A method for aerial refueling positioning using a target and dual cameras

Abstract

This invention provides a method for aerial refueling positioning using targets and dual cameras. A microcontroller controls a target system consisting of a liquid crystal light valve, a button battery, and a reflector to actively flash light. Three targets are fixed to the refueling hose, and two targets are fixed to the receiver aircraft's fuel tank. Two cameras are mounted on the refueling aircraft to continuously photograph the five targets. Based on the camera's intrinsic and extrinsic parameters, the pixel positions of the targets in the photographs, and the target distances, linear and nonlinear equations are constructed. Solving these equations yields the three-dimensional coordinates of the five targets. Averaging these coordinates yields the coordinates of the hose center and the fuel tank center, as well as their deviation vector. Finally, the vertical vector of the hose plane is calculated using the three targets on the hose. Through decomposition, the three-dimensional error vector required for hose control is obtained, completing the error measurement during the aerial refueling process.

Description

Technical Field

[0001] This invention relates to the field of optical positioning measurement, and more specifically, to a method for achieving aerial refueling positioning using a target and dual cameras. Background Technology

[0002] Traditional domestic aerial refueling positioning mainly relies on the pilot's superb skills to gradually synchronize the tanker and receiver aircraft while simultaneously achieving a hard docking of the refueling probe with the receiver aircraft's inlet. With international technological advancements, the use of retractable and horizontally / vertically movable flexible hoses for aerial refueling is becoming a hot research topic in domestic dual-use technologies. Besides the refueling docking position control algorithm, the most crucial aspect is aerial positioning and measurement, namely measuring the relative position, relative distance, and relative attitude of the two aircraft. Using a combination of targets and visible light industrial cameras for close-range measurement and control is currently a viable and relatively reliable solution. Based on this background, this invention proposes a method for aerial refueling positioning and calculation using a minimum configuration of five targets and two cameras. This method boasts advantages such as technological maturity and simplicity, high reliability, ease of implementation, and low economic cost, thus possessing high engineering practicality and market value.

[0003] It should be noted that the information in the background section above is only used to enhance the understanding of the background of the present invention, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0004] The purpose of this invention is to provide a method for aerial refueling positioning using a target and dual cameras, thereby overcoming the difficulties in positioning calculation and insufficient calculation rate during the refueling process caused by related technical limitations.

[0005] According to one aspect of the present invention, a method for aerial refueling positioning using a target and dual cameras is provided, comprising the following five steps:

[0006] Step S10: A control circuit is composed of an ultra-low power microcontroller. A target module system is composed of a button battery, a liquid crystal light valve, a reflector, and a control circuit. The microcontroller program sets the target flashing period and controls the target to form alternating bright and dark light spots for output.

[0007] Step S20: According to the design method of the target module system described above, five targets are manufactured. Targets one, two, and three are fixed with screws to the front end of the retractable fuel hose of the fuel dispenser at an equilateral triangle position centered on the fuel inlet, and the distance between targets one and two is measured. Targets four and five are fixed with screws at symmetrical positions on both sides of the center of the fuel tank inlet of the receiving machine. Then, two visible light industrial cameras are installed on the fuel dispenser. Based on the target flashing period, the continuous shooting frequency of the two cameras is set to the same value, and continuous images are taken of the five targets. The two cameras are connected to a computer to read the image data. Based on the target highlights in the camera images, the pixel positions of the five targets in the two camera images are located. Then, the pixel positions of the five targets in the two camera images are detected.

[0008] Step S30: Based on the pixel position of the first target in the two camera images, calculate the lateral pixel relative distance of the first target to the two cameras and the longitudinal pixel relative distance of the first target to the two cameras respectively; calculate the lateral comprehensive coefficient and the longitudinal comprehensive coefficient of the camera based on the known calibrated internal and external parameters of the industrial camera lens; finally, calculate the comprehensive position coefficient of the first target relative to the two cameras based on the lateral pixel relative distance of the first target to the two cameras, the longitudinal pixel relative distance of the two cameras, and the camera parameters.

[0009] In step S40, the combined position coefficients of the two cameras related to the second, third, fourth, and fifth targets are obtained according to the same steps as in S30. Based on the combined position coefficients of the first, second, and third targets relative to the first and second cameras, and the distance parameters between the first and second targets, two sets of nonlinear equations for the center of the refueling hose are constructed. The fsolve function of Matlab software is used to solve the equations, and two sets of solutions for the three-dimensional coordinates of the positions of the first, second, and third targets are obtained. The average value is then taken to obtain the nonlinear solution for the three-dimensional coordinates of the hose center.

[0010] Step S50: Based on the combined position coefficients of the first target relative to the first and second cameras, construct a system of four linear equations for the first target. Solve the system using the `solve` function in MATLAB to obtain four linear solutions for the first target. Then, calculate the average to obtain the linear mean solution of the three-dimensional coordinates of the first target. Then, construct a system of linear equations using the same method to obtain the linear mean solutions of the three-dimensional coordinates of the second, third, fourth, and fifth targets. Then, average the linear mean solutions of the three-dimensional coordinates of the first, second, and third targets to obtain the linear solution of the three-dimensional coordinates of the hose center. Average the linear mean solutions of the three-dimensional coordinates of the fourth and fifth targets to obtain the linear solution of the three-dimensional coordinates of the fuel tank inlet center.

[0011] Step S60: Based on the linear solution of the three-dimensional coordinates of the fuel tank inlet center and the nonlinear solution of the three-dimensional coordinates of the hose center, the three-dimensional vector of the relative position of the hose and fuel tank is calculated; based on the linear mean solution of the three-dimensional coordinates of the first, second, and third targets, the vertical vector of the hose cross-section is calculated; then, based on the decomposition of the vertical vector of the hose cross-section, the vertical component of the relative position three-dimensional vector and the hose planar component are obtained; then, based on the linear mean solution of the three-dimensional coordinates of the second and third targets, the x-axis component of the hose plane is calculated; then, based on the decomposition of the hose planar component and the hose x-axis component, the x-axis component and y-axis component of the relative position hose plane are obtained; finally, according to the continuous shooting frequency of the camera, the data of the x-axis component, y-axis component, and vertical component of the relative position three-dimensional vector are transmitted to the fuel hose control system as the deviation in three directions relative to the front of the hose, providing the error measurement information required for control, and realizing refueling positioning and control.

[0012] Furthermore, the target module system consists of a liquid crystal light valve (including a reflector and a button battery) and a control circuit board. The liquid crystal light valve and the control circuit board are flexibly connected, and the liquid crystal light valve is located on the underside of the control circuit board.

