Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

67 results about "Maximum a posteriori estimator" patented technology

A maximum a posteriori estimation approach is used to evaluate the optimal values for the estimates of the parameters.

Underwater robot autonomous navigation method based on energy consumption perception

The invention discloses an autonomous navigation method for an underwater robot based on energy consumption perception, which comprises the following steps of: acquiring underwater environment parameters and robot state parameters by fusing multi-sensor data, constructing a factor graph model containing pose nodes and energy consumption nodes by adopting a graph-based SLAM (Simultaneous Localization and Mapping) algorithm; a local area environment energy consumption map is obtained by fusing maximum posteriori estimation of poses and energy consumption with robot state parameters, the local energy consumption map is updated in real time according to extension of a robot exploration area, if the current local energy consumption map covers a global target position, a global energy consumption optimal path is calculated according to the path length and energy consumption, and the optimal energy consumption is calculated according to the path length and the energy consumption. And if the current local energy consumption map does not cover the global target position, triggering local path re-planning, generating an energy consumption compensation strategy, and if the current path is invalid due to major change of the environment, re-executing global planning.
Owner:GUANGZHOU MARITIME INST

Underwater vision-inertia-acoustic odometer positioning method fused with DVL

The invention discloses an underwater vision-inertia-acoustic odometer positioning method fused with a DVL, and belongs to the technical field of robot navigation positioning. According to the method, firstly, an underwater image acquired by a camera and carrier motion data acquired by an IMU and a DVL are used as input information to carry out time alignment and space coordinate system conversion; performing feature point extraction on the image to construct a visual re-projection residual error, and dynamically adjusting a weight ratio of visual information based on the number of feature points and the re-projection residual error; calculating IMU data increment between adjacent frames based on pre-integration to construct an inertial residual error; calculating the displacement change between image frames according to the speed information measured by the DVL to construct a DVL residual error; establishing a nonlinear least square optimization function based on Bayesian maximum posteriori estimation; vision, inertia and acoustic speed measurement data are fused at the rear end through a sliding window mechanism, a multi-constraint joint optimization model is constructed, and the robustness and accuracy of a vision-inertia-acoustic odometer system in the underwater environment are improved.
Owner:HOHAI UNIV

Low-dose CT reconstruction method and system based on global optimization iteration deep learning

The invention discloses a low-dose CT reconstruction method and system based on global optimization iteration deep learning, and the method comprises the steps: collecting CT cross section data of a normal dose, and generating low-dose CT data; the method comprises the following steps: establishing a dual-domain analysis compression iteration model, constructing maximum posteriori estimation based on a CT projection chordal graph data composite Poisson noise generation mechanism, subdividing CT projection data into a projection domain and a chordal graph domain, sequentially updating the projection data, the chordal graph data and image data, and constructing an iteration algorithm to realize CT global optimization modeling. Expanding an iterative algorithm into a trained reconstruction network; inputting the paired data into the expanded network for training, and storing a model with the minimum output result loss; a sub-network and an attention mechanism are added for CT data features, jump connection weights are finely adjusted, middle layer neural network parameters are finely adjusted and updated through middle supervision, and a global optimization iteration deep learning model NGACI-Net is obtained. According to the method, the problems that the low-dose CT image quality is poor, the reconstruction efficiency is low, and generalization and interpretability are lacked are solved.
Owner:XI AN JIAOTONG UNIV

Multi-source rainfall prediction method and device integrating GNSS and communication satellite

