522 results about "Quaternion" patented technology

The invention provides a brushless motor attitude detection and balance control method and system based on a gyroscope, and relates to the technical field of control, and the method comprises the steps: collecting angular velocity and acceleration data through a six-axis gyroscope, carrying out the noise reduction through wavelet transform, and carrying out the attitude calculation through the combination of an extended Kalman filter and a quaternion algorithm. A rotor position signal is obtained through a magnetic encoder, nonlinear compensation is carried out, and rotating speed data are calculated. A motor state is modeled by adopting a long-short-term memory network, a double-layer adaptive fuzzy neural network controller is constructed, and attitude error compensation and rotation speed fluctuation suppression are realized. A controller model is optimized through particle swarm optimization and a genetic algorithm, a compensation current vector is corrected in real time, and the control precision and stability of the brushless motor are improved. According to the method, the operation efficiency and the dynamic response capability of the brushless motor are effectively improved.

The invention provides a deep learning-based rapid ore sorting method under adaptive illumination intensity, and relates to the technical field of ore sorting, the deep learning-based rapid ore sorting method comprises the following steps: obtaining illumination intensity distribution information and ore three-dimensional point cloud data, and calculating ore space orientation parameters based on quaternion attitude estimation; inputting the obtained data into a layered illumination compensation network, and obtaining an illumination normalized enhanced image through adaptive brightness mapping and local light spot correction in combination with reflectivity recovery and non-local mean filtering; inputting the illumination normalized enhanced image and the ore three-dimensional point cloud data into a multi-branch feature extraction network, constructing feature high-order association, and adopting adaptive feature reconstruction and conditional random field optimization features; and inputting the reconstructed features into a classification model based on active learning, optimizing a feature manifold space by using a self-organizing mapping network, carrying out sample labeling in combination with active sampling of adaptive entropy threshold screening and uncertainty perception, and realizing ore classification and sorting through random forest integration.

The invention discloses a vehicle positioning method and system based on UWB and vehicle constraint, and relates to the field of vehicle positioning, and the method comprises the steps: deploying a UWB reference station in a well mining environment, and enabling a vehicle to be provided with a UWB tag; obtaining distance measurement data of at least two UWB reference stations, and detecting and compensating the distance measurement data to obtain a vehicle UWB position; collecting vehicle-mounted IMU (Inertial Measurement Unit) data, and obtaining a vehicle attitude quaternion through attitude calculation; collecting vehicle wheel speed meter data, and correcting the vehicle wheel speed meter data according to vehicle motion constraint conditions; according to the vehicle UWB position, the vehicle attitude quaternion and the corrected vehicle wheel speed meter data, the vehicle position is obtained through an adaptive Kalman filtering fusion algorithm. Aiming at low positioning precision caused by easy accumulation of errors and drifting of a traditional inertial navigation system in a well mine closed space in the prior art, the invention reduces the influence of a multipath effect in the traditional inertial navigation system on the positioning precision.

The invention relates to the technical field of visual identification, and discloses a belt tearing detection method, device and equipment based on visual identification, and the method comprises the steps: carrying out the dual-channel image collection and preprocessing of the surface of a conveying belt, and obtaining a multi-channel preprocessing image; extracting a thermal difference feature and a texture structure feature of the multi-channel preprocessed image through a double-flow feature network, and inputting the thermal difference feature and the texture structure feature into a quaternion material deformation analysis model for deformation gradient tensor and invariant parameter calculation to obtain a tear feature description vector; performing hierarchical progressive identification and three-dimensional reconstruction analysis to obtain target tearing feature data; and risk assessment is carried out based on the target tear feature data to obtain a tear grade classification result, the interference of ambient temperature drift and non-uniform illumination is effectively eliminated, the calculation efficiency and accuracy of tear detection are improved, and false alarms and missing alarms are reduced.