[0013] Furthermore, the control circuit board controls the intermittent bright and dark light spot output of the liquid crystal shutter in the following way: according to the characteristics of the liquid crystal shutter, the power input terminal of the liquid crystal shutter is set to a high level, and a low level is set when a light spot needs to be formed and maintained for a period of time, thus forming a stroboscopic light spot similar to an active light emission.

[0014] Furthermore, the target flashing period is set to T, and the distance between targets one and two is measured and denoted as d. 12 The pixel position of the first target captured by the camera in the first camera image is denoted as (u). 11i ,v 11i ), where i is an integer representing the pixel position at a time interval of T, which is (u 11i ,v 11i The value represents the pixel position of the first target in the first camera image at time t = Ti.

[0015] Furthermore, the pixel position of the j-th target obtained by the camera in the k-th camera image is denoted as (u jki ,v jki ), that is (u jki ,v jki ) represents the pixel position of the j-th target in the k-th camera image at time t = Ti, where j = 2, 3, 4, 5 and k = 1, 2.

[0016] Furthermore, based on the known calibrated internal and external parameters of the industrial camera lens and the pixel positions of the target in the first and second camera images, the lateral pixel relative distances of the first target to the first and second cameras are calculated as follows:

[0017] m a11i =(u 11i -u 10 );

[0018] n a11i =(v 11i -v 10 );

[0019] m a12i =(u 12i -u 20 );

[0020] n a12i =(v 12i -v 20 );

[0021] Among them u 10 v 10 This represents the x and y coordinates of the pixel position of the center of the first camera's image coordinate system in the pixel coordinate system; u 20 v 20 This represents the x-coordinate and y-coordinate of the pixel position of the center of the second camera's image coordinate system in the pixel coordinate system; m a11i n is the lateral pixel relative distance between the first target and the first camera; a11i The vertical pixel distance between the first target and the first camera is m. a12i n is the lateral pixel relative distance between the first target and the second camera; a12i The vertical pixel relative distance between the first target and the second camera.

[0022] Furthermore, based on the known internal and external parameters of the two calibrated industrial camera lenses, the camera's lateral and longitudinal composite coefficients are calculated as follows:

[0023]

[0024]

[0025] Where F1 is the camera's focal length, c x1 c y1The parameters are the actual physical dimensions of each pixel in the x and y directions of the camera image. These parameters are internal camera parameters and can be obtained from the camera manufacturer or through self-measurement and calibration; the acquisition process will not be elaborated here. m1 is the camera's horizontal composite coefficient, and n1 is the camera's vertical composite coefficient. Here, we select two cameras with the same lens model, so the above parameters are the same for both cameras.

[0026] Furthermore, based on the known calibrated internal and external parameters of the industrial camera lens, and the lateral and vertical pixel relative distances between the first target and the first camera, the comprehensive position coefficient vector of the first target relative to the first camera is calculated as follows:

[0027] g 111i =m a11i r 131 -m1r 111 ;

[0028] g 112i =m a11i r 132 -m1r 112 ;

[0029] g 113i =m a11i r 133 -m1r 113 ;

[0030] g 114i =-m a11i t 13 +m1t 11 ;

[0031] g 115i =n 11ai r 131 -n1r 121 ;

[0032] g 116i =n 11ai r 132 -n1r 122 ;

[0033] g 117i =n 11ai r 133 -n1r 123 ;

[0034] g 118i =-n 11ai t 13 +n1t 12 ;

[0035] Where t 11 ,t 12,t 13 The three axial translation parameters between the camera coordinate system and the world coordinate system of the first camera; r 111 ,r 112 ,r 113 ,r 121 ,r 122 ,r 123 ,r 131 ,r 132 ,r 133 For the nine rotation parameters of the first camera. 111i g 112i g 113i g 114i g 115i g 116i g 117i g 118i The comprehensive position coefficient of the first camera associated with the first target.

[0036] Furthermore, based on the known calibrated internal and external parameters of the industrial camera lens, and the lateral and vertical pixel relative distances between the first target and the second camera, the comprehensive position coefficients of the first target relative to the second camera are calculated as follows:

[0037] g 121i =m a12i r 231 -m1r 211 ;

[0038] g 122i =m a12i r 232 -m1r 212 ;

[0039] g 123i =m a12i r 233 -m1r 213 ;

[0040] g 124i =-m a12i t 223 +m1t 221 ;

[0041] g 125i =n 12ai r 231 -n1r 221 ;

[0042] g 126i =n 12ai r 232 -n1r 222 ;

[0043] g127i =n 12ai r 233 -n1r 223 ;

[0044] g 128i =-n 12ai t 23 +n1t 22 ;

[0045] Where t 21 ,t 22 ,t 23 The three axial translation parameters between the camera coordinate system and the world coordinate system of the second camera; r 211 ,r 212 ,r 213 ,r 221 ,r 222 ,r 223 ,r 231 ,r 232 ,r 233 Nine rotation parameters for the second camera; g 121i g 122i g 123i g 124i g 125i g 126i g 127i g 128i This is the combined position coefficient of the first target relative to the second camera.

[0046] Furthermore, following the same steps as in S30, the combined position coefficients of the two cameras related to the second, third, fourth, and fifth targets are obtained; let g be the denoted g. wp1i g wp2i g wp3i g wp4i g wp5i g wp6i g wp7i g wp8i Let w be the comprehensive position coefficient of the p-th camera related to the w-th target; where w = 2, 3, 4, 5 and p = 1, 2.

[0047] In one exemplary embodiment of the present invention, based on the combined position coefficients of the first, second, and third targets relative to the first camera, and the distance parameter between the first and second targets, a first set of nonlinear equations for the center of the refueling hose is constructed. The fsolve function of Matlab software is used to solve these equations to obtain two sets of solutions for the three-dimensional coordinates of the positions of the first, second, and third targets. The average value is then taken to obtain the nonlinear solution for the three-dimensional coordinates of the hose center, including:

[0048] g 111i x1+g 112i y1+g113i z1=g 114i ;

[0049] g 115i x1+g 116i y1+g 117i z1=g 118i ;

[0050] g 211i x2+g 212i y2+g 213i z2=g 214i ;

[0051] g 215i x2+g 216i y2+g 217i z2=g 218i ;

[0052] g 311i x3+g 312i y3+g 313i z3 = g 314i ;

[0053] g 315i x3+g 316i y3+g 317i z3 = g 318i ;

[0054]

[0055]

[0056]

[0057]

[0058] Where x i ,y i ,z i (i = 1, 2, 3) represents the solution to the first set of nonlinear equations for the three-dimensional coordinates of the i-th target; x bi ,y bi ,z bi (i = 1, 2, 3) uses the same method to solve the second set of nonlinear equations for the i-th target, x c ,y c ,z c The nonlinear solution is the three-dimensional coordinates of the hose center.

