Vehicle-mounted SAR motion compensation method and device under large measurement error

Abstract

The application discloses a vehicle-mounted SAR motion compensation method and device under large measurement error, relates to the field of vehicle-mounted synthetic aperture radar imaging technology, and solves the problem that low-precision inertial navigation equipment is difficult to meet the demand of vehicle-mounted SAR motion compensation and high-precision inertial navigation equipment is expensive in the prior art. The method comprises the following steps: accurately estimating motion error by using low-precision inertial navigation data and radar echo data; on the basis of obtaining the accurate motion error, dividing original data into sub-blocks, and performing motion compensation and range migration correction in the sub-blocks, so that the focusing effect of scene edge points is improved, and the operation efficiency is improved compared with the back projection algorithm.

Description

Technical Field

[0001] This invention relates to the field of vehicle-mounted synthetic aperture radar imaging technology, and in particular to a method and device for motion compensation of vehicle-mounted SAR under large measurement errors. Background Technology

[0002] Synthetic Aperture Radar (SAR) imaging technology is not constrained by time, weather conditions, or distance, and can effectively acquire ground information and obtain high-resolution two-dimensional images, making it widely used in military and civilian fields. Meanwhile, millimeter-wave radar, as one of the core sensors in autonomous driving systems, suffers from angular resolution limitations due to the actual antenna aperture, making it difficult to meet the spatial resolution requirements of driving environments. Against this backdrop, SAR imaging technology combined with millimeter-wave radar can provide high-resolution two-dimensional imaging of driving scenarios. Its all-weather, all-day operation further enhances the vehicle's ability to stably perceive its surroundings, outputting relatively clear image information of the surrounding environment even in harsh conditions. Therefore, vehicle-mounted SAR imaging technology has become one of the most promising technologies for achieving high-resolution imaging in driving scenarios and has received widespread attention in recent years. Given the characteristics of real-world vehicle trajectories, the actual trajectory of a vehicle is not an ideal straight line and contains significant motion errors. Furthermore, due to cost constraints, the vehicle's built-in navigation equipment cannot meet the accuracy requirements of vehicle-mounted SAR motion compensation. To enhance the practicality of vehicle-mounted SAR, it is necessary to study motion compensation methods for vehicle-mounted SAR under conditions of large measurement errors.

[0003] Existing Solution: Nanjing University of Aeronautics and Astronautics disclosed a vehicle-mounted SAR imaging method under non-ideal trajectories in a conference paper titled "Automotive Millimeter WaveSAR Imaging in Curved Trajectory" (doi:10.1109 / ICSPS58776.2022.00090). This method acquires vehicle motion information through a high-precision integrated navigation device and combines a back projection algorithm to complete spatially varying error compensation and focusing imaging. Although this method significantly reduces the complexity of the algorithm by selecting a reasonable sub-aperture imaging grid, it still has two problems. First, it relies on an additional high-precision integrated navigation device, which increases the cost of vehicle-mounted SAR motion compensation and limits its practical application. Second, although this method improves computational efficiency compared to the traditional back projection algorithm, the computational load is still very large due to the limitations of the algorithm itself, which cannot meet the real-time requirements of vehicle-mounted SAR.

[0004] Existing vehicle-mounted SAR motion compensation technologies suffer from several drawbacks: they require additional high-precision integrated navigation equipment, resulting in high implementation costs. Furthermore, vehicle-mounted SAR motion errors exhibit significant spatial variation in the scene, leading to a large computational burden when using back projection algorithms to compensate for spatially varied errors. Summary of the Invention

[0005] This invention provides a method and device for motion compensation of vehicle-mounted SAR under large measurement errors. It solves the problems of high implementation cost and large spatial variation of motion error scene in the prior art, and the large amount of calculation required to compensate for spatial variation error using the back projection algorithm. It improves the focusing effect of scene edge points and also improves the computational efficiency compared with the back projection algorithm.

[0006] In a first aspect, the present invention provides a method for motion compensation of vehicle-mounted SAR under large measurement errors, the method comprising:

[0007] Low-precision inertial navigation data and radar echo data are acquired, and the motion error is coarsely estimated based on the low-precision inertial navigation data and the radar echo data to obtain a coarse estimate value.

[0008] Using the radar echo data compensated by the low-precision inertial navigation data and the coarse estimate, the motion error is precisely estimated to obtain the precise estimate.

[0009] The precise motion error is obtained by using interpolation processing to calculate the low-precision inertial navigation data, the coarse estimate, and the fine estimate;

[0010] The radar echo data is divided into sub-block center points, and the division conditions are determined to obtain the data block set after division.

[0011] The precise motion error is used to perform consistent compensation on the divided data block set to obtain a consistent compensated data block set;

[0012] The consistent compensation data block set is transformed to the range frequency domain, and the accurate motion error is used to perform consistent phase compensation on the transformed data block to obtain the compensation data block set.

[0013] The compensation data block set is transformed and compensated in the range and azimuth directions using the Fourier transform formula to obtain the transformed compensation data block set. The transformed compensation data block set is then compressed in the azimuth direction to obtain the compressed compensation data block set.

