69 results about "Nonlinear filter" patented technology

The invention relates to the technical field of underwater navigation, and discloses a real-time attitude estimation method of an underwater inertial navigation system. The method comprises the steps that the underwater vehicle collects multi-axis angular velocity and linear acceleration data through an inertial measurement unit and uploads the data to a central processing module, and the central processing module conducts kinematics modeling on the data to generate attitude change initial feature vectors; receiving water flow velocity data of a Doppler velocimeter and pressure data of a depth sensor, and performing flow field interference compensation optimization on the attitude change initial feature vector based on the data to generate an anti-interference attitude feature vector; fusing the anti-interference attitude feature vector and a historical attitude estimation result to generate space-time fusion attitude state representation, and then resolving a real-time Euler angle through a nonlinear filter; and matching a corresponding attitude stability control instruction from the navigation strategy library according to the real-time Euler angle, and issuing the attitude stability control instruction to a propeller execution unit. The method can effectively cope with underwater complex environment interference and guarantee stable operation of the underwater vehicle.

The invention provides an accurate calculation method and system for the axial pretightening force of an ultrasonic bolt, and the method combines a high-order statistic theory with a nonlinear filtering technology, and solves the problem that the measurement precision of a conventional method is insufficient under the conditions of strong noise and nonlinearity. Particle filtering and Kalman filtering are combined, so that the problem that a traditional Kalman filtering method easily causes sensitivity of an initial value of filtering divergence is solved; a double-Gaussian attenuation model is adopted to fit an echo envelope, the problem that a single Gaussian or single index model cannot adapt to envelope distortion caused by stress change is solved, and a nonlinear state space model is established to adapt to nonlinear echo signal characteristics under dynamic stress; gaussian noise is effectively suppressed by using a third-order cumulant cross-correlation algorithm, the time difference resolution in a low signal-to-noise ratio environment is remarkably improved, and the Gaussian noise and asymmetric interference are effectively suppressed.

The invention provides an OBD message analysis and attitude rendering method and system based on a vehicle dynamics model, and belongs to the field of vehicle electronics and control. The method comprises the steps that a message is analyzed through an OBD interface, and vehicle operation parameters are obtained; establishing a unified time base and carrying out resampling and synchronous processing on the data of each channel; constructing a fifteen-degree-of-freedom vehicle dynamics model comprising a vehicle body six-degree-of-freedom pose, a vehicle body angular velocity, a four-wheel rotation angular velocity and unsprung mass vertical displacement for state prediction; designing an observation function to map a model state to an OBD observation space, and adopting an observability enhancement strategy; predicting and observing data are fused by using a nonlinear filter, and abnormal observation gating is carried out based on a mahalanobis distance and an amplitude threshold value; compensating system time delay and adaptively adjusting observation noise; and performing three-dimensional real-time rendering on the attitude information. According to the method, high-precision and high-robustness vehicle attitude estimation can be realized only by depending on the OBD message, and the system cost is effectively reduced.

In various examples, the present disclosure relates to using temporal filters for automated real-time classification. The technology described herein improves the performance of a multiclass classifier that may be used to classify a temporal sequence of input signals—such as input signals representative of video frames. A performance improvement may be achieved, at least in part, by applying a temporal filter to an output of the multiclass classifier. For example, the temporal filter may leverage classifications associated with preceding input signals to improve the final classification given to a subsequent signal. In some embodiments, the temporal filter may also use data from a confusion matrix to correct for the probable occurrence of certain types of classification errors. The temporal filter may be a linear filter, a nonlinear filter, an adaptive filter, and / or a statistical filter.

The embodiment of the invention provides a signal processing method and device, a model training method and device, equipment and a storage medium. The method comprises the steps of obtaining a mixed sound signal and a reference sound signal; filtering a linear echo component in the mixed sound signal based on the reference sound signal to obtain a first filtered signal; obtaining an output signal through a pre-trained nonlinear filtering model and the first filtering signal, and sending the output signal to the far-end terminal equipment; wherein the output signal filters out a nonlinear echo component of a target energy quantity level relative to the first filtering signal, the target energy quantity level is smaller than a preset energy level, and the preset energy level is determined based on the energy level of the human voice component in the first filtering signal. The nonlinear component in the sample filtering signal is filtered based on the nonlinear filtering model, and meanwhile, the filtering amount of the nonlinear component is controlled, so that excessive suppression of human voice components is avoided, and the voice communication quality in a complex environment is improved.

