37 results about "Motion parameter estimation" patented technology

The invention relates to the technical field of radar signal processing, and discloses a method for calibrating radar echo signals on a mobile platform, which comprises the following steps of: acquiring original echo signals of a radar on the mobile platform under different working conditions and preprocessing the original echo signals, acquiring multi-source IMU sensor and differential GPS information and performing time sequence fusion; carrying out platform motion parameter estimation on the preprocessed echo signal; and based on a preset reference signal model, performing feature extraction on the echo signal, and obtaining feature parameters for calibration through time-frequency processing and spatial domain adaptive filtering. Multi-source IMU (inertial measurement unit) data and differential GPS (global positioning system) data are fused to obtain six-degree-of-freedom continuous motion parameters of a mobile platform, and time synchronization and preprocessing of echo signals are combined to realize dynamic calibration of original radar echo signals. And the problem of amplitude, phase and Doppler frequency deviation of the echo signal caused by the attitude and the motion state of the mobile platform is solved.

The invention provides a target third-order motion parameter estimation method and device based on high-order phase fusion, and the method comprises the steps: building a third-order polynomial phase model based on a radar echo signal, and carrying out the high-order phase fusion estimation, and obtaining an estimation value of a third-order motion parameter; performing third-order phase compensation based on the estimated value of the third-order motion parameter to obtain an estimated value of a second-order motion parameter; performing phase compensation on the slow time complex sequence based on the estimated value of the third-order motion parameter and the estimated value of the second-order motion parameter to obtain a compensation sequence; and converting the compensation sequence into a single-frequency slow-time complex sequence, and estimating motion parameters of the to-be-measured target based on the single-frequency slow-time complex sequence. Through step-by-step compensation of the third-order phase term and the second-order phase term, coherent focusing of target energy can be realized under the condition of not depending on dense grid search, so that the robustness and precision of maneuvering target detection are improved, and meanwhile, the requirements of a vehicle-mounted system on real-time performance and computing power cost are considered.

The invention discloses a star point motion parameter estimation method for a high-dynamic star sensor. The method specifically comprises the following steps: (1) acquiring a star map under a dynamic condition through a fixed star sensor; (2) carrying out filtering preprocessing on the fixed star sensor image obtained in the step (1); (3) carrying out threshold segmentation on the image processed in the step (2) and carrying out connected domain extraction to obtain rough star points; (4) selecting the brightest star point in the star points obtained in the step (3); (5) obtaining a sub-pixel star point track by using the brightest star point adjacent area obtained in the step (4) and adopting a Zernike moment sub-pixel edge extraction method; and (6) utilizing the sub-pixel star point trajectory obtained in the step (5), and adopting least square fitting to obtain star point motion parameters. The method has the beneficial effects that the precision of star point trailing length and angle estimation is improved, so that the extraction precision of star points in a dynamic environment is improved, and the precision of a fixed star sensor is improved.

The application discloses a kind of high-efficiency motorized weak target long-time coherent accumulation detection methods, belong to radar target detection field;Including the three-scale decomposition of motorized target radial motion parameter variable;In target blur scale motion parameter variable search space eliminates Doppler frequency blur effect;Through Keystone conversion correction first-order range migration caused by motorized target baseband speed;In target coarse scale baseband motion parameter variable search space, estimate motorized target coarse scale baseband motion parameter by GIFT conversion;In target fine scale baseband motion parameter variable search space, estimate motorized target fine scale baseband motion parameter, obtain temporary range-doppler spectrum;According to motorized target motion parameter variable three-scale decomposition model, estimate the complete motion parameter of motorized target;Based on motorized target motion parameter estimation, obtain motorized target range-doppler spectrum.The method has the characteristics of low computational complexity, Doppler blur effective compensation, high detection performance.

The application provides a multi-target motion parameter estimation method and device based on MIMO radar, which comprises: performing down-conversion processing on echo signals observed by each receiving array of a MIMO radar system to obtain identification results of baseband signals corresponding to each receiving array; constructing a covariance matrix corresponding to each receiving array based on the identification results; performing eigenvalue decomposition on each covariance matrix, and determining a noise subspace corresponding to each receiving array based on the results of the eigenvalue decomposition of each receiving array; and determining an estimation result of a motion parameter of each target to be measured based on the noise subspace corresponding to each receiving array and an array response matrix; wherein the array response matrix comprises array response characteristics of the receiving array and signal characteristics of the echo signals observed by the receiving array. The application can effectively reduce the calculation load in a multi-target scene, thereby overcoming the limitation of direct estimation methods in the application of multi-target scenes.