[0014] The compressed compensation data block set is subjected to azimuth inverse Fourier transform to obtain multiple range-frequency domain azimuth-time domain signals.

[0015] The multiple distance, frequency, azimuth, and time domain signals are spliced ​​together to obtain the final data.

[0016] In conjunction with the first aspect, in one possible implementation, the step of coarsely estimating the motion error based on the low-precision inertial navigation data and the radar echo data to obtain a coarse estimate includes:

[0017] A motion error model is established, and the instantaneous slant range error expression of any point target in the scene is obtained through analysis, thus obtaining the instantaneous slant range error of the target;

[0018] The radar echo data is compensated using the low-precision inertial navigation data to obtain the initial compensated radar echo data.

[0019] The initial compensation radar echo data is divided into blocks along the range dimension. The azimuth frequency modulation error of the center point of each sub-block is obtained by using image offset autofocus frequency modulation estimation. The azimuth frequency modulation error of the center point of each sub-block is integrated twice to obtain the slant range error of the center point of each sub-block.

[0020] By combining the instantaneous slant range error of the target with the expression for the instantaneous slant range error of any point target in the scene, the motion error equation is obtained, and the motion error equation is solved to obtain the coarse estimate.

[0021] In conjunction with the first aspect, in one possible implementation, the step of using the radar echo data compensated by the low-precision inertial navigation data and the coarse estimate to perform a fine estimate of the motion error, thereby obtaining a fine estimate, includes:

[0022] The slant range error is used to perform consistent compensation on the center point of the sub-block to obtain secondary compensated radar echo data. Range migration correction is then performed on the secondary compensated radar echo data to obtain corrected radar echo data.

[0023] The azimuth frequency error of the center point of each corrected sub-block is obtained by using the image bias autofocus frequency estimation. The azimuth frequency error of the center point of each corrected sub-block is integrated twice to obtain the slant distance error of the center point of each corrected sub-block.

[0024] By combining the instantaneous slant distance error of the target with the slant distance error of the center point of each corrected sub-block, the corrected motion error equation is obtained, and the corrected motion error equation is solved to obtain a precise estimate.

[0025] In conjunction with the first aspect, in one possible implementation, the partitioning conditions include:

[0026] After consistent compensation is applied to the divided data, the remaining slant distance error caused by the spatial variation of the distance to the edge points is less than the first threshold.

[0027] The distance migration difference within the center point of the sub-block is less than the second threshold.

[0028] In conjunction with the first aspect, in one possible implementation, the step of using the precise motion error to perform consistent compensation on the divided data block set to obtain a consistent compensated data block set includes: performing envelope compensation and phase compensation on the divided data block set.

[0029] The envelope compensation function is expressed as:

[0030]

[0031] The phase compensation function is expressed as:

[0032]

[0033] Where η represents the azimuth slow time; fr represents the distance frequency; c represents the speed of light; ΔR represents the azimuth slow time; Rref represents the slant distance to the center point of the sub-block; λ represents the wavelength; t represents the fast time; exp represents the exponential operation with base e; j represents the imaginary part; π represents pi; and ΔR(η,Rref) represents the slant distance error of the center point.

[0034] In conjunction with the first aspect, in one possible implementation, the step of transforming and compensating the compensation data block set in the range and azimuth directions using the Fourier transform formula to obtain the transformed compensation data block set includes:

[0035] Perform inverse Fourier transform on the compensation data block set in the distance direction and Fourier transform on the azimuth direction to obtain the first transformed dataset.

[0036] The first transformed dataset is subjected to distance migration correction to obtain the first corrected transformed dataset;

[0037] The first correction transform dataset is then subjected to Fourier transform on the compensation data block set in the distance direction, and inverse Fourier transform on the compensation data block set in the azimuth direction to obtain the second transform dataset.

[0038] The remaining phase compensation is performed on the transformed data in the second transformed dataset using the precise motion error to obtain the transformed compensated data block set.

[0039] In conjunction with the first aspect, in one possible implementation, the residual phase compensation function is expressed as:

[0040]

[0041] Where η represents the azimuth slow time; fr represents the distance frequency; Rref represents the slant distance to the center point of the sub-block; λ represents the wavelength; exp represents the exponential operation with base e; j represents the imaginary part; π represents pi; and ΔR(η,Rref) represents the slant distance error of the center point.

[0042] Secondly, the present invention provides a vehicle-mounted SAR motion compensation device under large measurement errors, the device comprising:

[0043] The coarse estimation calculation module is used to acquire low-precision inertial navigation data and radar echo data, and to perform a coarse estimation of motion error based on the low-precision inertial navigation data and the radar echo data to obtain a coarse estimation value.

[0044] The fine estimation calculation module is used to perform a fine estimation of the motion error using the low-precision inertial navigation data and the coarse estimation value, and obtain the fine estimation value.

[0045] The precise error calculation module is used to calculate the precise motion error by using interpolation processing on the low-precision inertial navigation data, the coarse estimate, and the fine estimate.

[0046] The data partitioning module is used to divide the radar echo data into sub-block center points, determine the partitioning conditions, and obtain the partitioned data block set.

[0047] The envelope compensation module is used to perform consistent compensation on the divided data block set using the precise motion error, so as to obtain a consistent compensated data block set.