The invention provides a multi-source rainfall prediction method and device fusing GNSS and communication satellites, and the method comprises the steps: building a plurality of satellite-ground microwave links which are used for transmitting GNSS data and communication satellite data; gNSS data and communication satellite data on the multiple satellite-ground microwave links are acquired; performing atmospheric rainfall inversion on the plurality of satellite-ground microwave links based on GNSS data, and constructing a two-dimensional water vapor field according to an inversion result; performing rainfall intensity inversion on the plurality of satellite-ground microwave links based on communication satellite data, and constructing a two-dimensional rainfall field according to an inversion result; and fusing the two-dimensional water vapor field and the two-dimensional rainfall field by adopting Bayesian-maximum posteriori estimation to obtain a fusion field, and converting the fusion field to obtain a rainfall prediction field. According to the method, the rainfall field and the water vapor field are fused for rainfall prediction, so that the whole rainfall process can be accurately predicted.
Owner:WUHAN MEASURING FUTURE TECH CO LTD

Vehicle state monitoring method and system based on deep learning

The invention relates to the technical field of vehicle intelligent management, and discloses a vehicle state monitoring method and system based on deep learning, and the method comprises the steps: analyzing the physical dependence relation among vibration abnormality detection, loosening positioning and fatigue evaluation monitoring tasks based on structural mechanics constraints, and carrying out the monitoring of the vibration abnormality detection, the loosening positioning and the fatigue evaluation; generating a task dependence directed graph; inputting the multi-modal monitoring data matrix into a shared encoder of a multi-task learning network to generate a unified representation vector; respectively inputting the unified representation vector into a vibration anomaly detection decoder, a loose positioning decoder and a fatigue evaluation decoder, and outputting an initial prediction result of each task; and establishing a conditional probability model between task outputs based on the constraint relationship defined by the task dependent directed graph, and generating a consistency diagnosis result meeting physical constraints through maximum posteriori estimation. The technical problems of one-sided diagnosis results, mutual contradiction and low calculation efficiency are solved.
Owner:吉林明瑞科技有限公司

Underground coal mine goaf gap space sensing method based on multiple detection technology

The invention discloses an underground coal mine goaf gap space sensing method based on a multiple detection technology, and the method comprises the steps: obtaining a three-dimensional dielectric constant body through a ground penetrating radar, and determining the space position of a goaf gap; resistivity tomography is carried out through the optimally arranged electrode network, and a three-dimensional resistivity body is inverted; constructing a three-dimensional wave velocity body by adopting sound wave detection, and identifying a velocity low-value region; a dielectric constant body, a resistivity body and a wave velocity body obtained by the three technologies are uniformly registered to a cubic grid with a preset resolution and normalized, a caving zone void probability body is constructed by adopting adaptive weighted fusion based on information entropy and a Bayesian maximum posteriori estimation framework, and finally a three-dimensional void model is generated and the total volume is calculated. Through multi-source data fusion of ground penetrating radar, resistivity tomography and sound wave detection, the limitation of a single technology in goaf space detection is solved, the precision of gap detection is improved, and the method is suitable for goaf space perception and treatment.
Owner:中煤能源研究院有限责任公司 +1

Signal processing method and device of wind measurement laser radar

The invention discloses a signal processing method and device for a wind measurement laser radar, and the method comprises the steps: carrying out the windowing interception of an original laser echo signal, and obtaining an intercepted signal; performing dual-tree complex wavelet decomposition transformation on the intercepted signal to obtain each level of approximation wavelet and each level of detail wavelet; performing correlation operation and soft threshold filtering on each level of detail wavelet to obtain a corrected detail wavelet; combining the highest-level approximation wavelet with the corrected detail wavelet to obtain a preliminary noise reduction signal; performing Bayesian maximum posteriori estimation filtering on the preliminary noise reduction signal to obtain a final noise reduction signal; performing dual-tree complex wavelet inverse transformation on the final noise reduction signal to obtain a noise reduction time domain signal; performing fast Fourier transform on the denoised time domain signal to obtain a frequency domain signal; and performing frequency spectrum correction on the frequency domain signal, and then extracting to obtain Doppler frequency. According to the invention, noise filtering of the signals of the wind measurement laser radar is realized, and the Doppler frequency extraction accuracy is improved.
Owner:HUBEI JIUZHIYANG INFRARED SYST CO LTD