The invention discloses long-time monitoring based on kinematics and mechanics priori under complex working conditions of AI + vision + IMU fusion, and belongs to the technical field of large-scale civil structure health monitoring, and the method comprises the following steps: firstly, respectively collecting image data and motion data of a civil structure by using a vision camera and an IMU sensor; decoupling the rotation motion of the camera through a mathematical model based on kinematics and mechanics priori knowledge; only one control point is used, decoupling information and collected data are combined, and six-degree-of-freedom motion parameters are estimated through a measurement equation optimized through AI; and finally calculating displacement information and performing error compensation. The system comprises a data acquisition module, a rotary motion decoupling module, a six-degree-of-freedom motion parameter estimation module and a displacement measurement and compensation module. According to the invention, the hardware cost is reduced, the measurement flexibility and adaptability are improved, and the method is suitable for long-time monitoring under complex working conditions.

This invention provides a radar motion parameter estimation method and apparatus based on information criterion values. It acquires raw radar echo data, performs range processing and target extraction to obtain a target slow-time complex sequence. Based on the target slow-time complex sequence, it constructs multi-order motion candidate models, calculates the residual sum of squares, and calculates the information criterion value based on the residual sum of squares. The optimal model order is determined based on the information criterion value, and the target radial motion parameters are inverted and output based on the phase coefficients corresponding to the optimal model order. Instead of pre-fixing the use of a first-order, second-order, or third-order model, it determines the optimal model order after constructing multi-order candidate models and evaluating them using the calculated information criterion value. This approach adaptively determines the appropriate motion model order for the target based on the phase evolution characteristics of the signal itself, and completes motion parameter estimation accordingly. This ensures estimation accuracy while suppressing unnecessary model complexity, thus improving its engineering application value.

The present disclosure provides a kind of long-distance high-speed weak target detection method and device based on DP-RFT, belong to radar detection technical field.In the target detection process, first, dynamic programming (DP) algorithm is used to estimate target motion parameters, and the DP parameter estimation result is used as the prior information of Radon Fourier transform (RFT) algorithm to solve the blind speed sidelobe problem of RFT long-time accumulation;Then, using the target motion parameters obtained by the DP algorithm as the center, the RFT algorithm is used for coherent accumulation to obtain the target motion parameter estimation value for compensation.Then, the target echo after motion parameter compensation is synthesized to obtain the target high-resolution one-dimensional range image.The present application can estimate the target motion parameters and synthesize the target high-resolution one-dimensional range image in the case of range ambiguity, avoid blind speed sidelobe, reduce the amount of calculation, so as to realize the stable detection of long-distance high-speed weak target.

An unmanned aerial vehicle time-frequency image reconstruction method based on deep learning belongs to the technical field of radar signal processing and image reconstruction, solves the technical problem that unmanned aerial vehicle time-frequency images are difficult to identify and analyze under the condition of low signal-to-noise ratio, and comprises the following steps: S1, performing short-time Fourier transform on echo signals, and establishing an unmanned aerial vehicle radar echo model STFT; s2, designing a SelfNet model based on a convolutional neural network, and reconstructing an unmanned aerial vehicle time-frequency curve graph; s3, according to an unmanned aerial vehicle radar echo model STFT and a time-frequency curve output image, establishing an unmanned aerial vehicle time-frequency image data set, and performing training by using a SelfNet model; and S4, reconstructing the time-frequency image data set of the unmanned aerial vehicle by using the trained SelfNet model weight. The method can be used for actually measuring data, the precision of the time-frequency image of the unmanned aerial vehicle under the low signal-to-noise ratio can be improved, and follow-up multi-class classification and micro-motion parameter estimation of the unmanned aerial vehicle are facilitated.

The invention relates to the field of golf sensing, and discloses a golf swing data calculation method and system based on machine vision, and the method comprises the following steps: S1, obtaining an image sequence containing a golf club swing process, and processing the image sequence to obtain an initial motion track of a club head; s2, outputting a prediction rod type and an initial motion parameter estimation value through a first pre-training model; s3, processing the image sequence to construct a six-degree-of-freedom pose of the club head, and calculating based on the six-degree-of-freedom pose to obtain a group of swing data; the system comprises an initial analysis module, an initial analysis module and a data calculation module. According to the method, the two-dimensional positioning model and the three-dimensional geometric model which are accurately matched with the type of the golf club are adaptively and dynamically selected from the model library through the preliminarily predicted type of the golf club, so that pertinent geometric priori knowledge is obtained through subsequent refined calculation, and the effect of accurately analyzing different golf clubs is achieved.