A signal processing method, a model training method, an apparatus, a device, and a storage medium. The signal processing method comprises: acquiring a mixed sound signal and a reference sound signal (S101); on the basis of the reference sound signal, filtering a linear echo component in the mixed sound signal to obtain a first filtered signal (S102); and obtaining an output signal by means of a pre-trained nonlinear filtering model and the first filtered signal, and sending the output signal to a remote terminal device, wherein for the output signal, a nonlinear echo component of a target energy level is filtered out relative to the first filtered signal, the target energy level is less than a preset energy level, and the preset energy level is determined on the basis of the energy level of a human voice component in the first filtered signal (S103).A nonlinear component in a sample filtered signal is filtered out on the basis of a nonlinear filtering model, and the amount of the nonlinear component filtered out is controlled, avoiding over-suppression of the human voice component, and improving the voice communication quality in complex environments.

The invention provides a high-precision deformation monitoring method and system based on tight combination of Beidou and IMU, and a medium, and relates to the field of geological disaster monitoring and engineering structure safety monitoring and early warning. The method comprises the following steps: constructing a three-dimensional displacement increment between Beidou carrier TDCP epochs, collecting triaxial acceleration and angular velocity data by using an IMU, carrying out tight combination modeling on calculation results of the two, constructing a nonlinear filtering equation taking the displacement increment as an observed quantity, realizing complementation of low-frequency high precision and high-frequency anti-interference through weighted fusion, and obtaining a monitored displacement quantity. Through cooperation of displacement change calculation between Beidou epochs and IMU vibration signals, complementation of low-frequency high precision and high-frequency anti-interference is achieved, the real-time performance of calculating collapse and slip displacement is improved, the displacement magnitude estimation error is reduced, the data utilization rate and deformation monitoring precision are improved, seamless switching of dynamic and static monitoring modes is achieved at the same time, and the real-time performance of the system is improved. And the robustness in a complex environment is enhanced.

The invention relates to the technical field of digital twinning and system health state prediction, in particular to a steering system model construction method based on digital twinning. Comprising the steps of 1, constructing a high-fidelity physical mechanism model representing the dynamic characteristics of the steering system; acquiring a working condition data flow of the steering system in real time; constructing a state observer; resolving a residual signal through a state observer; step 2, constructing an augmented state vector; performing recursive estimation on the augmented state vector by adopting a nonlinear filtering algorithm; 3, injecting the online identification parameter sequence into the high-fidelity physical mechanism model in real time to realize adaptive updating of the physical mechanism model; collecting historical data of the online identification parameter sequence; training a degradation prediction model by using historical data; and predicting and outputting a degradation trend prediction time domain of the steering system. According to the method, the fidelity and credibility of the model are remarkably improved, and a solid foundation is laid for subsequent analysis and prediction.

The application discloses a kind of based on adaptive nonlinear filtering's aircraft multi-source data attitude evaluation system, it is related to chromatic aberration detection technical field, including multi-source data acquisition end, adaptive filtering fusion end, noise characteristic estimation end and attitude solution evaluation end, the multi-source data acquisition end, adaptive filtering fusion end, noise characteristic estimation end and attitude solution evaluation end pass through on-board bus communication protocol and carry out two-way data interaction between, and each end is electrically connected with central logic processing matrix.This based on adaptive nonlinear filtering's aircraft multi-source data attitude evaluation system, the present application is aimed at atmospheric data and inertial navigation data this kind of multi-source heterogeneous data, establishes the unified fusion framework based on unscented kalman filter, and introduces multiple fading factor matrix, solves the problem of insufficient tracking ability of traditional method under strong nonlinear and large maneuvering flight conditions, significantly improves the precision of attitude estimation.