Sparse Aperture ISAR Maneuvering Target Imaging and Calibration Method

The present invention discloses a sparse aperture ISAR maneuvering target imaging and calibration method, comprising the following steps: establishing a maneuvering target ISAR sparse aperture signal model, characterizing the maneuverability of the target by constructing a variable scale non-uniform Fourier two-dimensional dictionary containing MTRC; constructing a parameterized sparse imaging model by maximum a posteriori estimation, expressed as a cost function of joint target parameter estimation and sparse imaging; finding the target rotation parameter by the minimum entropy method, and implementing a rough estimation of the maneuvering target rotation parameter by performing parameter iteration by the coordinate descent method; and implementing high-quality imaging with MTRC correction and accurate parameter estimation by combining sparse ISAR imaging and target rotation parameter estimation, decomposing into sparse imaging and dictionary learning, and iterating and solving in sequence. The present invention realizes the use of sparse aperture echo data to solve the problem of high-resolution imaging of ISAR maneuvering targets, greatly increasing the implementation scope and application scenarios of radar imaging.
Owner:SOUTHEAST UNIV

Method and device for determining model parameters for a control strategy for a technical system with the aid of a Bayesian optimization method

Methods for ascertaining a control strategy for a technical system using a Bayesian optimization method. The control strategy is created based on model parameters of a control model and is executable. The method includes providing a quality function whose shape corresponds to a regression function and that evaluates a quality of a controlling of the technical system based on model parameters; carrying out a Bayesian optimization method based on the quality function in order to iteratively ascertain a model parameter set having model parameters within a model parameter domain that indicates the permissible value ranges for the model parameters; and determining the model parameter domain for at least one of the model parameters as a function of an associated maximum a posteriori estimated value of the quality function.
Owner:ROBERT BOSCH GMBH

Ultrasonic cross-metal channel estimation method based on improved block sparse Bayesian learning

The invention discloses an ultrasonic cross-metal channel estimation method based on improved block sparse Bayesian learning, and belongs to the field of ultrasonic cross-metal communication channel estimation. Constructing a pilot signal relationship between the receiving end and the transmitting end on the two sides of the metal wall; dividing a to-be-estimated metal channel impulse response into zero-value blocks or non-zero blocks of block sparse Bayesian according to whether the to-be-estimated metal channel impulse response is zero or not; the method comprises the following steps: iteratively updating hyper-parameters a, b, Bi and lambda of a non-zero block by adopting an expectation maximization algorithm, realizing structured sparse learning, calculating a maximum posteriori estimation result by utilizing a final hyper-parameter of the non-zero block, and taking the maximum posteriori estimation result as an estimation value of ultrasonic cross-metal channel pulse response. According to the method, a block sparse Bayesian learning algorithm is improved by utilizing the characteristics of metal channel pulse response, and compared with an original algorithm, the accuracy of channel estimation in a low signal-to-noise ratio environment is improved.
Owner:CHINA UNIV OF MINING & TECH

Visual inertia pressure fusion underwater pose estimation method based on bidirectional pre-integration

The invention discloses a visual inertial pressure fusion underwater pose estimation method based on bidirectional pre-integration, and the method comprises the steps: employing the obtained visual observation constraint of each key frame; the optimization problem constructed by the IMU pre-integration observation constraint and the pressure gauge pre-integration constraint can efficiently and accurately realize the scale of the system state parameters, the speed of the key frame, the zero offset of the acceleration and the initial value of the distance between the pressure gauge and the sea level, realize the initialization of the system state parameters, facilitate the subsequent parameter updating, and improve the system performance. According to the method, the initial parameter is taken as a starting point, the parameter is updated through the maximum posteriori estimation problem constructed by the IMU pre-integration residual error, the pressure gauge bidirectional pre-integration residual error and the visual reprojection residual error, and the underwater pose estimation can be accurately realized.
Owner:ZHEJIANG UNIV

Self-adaptive sparse foresight super-resolution imaging method based on echo restoration strategy