The application provides a spatial precession cone target micro-motion parameter estimation method based on deep learning under echo defect conditions; the method comprises the following steps: generating a time-frequency image dataset containing different defect degrees; pre-processing the time-frequency graph, including defect simulation, curve repair and peak feature enhancement; constructing a double-branch anti-defect parameter estimation network (DAPE-Net), which has a shared feature extraction backbone, a space-time attention mechanism, a double-branch feature decoupling structure and an independent regression head, realizes end-to-end estimation of precession period and precession angle from the defective time-frequency graph, and verifies the robustness and high precision of the model under various defect proportions through training and testing; experimental results show that under certain defect conditions, the model can accurately estimate the micro-motion period and precession angle of the target, and is significantly better than mainstream deep learning models, and is suitable for real-time estimation of radar target micro-motion parameters in complex electromagnetic environments.

The present disclosure provides a kind of based on coherent array detector moving target laser synthetic aperture sparse imaging method, comprising: using narrowband laser emission signal to irradiate the target to be detected, and the target to be detected exists rotation and translation;The multiple frames of laser complex images of the target to be detected are collected by laser local oscillator coherent array detector;On the basis of the motion parameter estimation of the target to be detected, the multiple frames of laser complex images are carried out Fourier inverse transform in digital domain, and the multiple frames of spatial sampling signals are obtained, and the multiple frames of spatial sampling signals are carried out phase compensation and motion compensation;The spatial sampling center is calculated, and the multiple frames of spatial sampling signals are spliced, and large range space sparse sampling signal is formed;By lengthening observation time, the sparsity of the large range space sparse sampling signal is reduced, and by compressed sensing processing, the high-resolution image reconstruction and image sidelobe suppression of large range space sparse sampling signal are realized.

This invention discloses a plate motion analysis method based on GNSS and SLR stations of the Eurasian Plate, belonging to the field of geodesy and surveying engineering technology. The method includes: acquiring raw coordinate time-series data from GNSS and SLR stations and preprocessing them; decomposing the preprocessed sequence using singular spectrum analysis (SSA) to reconstruct and separate trend, periodic, and noise terms; performing Fast Fourier Transform (FFT) on the reconstructed signal to extract frequency domain features; establishing a comprehensive model of linear velocity plus periodic correction by combining time and frequency domain features; and analyzing the motion characteristics of the Eurasian Plate based on this model to achieve motion trend prediction and anomaly detection. This invention, through the combination of SSA and FFT, significantly improves the accuracy of motion parameter estimation and the ability to identify weak signals, providing a reliable method for plate motion analysis and earthquake prediction.

The application discloses a SAR motion parameter estimation method based on accurate measurement of time of arrival and direction of arrival, which comprises the following steps: reading N LFM direction of arrival alpha n and corresponding measurement value of time of arrival t n , performing cubic polynomial fitting on the line of sight angle alpha n by using a least square method to smooth measurement error; performing quadratic polynomial fitting on the time of arrival t n by using a least square method to obtain a linear term coefficient p1 and a quadratic term coefficient p2; obtaining a radar pulse repetition interval T by using a statistical pulse time of arrival difference; updating a detection pulse serial number n; solving a radar initial distance r0; and solving a radar flight speed v and a flight direction beta by using the initial distance r0, the direction of arrival alpha n and the time of arrival t n . The application has the characteristic that a signal waveform of a SAR emission is synthesized with an LFM pulse, accurate estimation of SAR motion parameters is realized, the variation rate parameter of the line of sight angle is avoided, and the robustness of the algorithm is improved and the requirement for the direction finding accuracy is reduced.

The application relates to a medium earth orbit (MEO) multi-channel SAR ground moving target motion parameter estimation method and belongs to the field of radar signal processing. A signal model of a distance equation of a first channel and a second channel of MEO SAR based on third-order Taylor expansion is established, a first-order phase coefficient, a second-order phase coefficient, a third-order phase coefficient and an intermediate parameter are estimated based on a CPF method, along-course velocity and radial acceleration in the second-order phase coefficient are adjusted by combining multi-channel information, and along-course velocity, radial velocity, along-course acceleration and radial acceleration of the target are estimated based on a coupling relationship. The method solves the problem of difficult motion parameter estimation of a moving target caused by the coupling relationship.