The application belongs to the technical field of target tracking and fusion, and relates to a nonlinear filtering distributed target tracking method based on variational inference. Measurement noise parameters of a conventional nonlinear filtering distributed algorithm are assumed to be known, and the application extends the nonlinear filtering distributed algorithm based on variational inference to a single-target tracking scene under unknown measurement noise parameters. Under the distributed algorithm, weight likelihood parameters of each sensor to the target state tend to be consistent, so that the algorithm has strong robustness, small calculation amount, high flexibility and timeliness, and in combination with the variational inference method, the algorithm can achieve high-precision target tracking in a distributed multi-sensor scene under unknown measurement noise variance. The algorithm has close tracking precision to a distributed particle filtering algorithm with known measurement noise variance, and enhances practicability.

A noise reduction system for actively compensating background noise in a passenger transport area of a vehicle. The noise reduction system includes a nonlinearity filter unit having a model of a non-linear transfer function of the sound generator, wherein the nonlinearity filter unit is configured to receive the anti-noise signal and to generate a filtered anti-noise signal by applying a non-linear filter function on the anti-noise signal, which is based on the model of the non-linear transfer function in that the non-linear response of the sound generator is at least partially corrected when driven by the filtered anti-noise signal. Wherein the nonlinearity filter unit is further configured to output the filtered anti-noise signal to a sound generator.

The application provides an example in the technical field of signal processing, and provides an audio device, an echo cancellation method and an apparatus. The audio device comprises a loudspeaker, a microphone, and a processor connected with the microphone and the loudspeaker respectively. The processor is configured to: acquire an audio signal frame received by the microphone; perform echo cancellation processing on the audio signal frame; if a nonlinear filter suppression coefficient used in the echo cancellation processing is less than a coefficient threshold value, and a current call state is detected as a double-talk state, then add 1 to a count value; if the count value is greater than or equal to a count threshold value, then reduce the volume of the loudspeaker; if the count value is less than the count threshold value, then perform the acquisition of the audio signal frame received by the microphone; if the nonlinear filter suppression coefficient is greater than or equal to the coefficient threshold value, or the current call state is detected as a non-double-talk state, then clear the count value, and perform the acquisition of the audio signal frame received by the microphone. The application can effectively eliminate echo.

The application discloses a method, device and equipment for applying an improved nonlinear filtering algorithm to a UAV, and relates to the technical field of UAV application. The method comprises the following steps: based on each sensor arranged on the upper portion of the UAV, collecting GNSS data and IMU data in real time and synchronously according to a set sampling frequency and inputting the GNSS data and the IMU data into the improved nonlinear filtering algorithm; performing preliminary fusion on the GNSS data and the IMU data through an unscented Kalman filtering algorithm, and combining UT transformation and an adaptive mechanism to obtain a preliminary state estimation value of the UAV; inputting the preliminary state estimation value into a cubature Kalman filtering algorithm, updating and predicting the state of the UAV based on a cubature rule, and obtaining final positioning and attitude estimation results and applying the final positioning and attitude estimation results to the UAV. The application can greatly improve the positioning and attitude estimation precision of the UAV.

The present application integrates the idea of separating noise and true value into the PLKF framework, and proposes an unmanned aerial vehicle target positioning method based on two-stage unbiased pseudo-linear Kalman filtering (2S-UPLKF). The angle estimation based on EKF in the first stage can effectively decouple noise interference when the unmanned aerial vehicle is tracking at a long distance, providing a robust initial state for the system. The second stage constructs a noise-true value separation mechanism to eliminate the correlation between the observation matrix and the noise in the pseudo-linear equation from the principle level. This method not only inherits the advantage of efficient calculation of pseudo-linear filtering, but also realizes the active suppression of bias in the dynamic tracking process through two-stage collaborative optimization. Theoretical analysis and simulation experiments show that, compared with other nonlinear filtering algorithms, it can still effectively suppress error fluctuations under extreme conditions such as strong nonlinear observation, large angle noise and long distance observation, and has better tracking performance.