[0048] The consistency compensation module is used to transform the consistency compensation data block set to the range frequency domain respectively, and use the precise motion error to perform consistency phase compensation on the transformed data blocks to obtain the compensation data block set;

[0049] The data processing module is used to transform and compensate the compensation data block set in the range direction and azimuth direction respectively using the Fourier transform formula to obtain the transformed compensation data block set, and to compress the transformed compensation data block set in the azimuth direction to obtain the compressed compensation data block set.

[0050] The data transformation module is used to perform azimuth-directed inverse Fourier transform on the compressed compensation data block set to obtain multiple range-frequency domain azimuth-time domain signals.

[0051] The output module is used to splice together the multiple range-frequency-domain and azimuth-time-domain signals to obtain the final data.

[0052] Thirdly, the present invention provides a vehicle-mounted SAR motion compensation server under large measurement errors, the server comprising: a memory and a processor;

[0053] The memory is used to store computer-executable instructions;

[0054] The processor is used to execute the computer-executable instructions to implement a vehicle-mounted SAR motion compensation method under large measurement errors.

[0055] Fourthly, the present invention provides a computer-readable storage medium having executable instructions, wherein when a computer executes the executable instructions, it can realize a vehicle-mounted SAR motion compensation method under large measurement errors.

[0056] One or more technical solutions provided in this invention have at least the following technical effects or advantages:

[0057] (1) This invention reduces the motion compensation requirement for the measurement accuracy of the integrated navigation equipment and reduces the cost of motion compensation by first using low-precision inertial navigation data to perform consistent coarse compensation on the echo data, and then using two-step image offset autofocus to accurately estimate the motion error.

[0058] (2) This invention proposes a block motion compensation method, which assumes that the spatial variation of motion error is negligible within the center point of the sub-block, but must be considered between the center points of the sub-block. By performing motion compensation and distance migration correction in blocks, the imaging effect of scene edge points is improved. Attached Figure Description

[0059] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments of the present invention or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0060] Figure 1 A flowchart illustrating the steps of the vehicle-mounted SAR motion compensation method under large measurement errors provided in this embodiment of the invention;

[0061] Figure 2 This is a schematic diagram of the geometric model for vehicle-mounted non-ideal trajectory SAR imaging provided in an embodiment of the present invention;

[0062] Figure 3 A flowchart for target motion error estimation provided in an embodiment of the present invention;

[0063] Figure 4 A flowchart of the block compensation steps provided in an embodiment of the present invention;

[0064] Figure 5 This is a simulation diagram of the actual trajectory of the target motion and the low-precision inertial navigation record value provided in an embodiment of the present invention;

[0065] Figure 6A The target distribution of the simulation scene provided in the embodiments of the present invention;

[0066] Figure 6B The simulation results of point targets in the entire scene provided in the embodiments of the present invention;

[0067] Figure 7A The simulation results for the scene center point and edge point provided by the conventional method in the embodiments of the present invention;

[0068] Figure 7B The simulation results of the scene center point and edge point provided by the method of the present invention are provided in the embodiments of the present invention;

[0069] Figure 8 The imaging results of the experimental data provided in the embodiments of the present invention;

[0070] Figure 9A Partial detail diagram of the conventional method provided in the embodiments of the present invention;

[0071] Figure 9B The diagram shows partial details of the method proposed in this invention, provided as an embodiment of the invention. Detailed Implementation

[0072] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0073] This invention provides a method for motion compensation of vehicle-mounted SAR under large measurement errors, such as... Figure 1 As shown, the method includes the following steps S101 to S109.

[0074] S101: Acquire low-precision inertial navigation data and radar echo data. Based on the low-precision inertial navigation data and the radar echo data, perform a coarse estimation of the motion error to obtain a coarse estimate value. The radar echo data is a two-dimensional matrix, divided into range and azimuth directions.

[0075] Specifically, step S101 includes the following steps S1011 to S1014.

[0076] S1011, Establish the motion error model, analyze the motion error expression, and obtain the instantaneous slant range error of the target. Specifically, the vehicle-mounted non-ideal trajectory SAR imaging model is as follows: Figure 2As shown, the radar is at altitude h and operates in frontal side-view mode. During imaging, the car undergoes variable speed movement along the Y-axis (azimuth direction), while experiencing significant disturbance along the X-axis. The altitude remains constant. The actual trajectory is represented by a solid red line, and the ideal trajectory by a dashed blue line. At a certain moment, the radar's true position is A'[x]. r (η),y r The ideal position is A[0,vη,h], where η represents the azimuth time, v represents the vehicle speed during ideal motion, and h represents the radar altitude. The motion error can be expressed as the deviation of the actual trajectory from the ideal trajectory, denoted as [Δx(η),Δy(η),0]. The shortest slant distance of the point target P is R. s Subject to horizontal beamwidth θ bw Due to limitations, the radar can only receive the echo signal of the point target P during the period from C to C'.

[0077] During the time interval C to C', if the motion errors of point target P in the range and azimuth directions are projected onto the slant range, then the instantaneous slant range error can be expressed as:

[0078]

[0079] in, θ represents the angle between the projection of PA on the ground and the X-axis, and θ represents the complementary angle of the downward angle of point P relative to point A.