The invention discloses a neural network inertial attitude estimation method fusing physical information constraints. The method comprises the following steps: acquiring IMU data and attitude quaternion values at corresponding moments; performing windowing processing on the IMU data, and screening an attitude quaternion value as an attitude quaternion pseudo-true value based on physical consistency; training the constructed neural network prediction model; optimizing the neural network prediction model through a model loss function, wherein the model loss function comprises a physical constraint loss function and a supervision loss function; and during target carrier attitude estimation, inputting IMU data of a corresponding window into the trained neural network prediction model, and calculating to obtain a target carrier attitude quaternion estimated value of the corresponding window. According to the method, the generalization ability of the neural network is enhanced through physical information constraint, the precision and robustness of inertial attitude estimation are improved, and the method is particularly suitable for inertial navigation application in a high-dynamic environment.

The invention belongs to the field of robot control, and particularly relates to a robot self-adaptive impeller welding control method and system fused with multi-modal data, and the method comprises the steps: obtaining an impeller workpiece image and three-dimensional point cloud data through an industrial vision subsystem, and extracting a three-dimensional welding seam position point set representing a welding seam track; generating a candidate welding path set based on welding seam geometric feature analysis; in combination with welding seam morphology and process requirements, optimal welding process parameters are predicted through a regression model, and a welding gun attitude angle is calculated through quaternion conversion according to a curved surface normal vector; in the welding process, the relative position deviation of a robot and a workpiece and the parameter deviation of welding current / voltage and a planned value are monitored in real time, when the deviation exceeds a preset tolerance interval, a PID control algorithm is adopted to dynamically adjust robot motion parameters or welding process parameters, and closed-loop control over the welding process is achieved; according to the method, precise planning of the welding path and self-adaptive regulation and control of the process are achieved, and the welding quality and the forming consistency are remarkably improved.

The invention discloses a wind power generation tower pose monitoring method based on a GNSS / MEMS IMU / magnetometer combination, and the method comprises the steps: carrying out the estimation of a state vector containing the position, speed, attitude error, sensor zero offset and scale factor error through error state Kalman filtering based on the GNSS / MEMS IMU combination, and carrying out the correction and fusion of GNSS and IMU data through a lever arm effect; a complementary filtering algorithm is adopted, an error vector is constructed by combining measured values of an accelerometer and a magnetometer, a gyro angular velocity is corrected through a PI controller, an attitude quaternion is recurred, a course angle is calculated, and the course angle is used as an auxiliary observed quantity to update filtering. According to the invention, by adding course information provided by the MEMS IMU and the magnetometer, the position and the attitude are comprehensively estimated, and high-precision six-degree-of-freedom pose monitoring is realized.

The invention provides a robot teleoperation method and a robot. The robot teleoperation method comprises the steps that an acceleration command, an angular acceleration command and an expected attitude quaternion of an end effector are determined based on the current linear acceleration, the current angular acceleration and the current attitude quaternion of a control end; the current position and the current speed fed back by the end effector are obtained, a torque control signal for controlling the position of the end effector is determined based on the acceleration command, the current position and the current speed, and end acceleration feedforward is obtained; determining a torque control signal for controlling the attitude of the end effector based on the angular acceleration command and the expected attitude quaternion to obtain attitude feedforward; and a joint torque instruction of a mechanical arm joint corresponding to the end effector is determined based on the end acceleration feedforward and the posture feedforward, and the joint torque instruction is sent to a corresponding joint driver so that the end effector can synchronously act along with the control end. According to the invention, the problem of attitude lag of the end effector can be avoided, and the tracking error of the end effector is reduced.

The invention discloses a pose estimation method and device based on point pair feature matching, a medium and a product. The method comprises the following steps: performing coarse registration on a fusion scene point cloud and each template point cloud, and calculating point pair features; according to the corresponding hash key values, retrieving from each hash table to establish registration point pairs; calculating a translation vector and a rotation quaternion matched with each registration point pair according to the reference point coordinate system and the template coordinate system, and performing clustering processing on each registration point pair to obtain each initial corresponding set; screening out a target pose candidate transformation matrix and a matched target template point cloud from the pose candidate transformation matrixes matched with the initial corresponding sets respectively; a target optimization function between the fusion scene point cloud and the target template point cloud is constructed, iterative optimization is performed on parameters in the target pose candidate transformation matrix, and the target pose of the fusion scene point cloud is obtained at the end of iteration, so that the precision, real-time performance and reliability of six-dimensional pose estimation are improved.