The invention discloses a method, device and equipment for applying an improved nonlinear filtering algorithm to an unmanned aerial vehicle, and relates to the technical field of unmanned aerial vehicle application, and the method comprises the steps: synchronously collecting GNSS data and IMU data in real time according to a set sampling frequency based on each sensor disposed on the unmanned aerial vehicle, and inputting the GNSS data and IMU data to the improved nonlinear filtering algorithm; carrying out preliminary fusion on the GNSS data and the IMU data through an unscented Kalman filtering algorithm, and obtaining a preliminary state estimation value of the unmanned aerial vehicle in combination with UT conversion and an adaptive mechanism; and inputting the initial estimation value into a volume Kalman filtering algorithm, updating and predicting the state of the unmanned aerial vehicle based on volume rule calculation, obtaining a final positioning and attitude estimation result, and applying the final positioning and attitude estimation result to the unmanned aerial vehicle. According to the invention, the precision of unmanned aerial vehicle positioning and attitude estimation can be greatly improved.

The present invention discloses a method for multi-node collaborative navigation of a flexible lander under limited navigation information, belonging to the field of deep space exploration technology. The present invention is implemented as follows: based on the current altitude and the associated observation information between nodes, an indirect observation equation for navigation landmarks is constructed based on homography transformation or epipolar geometry to build a transmission path for observation information between nodes; the Fisher information matrix is ​​used to analyze node observability, and multi-node information interaction conditions are established with node observability and the flexible lander altitude as indicators; the information interaction conditions are used to determine the information transmission direction between nodes that makes the positions of all nodes observable and the minimum transmitted observation information in the corresponding information transmission direction; a single node uses the observation information obtained by the node itself and the observation information transmitted by other nodes to estimate the node state through a nonlinear filtering algorithm, and all nodes synchronously perform state estimation according to the single node state estimation method to achieve multi-node collaborative navigation under limited navigation information.

The present application relates to the technical field of oil and gas exploration, and particularly relates to a method and system for determining nonlinear isochronous stratigraphic slices. The method comprises the following steps: obtaining original seismic data; performing seismic data preprocessing on the original seismic data to obtain preprocessed seismic data; performing adaptive nonlinear filtering processing on the preprocessed seismic data to obtain adaptive nonlinear filtering data; performing nonlinear seismic attribute extraction on the adaptive nonlinear filtering data to obtain a nonlinear seismic attribute field; obtaining geological data and inversion target data; constructing an initial geological model according to the geological data to obtain the initial geological model; and performing residual error calculation according to the inversion target data and the initial geological model to obtain a residual error calculation result. The present application can better handle nonlinear geological structure complexity, improve the accuracy and reliability of isochronous stratigraphic slices, and provide more accurate geological information for oil and gas exploration and development.

This invention discloses a robust relative navigation method for aircraft based on a hybrid distribution under non-Gaussian noise, belonging to the technical field of computation, estimation, or counting. To address the problem of filter divergence caused by non-stationary heavy-tail noise in time-varying environments during relative navigation, this invention introduces a Dirichlet random mixture vector fusion of Gaussian, Student's t, and multivariate K-distributions, proposing a Gaussian-Student's t-multivariate K-distribution modeling of measurement likelihood. Then, by minimizing the KLD of the true posterior probability density function and the approximate posterior probability density function using variational Bayesian techniques, the approximate posterior estimates of the aircraft's relative motion state and filter parameters are obtained, yielding the target's state information relative to the aircraft and solving for relative position and velocity. Finally, a nonlinear filter based on the Gaussian-Student's t-multivariate K-distribution is derived to improve relative navigation accuracy for angle-only relative navigation of aircraft in time-varying environments.