[0080] like Figure 1 As shown, the true position of the radar at a certain moment is A'[x] r (η),y r The ideal position is A[0, vη, h], and both the actual and ideal positions change with the azimuth and time η (generally 2000 per second). Both θ and η change with η and are related to the position of the target point. Equation (1.1) fully reflects the spatial variability and coupling of motion error. Equation (1.1) is satisfied at any orientation and time η of the ideal trajectory and the real trajectory.

[0081] S1012 uses low-precision inertial navigation data to compensate for radar echo data, obtaining initially compensated radar echo data. Specifically, envelope compensation is performed in the two-dimensional time domain, and phase compensation is performed in the range-frequency domain and azimuth-time domain.

[0082] The envelope compensation function is expressed as:

[0083]

[0084] The phase compensation function is expressed as:

[0085]

[0086] Where η represents the azimuth slow time; fr represents the distance frequency; c represents the speed of light; Rref represents the slant distance to the center point of the sub-block; λ represents the wavelength; t represents the fast time; exp represents the exponential operation with base e; j represents the imaginary part; π represents pi; and ΔR(η,Rref) represents the slant distance error of the center point.

[0087] S1013, the initial compensation radar echo data is divided into blocks along the range dimension. The azimuth frequency modulation error of the center point of each sub-block is obtained using image offset autofocus frequency modulation estimation. The azimuth frequency modulation error of the center point of each sub-block is then integrated twice to obtain the slant range error of the center point of each sub-block. Specifically, the initial compensation radar echo data is divided into blocks along the range dimension, and the azimuth frequency modulation error Δk of the center point of each sub-block is obtained based on image offset autofocus frequency modulation estimation. a First, for Δk a By integrating once and removing the linear components, and then integrating twice, we can obtain the nonlinear phase error φ(η) at the center time of each sub-block center point caused by the motion error. Then, the slant distance error of the sub-block center point can be expressed as ΔR(η)=λφ(η) / (4π).

[0088] S1014, combining the instantaneous slant range error of the target with the expression for the instantaneous slant range error of any point target in the scene, the motion error equation is obtained, and the motion error equation is solved to obtain a coarse estimate. Specifically, a series of equations related to motion error are constructed by combining the projection expansion of the instantaneous slant range error as follows:

[0089]

[0090] Where n represents the number of range blocks, and at time η0, the corresponding θ and θ can be calculated based on the ideal radar position and the position of the center point of the sub-block. Theoretically, n=2 is sufficient to obtain the unique solution to the above system of equations. In practice, the method of dividing the system into multiple sub-block center points can be used to reduce the influence of noise and bad values ​​(while also taking into account the computational complexity), and the least squares method can be used to obtain rough estimates of Δx and Δy.

[0091] S102 uses radar echo data compensated by low-precision inertial navigation data and coarse estimation values ​​to perform fine estimation of motion error and obtain fine estimation value.

[0092] Specifically, step S102 includes:

[0093] S1021, using slant range error to uniformly compensate the sub-block center point, obtains secondary compensated radar echo data. Range migration correction is then applied to this secondary compensated radar echo data to obtain corrected radar echo data. Specifically, the range migration correction function is expressed as follows:

[0094]

[0095] Where c represents the speed of light; λ represents the wavelength; R ref f represents the slant distance from the center point of the sub-block; a Indicates azimuth frequency. v represents the car's speed under ideal motion; θ′ represents the instantaneous angle of view; k represents the frequency modulation frequency of the transmitted linear frequency modulated signal; t represents the distance-to-frequency fast time.

[0096] S1022, the azimuth frequency error of the center point of each corrected sub-block is obtained by using the image bias self-focusing frequency estimation, and the azimuth frequency error of the center point of each corrected sub-block is integrated twice to obtain the slant distance error of the center point of each corrected sub-block.

[0097] S1023, combining the instantaneous slant range error of the target with the slant range error of the center point of each corrected sub-block, the corrected motion error equation is obtained, and the corrected motion error equation is solved to obtain the precise estimate.

[0098] S103, interpolation is used to calculate the precise motion error from the low-precision inertial navigation data, coarse estimates, and fine estimates. Specifically, the data repetition frequencies (simply put, the number of data updates per second) obtained from the low-precision inertial navigation data, coarse estimates Δx and Δy, and fine estimates Δx and Δy are all less than the data rate ΔR(η) used for compensation, requiring interpolation. This reduces the measurement error.

[0099] For example, such as Figure 3 The diagram shows a flowchart of motion error estimation provided by the present invention. Specifically, when using the method provided by the present invention, the radar echo data is first compensated using low-precision inertial navigation data. Then, the image offset sub-focus is used for the first time to obtain the coarse estimates of the residual error Δx and Δy after inertial navigation compensation (because the inertial navigation data is inaccurate; if the inertial navigation data were accurate, the precise motion error could be obtained through inertial navigation).

[0100] Then, the slant range error obtained from the coarse estimation is used to perform another uniform compensation on the echo data after the first uniform compensation.