[0059] In one exemplary embodiment of the present invention, based on the combined position coefficients of the first target relative to the first and second cameras, a system of four linear equations for the first target is constructed. These equations are then solved using the `solve` function in MATLAB, yielding four linear solutions for the first target, including:

[0060] g 111i x1+g 112i y1+g 113i z1=g 114i ;

[0061] g 115i x1+g 116i y1+g 117i z1=g 118i ;

[0062] g 121i x1+g 122i y1+g 123i z1=g 124i ;

[0063] g 125i x1+g 126i y1+g 127i z1=g 128i ;

[0064] Where x l1i ,y l1i ,z l1i (i = 1, 2, 3, 4) represents any three of the four linear equations mentioned above. The solve function in MATLAB is used to solve each of them to obtain the four linear solutions of the first target, which represents the i-th linear solution of the first target.

[0065] In one exemplary embodiment of the present invention, the mean of the four linear solutions for the first target is then calculated to obtain the linear mean solution of the three-dimensional coordinates of the first target; then, the same method is used to construct a system of linear equations to solve for the linear mean solutions of the three-dimensional coordinates of the second, third, fourth, and fifth targets; then, the linear mean solutions of the three-dimensional coordinates of the first, second, and third targets are averaged again to obtain the linear solution of the three-dimensional coordinates of the hose center; the linear mean solutions of the three-dimensional coordinates of the fourth and fifth targets are averaged again to obtain the linear solution of the three-dimensional coordinates of the fuel tank inlet center, which includes:

[0066]

[0067]

[0068]

[0069] Where x l1 ,y l1 ,z l1This is the linear mean solution for the three-dimensional coordinates of the first target. lj ,y lj ,z lj (j=1,2,3,4,5) represents the linear mean solution for the three-dimensional coordinates of the second, third, fourth, and fifth targets obtained using the same method. Where x l ,y l ,z l The solution is the linear three-dimensional coordinates of the hose center. r ,y r ,z r The solution is the linear solution for the three-dimensional coordinates of the center of the fuel tank inlet.

[0070] In one exemplary embodiment of the present invention, the solution for the vertical component of the relative position three-dimensional vector and the x-axis component and y-axis component of the relative position hose plane includes:

[0071] e x =x r -x c ,e y =y r -y c ,e z =z r -z c ;

[0072] a1=x l2 -x l1 a2=y l2 -y l1 a3=z l2 -z l1 ;

[0073] b1 = x l3 -x l1 b2=y l3 -y l1 b3=z l3 -z l1 ;

[0074] c1=a2b3-a3b2, c2=a3b1-a1b3, c3=a1b2-a2b1;

[0075]

[0076] w3 = e x d1+e y d2+e z d3;

[0077] f1 = e x -w3d1,f2=e y -w3d2,f3=e z -w3d3;

[0078]

[0079] w1 = f1g1 + f2g2 + f3g3;

[0080]

[0081] Among them (e) x ,e y ,e z (d1, d2, d3) represents the three-dimensional vector of the relative position of the hose and the tank; (d1, d2, d3) represents the vertical vector of the hose cross-section; (a1, a2, a3), (b1, b2, b3), and (c1, c2, c3) are intermediate variables in the process of calculating the vertical vector of the hose cross-section. w3 represents the vertical component of the relative position three-dimensional vector; (f1, f2, f3) represents the planar component of the hose; (g1, g2, g3) represents the x-axis component of the hose plane, d w The distance between the second and third targets is w1. w1 is the x-axis component of the relative position hose plane, and w2 is the y-axis component of the relative position hose plane.

[0082] Beneficial effects

[0083] This invention provides a method for aerial refueling positioning using a target and dual cameras, and its main innovations are as follows:

[0084] The first method proposed is to use five targets and two cameras to realize optical measurement and positioning calculation of the aerial refueling process of the aircraft, which solves the problems of high measurement accuracy requirements during the refueling process and high measurement data output frequency requirements due to the high speed of the aircraft.

[0085] Secondly, a simplified implementation method using five targets and two cameras to achieve localization calculation is proposed, which has the advantages of low economic cost and fast calculation speed.

[0086] Thirdly, it provides a measurement and solution method that combines linear and nonlinear equations. The solution of the linear equations is used to calculate the positioning of the oil hose plane and the center point of the oil tank; the nonlinear equations are used to solve for the positioning of the hose center point; and a method for calculating the three-dimensional control error relative to the hose plane based on the coordinates of five targets is given, providing the most direct error measurement information for hose control.

[0087] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit the invention. Attached Figure Description

[0088] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and, together with the description, serve to explain the principles of the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0089] Figure 1 This is a flowchart of a method for aerial refueling positioning using a target and dual cameras provided by the present invention;

[0090] Figure 2 This is a schematic diagram of the target module system device structure of the method provided in the embodiment of the present invention;

[0091] Figure 3 This is a schematic diagram of the liquid crystal light valve of the method provided in the embodiment of the present invention;

[0092] Figure 4 This is a schematic diagram of the liquid crystal light valve flickering method provided in the embodiments of the present invention;

[0093] Figure 5 This is a schematic diagram of the relative positions of the target camera and the fuel tank hose during the refueling process provided in the embodiments of the present invention. Detailed Implementation

[0094] Exemplary embodiments will now be described more fully with reference to the accompanying drawings. However, these exemplary embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make the invention more comprehensive and complete, and to fully convey the concept of the exemplary embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of the invention. However, those skilled in the art will recognize that the technical solutions of the invention may be practiced with one or more of these specific details omitted, or other methods, components, apparatus, steps, etc., may be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of the invention.

[0095] This invention provides a method for aerial refueling positioning using targets and dual cameras. A microcontroller controls a target system consisting of a liquid crystal light valve, a button battery, and a reflector to actively flash light. Three targets are fixed to the refueling hose, and two targets are fixed to the receiver aircraft's fuel tank. Two cameras are mounted on the refueling aircraft to continuously photograph the five targets. Based on the camera's intrinsic and extrinsic parameters, the pixel positions of the targets in the photographs, and the target distances, linear and nonlinear equations are constructed. Solving these equations yields the three-dimensional coordinates of the five targets. Averaging these coordinates yields the coordinates of the hose center and the fuel tank center, as well as their deviation vector. Finally, the vertical vector of the hose plane is calculated using the three targets on the hose. Through decomposition, the three-dimensional error vector required for hose control is obtained, completing the error measurement during the aerial refueling process.