The invention discloses an under-actuated ship finite time formation control method based on a historical data triggering mechanism, and the method comprises the steps: constructing a ship formation virtual control law through introducing a power nonlinear feedback term according to a formation tracking error, and introducing a first-order nonlinear filter to carry out the filtering processing of the virtual control law, performing approximation processing on unknown function items in the error dynamic equation by adopting a radial basis function neural network to obtain an optimized error dynamic equation; and constructing an information guidance dynamic event triggering mechanism based on historical triggering data, combining a formation cooperative control law and a parameter updating law, and realizing under-actuated ship finite time formation control based on the historical data triggering mechanism. The method solves the problems that the dynamic balance between the control performance and the communication efficiency is difficult to realize and the finite time quick response requirement of the time-sensitive task cannot be met due to the fact that the data weight dynamic adjustment characteristic of the historical triggering moment is not fully fused in the existing method.

This invention belongs to the field of robot control technology, specifically relating to a global tracking control method for flexible joint robots based on obstacle functions. The method includes: S1: constructing and reconstructing the dynamic model of the flexible joint robot to adapt it to a reverse-engineering control framework; S2: designing a nonlinear filter that integrates obstacle functions, and using the obstacle functions to limit the filtering error within a preset range; S3: based on the nonlinear filter, recursively designing a controller using a reverse-engineering method, constructing an energy function, and performing stability analysis to ensure that all signals in the closed-loop system are globally consistent and bounded. This invention enables the control method to significantly improve tracking accuracy and system robustness while maintaining low computational complexity. It is suitable for robot systems with compliant actuation characteristics and has good engineering application value.

The invention discloses a time-varying optimization formation control method for networked robots with a layered architecture, and the method comprises the steps: collecting the pose and kinetic parameter information of the networked robots, carrying out the equation conversion based on kinematics and kinetic models, and constructing a chain control form capable of describing the state change; track correction is realized by setting a cost function and optimizing a signal generator; a nonlinear filtering function and a virtual controller are introduced to suppress position and speed errors; a radial basis function neural network is adopted to carry out approximation on an unknown nonlinear dynamic state, an adaptive control law is constructed in combination with error feedback, and time-varying optimization of the networked robot formation is realized. According to the method, the convergence speed, stability and robustness of the system can be effectively improved in dynamic and uncertain environments, it is guaranteed that the robot group achieves accurate formation control in complex tasks, and the method has high engineering application value.

This invention belongs to the field of measurement and control, and relates to a fixed-time preset performance control method, device, computer equipment, and storage medium for a macro-micro composite motion platform. The method includes: constructing a dynamic system model of the macro-micro composite motion platform; transforming the dynamic system model into a state-space equation suitable for a backstepping recursive control framework; designing a fixed-time preset performance function to constrain the tracking error signal within a preset range within a fixed time; constructing a nonlinear filter to simplify the control design process and establishing a switching function to resolve coupling terms in the system; designing an adaptive fixed-time preset performance tracking control algorithm and proving the stability of the closed-loop system based on Lyapunov stability theory. By designing the controller using an adaptive backstepping recursive control framework, the control design process is simplified, and the computational load is reduced; the tracking error signal of the closed-loop system converges to the preset range within a fixed time, improving the transient performance and control accuracy of the system.

The invention particularly relates to a sea clutter pulse interference suppression method for cascaded Hammerstein robust filtering, and the method comprises the steps: receiving an echo signal, and carrying out the preprocessing of the echo signal, so as to obtain a direct signal and a sea clutter signal, which are separated from each other; a cascaded Hammerstein filter is used to carry out multi-stage filtering processing on the separation signal, and a non-linear mapping output signal is obtained; wherein the cascaded Hammerstein filter comprises a first nonlinear filtering module, a second linear filtering module and a third nonlinear filtering module which are arranged in a cascaded manner; and carrying out filtering processing on the nonlinear mapping output signal through a robust filter to obtain an output signal after interference suppression. According to the method, nonlinear distortion and non-Gaussian pulse interference can be effectively suppressed in a complex marine environment, and the anti-interference capability, the signal fidelity and the adaptivity of a radar and a communication system are improved.