[0101] Then, distance migration correction is performed, and then the image bias subfocus is used a second time to obtain the precise estimates of Δx and Δy.

[0102] Finally, by using low-precision inertial navigation data, coarse estimates Δx and Δy, and fine estimates Δx and Δy, the data is interpolated to obtain the accurate motion error.

[0103] S104: Divide the radar echo data into sub-block center points and determine the division conditions to obtain the subdivided data block set.

[0104] Specifically, in step S104, the partitioning conditions include: (1) after consistent compensation of the partitioned data, the remaining slant distance error caused by the spatial variation of the distance to the edge point is less than the first threshold; (2) the distance migration difference within the center point of the sub-block is less than the second threshold.

[0105] For example, when dividing radar echo data into sub-block center points, the division of range-direction sub-block center points needs to meet two conditions: after consistent compensation, the remaining slant range error caused by the spatial variation of range-direction edge points is less than 1.5ρ. r The distance migration difference within the center point of the sub-block is less than ρ. r / 4, where ρ r This indicates the distance resolution. The division of azimuth sub-block center points requires meeting one condition: after consistent compensation, the remaining slant range error at azimuth edge points due to spatial variation is less than 1.5ρ. r .

[0106] S105, use the precise motion error to perform consistent compensation on the divided data block set to obtain a consistent compensated data block set.

[0107] Specifically, in step S105, the data block set after division is uniformly compensated using the precise motion error to obtain a uniformly compensated data block set, including: envelope compensation and phase compensation of the data block set after division. Specifically, envelope compensation and phase compensation are calculated using the above formulas (1.2) and (1.3).

[0108] S106, transform the consistent compensation data block set to the range frequency domain respectively, and use the precise motion error to perform consistent phase compensation on the transformed data block to obtain the compensation data block set.

[0109] S107. The Fourier transform formula is used to transform and compensate the compensation data block set in the range and azimuth directions respectively to obtain the transformed compensation data block set. The transformed compensation data block set is then compressed in the azimuth direction to obtain the compressed compensation data block set.

[0110] Specifically, in step S107, the Fourier transform formula is used to transform and compensate the compensation data block set in the range direction and azimuth direction respectively to obtain the transformed compensation data block set, including the following steps S1071 to S1075.

[0111] S1071, Perform inverse Fourier transform on the compensation data block set in the distance direction and Fourier transform on the azimuth direction to obtain the first transformed dataset.

[0112] S1072, perform distance migration correction on the first transformed dataset to obtain the first corrected transformed dataset. Specifically, data correction is performed using formula (1.5) of the correction function.

[0113] In step S1073, the first correction transform dataset is subjected to Fourier transforms on the compensation data block set in the range direction and inverse Fourier transforms on the compensation data block set in the azimuth direction to obtain the second transform dataset. The compensation performed in steps S105, S1061, S1062, and S1063 is all for the center point in the scene; in fact, ΔR(η,R) ref It varies with distance.

[0114] S1074, using precise motion errors, residual phase compensation is performed on the transformed data in the second transformed dataset to obtain the transformed compensated data block set. Specifically, the residual phase compensation function is expressed as:

[0115]

[0116] Where η represents the azimuth slow time; fr represents the distance frequency; Rref represents the slant distance to the center point of the sub-block; λ represents the wavelength; exp represents the exponential operation with base e; j represents the imaginary part; π represents pi; and ΔR(η,Rref) represents the slant distance error of the center point.

[0117] Specifically, in step S106, the azimuth compression function is expressed as follows:

[0118]

[0119] Where fr represents the range frequency; fa represents the azimuth frequency; R ref This represents the slant distance from the center point of the sub-block.

[0120] S108 performs azimuth-direction inverse Fourier transform on the compressed compensation data block set to obtain multiple range-frequency domain azimuth-time domain signals.

[0121] S109 splices multiple range, frequency, azimuth, and time domain signals to obtain the final data.

[0122] In a specific embodiment provided by the present invention, such as Figure 4 As shown, firstly, the radar echo data is divided into sub-block center points to obtain the divided data block set.

[0123] Then, the precise motion error is used to perform consistent compensation on the divided data block set to obtain a consistent compensated data block set.

[0124] The consistent compensation data block set is then transformed to the range frequency domain, and consistent phase compensation is performed on the transformed data block using the precise motion error to obtain the compensation data block set.

[0125] Then, the Fourier transform formula is used to transform and compensate the compensation data block set in the range and azimuth directions respectively to obtain the transformed compensation data block set. The transformed compensation data block set is then compressed in azimuth to obtain the compressed compensation data block set.

[0126] Subsequently, the compressed compensation data block set was subjected to azimuth inverse Fourier transform to obtain multiple range-frequency domain azimuth-time domain signals.

[0127] Finally, the data of the center points of all sub-blocks are concatenated to obtain the final data.

[0128] This invention can obtain vehicle-mounted SAR images with good focusing effect under non-ideal trajectories without relying on high-precision integrated navigation equipment, reducing the implementation cost of high-precision motion compensation for vehicle-mounted SAR and making it more widely applicable.

[0129] This invention provides a specific simulation embodiment.

[0130] Set the simulation parameters as shown in Table 1.