The invention discloses a self-adaptive sparse foresight super-resolution imaging method based on an echo restoration strategy, which comprises the following steps: firstly, establishing a scanning radar super-resolution model, then, based on a regularization theory, using an L1 norm as a constraint term to improve the angular resolution of a sparse target, and then, using a multi-scale representation characteristic of wavelet variation to separate signals and noise in echoes, so as to realize the self-adaptive sparse foresight super-resolution imaging. The problem that an existing L1-IRN method is poor in performance under the low signal-to-noise ratio is solved while the echo signal-to-noise ratio is improved, a new optimization cost function is constructed based on preprocessed echoes, finally, a sparse estimation problem is converted into a maximum posteriori estimation problem based on the Bayesian theory, a self-adaptive iteration weight is obtained, and foresight super-resolution imaging is achieved. The echo repairing strategy of the method can weaken the influence of noise on super-resolution imaging, adaptive sparse foresight super-resolution imaging is realized under the condition of extremely low signal-to-noise ratio, a high-quality imaging result is obtained, meanwhile, the efficiency of a foresight two-dimensional super-resolution imaging algorithm is improved, and high robustness is achieved.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA

Tide-corrected coastline datum plane determination method

The invention provides a coastline reference plane determination method based on tide correction, and belongs to the technical field of coastline detection.The method comprises the steps that a high-resolution satellite image and airborne laser radar data are collected and input into a sea-land boundary recognition model constructed by a neural architecture growth algorithm based on attention guidance, and a refined instantaneous coastline is extracted; a coastal tide numerical model containing a shallow moisture tide term and a composite partial tide term is established, a coastal refined tide forecast tide level sequence is obtained through a variational data assimilation optimal control algorithm, and a refined instantaneous coastline is associated with a corresponding forecast tide level to establish a spatial distribution function; the coastline position corresponding to the lowest tide level of the multi-temporal remote sensing data inversion theory is integrated through the Bayesian maximum posteriori estimation method to serve as the coastline reference plane, and the technical problem that the determination precision of the coastline reference plane is affected by dynamic changes of tides and water density differences, and consequently spatial positioning is inaccurate is solved.
Owner:自然资源部天津海洋中心(自然资源部天津海洋预报台)

Zero sample transfer learning-based structural damage identification Bayesian model rapid solving method

The invention discloses a structure damage identification Bayesian model rapid solving method based on zero sample transfer learning, and belongs to the technical field of structure damage identification and machine learning. The problem that the generalization performance of a structural damage identification Bayesian model is reduced is solved. The method comprises the following steps: constructing a Bayesian model for structural damage identification; constructing and training an adversarial sparse auto-encoder in combination with the sparse auto-encoder and adversarial learning, wherein the adversarial sparse auto-encoder comprises a source domain sparse auto-encoder module, a target domain sparse auto-encoder module and a domain discriminator module; a PhyCNN proxy model is constructed; generating a source domain label data set, and constructing a source domain PhyCNN proxy model; designing a parameter freezing migration strategy, and constructing a target domain PhyCNN proxy model; and sampling a posterior sample based on the target domain PhyCNN proxy model, updating the Bayesian model for structural damage identification by using the posterior sample, and calculating the maximum posterior estimated value of the structural damage parameter to realize positioning and quantification of the structural damage.
Owner:HARBIN INST OF TECH

MRI paraspinal muscle segmentation method based on bayesian fusion and probability shape modeling

The application discloses a MRI paravertebral muscle segmentation method based on Bayesian fusion and probability shape modeling, and comprises the following steps: a) a Fourier-Gaussian process probability shape model is established to parameterize the muscle shape, the Fourier-Gaussian process probability shape model comprises radial modeling based on a Fourier basis function to represent a muscle cross-section profile, and axial modeling based on a Gaussian process to capture the axial variation law of the muscle shape; b) a muscle edge feature model based on a convolutional neural network is constructed to perform multi-scale feature extraction on the muscle edge in the MRI image; and c) a Bayesian segmentation framework is designed, and based on a maximum posterior estimation principle, online segmentation of the paravertebral muscle in the MRI image is realized. The application is customized according to the characteristics of the paravertebral muscle segmentation task through the idea of fusing high-quality artificial labeling and a powerful learning model, so that the application can achieve high segmentation and reconstruction quality under a small amount of interaction, and has strong shape change description capability.
Owner:PEKING UNIV