The invention discloses a fusion filtering attitude estimation method based on EKF and Mahony, and the method comprises the steps: firstly obtaining the original data information of an IMU sensor at the current moment, carrying out the normalization of the attitude information of an accelerometer and a magnetometer, carrying out the error calculation, and compensating the error to the angular velocity of a gyroscope through a PI controller, the method comprises the following steps: performing Mahony filtering correction on an angular velocity and a corrected attitude quaternion to obtain an angular velocity subjected to Mahony filtering correction and a corrected attitude quaternion, constructing a state equation of an EKF by using the corrected attitude quaternion and a gyroscope zero offset value, and constructing an observation equation by using data of an accelerometer and a magnetometer so as to obtain an optimally estimated attitude quaternion and a gyroscope zero offset value; and finally, compensating the optimally estimated zero offset value of the gyroscope to the angular velocity correction process in the Mahony filtering process.

Provided in the present invention is a video stabilization method that deeply integrates optical flow and IMU data. The method comprises: acquiring and preprocessing data, computing optical flow, computing historical data of an original pose and a stable pose, predicting a stable camera pose, implementing video stabilization, and training a deep learning model, wherein computing the optical flow involves mixing the optical flow and translation of a camera between adjacent frames as parameters of a translational motion of the camera and inputting the parameters into a deep learning network; computing the historical data of the original pose and the stable pose and predicting the stable camera pose involve using an LSTM network to train and predict stabilization quaternion; and implementing video stabilization involves dividing a video frame into rectangular grids, transforming vertices of the grids by means of the quaternion output by the deep learning network, obtaining each pixel value in the video frame by means of performing quadratic interpolation on four vertices of the grid where the pixel value is located, and processing each pixel value to obtain a stabilized video frame. The method of the present invention can adapt to complex motion scenarios and provide a highly stable video output.

The invention relates to an all-humanoid robot motion redirection method and device based on motion capture equipment, and the method comprises the steps: converting a human body posture quaternion collected by the motion capture equipment into a rotation matrix, and achieving the precise mapping from a Cartesian space to a robot joint space in combination with an Euler angle decomposition strategy; a quaternion-disassembled joint angle is used as an initial solution, various physical constraints are introduced in real time according to the actual supporting state of the robot, and the lower limb joint angle is corrected online by constructing and solving an inverse kinematics problem with various constraints, so that the robot reproduces the action of an operator, and the joint angle of the lower limb can be corrected online. And the requirements of joint limiting, self-collision avoidance, dynamic balance and the like are strictly met, so that the accuracy of motion simulation and the stability of the system are effectively unified.

The invention discloses an LED spherical display screen-oriented geographic data dynamic adaptive display method, which comprises the following steps of: constructing a non-uniform grid mapping model based on LED spherical screen physical parameters (diameter, support structure cavity size and pixel row / column spacing); the method comprises the following steps: accurately mapping geographic data (converted into longitude and latitude coordinates in advance) from a longitude and latitude coordinate system to a spherical screen pixel space, and carrying out resampling adaptation on the data based on the resolution of a display; meanwhile, a quaternion rotation transformation algorithm and a dynamic mask technology are adopted, the shielding area of the supporting structure of the LED spherical screen is calculated in real time, invisible data are automatically cut in the three-dimensional rotation process, and the problems of coordinate mapping distortion, rotation tearing and shielding area adaptation delay existing in a traditional LED spherical display system are solved. According to the method, the sub-pixel-level mapping precision of the geographic data on the LED spherical screen and the self-adaptive shielding compensation response of real-time rotation are realized.