[0131] Table 1 Simulation Parameters

[0132] carrier frequency 77GHz radar pulse frequency 2000Hz Number of directional points 8000 ADC sampling rate 10MHz Distance Points 256 Heading angle accuracy 2° Radar Altitude 1.2m Speed ​​accuracy 0.05m / s Reference Slope Distance 10m Inertial navigation measurement data rate 50Hz

[0133] like Figure 5 As shown, the simulation results primarily consider the inertial navigation heading angle measurement error, velocity measurement error, and data frame rate, comparing the actual motion trajectory value with the low-precision inertial navigation recorded value.

[0134] like Figure 6A As shown, the distribution of point targets in the scene; Figure 6B For the imaging results of the entire scene, from Figure 6B As can be seen, all point targets in the scene are in focus.

[0135] Figure 7A The imaging results are from traditional methods. Figure 7B The image shows the imaging results of the method of this invention. Traditional methods refer to imaging after directly using large measurement error data for full aperture motion compensation. As can be seen from the figure, the scene center point PT1 in the traditional method cannot be fully focused due to the influence of measurement errors, and the scene edge points PT2 and PT3 are almost impossible to focus due to the simultaneous influence of measurement errors and residual spatial variation errors. In contrast, the method proposed in this invention can focus all point targets in the scene very well. The number of sampling units on both the horizontal and vertical axes is the number of sampling points after interpolation by eight times. Due to the influence of platform speed, pulse repetition frequency, and range resolution units, the length represented by each cell on the horizontal and vertical axes is different.

[0136] Figure 8These are the experimental results of the algorithm proposed in this invention driving along a non-ideal trajectory in a parking lot. Overall, the focusing effect is good, and the car outline is relatively clear. To further reveal details, during the experiment... Figure 8 Three corner reflectors were placed in the box (the car's windshield).

[0137] Figure 9A and Figure 9B The figures show the imaging results of the three corner reflectors using the traditional method and the method proposed in this invention. It can be seen from the figures that the method proposed in this invention has a good focusing effect, while the traditional method does not have an ideal imaging effect due to the influence of measurement error and scene spatial variation. The experimental results show that the method proposed in this invention is more suitable for vehicle motion compensation under large measurement error.

[0138] This invention has broad application prospects in the field of autonomous driving. Vehicle-mounted SAR imaging technology can significantly improve the angular resolution of traditional millimeter-wave radar, providing two-dimensional high-resolution images for driving scenarios. In harsh environments or under low light conditions, this technology can effectively compensate for the limited perception capabilities of cameras and lidar, demonstrating a more reliable perception level. After being calculated and analyzed by the vehicle's intelligent system, vehicle-mounted SAR images can accurately detect potential hazards and issue corresponding instructions to the driver through the control system, enabling the driver to detect and eliminate potential risks in advance, thereby significantly improving driving safety. Considering the characteristics of actual vehicle trajectories, the application of vehicle-mounted SAR technology to autonomous driving must address the problem of imaging under non-ideal trajectories. Compared with traditional methods, the method proposed in this invention does not require high-precision integrated navigation equipment; it can achieve high-precision motion compensation solely relying on the vehicle's own motion sensors, reducing the hardware cost of applying vehicle-mounted SAR imaging technology in autonomous driving. Furthermore, to address the problem of large spatial variation in motion errors in vehicle scenarios, the block motion compensation method proposed in this invention has a greater advantage in computational efficiency than existing back projection algorithms, making it more conducive to real-time imaging.

[0139] This invention provides a vehicle-mounted SAR motion compensation device under large measurement errors. The device includes: a coarse estimate calculation module, a fine estimate calculation module, a precise error calculation module, a data partitioning module, an envelope compensation module, a consistency compensation module, a data processing module, and an output module.

[0140] The coarse estimation module is used to acquire low-precision inertial navigation data and radar echo data, and to make a coarse estimate of the motion error based on the low-precision inertial navigation data and radar echo data to obtain a coarse estimate value.

[0141] The fine estimation calculation module is used to perform fine estimation of motion error using low-precision inertial navigation data and coarse estimation values ​​to obtain fine estimation values.

[0142] The precise error calculation module is used to calculate the precise motion error by interpolating low-precision inertial navigation data, coarse estimates, and fine estimates.

[0143] The data partitioning module is used to divide the radar echo data into sub-block center points, determine the partitioning conditions, and obtain the partitioned data block set.

[0144] The envelope compensation module is used to perform consistent compensation on the divided data block set using precise motion errors, so as to obtain a consistent compensated data block set.

[0145] The consistency compensation module is used to transform the consistency compensation data block set to the range frequency domain and use the precise motion error to perform consistency phase compensation on the transformed data block to obtain the compensation data block set.

[0146] The data processing module is used to transform and compensate the compensation data block set in the range and azimuth directions using the Fourier transform formula to obtain the transformed compensation data block set. The transformed compensation data block set is then compressed in the azimuth direction to obtain the compressed compensation data block set.

[0147] The data transformation module is used to perform azimuth-directed inverse Fourier transform on the compressed compensation data block set to obtain multiple range-frequency domain azimuth-time domain signals.