Method for quickly generating high-resolution carbon emission raster data

The invention discloses a method for quickly generating high-resolution carbon emission raster data, which comprises the following steps of: fusing national carbon emission panel data and night light remote sensing images, establishing a provincial-scale logarithmic regression model, extracting parameters of the logarithmic regression model and constructing a parameter characteristic matrix; a space-time constraint K-Means algorithm is adopted to perform clustering analysis on provincial administrative units throughout the country and divide the provincial administrative units into K carbon emission type areas, for each carbon emission area, an area regression model is established by using logarithmic carbon emission and night light data, and a clustering center parameter is used as Gaussian prior to obtain a clustering result; and performing parameter updating through Bayesian maximum posteriori estimation, estimating the carbon emission by using the model, performing correction on a pixel scale by adopting a zero error method, and finally generating a national carbon emission grid data set. The space-time constraint clustering and Bayesian maximum posteriori estimation method is introduced on the basis of carbon emission modeling, and balance of few models and high precision is achieved.
Owner:NANTONG UNIV

Remote control train protection system

The invention relates to the technical field of train safety protection, and discloses a remote control train protection system which comprises the following steps: acquiring multi-source environment data in front of a train through a sensing acquisition module; inputting the multi-source environment data into a perception fusion module for data preprocessing and unified format conversion; in the sensing fusion module, fusing data of different sensors based on a maximum posteriori estimation method; the trusted obstacle data stream is transmitted to a security judgment module, and a minimum feasible obstacle set is constructed; the current speed of the train is combined through a train state analysis module; if the conditions are met; and inputting the optimal control instruction data flow to a control output module. According to the system and the method, high-frequency and high-confidence perception of the train operation environment and the dynamic obstacles is realized through a multi-source perception acquisition mechanism fusing the laser radar, the binocular vision and the GNSS / IMU, and the basic accuracy of system front-end safety judgment is effectively improved.
Owner:RIZHAO PORT GRP CO LTD +1

Edge data adjustment method for element geometric measurement

The invention belongs to the related technical field of computer digital image processing, and particularly relates to an element geometric measurement-oriented edge data adjustment method, which comprises the following steps of: carrying out overlapped division on element edges in an image to obtain an edge fragment set; based on a finite Gaussian mixture model and by fusing wavelets and total variation, establishing an adjustment edge maximum posteriori estimation function corresponding to all the edge segments, minimizing the function, performing independent adjustment on the edge segments, and aggregating the adjusted edge segments to obtain a pre-adjustment edge; carrying out overlapping division on the pre-adjustment edge to obtain a pre-adjustment edge fragment set; and the obtained finite Gaussian mixture model is fused with second-order total variation regularization, an ideal edge maximum posteriori estimation function of each pre-adjustment edge fragment is established, the function is minimized, independent smooth correction is performed on the pre-adjustment edge fragments, all the edge fragments after smooth correction are aggregated, and a final adjustment edge is obtained. According to the method, the geometric element measurement precision is improved on the algorithm level.
Owner:HUAZHONG UNIV OF SCI & TECH

