Method, device and UAV platform for positioning marine targets based on SAR images

Abstract

The present invention discloses a method, device and unmanned aerial vehicle platform for positioning maritime targets based on SAR images. The method includes: performing target detection on the SAR image to estimate the target distance and the target Doppler position, and correcting the target squint angle based on the target Doppler position; determining an estimated value of the target distance position in the imaging coordinate system according to the corrected squint angle of the target and the range-Doppler equation; according to the target distance, taking the rapid estimated value of the distance position as the center, and solving the possible solution set of the target position that satisfies the target equidistant equation; solving the equal squint angle of the point set on the equidistant line in the possible solution set of the target position, and based on the corrected squint angle, solving the optimal coordinate point whose squint angle in the point set satisfies the convergence error; converting the optimal coordinate point into the target positioning coordinate in the launch coordinate system according to the rotation angle between the imaging coordinate system and the launch coordinate system. The present invention does not involve complex matrix inversion operations and iterative operations, and has a small amount of computation.

Description

Technical Field

[0001] The present invention belongs to the technical field of airborne radar target detection and tracking, and in particular relates to a method, device and unmanned aerial vehicle (UAV) platform for positioning marine targets based on SAR slant-range images. Background Art

[0002] When UAV-borne SAR, especially small UAV-borne SAR, was first developed, due to the complex imaging algorithm and the slow speed of the processor device, the UAV-borne SAR image could not be real-time imaged, so it was not applied to the UAV tracking and guidance process and was only used for auxiliary target recognition. With the development of SAR imaging algorithms and the continuous update and iteration of processor devices, SAR images can already be real-time imaged, and the image frame rate reaches 20 Hz. Target image tracking based on SAR has been widely applied. Since target tracking based on SAR images must solve the problem of target positioning, how to quickly locate targets based on SAR images has become a research hotspot.

[0003] For the target positioning technology based on SAR images of airborne platforms, there are mainly two mainstream positioning methods: One is to project the SAR image onto the ground first to obtain the ground range image of the target, and then perform target detection on the ground range image, and perform range-Doppler positioning based on the detected target position. This method involves a large amount of interpolation operations because it is necessary to obtain a fine ground range image of the target area, and thus has poor real-time performance due to the large amount of computation. The other is to first match the key points and GCP points in the SAR image, then construct a range-Doppler equation based on the GCP points, and through spectral correction and iterative algorithms, the precise motion equation of the entire missile-borne platform can be solved, the trajectory of the moving platform with errors can be corrected, and then based on the precise motion equation, the target positioning of any ground image can be solved. This method has high positioning accuracy, but involves matrix inversion and iterative operations, and the algorithm complexity is high, which is not suitable for UAV platforms with high requirements for time-consuming applications, especially small UAV platforms. In addition, when positioning marine targets, this method is also inapplicable when there are no GCP control points on the sea surface.

[0004] Therefore, the existing SAR target positioning algorithms are complex and rely on positioning devices such as high-precision GPS, making it difficult to meet the requirements of high-dynamic applications of UAV platforms, especially small UAV platforms with limited computing resources. Summary of the Invention

[0005] Based on the imaging geometry of terminal guidance SAR of the airborne platform, the present invention transforms the range-Doppler equation with multi-variable coupling in the launch coordinate system (guidance coordinate system) into the imaging coordinate system by transforming the coordinate system, realizing variable separation and dimension reduction.

[0006] In the first aspect of the present invention, a method for positioning marine targets is disclosed, and the method includes:

[0007] Obtain the SAR image of the target, perform target detection on the SAR image to estimate the target distance and the target Doppler position, and correct the target squint angle based on the target Doppler position;

[0008] According to the corrected squint angle of the target and the range-Doppler equation, determine the estimated value of the range position of the target in the imaging coordinate system;

[0009] According to the target distance, with the quick estimated value of the range position as the center and a predetermined iteration step size, solve the possible solution set of the target position that satisfies the target equidistant equation;

[0010] Solve the equal squint angle of the point set on the equidistant line in the possible solution set of the target position, and based on the corrected squint angle, solve the optimal coordinate point in the point set whose squint angle satisfies the convergence error;

[0011] According to the rotation angle between the imaging coordinate system and the emission coordinate system, convert the optimal coordinate point into the target positioning coordinate in the emission coordinate system.

[0012] In particular, the above method is used for real-time positioning of marine targets by an unmanned aerial vehicle (UAV) platform.

