process of regular flight, the algorithm for the formation of correction areas in the ISGI is executed. To implement the algorithm, at the start-up stage, the period of formation of the correction areas ΔT _{foc is} calculated.

Time for the calculated trajectory is determined by the formula:

where t _pt is the length of the calculated trajectory in time, K _T is the number of periods of creation of correction areas, t _add is the permissible time of a straight flight from the moment of an emergency, R _pov is the UAV turning radius, V _pv is the UAV flight speed when making a turn.

в уравнении (1) представляет время, затрачиваемое на совершение маневра по развороту БПЛА и определяется из формулы по нахождению длины окружности, в данном случае ее половины. Решая, уравнение (1) относительно ΔТ_фок, получим:Expression

in equation (1) represents the time taken to complete a maneuver to turn the UAV and is determined from the formula for finding the circumference, in this case half of it. Solving equation (1) with respect to ΔТ _foc , we obtain:

The calculation of the period of formation of the correction areas is a prerequisite for the functioning of the algorithm for the formation of correction areas in the block for the formation of the correction area 19.

The automatic and remote flight control system 2 ensures the UAV is interconnected with the ground control point through the information transmitter 4. For this, the multiplexer 13 generates information from the OES 7 and the reliable coordinates determination unit 6. The UAV coordinates are determined by the satellite SNA 12 and in block 6, due to the redundancy of the information received from the barometer 9, magnetometer 10 and SINS 11 are checked for reliability. At the ground control point, the operator, observing the image with geospatial reference, can make changes to the flight path through the receiver of remote control signals 7.

Meanwhile, in flight ISGI 2 with a period? T formation areas correction processing _jib carries a series of sequential images with the ECO 3. The pre-processed image information with the preparation unit 16 output is input to the unit creating the reference image, where the image of the binding geospatial information block 6 The processed image is stored in the storage unit of the reference image 20.

In the case of loss of a signal with satellites, failure of the SNA or the determination of the replacement of the navigation field by the reliable coordinates determining unit 6, the reference image storage loop is disabled. For UAV navigation, correction areas are used that are formed during flight in the normal mode or pre-prepared reference images of block 20.

Due to the increase in SINS errors, there is the possibility that, at a constant height, the UAV will not cross the reference area and will not be able to carry out correction. To form the necessary conditions for the capture of terrain by current ECO images in the control signal correction block 5, an algorithm for calculating the UAV flight altitude is used.

- координаты геометрических центров областей коррекции, Iⁱ - изображение областей коррекции с геопространственной привязкой, t_крит - расчетная длительность полета БПЛА в отсутствии спутникового сигнала, в течение которого требуется коррекция ИНС,

- координаты угловых точек периметра области коррекции,

- коэффициент, характеризующий долю от всей площади для каждой области коррекции, необходимой для определения минимального количества общих точек, позволяющих провести коррекцию,

- высота, на которой были получены изображения каждой области коррекции.Initial data is generated: X _S , Y _S , Z _S - coordinates of the start point, X _F , Y _F , H _F - coordinates of the point F of the start of the turn maneuver,

are the coordinates of the geometric centers of the correction areas, I ⁱ is the image of the correction areas with geospatial reference, t _crit is the estimated UAV flight duration in the absence of a satellite signal, during which the ANN correction is required,

- coordinates of the corner points of the perimeter of the correction area,

- coefficient characterizing the proportion of the total area for each correction area necessary to determine the minimum number of common points that allow for correction,

- the height at which images of each correction area were obtained.

The coordinates of the maneuver end point for a straight flight to the next route point are calculated.

The maximum deviation from the correction area is determined. The maximum error is calculated according to the technical characteristics of the SINS during the flight to the correction area.

A sufficient maximum height of H ₂ is determined by the formula:

where H _2max - sufficient maximum altitude above the correction area, m; H ₁ - the height of the image creation of the correction area, m; To _{1 the} minimum share of the reference image necessary for the correction,%; K ₂ - the minimum share of the common reference image from the current,%.

The coefficient K ₁ characterizes the sparseness of the common points of the underlying surface from a height of H ₁ . The minimum number of common points μ required for the correction from a height of H ₂ takes a value equal to μ = 3. For each correction area, the minimum area is calculated, which should fall into the ECO lens from a height and have three common points, for any random direction of intersection.