This invention discloses a method and apparatus for processing echoes and estimating motion parameters of uniformly accelerated moving targets. The method includes: acquiring the frequency-coded target echo signal of a uniformly accelerated moving target; performing baseband echo processing to obtain a pulse-compressed echo; performing range compensation and frequency domain transformation processing to obtain a fast-time Fourier transform echo; performing second-order non-standard Keystone transform processing to obtain a second-order non-standard transform signal; performing acceleration estimation and compensation processing to obtain a compensated echo; performing first-order non-standard Keystone transform processing to obtain a first-order non-standard transform signal; performing echo coherent accumulation processing to obtain a coherently accumulated echo; and performing motion parameter estimation to obtain motion parameter estimation results, including the target's initial range and target velocity. This invention achieves echo processing and motion parameter estimation, improves estimation accuracy, and reduces the computational load. This invention can be widely applied in the field of radar signal processing technology.

This application discloses a radar low-altitude target detection method and system based on least squares inversion, involving radar and data processing technology. The method includes: establishing the target's motion trajectory based on multipath echo signals received by the radar, and establishing a multipath observation model based on the motion trajectory; establishing residuals and a target function based on the multipath observation model; performing inversion iteration using least squares based on the established residuals and target function, solving the problem along the iteration direction based on the iteration step size; and outputting the inverted parameters for target tracking. This application achieves adaptive compensation for multipath effects by modeling multipath bias and jointly inverting parameters such as target range and elevation angle, thereby improving the accuracy of radar estimation of low-altitude target motion parameters.

The invention discloses an ISAR coherent translation compensation and imaging method based on a motion parameter estimation network, and relates to the technical field of imaging, and the method comprises the steps: receiving an echo of a space target inverse synthetic aperture radar; preprocessing the echo to obtain a compensated echo sequence; processing the compensated echo sequence by adopting a trained motion parameter estimation network model to obtain a distance grid offset rough estimation value of the central position of the target in each echo; multiplying the distance grid offset coarse estimation value by the distance resolution unit to obtain a distance coarse estimation value, fitting to obtain a translation parameter coarse estimation value, taking the translation parameter coarse estimation value as a preset imaging parameter to be optimized of an imaging model, and performing iterative optimization to obtain an optimized imaging parameter; obtaining a translation error term of each echo according to the optimized imaging parameters; and compensating the translation error term of each echo, and performing coherent accumulation on each compensated echo to obtain an imaging result. According to the invention, the imaging result is clearer and more accurate.

The application discloses a kind of motion estimation methods for fusing deep learning feature optical flow and binocular vision, comprising: based on controllable adaptive histogram equalization preprocessing to driving image dataset;Construct the optical flow feature extraction model based on deep learning, and the moving target is identified training;Distance measurement is carried out by binocular camera, and the target position is obtained;Vehicle motion speed is acquired.Compared with traditional optical flow speed measurement, this method is based on deep learning optical flow and binocular imaging principle, and the motion parameter estimation of carrier displacement and speed can be realized according to video data, which solves the problem that traditional optical flow estimation method is too sensitive and cannot be stably estimated when driving at night or in weak light environment, and further improves reliability.Meanwhile, this method avoids the cumulative error of traditional inertia sensor and the shortcomings of poor anti-interference ability and low update frequency of GPS positioning speed measurement.

The application discloses a radar signal processing technical field and a high-maneuvering target motion parameter estimation and coherent accumulation detection method. The high-maneuvering target motion parameter estimation and coherent accumulation detection method comprises the following steps: performing pulse compression, neighborhood cross-correlation processing and extracting a self-correlation item on a radar echo of a high-maneuvering target; designing a perception matrix, reconstructing a sparse signal by using a Bayesian compressive sensing algorithm and extracting high-order motion parameters of the target; performing high-order phase compensation, correcting a linear phase, obtaining coherent accumulation and performing CFAR detection; judging the number of high-order motion parameters of the target, completing coherent accumulation detection and outputting a detection result. The high-maneuvering target motion parameter estimation and coherent accumulation detection method has high estimation precision, low calculation complexity and can be effectively applied to single-target and multi-target scenes.