[0148] The output module is used to stitch together multiple range, frequency, azimuth, and time domain signals to obtain the final data.

[0149] The apparatus or module described in the above embodiments can be implemented by a computer chip or physical entity, or by a product with a certain function. For ease of description, the above apparatus is described by dividing it into various modules according to their functions. In implementing this invention, the functions of each module can be implemented in one or more software and / or hardware. Of course, a module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.

[0150] The methods, apparatus, or modules described in this invention can be implemented in a computer-readable program code manner. The controller can be implemented in any suitable manner, for example, as a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC 625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of a memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code manner, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, application-specific integrated circuits, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included within it for implementing various functions can also be considered as structures within the hardware component. Alternatively, the device used to implement various functions can be viewed as either a software module implementing the method or a structure within a hardware component.

[0151] Some modules in the apparatus described in this invention can be described in the general context of computer-executable instructions that are executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc., that perform a specific task or implement a specific abstract data type. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0152] This invention provides a vehicle-mounted SAR motion compensation server under large measurement errors. The server includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes the computer-executable instructions to implement a vehicle-mounted SAR motion compensation method under large measurement errors.

[0153] This invention provides a computer-readable storage medium having executable instructions, which enables a vehicle-mounted SAR motion compensation method under large measurement errors when the computer executes the executable instructions.

[0154] The aforementioned storage media include, but are not limited to, Random Access Memory (RAM), Read-Only Memory (ROM), Cache, Hard Disk Drive (HDD), or Memory Card. The memory can be used to store computer program instructions.

[0155] While this invention provides the method operation steps as described in the embodiments or flowcharts, more or fewer operation steps may be included based on conventional or non-inventive labor. The order of steps listed in this embodiment is merely one possible execution order among many and does not represent the only possible execution order. In actual device or client product execution, the methods shown in this embodiment or the accompanying drawings can be executed sequentially or in parallel (e.g., in a parallel processor or multi-threaded processing environment).

[0156] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary hardware. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product, or it can be embodied in the process of data migration. The computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, mobile terminal, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0157] The various embodiments described in this specification are presented in a progressive manner. Similar or identical parts between embodiments can be referred to interchangeably. Each embodiment focuses on its differences from other embodiments. All or part of this invention can be used in numerous general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, mobile communication terminals, multiprocessor systems, microprocessor-based systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices, etc.

[0158] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present invention.

Claims

1. A method for motion compensation of vehicle-mounted SAR under large measurement errors, characterized in that, include: Low-precision inertial navigation data and radar echo data are acquired, and the motion error is coarsely estimated based on the low-precision inertial navigation data and the radar echo data to obtain a coarse estimate value. The step of coarsely estimating the motion error based on the low-precision inertial navigation data and the radar echo data to obtain a coarse estimate includes: establishing a motion error model; obtaining the instantaneous slant range error expression for any point target in the scene through analysis; compensating the radar echo data using the low-precision inertial navigation data to obtain initial compensated radar echo data; dividing the initial compensated radar echo data into blocks along the range dimension; obtaining the azimuth frequency modulation error of the center point of each sub-block using image offset autofocus frequency modulation estimation; integrating the azimuth frequency modulation error of the center point of each sub-block twice to obtain the slant range error of the center point of each sub-block; combining the instantaneous slant range error of the target with the instantaneous slant range error expression for any point target in the scene to obtain a motion error equation; and solving the motion error equation to obtain the coarse estimate. Using the radar echo data compensated by the low-precision inertial navigation data and the coarse estimate, the motion error is precisely estimated to obtain the precise estimate. The precise motion error is obtained by using interpolation processing to calculate the low-precision inertial navigation data, the coarse estimate, and the fine estimate; The radar echo data is divided into sub-block center points, and the division conditions are determined to obtain the data block set after division. The precise motion error is used to perform consistent compensation on the divided data block set to obtain a consistent compensated data block set; The consistent compensation data block set is transformed to the range frequency domain, and the accurate motion error is used to perform consistent phase compensation on the transformed data block to obtain the compensation data block set. The compensation data block set is transformed and compensated in the range and azimuth directions using the Fourier transform formula to obtain the transformed compensation data block set. The transformed compensation data block set is then compressed in the azimuth direction to obtain the compressed compensation data block set. The compressed compensation data block set is subjected to azimuth inverse Fourier transform to obtain multiple range-frequency domain azimuth-time domain signals. The multiple distance, frequency, azimuth, and time domain signals are spliced ​​together to obtain the final data.

2. The vehicle-mounted SAR motion compensation method under large measurement errors according to claim 1, characterized in that, The process of using the radar echo data compensated by the low-precision inertial navigation data and the coarse estimate to perform a fine estimate of the motion error, resulting in a fine estimate, includes: The slant range error is used to perform consistent compensation on the center point of the sub-block to obtain secondary compensated radar echo data. Range migration correction is then performed on the secondary compensated radar echo data to obtain corrected radar echo data. The azimuth frequency error of the center point of each corrected sub-block is obtained by using the image bias autofocus frequency estimation. The azimuth frequency error of the center point of each corrected sub-block is integrated twice to obtain the slant distance error of the center point of each corrected sub-block. By combining the instantaneous slant distance error of the target with the slant distance error of the center point of each corrected sub-block, the corrected motion error equation is obtained, and the corrected motion error equation is solved to obtain a precise estimate.