[0096] Below, we will combine the appendix Figure 1 The present invention provides a further explanation and description of a method for aerial refueling positioning using a target and dual cameras. (Reference) Figure 1 As shown, this method for aerial refueling positioning using a target and dual cameras includes the following steps:

[0097] Step S10: A control circuit is composed of an ultra-low power microcontroller. A target module system is composed of a button battery, a liquid crystal light valve, a reflector, and a control circuit. The microcontroller program sets the target flashing period and controls the target to form alternating bright and dark light spots for output.

[0098] Specifically, the target module system equipment structure diagram is as follows: Figure 2 As shown, the target module consists of two parts: a liquid crystal light valve (including a reflector and a button battery) and a control circuit board. A schematic diagram of the liquid crystal light valve is shown below. Figure 3 As shown. Based on the characteristics of the liquid crystal shutter, by setting the power input terminal of the liquid crystal shutter to a high level and setting it to a low level when a light spot needs to be formed, and maintaining this low level for a period of time, a strobe effect similar to active light emission is created, such as... Figure 4 As shown. Preferably, in this example, the target flashing period is selected to be 5ms; there are general methods and programs for controlling the liquid crystal light valve to output the light spot using a microcontroller, which are not the focus of this invention and will not be described further here.

[0099] Step S20: According to the design method of the target module system described above, five targets are manufactured. Targets one, two, and three are fixed with screws to the front end of the retractable fuel hose of the fuel dispenser at an equilateral triangle position centered on the fuel inlet, and the distance between targets one and two is measured. Targets four and five are fixed with screws at symmetrical positions on both sides of the center of the fuel tank inlet of the receiving machine. Then, two visible light industrial cameras are installed on the fuel dispenser. Based on the target flashing period, the continuous shooting frequency of the two cameras is set to the same value, and continuous images are taken of the five targets. The two cameras are connected to a computer to read the image data. Based on the target highlights in the camera images, the pixel positions of the five targets in the two camera images are located. Then, the pixel positions of the five targets in the two camera images are detected.

[0100] Specifically, this can be broken down into the following four steps. Step 1: Based on the design method of the target module system described above, fabricate five targets. Fix each target to an equilateral triangle centered on the retractable fuel hose of the fuel dispenser using screws. Measure the distance between targets one and two, denoted as d. 12 .

[0101] The second step is to use screws to fix targets No. 4 and No. 5 at symmetrical positions on both sides of the center of the oil tank inlet of the receiving machine.

[0102] The third step involves setting up two visible light industrial cameras on the refueling machine. Based on the target flashing cycle, the continuous shooting frequency of the two cameras is set to the same value, and the five targets are continuously photographed.

[0103] Preferably, in this example of the invention, the continuous shooting frequency is selected as 400 frames per second, that is, one shot is taken every 2.5ms. In this case, T=0.0025 is selected, so that two shots are taken in the 5ms target flashing cycle. Therefore, the target flashing spot bright spot can be captured in the two 10ms target flashing cycles.

[0104] The fourth step involves connecting two industrial cameras to a computer to read the image data. Based on the target highlights in the camera images, the positions of five target pixels in the two camera images are located. Then, the pixel positions of the five targets in the two camera images are detected.

[0105] Since the target bright spot is circular, there are already mature and universal methods for locating the target pixel position in the photo based on the target flashing light spot bright spot in the camera photo, and the technical difficulty is not high; it is not the focus of protection and attention of this invention patent, so it will not be elaborated here.

[0106] The pixel position of the first target captured by the camera in the first camera image is denoted as (u). 11i ,v 11i), where i is an integer representing the pixel position at a time interval of 2.5ms, which is (u 11i ,v 11i The value represents the pixel position of the first target in the first camera image at time t = 0.0025i.

[0107] Similarly, the pixel position of the first target captured by the camera in the second camera image is denoted as (u). 12i ,v 12i ), that is (u 12i ,v 12i () represents the pixel position of the first target in the second camera image at time t = 0.0025i. (u) 21i ,v 21i () represents the pixel position of the second target in the first camera image at time t = 0.0025i. (u) 22i ,v 22i () represents the pixel position of the second target in the second camera image at time t = 0.0025i. (u) 31i ,v 31i () represents the pixel position of the third target in the first camera image at time t = 0.0025i. (u) 32i ,v 32i () represents the pixel position of the third target in the second camera image at time t = 0.0025i. (u) 41i ,v 41i () represents the pixel position of the fourth target in the first camera image at time t = 0.0025i. (u) 42i ,v 42i () represents the pixel position of the fourth target in the second camera image at time t = 0.0025i. (u) 51i ,v 51i () represents the pixel position of the fourth target in the first camera image at time t = 0.0025i. (u) 52i ,v 52i The ) represents the pixel position of the fourth target in the second camera image at time t = 0.0025i.

[0108] Step S30: Based on the pixel position of the first target in the two camera images, calculate the lateral pixel relative distance of the first target to the two cameras and the longitudinal pixel relative distance of the first target to the two cameras respectively; calculate the lateral comprehensive coefficient and the longitudinal comprehensive coefficient of the camera based on the known calibrated internal and external parameters of the industrial camera lens; finally, calculate the comprehensive position coefficient of the first target relative to the two cameras based on the lateral pixel relative distance of the first target to the two cameras, the longitudinal pixel relative distance of the two cameras, and the camera parameters.

[0109] Specifically, this can be broken down into the following five steps. Step 1: Based on the known calibrated internal and external parameters of the industrial camera lens and the pixel positions of the target in the first and second camera images, calculate the lateral pixel relative distance of the first target to the first and second cameras, and the longitudinal pixel relative distance of the first target to the first and second cameras, as follows:

[0110] m a11i =(u 11i -u 10 );

[0111] n a11i =(v 11i -v 10 );

[0112] m a12i =(u 12i -u 20 );

[0113] n a12i =(v 12i -v 20 );

[0114] Among them u 10 v 10 This represents the x and y coordinates of the pixel position of the center of the first camera's image coordinate system in the pixel coordinate system; u 20 v 20 This represents the x and y coordinates of the pixel position in the pixel coordinate system, indicating the center of the second camera's image coordinate system. These are internal parameters of the camera, which can be obtained from the camera manufacturer or through self-measurement and calibration; the process of obtaining them will not be elaborated here. a11i n is the lateral pixel relative distance between the first target and the first camera; a11i The vertical pixel distance between the first target and the first camera is m. a12i n is the lateral pixel relative distance between the first target and the second camera; a12i The vertical pixel relative distance between the first target and the second camera.

[0115] The second step involves calculating the camera's lateral and longitudinal composite coefficients based on the known internal and external parameters of the two calibrated industrial camera lenses, as follows:

[0116]

[0117]

[0118] Where f1 is the camera's focal length, c x1 c y1The parameters are the actual physical dimensions of each pixel in the x and y directions of the camera image. These parameters are internal camera parameters and can be obtained from the camera manufacturer or through self-measurement and calibration; the acquisition process will not be elaborated here. m1 is the camera's horizontal composite coefficient, and n1 is the camera's vertical composite coefficient. Here, we select two cameras with the same lens model, so the above parameters are the same for both cameras.