[0013] In a second aspect of the present invention, a marine target positioning device is disclosed, and the device includes:

[0014] A squint angle correction module, configured to correct the target squint angle by using the target distance and the target Doppler position estimated through target detection on the target SAR image;

[0015] A target range position estimation module, configured to determine the estimated value of the range position of the target in the imaging coordinate system according to the corrected squint angle of the target and the range-Doppler equation;

[0016] A first solving module, configured to solve the possible solution set of the target position that satisfies the target equidistant equation according to the target distance, with the estimated value of the range position as the center and a predetermined iteration step size;

[0017] A second solving module, configured to solve the equal squint angle of the point set on the equidistant line in the possible solution set of the target position, and based on the corrected squint angle, solve the optimal coordinate point in the point set whose squint angle satisfies the convergence error;

[0018] A coordinate conversion module, configured to convert the optimal coordinate point into the target positioning coordinate in the emission coordinate system according to the rotation angle between the imaging coordinate system and the emission coordinate system.

[0019] In a third aspect of the present invention, an unmanned aerial vehicle (UAV) platform is disclosed, including an SAR radar system and the marine target positioning device according to the above solution.

[0020] In a third aspect of the present invention, an airborne drone platform is disclosed, including: a memory for storing computer programs; at least one processor, communicatively coupled to the memory for executing the computer programs to implement the maritime target positioning method according to the above solution; in particular, the processor can be a DSP.

[0021] Compared with the prior art, when the present invention is applied to an airborne drone platform, especially a small airborne drone platform, since there is no complex matrix inversion operation and iterative operation involved in the target positioning process, the amount of computation is small, the solution is accurately controllable, and rapid and accurate positioning of maritime ship targets can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 It is a schematic diagram of SAR imaging and positioning geometry in the launch coordinate system (antenna beam coordinate system);

[0023] Figure 2 is Figure 1 a schematic diagram of the imaging coordinate system after rotation of the shown coordinate system;

[0024] Figure 3 It is a schematic diagram of the working process of the maritime target positioning method according to an embodiment of the present invention;

[0025] Figure 4 It is a schematic diagram of the composition of the maritime target positioning device according to an embodiment of the present invention;

[0026] Figure 5 It is a schematic diagram of the composition of the airborne drone platform according to an embodiment of the present invention;

[0027] Figure 6 It is a schematic diagram of the composition of the airborne drone platform according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] The following will describe in detail the specific embodiments of the present invention with reference to the accompanying drawings.

[0029] Figure 1 It is a schematic diagram of SAR imaging and positioning geometry in the launch coordinate system (antenna beam coordinate system). As shown in the figure, the position vector of the airborne radar is: P′=(x0′, y0′, z0′), and the position vector of the target is: T′=(x′, y′, z′). When the target is in the horizontal plane, such as on the sea surface, y = 0. In the terminal guidance section of the drone, the slant range of the target is less than 30 km, and the resulting sea level error is less than 10 m. Therefore, the error is within an acceptable range, and by continuously correcting the origin of the coordinate system and translating the coordinate system, the positioning error caused by the plane assumption can be controlled. The velocity vector of the airborne radar is: v′=(v x ′, v y′, v z ′).

[0030] Perform coordinate transformation on the emission coordinate system with the deviation angle α of the radar course speed to transform it into the radar imaging coordinate system with the target plane course speed as the x-axis. Among them, the deviation angle α of the radar course speed is:

[0031]

[0032] Among them, mid(v z ) and mid(v x ) are the instantaneous speeds of the center of the radar synthetic aperture.

[0033] The coordinate system after rotation is as shown in Figure 2 . At this time, that is, in the radar imaging coordinate system, the airborne radar velocity vector is: v = (v x , v y , v z ), where v z = 0. The airborne radar position vector is: P = (x0, y0, z0), and the target position vector is: T = (x, y, z).

[0034] Figure 3 It is a schematic diagram of the working process of the marine target positioning method according to an embodiment of the present invention. As shown in the figure, the method includes the following steps:

[0035] Step 301, obtain the SAR image of the target, perform target detection on the SAR image to estimate the target distance and the target Doppler position, and correct the target squint angle based on the target Doppler position;

[0036] Obtain the SAR slant range image of the target, perform target detection on the SAR slant range image, and obtain the distance estimate and Doppler estimate of the target.

[0037] Among them, the distance estimate R of the target is:

[0038]

[0039] Among them, nrn is the index of the target distance position detected in the SAR slant range image, F s is the echo complex baseband sampling rate, c is the speed of light, and R0 is the sampling distance of the closest point of the SAR image, which are known conditions.

[0040] The Doppler estimate of the target, that is, the Doppler frequency of the target is:

[0041]

[0042] where, nan is the target Doppler position index detected from the SAR slant range image; Na is the number of azimuth pulse accumulations during imaging, which is a constant; PRF is the system pulse repetition frequency, and f dc is the estimated target Doppler center frequency during imaging.

[0043] Then the corrected squint angle of the target is:

[0044]

[0045] where, λ is the system wavelength.