From the sensors of the OES 3, the image is fed to the inputs of the information preparation unit 16, the auto tracking unit and the coordinate correction 15 and the chronograph 15. The pre-processed image from the output of the unit 16 is fed to one of the inputs of the recognition unit by reference 18, and the reference image stored in storage unit of reference images 20. In order to reduce the time of comparison of the reference image with the current one, crude information of the inertial navigation system 11 is used, which is fed to one of the inputs of the block 20 and sets a certain range for enumerating the reference images.

In block 18, the degree of similarity of the current image with the standard is determined, and the result is transmitted to decision block 17. In the case of an unsatisfactory result, i.e. assessment of similarity below the established criterion, the corresponding signal is transmitted to block 16 for making changes to the preprocessing algorithm of the current image. This can be zooming, image rotation, a contour or two-level representation of the image, switching the spectral range from visible to infrared and vice versa, etc. In the case of a satisfactory identification assessment, the necessary signal is input to block 15, in which the identified object is captured and auto-tracked , determination of new coordinates in the field of view of the optical means. Simultaneously with the signal of block 15, a signal from the chronograph 14 is received in block 5 of the correction of control signals 14. In block 5, the time taken to identify the image is taken into account and a correction signal is generated for SINS 11 relative to the current coordinates of the recognized image. The process of image similarity analysis and decision making is repeated in each subsequent frame of the video sequence.

Thus, in the proposed new method, in addition to those listed in the closest analogue, the following additional steps are used:

1. At the stage of advance preparation of information, the period of formation of correction areas is calculated.

2. In a regular UAV flight with a period of formation of correction areas, images with geospatial reference are created.

3. Intelligent image processing. Coefficients are calculated that characterize the sparseness of the common points of the underlying surface and the actual fraction of the commonness of the reference image from the current, necessary to determine the sufficient UAV flight altitude.

A comparative analysis of the essential features of existing methods for determining UAV navigation parameters and the present method shows that the proposed method, based on the use of additional operations related to the formation of a period of correction areas, intelligent image processing and calculation of altitude coefficients, differs in that more information is provided by processing more information accurate determination of current navigation parameters in the absence of signals from the SNA.

Thus, the technical result of the invention is achieved — an increase in the accuracy of UAV positioning in the autonomous mode when using small-sized navigation sensitive elements, due to zeroing the growing SINS errors.

Sources of information:

1. Kupervasser O. Yu., Voronov V.V. Correction of inertial navigation system's errors by the help of video-based navigator based on Digital Terrarium Map.

2. Utility model RU No. 110070, IPC B64F 1/18; B64C 13/18; P05 B 1. 10, published: 11/10/2011.

3. Utility model RU No. 137016, IPC B64D 43/00, published: 01/27/2014.

4. Utility model RU No. 155323, IPC B64D 43/00, published: 09/27/2015.

Claims (1)

A method for positioning an unmanned aerial vehicle (UAV) in stand-alone mode, which consists in continuously determining the current coordinates with information-measuring devices of a low-precision strapdown inertial navigation system (SINS) of the UAV, periodically adjusting the current position using satellite navigation signals (SNA), and checking SNA data for reliability, automatically record areas or objects of observation according to the information of the optoelectronic system, form corrective corrections for the current coordinates of the UAV, characterized in that at the start-up stage, the period of formation of the data correction areas of the strapdown inertial system is calculated, during the flight processing, geospatial reference and save the images on board the UAV, form the initial data: the coordinates of the start point X _S , Y _S , Z _S , the coordinates of the start point of the maneuver on the turn X _F , Y _F , H _F , the coordinates of the geometric their centers of correction areas

, the maximum flight altitude H _2max , the minimum number of common correction points μ, the minimum share of the reference image K ₁ , the minimum share of the images K ₂ , calculate the coordinates of the end point of the maneuver for a straight flight to the next route point, for navigation determine the degree of similarity of the current image with as a standard, they capture and auto-track the identified object, generate a correction signal for SINS relative to the current coordinates, determine the new coordinates, repeat the process of analyzing the similarity and making decisions for each subsequent frame.

Images