Reservoir characterization method based on Viterbi algorithm

The invention relates to the technical field of reservoir characterization, in particular to a reservoir characterization method based on a Viterbi algorithm. The method comprises the following steps: step 1, describing a spatial change mode of a prior probability of a reservoir facies by adopting a Markov chain, and constructing a reservoir parameter joint probability distribution model with longitudinal continuity; 2, determining a maximum posteriori estimation problem of reservoir parameters by combining the reservoir parameter joint probability distribution model according to a maximum posteriori estimation theory; and step 3, based on a Viterbi algorithm, developing and solving the problem in the step 2, and establishing a corresponding reservoir characterization model. According to the method, on the basis of the Viterbi algorithm, rapid estimation of multi-target parameters of the hidden Markov model is achieved, a brand-new reservoir characterization model is established for the actual problem of quantitative prediction of reservoir properties, longitudinal geological continuity is introduced, and synchronous prediction of multiple reservoir parameters is achieved.
Owner:CHINA PETROLEUM & CHEMICAL CORP +1

Quantum image sensing method and device based on exposure modulation

The invention discloses a quantum image sensing method and device based on exposure modulation. The sensing end performs frame-by-frame exposure modulation on the same scene in continuous K frames, and regulates and controls the expected number of incident photons of each frame; the pixels output a 1-bit binary image in each frame and obey Poisson distribution. And the collection end takes K frames as a group to carry out maximum posteriori estimation to obtain reference intensity estimation of each pixel. In order to improve robustness and speed, an RED-PRO iterative algorithm based on FISTA acceleration is adopted in the reconstruction process, after gradient descent of a data consistency item, the data is projected to a pre-training de-noising device manifold, and a pixel adaptive step length is given based on Fisher information. According to the method, the signal-to-noise ratio is remarkably improved and artifacts are inhibited in a scene with low illumination, saturation and limited dynamic range, the 1-bit sensing circuit architecture does not need to be changed, and the existing QIS / SPAD sensor and the lens module can be compatible only through exposure modulation and software reconstruction.
Owner:ZHEJIANG UNIV

A bayesian model fast solution method for structural damage identification based on zero-shot transfer learning

The application discloses a kind of structural damage identification bayesian model fast solution method based on zero sample transfer learning, belong to structural damage identification and machine learning technical field.To solve the problem of generalization performance decline of structural damage identification bayesian model.The application includes constructing the bayesian model of structural damage identification;Combined with sparse auto-encoder and adversarial learning, construct and train the adversarial sparse auto-encoder, the adversarial sparse auto-encoder includes source domain sparse auto-encoder module, target domain sparse auto-encoder module and domain discriminator module;PhyCNN proxy model is constructed;Source domain label data set is generated, and source domain PhyCNN proxy model is constructed;Design parameter freezing transfer strategy, construct target domain PhyCNN proxy model;Based on target domain PhyCNN proxy model, sample posterior sample, update structural damage identification bayesian model using posterior sample, calculate the maximum posterior estimation value of structural damage parameter, realize the positioning and quantization of structural damage.
Owner:HARBIN INST OF TECH

Indoor pseudo satellite signal carrier-to-noise ratio positioning method and system based on commercial smart phone

The invention belongs to the technical field of pseudo satellite positioning, discloses an indoor pseudo satellite signal carrier-to-noise ratio positioning method based on a commercial smart phone, and solves the core problem through innovative design. The invention provides a positioning technology based on Bayesian reasoning alternating optimization dynamic estimation. According to an attenuation relationship between a carrier-to-noise ratio observation value and a distance base station, maximum posteriori estimation (MAP) is calculated through Bayesian reasoning, and an alternating estimation strategy is adopted to iteratively optimize an attenuation coefficient and a position until the attenuation coefficient and the position converge, so that the change of the attenuation coefficient is effectively dealt with, and the performance difference between smart phone devices is relieved; compared with a traditional weighted centroid method, the average positioning precision is improved by 30%. The technology provided by the invention is not only suitable for smart phones with different performances, but also can be used. The technology provides a universal solution for high-reliability positioning of the smart phone under the same-frequency interference condition.
Owner:WUHAN UNIV

IMU (Inertial Measurement Unit) and instantaneous wheel speed fused mapping method, device and equipment