3. The vehicle-mounted SAR motion compensation method under large measurement errors according to claim 1, characterized in that, The partitioning conditions include: After consistent compensation is applied to the divided data, the remaining slant distance error caused by the spatial variation of the distance to the edge points is less than the first threshold. The distance migration difference within the center point of the sub-block is less than the second threshold.

4. The vehicle-mounted SAR motion compensation method under large measurement errors according to claim 1, characterized in that, The step of using the precise motion error to perform consistent compensation on the divided data block set to obtain a consistent compensated data block set includes: performing envelope compensation and phase compensation on the divided data block set; The envelope compensation function is expressed as: The phase compensation function is expressed as: in, Indicates location and time. Indicates distance frequency; Represents the speed of light; Indicates the slant distance from the center point of the sub-block; Indicates wavelength; Indicates a fast time; This indicates exponential operations with base e. Indicates the imaginary part; Represents pi; This represents the slope distance error at the center point.

5. The vehicle-mounted SAR motion compensation method under large measurement errors according to claim 1, characterized in that, The process involves transforming and compensating the compensation data block set in both the range and azimuth directions using the Fourier transform formula, resulting in a transformed compensation data block set, including: Perform inverse Fourier transform on the compensation data block set in the distance direction and Fourier transform on the azimuth direction to obtain the first transformed dataset. The first transformed dataset is subjected to distance migration correction to obtain the first corrected transformed dataset; Perform a Fourier transform on the first corrected transform dataset in the distance direction and an inverse Fourier transform on the azimuth direction to obtain the second transform dataset; The remaining phase compensation is performed on the transformed data in the second transformed dataset using the precise motion error to obtain the transformed compensated data block set.

6. The vehicle-mounted SAR motion compensation method under large measurement errors according to claim 5, characterized in that, The remaining phase compensation function is expressed as: in, Indicates location and time. Indicates distance frequency; Indicates the slant distance from the center point of the sub-block; Indicates wavelength; This indicates exponential operations with base e. Indicates the imaginary part; Represents pi; This represents the slope distance error at the center point.

7. A vehicle-mounted SAR motion compensation device under large measurement errors, used to implement the vehicle-mounted SAR motion compensation method under large measurement errors as described in any one of claims 1 to 6, characterized in that, include: The coarse estimation calculation module is used to acquire low-precision inertial navigation data and radar echo data, and to perform a coarse estimation of motion error based on the low-precision inertial navigation data and the radar echo data to obtain a coarse estimation value. The step of coarsely estimating the motion error based on the low-precision inertial navigation data and the radar echo data to obtain a coarse estimate includes: establishing a motion error model; obtaining the instantaneous slant range error expression for any point target in the scene through analysis; compensating the radar echo data using the low-precision inertial navigation data to obtain initial compensated radar echo data; dividing the initial compensated radar echo data into blocks along the range dimension; obtaining the azimuth frequency modulation error of the center point of each sub-block using image offset autofocus frequency modulation estimation; integrating the azimuth frequency modulation error of the center point of each sub-block twice to obtain the slant range error of the center point of each sub-block; combining the instantaneous slant range error of the target with the instantaneous slant range error expression for any point target in the scene to obtain a motion error equation; and solving the motion error equation to obtain the coarse estimate. The fine estimation calculation module is used to perform a fine estimation of the motion error using the low-precision inertial navigation data and the coarse estimation value, and obtain the fine estimation value. The precise error calculation module is used to calculate the precise motion error by using interpolation processing on the low-precision inertial navigation data, the coarse estimate, and the fine estimate. The data partitioning module is used to divide the radar echo data into sub-block center points, determine the partitioning conditions, and obtain the partitioned data block set. The envelope compensation module is used to perform consistent compensation on the divided data block set using the precise motion error, so as to obtain a consistent compensated data block set. The consistency compensation module is used to transform the consistency compensation data block set to the range frequency domain respectively, and use the precise motion error to perform consistency phase compensation on the transformed data blocks to obtain the compensation data block set; The data processing module is used to transform and compensate the compensation data block set in the range direction and azimuth direction respectively using the Fourier transform formula to obtain the transformed compensation data block set, and to compress the transformed compensation data block set in the azimuth direction to obtain the compressed compensation data block set. The data transformation module is used to perform azimuth-directed inverse Fourier transform on the compressed compensation data block set to obtain multiple range-frequency domain azimuth-time domain signals. The output module is used to splice together the multiple range-frequency-domain and azimuth-time-domain signals to obtain the final data.

8. A vehicle-mounted SAR motion compensation server for large measurement errors, characterized in that, Including memory and processor; The memory is used to store computer-executable instructions; The processor is used to execute the computer-executable instructions to implement the vehicle-mounted SAR motion compensation method under large measurement errors as described in any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium has executable instructions, and when the computer executes the executable instructions, it can implement the vehicle-mounted SAR motion compensation method under large measurement errors as described in any one of claims 1-6.

Images