The application provides a target motion parameter estimation method and device based on a distributed MIMO radar, which comprises the following steps: based on a preset echo signal expression model, signal expression is performed on echo signals of a target to be measured observed by each receiving array of a MIMO radar system; wherein the echo signal expression model comprises an angle of arrival expression term, a time of arrival expression term and a Doppler frequency offset expression term, and the angle of arrival expression term, the time of arrival expression term and the Doppler frequency offset expression term are constructed based on a motion parameter term; the results of signal expression corresponding to each receiving array are subjected to down-conversion processing respectively, so as to obtain baseband signal expression results corresponding to each receiving array; and based on the baseband signal expression results corresponding to each receiving array, a maximum likelihood estimation method is used to determine an estimation result of the motion parameter of the target to be measured. The application can realize high-precision estimation of the motion parameter of the target to be measured.

The present application relates to a kind of radar maneuvering target parameter estimation methods under unknown time information, the echo signal of the maneuvering target of unknown time information is obtained by radar system, pulse compression is carried out to echo signal;Keystone transformation and RCM compensation function are constructed to carry out distance unit drift correction;The signal of the distance unit where maneuvering target is located is extracted and modeled as polynomial phase signal;Phase tracking is carried out to model based on extended Kalman filter, and tracking phase is intercepted using sliding time window, and the polynomial coefficient is inverted to reconstruct a new polynomial phase using the tracking phase of interception, and the similarity of tracking phase and reconstructed phase is compared to estimate time information;According to the time information obtained, the effective signal containing maneuvering target information is extracted, and phase tracking and inversion are carried out again to estimate motion parameters.The method is aimed at the motion parameter estimation of maneuvering target under unknown time information, and has outstanding advantages in calculation complexity and precision.

The application discloses a kind of robust frequency-modulated stepped frequency ISAR imaging radar velocity estimation methods, belong to radar technical field, first construct multiple frames frequency-modulated stepped frequency data model;Again, the velocity estimation and compensation of rough accuracy are realized using the same frequency point offset estimation of multi-frequency point synthesis;Then, the high-resolution imaging of the coarse resolution distance unit where the special point is located is obtained by the IFFT of intra-frame sub-pulse dimension, and the high-precision velocity estimation and compensation are realized using the high-resolution image of multiple special point synthesis;Finally, ISAR imaging is realized through the FFT between frames.The application realizes the estimation and compensation of velocity gradually using two-step method, first realizes the estimation and compensation of large motion amount by rough estimation, then realizes the residual motion estimation and compensation of high precision by synthesis high resolution;Offset estimation is realized using envelope cross correlation, motion parameter estimation is realized using offset fitting, and the amount of calculation is small;The robustness of algorithm is improved using the characteristics that multiple frequency points are not easily disturbed simultaneously, and the robustness of high-resolution velocity estimation is improved using the motion identity of multiple special points.

The invention relates to the technical field of unmanned aerial vehicle positioning, and discloses a method, a device and equipment for estimating motion parameters of a far-field intrusion unmanned aerial vehicle, and a medium, which are used for solving the technical problem that the motion parameter estimation accuracy of the far-field intrusion unmanned aerial vehicle is relatively low. The method comprises the following steps: when an invading unmanned aerial vehicle invades from a far field, acquiring first position information of the invading unmanned aerial vehicle through a plurality of preset anchor sensors; estimating the direction of arrival of the invading unmanned aerial vehicle according to the first position information; arranging a plurality of mobile sensors in a preset monitoring countering area according to the direction of arrival; acquiring second position information of the invading unmanned aerial vehicle through a mobile sensor; and determining motion parameters of the invading unmanned aerial vehicle according to the second position information.

The application discloses a kind of radar high-speed range expansion target's joint phase accumulation and detection method, to fully accumulate and utilize the radar echo energy of the purpose of the detected target, first, phase accumulation algorithm is combined with target detection model, constructs new detection statistics, then based on sparse representation theory, the optimization problem of the adaptive estimation of distance expansion target scattering point of different sizes is established and solved, finally the scattering point information obtained is applied to the detection model established, while realizing distance walk correction, radar echo energy multi-pulse phase accumulation, and the non-phase accumulation of the echo energy between each scattering point.The method of the application fully accumulates the energy between radar echo pulses and within pulse, effectively improves the detection probability and motion parameter estimation accuracy of high-speed range expansion target with unknown size, is suitable for air monitoring, traffic monitoring and other fields, and has wide application prospect.