The invention provides a mapping method, device and equipment fusing an IMU (Inertial Measurement Unit) and an instantaneous wheel speed, and relates to the technical field of vehicles. Image information of a current common frame and IMU information of the current common frame are acquired, and first pose estimation information of the current common frame is determined by adopting multi-sensor information fusion; and updating the first pose estimation information of a plurality of latest continuous common frames by adopting a maximum posteriori estimation theory. If the current common frame is not the key frame or the number of the current key frames does not reach the preset value, continuing to execute the steps of obtaining the image information of the current common frame and the IMU information of the current common frame and subsequent steps; if the preset value is reached, updating the first pose estimation information of the plurality of key frames and the position information of the landmark object through a maximum posteriori estimation theory; according to the method, the visual map in the vehicle driving process is constructed, so that the influence of image scratching and low extrinsic parameter precision can be relieved and even overcome, and the precision of constructing the visual map in the vehicle driving process is improved.
Owner:SAIC MOTOR

Field tomography measurement method and device for multi-physical quantities in combustion field based on multi-source light collaboration

The present invention belongs to the technical field of high-temperature flame combustion diagnosis and optical measurement, and specifically relates to a method and device for measuring multi-physical quantities of a combustion field based on multi-source light synergy. The present invention comprises the following steps: obtaining spontaneous radiation information, obtaining laser absorption spectrum information, constructing an objective function about the radiation source term distribution S and the radiation characteristic distribution k, solving the objective function F(S), calculating the temperature distribution T, solving the objective function F(k), judging whether the objective function converges, and calculating the soot particle concentration f v , constructing a probability density function of temperature, constructing a maximum a posteriori estimation expression, solving the maximum a posteriori estimation expression and outputting a measurement result. The present invention also provides a measuring device for implementing the above method. Through the above method, the technical problem that the single detection method based on laser has low spatial resolution and cannot realize high-precision coordinated measurement of multiple physical quantity fields such as temperature, gas component concentration, soot particle concentration and spectral radiation characteristics is solved.
Owner:HARBIN INST OF TECH

SLAM method based on maximum a posteriori estimation of underwater motion drift and noise

The present invention discloses a SLAM method for underwater motion drift and noise estimation based on maximum a posteriori estimation, comprising: 1) estimating the motion drift of an underwater robot based on EKF-SLAM; 1.1) initializing the robot SLAM system state; 1.2) constructing sampling points based on EKF-SLAM state estimation; 1.3) updating the covariance matrix based on the state estimation error; 2) updating the motion drift of the underwater robot based on maximum a posteriori probability; 3) estimating underwater noise parameters based on EKF-SLAM; 3.1) initializing the underwater noise parameters; 3.2) estimating the underwater noise parameter covariance matrix; 3.3) updating the covariance matrix based on the state estimation; and 4) updating the underwater noise parameters based on maximum a posteriori probability. This method improves the state prediction portion of the existing EKF-SLAM method, combines estimation of motion noise and observation noise in the system model, adaptively filters the Gaussian distribution noise variance of the system model, and then performs SLAM estimation, thereby improving underwater self-positioning accuracy by approximately 20%.
Owner:YANGZHOU UNIV

Plug-and-play image shadow edge removal method, device, equipment and medium

The present application relates to a plug-and-play image shadow edge removal method, apparatus, device, and medium. A brightness change coordinate system is constructed based on the brightness changes of pixels in a shadow image and a shadow-free image, and a Gaussian mixture model is used for modeling. The maximum a posteriori estimation problem for the position and belonging region of each point data in the brightness change coordinate system is used as a problem model. An iterative algorithm is used to solve the problem. During each iteration, the belonging region division is updated, the distribution parameters representing each updated belonging region are updated, and the position of each point data in the brightness change coordinate system is updated according to the updated distribution parameters to obtain a distribution-matched reference image. The edge region in the image is then optimized and updated to obtain the current iteration result, until a preset iteration termination condition is met to obtain a shadow-free image. This method can enhance the performance of shadow edge region recovery and obtain an image with better shadow removal effect.
Owner:NAT UNIV OF DEFENSE TECH