The invention discloses a ship dynamic collision avoidance decision-making method based on an improved DWA algorithm. The ship dynamic collision avoidance decision-making method comprises the steps of constructing sea chart environment data containing obstacle density through a grid method; establishing a dynamic elliptical ship domain based on a quaternion ship domain (QSD) model, and calculating a distance collision risk degree (NCRI) between ships based on the dynamic elliptical ship domain; improving a DWA algorithm evaluation function, and adding a path yaw evaluation function and a collision avoidance rule evaluation function on the basis of optimizing a speed and azimuth angle evaluation function; fusing the multiple evaluation functions by adopting a weight distribution method; and finally, generating a local collision avoidance optimization route based on environment data, NCRI and an improved DWA algorithm. According to the method, safety boundary dynamic adjustment, rule compliance constraint and ship control characteristic cooperation are realized, and the robustness and practicability of collision avoidance decision making under complex sea conditions are remarkably improved.

The invention discloses a precise and synchronous collaborative data signal analysis method based on dual-channel optical fiber sensing, and relates to the field of collaborative data signal analysis, polarization compensation and nonlinear noise decoupling analysis are carried out on time domain optical signals collected by independent A and B sensing channels to obtain polarization dimension-increasing coherent collaborative signals, and the collaborative data signal analysis is carried out on the time domain optical signals. Performing quaternion mapping and transformation compensation analysis to obtain a pure polarization state of strain immunization, performing fractional differential, HOSVD decomposition and time varying Jacobian matrix propagation disturbance based on the pure polarization state of strain immunization to obtain a non-equilibrium entanglement feature vector, combining quantum Bayesian update and a KLD weight model, generating a particle set through state sampling, and obtaining a non-equilibrium entanglement feature vector. And performing directional transfer entropy calculation on the particle set, constructing a weighted Dirichlet clustering model and directional causal energy flow analysis, comparing dual-channel energy values to obtain a channel fault judgment result, and performing channel fault judgment timely and accurately.

The invention relates to the technical field of attitude determination of a photoelectric pod of an unmanned aerial vehicle, in particular to an MEMS-IMU dual-mode correction attitude determination method for the photoelectric pod of the unmanned aerial vehicle. Comprising the following steps: establishing a damping system second-order transfer function model, and outputting an attenuation coefficient matrix; based on MEMS-IMU three-axis motion parameters, distinguishing a large maneuver turning state and a linear flight state; respectively compensating a roll angle, a pitch angle and a course angle of the MEMS-IMU by using the main inertial navigation according to different states; converting the corrected attitude angle into a quaternion, and optimizing and outputting a fused quaternion through a gradient descent method; taking the fused quaternion as an input, constructing a seven-dimensional state space model, and outputting an error compensation amount through Kalman filtering; dynamically adjusting the main inertial navigation weight according to the accelerated speed root mean square; and outputting a final attitude angle in combination with the error compensation amount and the weight, and verifying the final attitude angle through GNSS (Global Navigation Satellite System) position inversion. The method has the advantages that cross interference is avoided; therefore, the method is suitable for the photoelectric pod scene of the unmanned aerial vehicle.

The invention provides a train attitude precision keeping method and system based on inertial quaternion modeling, and belongs to the technical field of train operation positioning. Through a quaternion modeling train attitude feedback compensation model, auxiliary calibration of a gyroscope is realized by using an accelerometer, and calculation of an inertial attitude is realized; converting the acquired wheel speed sensor information into three-dimensional speed under a navigation coordinate system through attitude information resolved by a quaternion modeling train attitude feedback compensation model, and further integrating to obtain a position; and fusing the speed and position information of the wheel speed sensor after attitude conversion with the speed and position information obtained by inertial navigation dead reckoning to realize error correction of the inertial navigation system. According to the method, the problem that the train state sensing precision is degraded due to the fact that a satellite navigation / inertial navigation system positioning model is difficult to overcome inertial device error accumulation when a satellite fails is solved, the train attitude precision is kept, and then error divergence of a combined positioning system during satellite signal denial is effectively restrained.