[0119] The third step involves calculating the comprehensive position coefficient vector of the first target relative to the first camera based on the known calibrated internal and external parameters of the industrial camera lens, as well as the horizontal and vertical pixel relative distances between the first target and the first camera.

[0120] g 111i =m a11i r 131 -m1r 111 ;

[0121] g 112i =m a11i r 132 -m1r 112 ;

[0122] g 113i =m a11i r 133 -m1r 113 ;

[0123] g 114i =-m a11i t 13 +m1t 11 ;

[0124] g 115i =n 11ai r 131 -n1r 121 ;

[0125] g 116i =n 11ai r 132 -n1r 122 ;

[0126] g 117i =n 11ai r 133 -n1r 123 ;

[0127] g 118i =-n 11ai t 13 +n1t 12 ;

[0128] Where t 11 ,t 12 ,t13 The three axial translation parameters between the camera coordinate system and the world coordinate system of the first camera; r 111 ,r 112 ,r 113 ,r 121 ,r 122 ,r 123 ,r 131 ,r 132 ,r 133 These are the nine rotation parameters for the first camera. The above 12 parameters are the camera's external parameters, which can be obtained through measurement and calibration. Since there are many related general calibration methods, and they are not the focus of this invention, they will not be elaborated here. 111i g 112i g 113i g 114i g 115i g 116i g 117i g 118i The comprehensive position coefficient of the first camera associated with the first target.

[0129] The fourth step involves calculating the overall position coefficients of the first target relative to the second camera based on the known internal and external parameters of the calibrated industrial camera lens, as well as the lateral and vertical pixel distances between the first target and the second camera.

[0130] g 121i =m a12i r 231 -m1r 211 ;

[0131] g 122i =m a12i r 232 -m1r 212 ;

[0132] g 123i =m a12i r 233 -m1r 213 ;

[0133] g 124i =-m a12i t 223 +m1t 221 ;

[0134] g 125i =n 12ai r 231 -n1r 221 ;

[0135] g 126i =n 12ai r232 -n1r 222 ;

[0136] g 127i =n 12ai r 233 -n1r 223 ;

[0137] g 128i =-n 12ai t 23 +n1t 22 ;

[0138] Where t 21 ,t 22 ,t 23 The three axial translation parameters between the camera coordinate system and the world coordinate system of the second camera; r 211 ,r 212 ,r 213 ,r 221 ,r 222 ,r 223 ,r 231 ,r 232 ,r 233 Nine rotation parameters for the second camera; g 121i g 122i g 123i g 124i g 125i g 126i g 127i g 128i This is the combined position coefficient of the first target relative to the second camera.

[0139] In step S40, the combined position coefficients of the two cameras related to the second, third, fourth, and fifth targets are obtained according to the same steps as in S30. Based on the combined position coefficients of the first, second, and third targets relative to the first and second cameras, and the distance parameters between the first and second targets, two sets of nonlinear equations for the center of the refueling hose are constructed. The fsolve function of Matlab software is used to solve the equations, and two sets of solutions for the three-dimensional coordinates of the positions of the first, second, and third targets are obtained. The average value is then taken to obtain the nonlinear solution for the three-dimensional coordinates of the hose center.

[0140] Specifically, this can be broken down into the following three steps. First, following the same steps as in S30, obtain the combined position coefficients of the two cameras related to the second, third, fourth, and fifth targets; denoted as g... 211i g 212i g 213i g 214i g 215i g 216i g 217i g 218iLet g be the combined position coefficient of the first camera relative to the second target; 221i g 222i g 223i g 224i g 225i g 226i g 227i g 228i Let g be the combined position coefficient of the second camera relative to the second target; 311i g 312i g 313i g 314i g 315i g 316i g 317i g 318i The combined position coefficient of the first camera relative to the third target; denoted as g 321i g 322i g 323i g 324i g 325i g 326i g 327i g 328i Let g be the combined position coefficient of the second camera relative to the third target; 411i g 412i g 413i g 414i g 415i g 416i g 417i g 418i The combined position coefficient of the first camera relative to the fourth target; denoted as g 421i g 422i g 423i g 424i g 425i g 426i g 427i g 428i Let g be the combined position coefficient of the second camera relative to the fourth target; 511i g 512i g 513i g 514i g 515i g 516i g 517i g 518i The combined position coefficient of the first camera relative to the fifth target; denoted as g 521i g 522i g 523i g 524i g 525i g 526i g 527i g 528iThe combined position coefficients of the second camera relative to the fifth target.

[0141] The second step involves constructing the first set of nonlinear equations for the center of the refueling hose based on the combined position coefficients of the first, second, and third targets relative to the first camera, as well as the distance parameters between the first and second targets. These equations are then solved using the fsolve function in Matlab software, as follows:

[0142] g 111i x1+g 112i y1+g 113i z1=g 114i ;

[0143] g 115i x1+g 116i y1+g 117i z1=g 118i ;

[0144] g 211i x2+g 212i y2+g 213i z2=g 214i ;

[0145] g 215i x2+g 216i y2+g 217i z2=g 218i ;

[0146] g 311i x3+g 312i y3+g 313i z3 = g 314i ;

[0147] g 315i x3+g 316i y3+g 317i z3 = g 318i ;

[0148]

[0149]

[0150]

[0151] Where x i ,y i ,z i (i = 1, 2, 3) represents the solution to the first set of nonlinear equations for the three-dimensional coordinates of the i-th target.

[0152] Third, using the same method, solve the second set of nonlinear equations for the i-th target, and remember x. bi ,y bi,z bi (i = 1, 2, 3); then the average value is calculated as follows:

[0153]

[0154] Where x c ,y c ,z c The nonlinear solution is the three-dimensional coordinates of the hose center.

[0155] Step S50: Based on the combined position coefficients of the first target relative to the first and second cameras, construct a system of four linear equations for the first target. Solve the system using the `solve` function in MATLAB to obtain four linear solutions for the first target. Then, calculate the average to obtain the linear mean solution of the three-dimensional coordinates of the first target. Then, construct a system of linear equations using the same method to obtain the linear mean solutions of the three-dimensional coordinates of the second, third, fourth, and fifth targets. Then, average the linear mean solutions of the three-dimensional coordinates of the first, second, and third targets to obtain the linear solution of the three-dimensional coordinates of the hose center. Average the linear mean solutions of the three-dimensional coordinates of the fourth and fifth targets to obtain the linear solution of the three-dimensional coordinates of the fuel tank inlet center.