Off-grid model-based iterative Bayesian focusing sound source localization method

The invention relates to the technical field of sound source localization, in particular to an off-grid model-based iterative Bayesian focusing sound source localization method, which comprises the following steps of: acquiring a sound pressure signal acquired by a microphone array; discretizing the sound source plane through a preset step length, and obtaining a discrete grid according to the area of the sound source plane; assuming that the actual sound source position is not on the grid point, constructing an off-grid forward model; modeling a Green function of the off-grid forward model by using first-order Taylor approximation; performing maximum posteriori estimation on the off-grid forward model; and deducing the maximum posteriori estimation of the IBF algorithm under the off-grid forward model through an expectation maximization algorithm to obtain the estimated position of the sound source. According to the method, the problems that in the current sound source target detection process, a sound source is difficult to accurately fall on pre-divided grid points and the existence base is not matched are solved.
Owner:CHANGZHOU UNIV

An accurate learning method for Gaussian variational inference of UAV trajectories

The Gaussian variational inference accurate learning method of the UAV trajectory disclosed by the present invention comprises the following steps: step S1: constructing a mathematical model of the UAV according to the aerodynamic principle and kinematic principle of the UAV in actual flight; step S2: based on Gaussian variational inference, linking the true Bayesian posterior and the approximate posterior through KL divergence, converting the state estimation problem into the approximate problem of the posterior, and then converting the calculation of the KL divergence into the optimization problem of the loss function by defining the loss function; step S3: using Newton's method to update the loss function and using the natural gradient descent method to iteratively optimize the mean and the inverse covariance matrix, so that the loss function V(q) is reduced to the minimum to achieve the purpose of effective approximate posterior. Batch state estimation of nonlinear systems is realized. Compared with the maximum a posteriori estimation method, the state estimation under the condition of Gaussian variational inference has higher accuracy and can more accurately estimate the flight trajectory of the UAV.
Owner:NANJING UNIV OF AERONAUTICS & ASTRONAUTICS

Lifting appliance attitude determination method, device and equipment, readable storage medium and product

The invention provides a lifting appliance attitude determination method, device and equipment, a readable storage medium and a product, and the method comprises the steps: obtaining a template point cloud associated with a target lifting appliance, the template point cloud being a point cloud when the target lifting appliance is in a normal pose; determining a fitting plane of the target lifting appliance in a normal pose based on the template point cloud; acquiring a real-time point cloud acquired by at least one preset laser radar, and determining distance information between the real-time point cloud and the fitting plane; maximum posteriori estimation information corresponding to the target lifting appliance is constructed according to the distance information, and current pose information of the target lifting appliance is determined based on the maximum posteriori estimation information. Therefore, the current pose information of the target lifting appliance can be accurately identified on the basis of not arranging a flag bit on the target lifting appliance.
Owner:SANY MARINE HEAVY INDUSTRY CO LTD

Electromagnetic signal enhancement processing module

The invention relates to the technical field of electronic communication and geological exploration, and discloses an electromagnetic signal enhancement processing module, which comprises a signal input unit used for receiving an electromagnetic signal from an external sensor and converting the electromagnetic signal into a digital signal suitable for subsequent processing; the signal preprocessing unit is used for carrying out preliminary conditioning on the input signal so as to adapt to subsequent processing; the wavelet transformation module is used for performing multi-scale time-frequency domain analysis on the input signal to extract different frequency band information of the signal; and the adaptive filtering and Kalman filtering module is used for carrying out noise suppression and dynamic optimization on the signal and enhancing the signal quality. According to the method, the electromagnetic signal is efficiently and accurately enhanced in a complex noise environment by combining wavelet transform, adaptive filtering and maximum posteriori estimation technologies, and the signal quality and the signal-to-noise ratio are remarkably improved.
Owner:SHANGHAI PUDONG NEW AREA CHUANHE WATER RESOURCES LAYOUT DESIGN INST