[0046] Step 302: Determine a quick estimated value of the distance position (x position) of the target in the imaging coordinate system according to the corrected squint angle of the target and the range-Doppler equation;

[0047] Based on the slant range vector of the radar and the target and the radar velocity vector, solve the theoretical squint angle. Among them, the theoretical slant range vector of the target is:

[0048] R = (x - x0, 0 - y0, z - z0)

[0049] Therefore, there is the following formula:

[0050]

[0051] Simplify the above formula:

[0052] R T ·v = (x - x0)v x + (0 - y0)v y + (z - z0)v z = |R|·|v|sinθ

[0053] R T ·v = xv x - x0v x - y0v y + zv z - z0v z = |R|·|v|sinθ

[0054] R T ·v = xv x + zv z - x0v x - y0v y - z0v z = |R|·|v|sinθ

[0055] R T ·v = xv x + zv z -(x0v x + y0vy +z0v z ) = |R|·|v|sinθ

[0056] Definition: P T v = x0v x +y0v y +z0v z , then the above formula is transformed into:

[0057] R T ·v = xv x +zv z -P T v = |R|·|v|sinθ

[0058] Transform the above formula to get:

[0059] xv x +zv z = |R|·|v|sinθ + P T v

[0060] In the imaging coordinate system, the radar velocity v z is approximately 0, so the estimated target distance position can be obtained:

[0061]

[0062] Step 303: According to the target distance, with the rapid estimated value of the distance position as the center and a predetermined iteration step size, solve the possible solution set of the target position that satisfies the target equidistant equation;

[0063] Based on the rapid estimated value of the target distance position, with the estimated value as the center, select the iteration step size, and based on the detected target slant range, solve the target solution set, and the target solution set satisfies the equidistant equation.

[0064] First, construct the equidistant equation of the target:

[0065] (x - x0) 2 +(y - y0) 2 +(z - z0) 2 = R 2

[0066] For marine targets, y = 0, so the above equation is simplified to:

[0067] (x - x0) 2 +y0 2 +(z - z0) 2 = R 2

[0068] Set the range of the target distance:

[0069]

[0070] Among them, n is an integer, which controls the number of points for range solution, and Δx is the step size for target discrete solution, used to control the solution accuracy. Substitute into the equation to solve for the possible positions of the target:

[0071]

[0072] The possible solution set of the target position is obtained as:

[0073] Step 304: Solve the equal-slope viewing angle of the point set on the equal-distance line in the possible solution set of the target position, and based on the corrected slope viewing angle, solve the optimal coordinate points in the point set where the slope viewing angle satisfies the convergence error;

[0074] Based on the possible solution set of the target, solve the equal-Doppler line (equal-slope viewing angle) of the point set on the equal-distance line, and based on the estimated slope viewing angle of the target, given the convergence error of the solution calculation, solve the optimal coordinate points in the point set that satisfy the error.

[0075] Among them, based on the possible solution set of the target, solve the best position of the target, and the optimization function is:

[0076]

[0077] Step 305: According to the rotation angle between the imaging coordinate system and the emission coordinate system, convert the optimal coordinate points into the target positioning coordinates in the emission coordinate system.

[0078] For the best target position obtained by solving in the imaging coordinate system, based on the rotation angle α between the imaging coordinate system and the emission coordinate system (i.e., the radar course speed deviation angle α mentioned above), perform coordinate inverse rotation to obtain the target positioning coordinates of the target in the emission coordinate system:

[0079]

[0080] Figure 4 It is a schematic diagram of the composition of the marine target positioning device 400 according to an embodiment of the present invention. As shown in the figure, the device includes:

[0081] A slope viewing angle correction module 401, configured to correct the target slope viewing angle by using the target distance and the target Doppler position estimated by target detection of the target SAR image;

[0082] A target distance position estimation module 402, configured to determine the estimated value of the target distance position in the imaging coordinate system according to the corrected slope viewing angle of the target and the range-Doppler equation;

[0083] The first solving module 403 is configured to solve a possible solution set of the target position that satisfies the target equidistant equation according to the target distance, with the distance position estimated value as the center and a predetermined iteration step size.

[0084] The second solving module 404 is configured to solve the isoclinal view angle of the point set on the equidistant line in the possible solution set of the target position, and solve the optimal coordinate point in the point set whose skew view angle satisfies the convergence error based on the corrected skew view angle.

[0085] The coordinate conversion module 405 is configured to convert the optimal coordinate point into the target positioning coordinate in the emission coordinate system according to the rotation angle between the imaging coordinate system and the emission coordinate system.

[0086] Figure 5 It is a schematic diagram of the composition of the airborne platform 500 according to an embodiment of the present invention. As shown in the figure, the airborne platform includes an SAR radar system and the above-mentioned marine target positioning device.