[0156] This can be broken down into the following three steps. The first step is to construct a system of four linear equations for the first target based on the combined position coefficients of the first target relative to the first and second cameras, as follows:

[0157] g 111i x1+g 112i y1+g 113i z1=g 114i ;

[0158] g 115i x1+g 116i y1+g 117i z1=g 118i ;

[0159] g 121i x1+g 122i y1+g 123i z1=g 124i ;

[0160] g 125i x1+g 126i y1+g 127i z1=g 128i ;

[0161] The second step is to randomly select three of the four linear equation systems mentioned above and solve them using the `solve` function in MATLAB, obtaining four linear solutions for the first target, denoted as x. l1i ,y l1i ,z l1i(i = 1, 2, 3, 4), which represents the i-th linear solution of the first target.

[0162] The third step is to average the four linear solutions for the first target as follows:

[0163]

[0164] Where x l1 ,y l1 ,z l1 This is the linear mean solution of the three-dimensional coordinates of the first target.

[0165] The fourth step involves using the same method to solve for the linear mean solution of the three-dimensional coordinates of the second, third, fourth, and fifth targets, denoted as x. lj ,y lj ,z lj (j = 1, 2, 3, 4, 5).

[0166] The fifth step involves averaging the three-dimensional coordinates of the first, second, and third targets to obtain a linear solution for the three-dimensional coordinates of the hose center. The linear solution for the three-dimensional coordinates of the hose center is as follows:

[0167]

[0168] Where x l ,y l ,z l The solution is the linear solution for the three-dimensional coordinates of the hose center.

[0169] Step 6: Average the linear mean solutions of the three-dimensional coordinates of the fourth and fifth targets again to obtain the linear solution of the three-dimensional coordinates of the fuel tank inlet center as follows:

[0170]

[0171] Where x r ,y r ,z r The solution is the linear solution for the three-dimensional coordinates of the center of the fuel tank inlet.

[0172] Step S60: Based on the linear solution of the three-dimensional coordinates of the fuel tank inlet center and the nonlinear solution of the three-dimensional coordinates of the hose center, the three-dimensional vector of the relative position of the hose and fuel tank is calculated; based on the linear mean solution of the three-dimensional coordinates of the first, second, and third targets, the vertical vector of the hose cross-section is calculated; then, based on the decomposition of the vertical vector of the hose cross-section, the vertical component of the relative position three-dimensional vector and the hose planar component are obtained; then, based on the linear mean solution of the three-dimensional coordinates of the second and third targets, the x-axis component of the hose plane is calculated; then, based on the decomposition of the hose planar component and the hose x-axis component, the x-axis component and y-axis component of the relative position hose plane are obtained; finally, according to the continuous shooting frequency of the camera, the data of the x-axis component, y-axis component, and vertical component of the relative position three-dimensional vector are transmitted to the fuel hose control system as the deviation in three directions relative to the front of the hose, providing the error measurement information required for control, and realizing refueling positioning and control.

[0173] Specifically, it can be broken down into the following seven steps. Step 1: Based on the linear solution of the three-dimensional coordinates of the fuel tank inlet center and the nonlinear solution of the three-dimensional coordinates of the hose center, the three-dimensional vector of the relative position of the hose and fuel tank is calculated as follows:

[0174] e x =x r -x c ,e y =y r -y c ,e z =z r -z c ;

[0175] Among them (e) x ,e y ,e z ) is a three-dimensional vector representing the relative position of the hose and the oil tank.

[0176] The second step involves calculating the vertical vector of the hose cross-section based on the linear mean solution of the three-dimensional coordinates of the first, second, and third targets, as follows:

[0177] a1=x l2 -x l1 a2=y l2 -y l1 a3=z l2 -z l1 ;

[0178] b1 = x l3 -x l1 b2=y l3 -y l1 b3=z l3 -z l1 ;

[0179] c1=a2b3-a3b2, c2=a3b1-a1b3, c3=a1b2-a2b1;

[0180]

[0181] Where (d1,d2,d3) is the vertical vector of the hose cross section; (a1,a2,a3), (b1,b2,b3), and (c1,c2,c3) are intermediate variables in the process of calculating the vertical vector of the hose cross section.

[0182] The third step involves decomposing the vertical vector of the hose cross-section to obtain the vertical components of the relative position three-dimensional vector and the planar components of the hose, as follows:

[0183] w3 = e x d1+e y d2+e z d3;

[0184] f1 = e x -w3d1,f2=e y -w3d2,f3=e z -w3d3;

[0185] Where w3 is the vertical component of the relative position three-dimensional vector; (f1,f2,f3) is the planar component of the hose;

[0186] Fourth step: Based on the linear mean solution of the three-dimensional coordinates of the second and third targets, the x-axis components of the hose plane are calculated as follows:

[0187]

[0188] Where (g1, g2, g3) are the x-axis components of the flexible tube plane, d w This represents the distance between the second and third targets.

[0189] Step 5: Based on the planar components and x-axis components of the hose plane, the relative position hose plane x-axis components and y-axis components are obtained as follows:

[0190] w1 = f1g1 + f2g2 + f3g3;

[0191]

[0192] Where w1 is the x-axis component of the relative position hose plane, and w2 is the y-axis component of the relative position hose plane.

[0193] The sixth step involves transmitting the data of the relative position hose plane x-axis component w1, relative position hose plane y-axis component w2, and relative position three-dimensional vector vertical component w3 to the oil hose control system at the camera's continuous shooting frequency. This data serves as the deviation in three directions relative to the front of the hose, providing the error measurement information required for control and enabling refueling positioning and control.

[0194] Specifically, in this case, the output frequency was set to 400, meaning that an error measurement message was output every 0.0025 seconds, thereby realizing the real-time calculation and output of continuous positioning measurement data.

[0195] This invention primarily focuses on positioning measurement and error information calculation during the refueling process; the specific refueling control algorithm is not the focus or protection of this invention and will not be elaborated further. This completes the continuous calculation of spatial optical measurement and error positioning between the hose and fuel tank during the in-flight refueling process of an aircraft using two cameras and five targets.

[0196] As can be seen from the above examples, the method provided by this invention is entirely feasible. Moreover, by using two cameras and a total of five targets, it can meet the refueling positioning requirements of most industrial and military scenarios. It is also a simplified minimum configuration implementation, with fewer targets and cameras, which has the advantages of low economic cost, fast calculation speed, and data output frequency according to refueling needs, thus having high engineering promotion value.