The invention discloses a space target rotation motion parameter estimation method and device based on multi-modal feature fusion. The method comprises the steps of obtaining a to-be-processed image sequence and an internal high-frequency sight vector sequence corresponding to each frame of image in the to-be-processed image sequence; processing by adopting the trained residual network and a first layer in the trained double-layer long short-term memory network to obtain a visual feature vector after each frame of image is fused and an intra-frame motion feature vector of each frame of image, projecting the visual feature vector and the intra-frame motion feature vector to the same dimension space, and then splicing to obtain a visual feature vector and an intra-frame motion feature vector; obtaining a multi-modal feature vector fusing visual information and motion information of each frame of image; and performing macroscopic motion time sequence modeling on the multi-modal feature vector corresponding to the image sequence to be processed by adopting a second layer in the trained double-layer long-short-term memory network to obtain an aggregation total feature, and performing regression processing on the aggregation total feature to obtain an estimated rotation motion vector. According to the invention, the accuracy of motion parameter estimation can be improved.

The invention relates to a method and device for estimating motion parameters of a moving target, electronic equipment and a medium, and belongs to the technical field of wireless communication signal processing.The method comprises the steps that an MISO system receiving signal model between the moving target and an antenna array is built; the method comprises the following steps: decomposing an MISO system receiving signal model into a plurality of factor nodes, constructing a factor graph based on the plurality of factor nodes, and dividing the factor graph into a plurality of function modules; and performing parameter estimation on a functional module on the factor graph by applying a hybrid message passing algorithm combined with a BP-EP-MF rule to obtain the estimated distance, angle, radial speed and tangential speed from the moving target to the antenna array. According to the method, the MISO system receiving signal model is decomposed into a plurality of factors for mixed message transmission, so that the calculation complexity of the MISO system receiving signal model is reduced.

The application discloses a target motion parameter estimation method and system based on a bistatic Doppler time-frequency curve; a plane rectangular coordinate system is established with a receiver position as an origin, a transmitter position is a known fixed coordinate, and a target is assumed to be a uniform linear motion; according to analytical expressions of distances between the transmitter and the target and between the receiver and the target, a bistatic equivalent path length is constructed, and an analytical expression of Doppler variation with time is derived based on a Doppler physical model; a target Doppler time-frequency curve is sampled at multiple time points to obtain a Doppler measurement vector, and outputs of the analytical model at corresponding time points are combined to form a model vector, so that an observation equation with target motion parameters as unknown quantities is formed; and a nonlinear least square algorithm is used to estimate the target motion parameters. The application has the advantages of not needing distance measurement, complete theoretical basis, high estimation precision and easy engineering implementation, and can be used for multi-station radar target trajectory matching and information fusion.

The application discloses a moving target CPA through feature detection method based on Doppler-Warping transformation and belongs to the technical field of underwater sound. The application combines the high gain characteristic of Doppler-Warping transformation and the real-time processing capability of a time sliding window, and proposes a moving target CPA through feature detection method. In each time window, firstly, a target motion model is established according to three motion parameters of a CPA time, a CPA distance and a target motion speed, and a Doppler-Warping operator corresponding to different parameter combinations is constructed; then, the time domain signal in the window is subjected to Doppler-Warping transformation by using the operator; finally, the optimal parameters are determined by searching for the minimum value of a spectrum entropy, and whether the line spectrum belongs to the moving target passing through the CPA is judged according to whether the signal-to-noise ratio gain before and after the transformation exceeds a set threshold value. The application can effectively extract the CPA through feature of the moving target and meet the requirements of target detection and motion parameter estimation in engineering.

The application discloses an ISAR phase correlation translational motion compensation and imaging method based on a motion parameter estimation network, relates to the field of imaging technology, and comprises the following steps: receiving echo of a spatial target inverse synthetic aperture radar; pre-processing the echo to obtain a compensated echo sequence; processing the compensated echo sequence by using a trained motion parameter estimation network model to obtain a distance grid offset coarse estimation value of a target center position of each echo; multiplying the distance grid offset coarse estimation value by a distance resolution unit to obtain a distance coarse estimation value, fitting the distance coarse estimation value to obtain a translational motion parameter coarse estimation value, taking the translational motion parameter coarse estimation value as an optimized imaging parameter of a preset imaging model, iteratively optimizing the translational motion parameter coarse estimation value to obtain an optimized imaging parameter; obtaining translational motion error terms of each echo according to the optimized imaging parameter; compensating the translational motion error terms of each echo to obtain an imaging result by coherently accumulating the compensated echoes.