The invention relates to the technical field of inertial sensing attitude monitoring, and particularly discloses a body motion attitude monitoring method and system based on an inertial sensing technology, and the method comprises the steps: collecting the original data of a body motion attitude, outputting an attitude quaternion after fusion and an angular velocity vector after calibration, carrying out the dynamic attitude feature extraction, and carrying out the calibration. And outputting a body motion posture feature vector, carrying out stability margin modeling, obtaining a stability margin scalar, determining a tumble probability update value, carrying out tumble evolution prediction, combining the tumble probability update value, outputting a tumble probability and a probability change trend, obtaining a tumble early warning level, analyzing adaptive calibration parameters and carrying out feedback. The problems that in the prior art, multi-sensor data cannot be effectively integrated, so that the information utilization efficiency is low, and the potential of multi-modal data fusion is not fully exerted are solved, and the monitoring accuracy can be continuously optimized.

The invention discloses a learning-based composite layered anti-interference control system and method suitable for a large-windward-area unmanned aerial vehicle, is used for realizing robust wind resistance control of the large-windward-area unmanned aerial vehicle, and aims at the flight characteristics of large windward area and sensitivity to external wind field change in a vertical take-off and landing stage. A wind speed estimation method without an additional sensor is provided, modeling of external disturbing force is realized in combination with a Gaussian process regression algorithm, and compensation of the disturbing force is realized by using quaternion-based model prediction control, so that the influence of an external change wind field on the dynamics of the large-windward-area unmanned aerial vehicle is reduced, and the trajectory tracking precision is remarkably improved. A backstepping controller based on SO (3) is designed in an attitude control loop, a nonlinear disturbance observer is integrated, and active compensation of external disturbance torque and robust tracking control of attitude are realized. Based on the technical characteristics, a complete dynamic control link capable of estimating and feeding back compensation disturbance in real time is further constructed.

The invention relates to the technical field of monitoring and correction, in particular to a ceilometer attitude correction method based on a buoy. The method comprises the following steps: step 1, collecting attitude data of a buoy for multiple times, and constructing a reference attitude quaternion according to the attitude data of the buoy collected each time; 2, extracting a wave dominant frequency component and a high-frequency residual component from a reference attitude quaternion sequence formed by all reference attitude quaternions by using discrete Fourier transform, and calculating a comprehensive attitude quaternion; 3, calculating a buoy wave disturbance variable by using wave spectrum analysis based on sea condition data obtained through real-time monitoring; and 4, performing attitude correction on the measured value of the ceilometer based on the buoy wave disturbance variable and the comprehensive attitude quaternion. According to the method, high-precision estimation of the attitude of the buoy under sea surface multi-band wave disturbance is realized, and real-time sea condition analysis and ceilometer optical axis correction are combined.

The invention provides a planning control method and device for a space target pointing attitude by a remote sensing satellite, and the method comprises the steps: obtaining an attitude quaternion from a satellite body system of the remote sensing satellite to a load coordinate system, acquiring a satellite orbit state and a space target orbit state at the current moment; determining the intersection moment of the remote sensing satellite and the space target through an orbit pre-pushing model; and determining the shooting time of a linear array camera based on the intersection time of the remote sensing satellite and the space target, and obtaining the space unit vector of three axes of a satellite load coordinate system corresponding to the expected attitude of the remote sensing satellite at the shooting time. According to the invention, the imaging of the line-scan digital camera is more accurate, the control precision of the pointing attitude of the remote sensing satellite is improved, and the problem of poor control effect of the remote sensing satellite in the prior art is solved.

Provided is a generalization generation method of industrial robot pose trajectories supporting changeable operation path points, which relates to the field of robot trajectory planning. The method includes: acquiring a plurality of groups of robot end pose trajectories; aligning time steps of the pose trajectories via a multi-dimensional dynamic time warping algorithm; constructing a Gaussian mixture model of robot pose trajectories in combination with a variational Bayesian method and an attitude quaternion tangent space mapping method; calculating reference pose trajectory distribution via a Gaussian mixture regression method; performing kernelized representation on pose trajectory distribution, and solving an optimal hyperparameter of a kernel function by minimizing a root mean square error of a reproduced reference trajectory; and generating the robot end pose trajectories adapted to the operation path points by updating the reference pose trajectory distribution.