Claims

1. A method for aerial refueling positioning using a target and dual cameras, comprising the following steps: Step S10: A control circuit is composed of an ultra-low power microcontroller. A target module system is composed of a button battery, a liquid crystal light valve, a reflector, and a control circuit. The microcontroller program sets the target flashing period and controls the target to form alternating bright and dark light spots for output. Step S20: According to the design method of the target module system described above, five targets are manufactured. Targets one, two, and three are fixed with screws to the front end of the retractable fuel hose of the fuel dispenser at an equilateral triangle position centered on the fuel inlet, and the distance between targets one and two is measured. Targets four and five are fixed with screws at symmetrical positions on both sides of the center of the fuel tank inlet of the receiving machine. Then, two visible light industrial cameras are mounted on the fuel dispenser. Based on the target flashing period, the continuous shooting frequency of the two cameras is set to the same value, and continuous images are taken of the five targets. The two cameras are connected to a computer to read the image data. Based on the target highlights in the camera images, the pixel positions of the five targets in the two camera images are located. Then, the pixel positions of the five targets in the two camera images are detected. Step S30: Based on the pixel position of the first target in the two camera images, calculate the lateral pixel relative distance of the first target to the two cameras and the longitudinal pixel relative distance of the first target to the two cameras respectively; calculate the lateral comprehensive coefficient and the longitudinal comprehensive coefficient of the camera based on the known calibrated internal and external parameters of the industrial camera lens; finally, calculate the comprehensive position coefficient of the first target relative to the two cameras based on the lateral pixel relative distance of the first target to the two cameras, the longitudinal pixel relative distance of the two cameras, and the camera parameters. Step S40: Following the same steps as in S30, the combined position coefficients of the two cameras related to the second, third, fourth, and fifth targets are obtained. Based on the combined position coefficients of the first, second, and third targets relative to the first and second cameras, and the distance parameters between the first and second targets, two sets of nonlinear equations for the center of the refueling hose are constructed. The fsolve function of Matlab software is used to solve these equations, resulting in two sets of solutions for the three-dimensional coordinates of the positions of the first, second, and third targets. The average value is then taken to obtain the nonlinear solution for the three-dimensional coordinates of the hose center. Step S50: Based on the combined position coefficients of the first target relative to the first and second cameras, construct a system of four linear equations for the first target. Solve these equations using the `solve` function in MATLAB to obtain four linear solutions for the first target. Then, calculate the average of these solutions to obtain the linear mean solution for the three-dimensional coordinates of the first target. Use the same method to construct a system of linear equations and solve for the linear mean solutions for the three-dimensional coordinates of the second, third, fourth, and fifth targets. Then, average the linear mean solutions for the three-dimensional coordinates of the first, second, and third targets again to obtain the linear solution for the three-dimensional coordinates of the hose center. Finally, average the linear mean solutions for the three-dimensional coordinates of the fourth and fifth targets again to obtain the linear solution for the three-dimensional coordinates of the fuel tank inlet center. Step S60: Based on the linear solution of the three-dimensional coordinates of the fuel tank inlet center and the nonlinear solution of the three-dimensional coordinates of the hose center, the three-dimensional vector of the relative position of the hose and fuel tank is calculated; based on the linear mean solution of the three-dimensional coordinates of the first, second, and third targets, the vertical vector of the hose cross section is calculated; then, based on the decomposition of the vertical vector of the hose cross section, the vertical component of the relative position three-dimensional vector and the hose planar component are obtained; then, based on the linear mean solution of the three-dimensional coordinates of the second and third targets, the x-axis component of the hose plane is calculated; then, based on the decomposition of the hose planar component and the hose x-axis component, the x-axis component of the relative position hose plane and the y-axis component of the relative position hose plane are obtained; finally, according to the continuous shooting frequency of the camera, the data of the x-axis component of the relative position hose plane, the y-axis component of the relative position hose plane, and the vertical component of the relative position three-dimensional vector are transmitted to the fuel hose control system as the deviation in three directions relative to the front of the hose, providing the error measurement information required for control, and realizing refueling positioning and control; The distance between targets one and two is measured and denoted as d. 12 Set the target blinking period to T, and denote the pixel position of the j-th target in the k-th camera image as (u). jki ,v jki ), that is (u jki ,v jki Let represent the pixel position of the j-th target in the k-th camera image at time t = Ti, where j = 1, 2, 3, 4, 5, and k = 1, 2. Further, based on the known calibrated internal and external parameters of the industrial camera lens and the pixel positions of the target in the first and second camera images, the lateral pixel relative distances of the first target to the first and second cameras, and the longitudinal pixel relative distances of the first target to the first and second cameras are calculated, including: m a11i =(u 11i -u 10 ); n a11i =(v 11i -v 10 ); m a12i =(u 12i -u 20 ); n a12i =(v 12i -v 20 ); Where u 10 v 10 This represents the x and y coordinates of the pixel position of the center of the first camera's image coordinate system in the pixel coordinate system; u 20 v 20 This represents the x-coordinate and y-coordinate of the pixel position of the center of the second camera's image coordinate system in the pixel coordinate system; m a11i n is the lateral pixel relative distance between the first target and the first camera; a11i The vertical pixel relative distance between the first target and the first camera; m a12i n is the lateral pixel relative distance between the first target and the second camera; a12i Let be the vertical pixel relative distance between the first target and the second camera; further, based on the known calibrated internal and external parameters of the two industrial camera lenses, the camera's lateral and vertical composite coefficients are calculated as follows: Where f1 is the camera's focal length, c x1 c y1 Let m1 be the actual physical size parameters of each pixel in the x and y directions of the camera image, and n1 be the camera's lateral composite coefficient and vertical composite coefficient. Since the two cameras use the same lens model, the above parameters are the same for both cameras. Furthermore, based on the known calibrated internal and external parameters of the industrial camera lens, and the lateral and vertical pixel relative distances between the first target and the first camera, the composite position coefficient vector of the first target relative to the first camera is calculated, which includes: g 111i =m a11i r 131 -m1r 111 ; g 112i =m a11i r 132 -m1r 112 ; g 113i =m a11i r 133 -m1r 113 ; g 114i =-m a11i t 13 +m1t 11 ; g 115i =n 11ai r 131 -n1r 121 ; g 116i =n 11ai r 132 -n1r 122 ; g 117i =n 11ai r 133 -n1r 123 ; g 118i =-n 11ai t 13 +n1t 12 ; Where t 11 ,t 12 ,t 13 The three axial translation parameters between the camera coordinate system and the world coordinate system of the first camera; r 111 ,r 112 ,r 113 ,r 121 ,r 122 ,r 123 ,r 131 ,r 132 ,r 133 Nine rotation parameters for the first camera; g 111i g 112i g 113i g 114i g 115i g 116i g 117i g 118i The comprehensive position coefficient of the first target relative to the first camera is given. Similarly, based on the known calibrated internal and external parameters of the industrial camera lens, and the lateral and vertical pixel relative distances of the first target relative to the second camera, the comprehensive position coefficient of the first target relative to the second camera is calculated, including: g 121i =m a12i r 231 -m1r 211 ; g 122i =m a12i r 232 -m1r 212 ; g 123i =m a12i r 233 -m1r 213 ; g 124i =-m a12i t 223 +m1t 221 ; g 125i =n 12ai r 231 -n1r 221 ; g 126i =n 12ai r 232 -n1r 222 ; g 127i =n 12ai r 233 -n1r 223 ; g 128i =-n 12ai t 23 +n1t 22 ; Where t 21 ,t 22 ,t 23 The three axial translation parameters between the camera coordinate system and the world coordinate system of the second camera; r 211 ,r 212 ,r 213 ,r 221 ,r 222 ,r 223 ,r 231 ,r 232 ,r 233 Nine rotation parameters for the second camera; g 121i g 122i g 123i g 124i g 125i g 126i g 127i g 128i Let g be the combined position coefficient of the first target relative to the second camera; following the same steps in S30, the combined position coefficients of the second, third, fourth, and fifth targets relative to the two cameras are obtained; let g be the denoted g. wp1i g wp2i g wp3i g wp4i g wp5i g wp6i g wp7i g wp8i Let w be the comprehensive position coefficient of the p-th camera related to the w-th target; where w = 2, 3, 4, 5 and p = 1, 2.