[0087] Figure 6 It is a schematic diagram of the composition of the airborne platform 600 according to another embodiment of the present invention. As shown in the figure, the airborne platform includes: a memory for storing a computer program; at least one processor, which is data communicatively coupled to the memory and is configured to execute the computer program to implement the marine target positioning method described in the above solution. Among them, the processor can be a DSP.

[0088] The present invention aims at the SAR imaging target positioning requirements of the airborne platform, especially the small airborne platform, and especially the need for accurate SAR image positioning of large marine ships. Based on the SAR imaging geometric configuration of the airborne platform, by converting the coordinate system, the range-Doppler equation with multi-variable coupling in the emission coordinate system is converted into the imaging coordinate system, realizing variable separation and dimensionality reduction.

[0089] At the same time, under sea level conditions, by constructing a rapid estimation of the target distance coordinate and based on the equidistant line and equi-Doppler line equations, the solution set of the target coordinate is realized.

[0090] Finally, based on the estimated target skew view angle, the range and azimuth coordinates of the target are solved, and through coordinate conversion, it is re-converted into the emission coordinate system to realize the determination of the target positioning coordinate.

[0091] The above positioning process does not involve complex matrix inversion operations and iterative operations, has a small amount of computation, the solution is accurately controllable, and can realize the rapid and accurate positioning of the target.

[0092] The above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention and not to limit them. Although the embodiments of the present invention have been described in detail with reference to the above preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the embodiments of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for positioning a maritime target, characterized in that, The method includes: Obtaining a target SAR image, performing target detection on the target SAR image to estimate the target distance and the target Doppler position, and correcting the target squint angle based on the target Doppler position; Determining an estimated value of the target's distance position in the imaging coordinate system according to the corrected squint angle of the target and the range-Doppler equation; According to the target distance, with the estimated value of the distance position as the center and a predetermined iteration step size, solving a possible solution set of the target position that satisfies the target equidistant equation; Solving the equal squint angle of the point set on the equidistant line in the possible solution set of the target position, and based on the corrected squint angle, solving the optimal coordinate point in the point set whose squint angle satisfies the convergence error; Converting the optimal coordinate point into the target positioning coordinate in the emission coordinate system according to the rotation angle between the imaging coordinate system and the emission coordinate system.

2. The maritime target positioning method according to claim 1, characterized in that Performing target detection on the target SAR image to estimate the target Doppler position as: Among them, f d is the target Doppler frequency, nan is the target Doppler position index detected in the SAR slant range image, Na is the number of azimuth pulse accumulations during imaging, PRF is the radar system pulse repetition frequency, and f dc is the estimated target Doppler center frequency during imaging.

3. The maritime target positioning method according to claim 1, characterized in that, Performing target detection on the target SAR image to estimate the target distance as: Wherein, R is the target distance, nrn is the index of the target distance position detected in the SAR image, F s is the echo complex baseband sampling rate, c is the speed of light, and R0 is the sampling distance of the closest point in the SAR image.

4. The method for positioning a maritime target according to claim 2, wherein, The corrected target squint angle θ is: Where λ is the radar system wavelength and v is the motion velocity vector of the radar system.

5. The maritime target positioning method according to claim 1, characterized in that, In the imaging coordinate system, the estimated value of the target's distance position is: Where, R is the target distance, v is the motion velocity vector of the radar system, θ is the corrected target slant angle, P is the spatial coordinate of the radar system, and v x is the velocity of the radar system along the x-axis direction.

6. The maritime target positioning method according to any one of claims 1-5, characterized in that, This method is used for real-time positioning of maritime targets by an unmanned aerial vehicle (UAV) platform.

7. An offshore target positioning device, characterized in that, The device includes: A squint angle correction module, configured to correct the target squint angle by using the target distance and the target Doppler position estimated through target detection on the target SAR image; A target distance position estimation module, configured to determine an estimated value of the target's distance position in the imaging coordinate system according to the corrected squint angle of the target and the range-Doppler equation; A first solving module, configured to solve a possible solution set of the target position that satisfies the target equidistant equation according to the target distance, with the estimated value of the distance position as the center and a predetermined iteration step size; A second solving module, configured to solve the equal squint angle of the point set on the equidistant line in the possible solution set of the target position, and based on the corrected squint angle, solve the optimal coordinate point in the point set whose squint angle satisfies the convergence error; A coordinate conversion module, configured to convert the optimal coordinate point into the target positioning coordinate in the emission coordinate system according to the rotation angle between the imaging coordinate system and the emission coordinate system.

8. An unmanned aerial vehicle (UAV) platform, including an SAR radar system and the maritime target positioning device according to claim 7.

9. An unmanned aerial vehicle (UAV) platform, comprising: A memory, configured to store a computer program; At least one processor, data communicatively coupled to the memory, configured to execute the computer program to implement the maritime target positioning method according to any one of claims 1-6; wherein, the processor is a DSP.

Images