The invention relates to the technical field of attitude control, in particular to a moving object attitude control method based on an improved LKF. Predicting prior attitude estimation at the next moment; calculating a process error covariance matrix; calculating a measurement attitude; calculating a quaternion measurement error covariance matrix; constructing feature vectors of acceleration interference and geomagnetic interference amplitude levels, and calculating an interference amplitude level value by using a Gaussian function model; updating the process error covariance matrix and the measurement error covariance matrix by using the interference amplitude level value; calculating a priori attitude estimation error covariance matrix by using the updated process error covariance matrix; calculating a Kalman gain matrix by using the priori attitude estimation error covariance matrix and the updated measurement error covariance matrix; calculating attitude posteriori estimation at the next moment; and performing attitude estimation at the next moment. According to the method, the problems of insufficient estimation precision and poor interference suppression capability of the attitude observation error covariance matrix of the existing LKF are solved.

The invention discloses a target tracking method and device based on a rolling and pitching type holder, and the method comprises the steps: building a polynomial response model through adaptive color calibration, and achieving the dynamic color correction; an improved YOLOv8 model is combined with a bidirectional feature pyramid network to carry out multi-scale target detection, and template matching is assisted to improve a low-confidence-coefficient scene recognition rate; predicting the position and the speed of the target by using Kalman filtering to cope with a shielding condition; and high-precision smooth rotation of the holder is realized through quaternion attitude solution in combination with PID (Proportion Integration Differentiation) and FOC (Fiber Operating Control) double-loop cooperative control. The device adopts a rolling and pitching type holder structure and comprises a main controller, an angle sensor, a visual sensor and a magnetic encoder, the mechanical structure is simplified, the weight is reduced, and higher response speed and higher control precision are achieved. The target tracking accuracy, stability and anti-interference capability of the intelligent mobile platform in a complex environment are remarkably improved, and the method is suitable for various dynamic scenes such as aerial photography and industrial inspection.

The invention discloses a radiation correction method and system for an airborne hyperspectral imager. The radiation correction method and system are used for improving the radiation correction precision in complex weather. The method comprises the following steps: calculating a theoretical sun position based on unmanned aerial vehicle initial coordinates, ephemeris data and atmospheric parameters; performing course and differential compensation on a solar azimuth angle by combining IMU course and wind speed and direction data, performing refraction correction on an elevation angle, and outputting a high-precision sun position parameter; hyperspectral images and irradiance data are synchronously acquired, quaternion interpolation is combined with IMU data to realize space-time alignment, and attitude disturbance is eliminated; a hemispherical space irradiance model is constructed based on alignment data, cloud layer modes (clear sky, thin cloud and thick cloud) are divided according to atmospheric optical thickness variance, and Lambert body proportion correction, a discrete longitudinal standard method or a deep convolutional spectral network are respectively adopted for radiation correction. The method remarkably improves the correction precision and generalization ability, supports large-area image splicing, and is suitable for quantitative application of geological exploration, environment monitoring and the like.

The present invention relates to the technical field of seismic exploration. Provided are a three-component seismic data vector anti-aliasing interpolation reconstruction method and an apparatus. The method comprises: representing three-component seismic data in a time domain as pure quaternions for vector joint; transforming the pure quaternion data in a time-space domain to a frequency-space domain to obtain frequency-space domain data; transforming the frequency-space domain data into a frequency-wave number domain; in the frequency-wave number domain, using an inclination angle scanning strategy to construct a mask matrix; performing vector anti-aliasing interpolation reconstruction on the frequency-space domain data and the mask matrix, so as to obtain a frequency-space domain reconstruction result; and inversely transforming the frequency-space domain reconstruction result to the time-space domain to obtain a time domain three-component three-dimensional data anti-aliasing reconstruction result. The present invention can effectively solve the problem of vector anti-aliasing reconstruction for three-component seismic data that contain both irregular and regular missing traces, thus improving the accuracy of effective wave event identification.