2. The method for aerial refueling positioning using a target and dual cameras according to claim 1, characterized in that, Based on the combined position coefficients of the first, second, and third targets relative to the first camera, and the distance parameters between the first and second targets, a first set of nonlinear equations for the center of the refueling hose is constructed. The fsolve function in Matlab is used to solve these equations, obtaining two sets of solutions for the three-dimensional coordinates of the first, second, and third targets. The average value is then taken to obtain the nonlinear solution for the three-dimensional coordinates of the hose center, including: g 111i x1+g 112i y1+g 113i z1=g 114i ; g 115i x1+g 116i y1+g 117i z1=g 118i ; g 211i x2+g 212i y2+g 213i z2=g 214i ; g 215i x2+g 216i y2+g 217i z2=g 218i ; g 311i x3+g 312i y3+g 313i z3=g 314i ; g 315i x3+g 316i y3+g 317i z3=g 318i ; Where x i ,y i ,z i (i = 1, 2, 3) represents the solution to the first set of nonlinear equations for the three-dimensional coordinates of the i-th target. Where x bi ,y bi ,z bi (i = 1, 2, 3) uses the same method to solve the second set of nonlinear equations for the i-th target, x c ,y c ,z c The nonlinear solution for the three-dimensional coordinates of the hose center, d 12 This is the distance between targets one and two, which is the side length of the equilateral triangle.

3. The method for aerial refueling positioning using a target and dual cameras according to claim 1, characterized in that, Based on the combined position coefficients of the first target relative to the first and second cameras, a system of four linear equations for the first target is constructed. These equations are solved using the `solve` function in MATLAB to obtain four linear solutions for the first target. The mean of these four linear solutions is then calculated to obtain the linear mean solution for the three-dimensional coordinates of the first target. The same method is then used to construct a system of linear equations to obtain the linear mean solutions for the three-dimensional coordinates of the second, third, fourth, and fifth targets. The linear mean solutions for the three-dimensional coordinates of the first, second, and third targets are then averaged again to obtain the linear solution for the three-dimensional coordinates of the hose center. Finally, the linear mean solutions for the three-dimensional coordinates of the fourth and fifth targets are averaged again to obtain the linear solution for the three-dimensional coordinates of the fuel tank inlet center. g 111i x1+g 112i y1+g 113i z1=g 114i ; g 115i x1+g 116i y1+g 117i z1=g 118i ; g 121i x1+g 122i y1+g 123i z1=g 124i ; g 125i x1+g 126i y1+g 127i z1=g 128i ; Where x l1i ,y l1i ,z l1i (i = 1, 2, 3, 4) represents any three of the four linear equations mentioned above. The solve function in MATLAB is used to solve each of them to obtain the four linear solutions of the first target, which represent the i-th linear solution of the first target. Where x l1 ,y l1 ,z l1 The linear mean solution for the three-dimensional coordinates of the first target; x lj ,y lj ,z lj (j=1,2,3,4,5) represents the linear mean solution for the three-dimensional coordinates of the second, third, fourth, and fifth targets obtained using the same method; where x l ,y l ,z l The linear solution for the three-dimensional coordinates of the hose center; x r ,y r ,z r The solution is the linear solution for the three-dimensional coordinates of the center of the fuel tank inlet.

4. The method for aerial refueling positioning using a target and dual cameras according to claim 1, characterized in that, Based on the linear solution of the three-dimensional coordinates of the tank inlet center and the nonlinear solution of the three-dimensional coordinates of the hose center, the three-dimensional vector of the relative position of the hose and tank is solved; based on the linear mean solution of the three-dimensional coordinates of the first, second, and third targets, the vertical vector of the hose cross-section is solved; then, based on the decomposition of the vertical vector of the hose cross-section, the vertical component and the hose planar component of the relative position three-dimensional vector are obtained; then, based on the linear mean solution of the three-dimensional coordinates of the second and third targets, the x-axis component of the hose plane is obtained; then, based on the decomposition of the hose planar component and the hose plane x-axis component, the relative position hose plane x-axis component and the relative position hose plane y-axis component are obtained, including: e x =x r -x c ,e y =y r -y c ,e z =z r -z c ; a1=x l2 -x l1 a2=y l2 -y l1 a3=z l2 -z l1 ; b1=x l3 -x l1 ,b2=y l3 -y l1 ,b3=z l3 -z l1 ; c1=a2b3-a3b2, c2=a3b1-a1b3, c3=a1b2-a2b1; w3=e x d1+e y d2+e z d3: f1=e x -w3d1,f2=e y -w3d2,f3=e z -w3d3; w1 = f1g1 + f2g2 + f3g3; Among them (e) x ,e y ,e z (d1, d2, d3) represents the three-dimensional vector of the relative position of the hose and the tank; (d1, d2, d3) represents the vertical vector of the hose cross-section; (a1, a2, a3), (b1, b2, b3), and (c1, c2, c3) are intermediate variables used in calculating the vertical vector of the hose cross-section; w3 represents the vertical component of the relative position three-dimensional vector; (f1, f2, f3) represents the planar component of the hose; (g1, g2, g3) represents the x-axis component of the hose plane, d w w1 represents the distance between the second and third targets; w2 represents the x-axis component of the relative position hose plane and w1 represents the y-axis component of the relative position hose plane.

Images