The invention relates to the technical field of performance testing, and discloses a performance testing method, device and equipment for a graphics card, and the method comprises the steps: carrying out the quaternion stage decoupling of a rendering pipeline of the graphics card, and obtaining a vertex shader decoupling unit, a geometric shader decoupling unit, a rasterization unit decoupling unit and a pixel shader decoupling unit; stage isolation performance tests are respectively carried out to obtain first performance test data of the vertex shader, second performance test data of the geometric shader, third performance test data of the rasterization unit and fourth performance test data of the pixel shader; the first performance test data, the second performance test data, the third performance test data and the fourth performance test data are comprehensively analyzed to obtain performance bottleneck positioning results of all stages of the rendering pipeline, the coupling states of all the stages of the rendering pipeline can be effectively separated, and accurate adjustment and control over the running states of all the stages are achieved.

The invention discloses a point cloud reconstruction method and system based on perception and dynamic initialization, and relates to the technical field of three-dimensional reconstruction, and the method comprises the steps: extracting point cloud data with position and color information from a multi-view image, converting the point cloud data into a CUDA tensor, and carrying out the processing, thereby achieving the dynamic initialization of 3D Gaussian distribution and parameters; a spherical harmonic function is adopted to process color features, an adjacent point average distance is adopted to determine initial Gaussian scaling, a quaternion is adopted to initialize a rotation angle, and an inverse sigmoid is adopted to set an initial opacity value, so that the initialization of a 3D model is optimized; a rendered image is generated by using a 3D-to-2D sputtering technology, and 3D Gaussian distribution parameters are optimized through pixel-level L1 loss, local structure-level SSIM loss and perceptual loss LPIPS. According to the method, 3D Gaussian distribution is dynamically initialized, and perception loss optimization is introduced, so that the problem of time overhead of a 3DGS method in an iteration process is solved, and the problem that the 3DGS method is easy to have a fuzzy detail or distortion phenomenon in scenes such as complex textures and high-frequency regions is solved.

The invention discloses a manned aircraft / unmanned aerial vehicle pursuit deep learning control method simulating a pigeon positive and negative feedback mechanism. The method comprises the following steps: step 1, giving mathematical models of a manned aircraft and an unmanned aerial vehicle and performing environment initialization; step 2, building a quaternion PID controller of the unmanned aerial vehicle; 3, a manned aerial vehicle / unmanned aerial vehicle hierarchical decision-making architecture is established; step 4, mathematical modeling of a pigeon positive and negative feedback mechanism; 5, designing a manned aerial vehicle / unmanned aerial vehicle cluster reward function simulating positive and negative feedback of the pigeon flock; 6, planning and designing an escape unmanned aerial vehicle path; step 7, predicting an escape unmanned aerial vehicle path; and 8, generating a manned aerial vehicle / unmanned aerial vehicle cluster pursuit learning strategy based on the actor-commentator network. According to the invention, an efficient, flexible and extensible bionic intelligent pursuit control framework is constructed, and the pursuit task execution efficiency and cooperative combat capability of the manned aerial vehicle / unmanned aerial vehicle heterogeneous cluster in a complex dynamic environment are improved.

The invention relates to the field of array signal processing, and particularly discloses a DOA joint estimation method based on a quaternion polarization sensitive array. Electromagnetic wave dual polarization components are captured through the orthogonal dipole and the loop antenna, and a quaternion observation matrix is constructed; calculating a quaternion covariance matrix by adopting a sliding window mechanism, and separating a signal / noise subspace by adopting quaternion singular value decomposition; and constructing a spatial spectrum function in combination with a quaternion steering vector, and realizing joint estimation of an azimuth angle and a pitch angle through two-dimensional search. According to the method, the unified characterization capability of quaternions on polarization-airspace information is fully utilized, and the DOA estimation precision and the anti-interference performance of the multi-polarization signal are remarkably improved.