The invention discloses a novel dynamic surface funnel control method for a high-speed train with time delay and saturation, and relates to the technical field of high-speed train control, and the method comprises the steps: constructing a high-speed train longitudinal dynamic model with input time delay and input saturation; defining a displacement tracking error and a speed tracking error of the high-speed train; respectively setting funnel boundaries for the displacement tracking error and the speed tracking error based on a dynamic funnel boundary function; constructing a Lyapunov function, and constructing a virtual control law and a nonlinear filter; determining an actual control law of the high-speed train according to the high-speed train longitudinal dynamic model with input time lag and input saturation, the filtered virtual control law and the speed tracking error; the influence of input time lag on system performance is effectively solved, high-precision tracking control of the train is achieved under the non-saturation condition, meanwhile, the system can still be kept stable when saturation occurs, and therefore the control reliability and stability of the high-speed train under different working conditions are improved.

The invention relates to the technical field of light beam drift prevention, in particular to an interference accurate estimation and robust control method for a laser stable pointing system, which comprises the following steps of: processing an output signal through a zero-phase filtering technology based on the output signal of the laser stable pointing system to obtain a controlled object mathematical model; based on the controlled object mathematical model, a feedback controller is obtained through design; based on an input signal of the laser stable pointing system, a controller output quantity is obtained through a feedback controller; the controller output quantity passes through an enhanced adaptive robust controller to obtain a filtering disturbance estimated value; the enhanced adaptive robust controller comprises a Kalman filter, an adaptive robust controller and a nonlinear filter. The MR-H infinity control designed by the invention can significantly enhance the tracking performance and adaptability of the system, and the EARC provided by the invention can accurately estimate disturbance in a strong measurement noise environment. The control strategy of the invention can effectively suppress the adverse effect of strong sensor noise on the system, and significantly improve the tracking precision and anti-interference performance of the system.

The embodiment of the invention provides a satellite receiver signal processing method and device, and the method comprises the steps: carrying out the demodulation processing of a received signal, and obtaining a baseband signal; performing correlation calculation processing on the baseband signal and a local correlation sequence to obtain a correlation value; constructing a signal space based on the correlation value; performing linear filtering processing on the correlation value of the signal space to obtain a linearly filtered signal space; based on the signal space after linear filtering, the maximum correlation value under the same Doppler frequency and different code phase delays in the first dimension is searched, the sum of the distances between the maximum correlation value and the correlation values under the same Doppler frequency and other code phase delays is calculated, and the maximum distance is selected from the sum of the distances corresponding to the Doppler frequencies; performing nonlinear filtering processing on the correlation value of the Doppler frequency corresponding to the maximum distance to obtain a correlation value after nonlinear filtering; and determining whether a target signal exists based on the correlation value after nonlinear filtering. The performance of capturing the target signal can be improved.

The present invention provides a maneuvering target tracking method based on hybrid inference. This method, based on variational Bayesian and sequential Monte Carlo hybrid inference, first designs a proposed distribution based on initial variational parameters. This proposed distribution is sampled to generate samples. The generated samples are then used to calculate a surrogate evidence lower bound. Finally, the variational parameters are re-optimized using stochastic gradient descent, resulting in continuous iterative updating of the sampled particles and variational parameters. This method transforms nonlinear state estimation and maneuvering model parameter identification into an evidence lower bound optimization problem. Stochastic gradient descent is used to achieve joint optimization of target state estimation and model parameter identification, enabling simultaneous state estimation and model parameter identification. This effectively addresses nonlinear filtering issues and improves nonlinear maneuvering target tracking performance. The method is advantageous in that it can adapt to nonlinear dynamic systems with unknown model parameters.

The present application belongs to the technical field of image processing, and particularly relates to a kind of image sharpness evaluation methods for long focal length visible light imaging system. The present application first adopts median filter algorithm to pre-process image, and the algorithm is a typical representative of nonlinear filtering algorithm, which can eliminate isolated noise points, and can effectively protect edge contour information while removing random noise. This feature is very consistent with the high demand of edge information for sharpness function. Then, the pre-processed image is decomposed by NSST (non-subsampled shearlet transform), to obtain low-frequency components and high-frequency subbands, and a robust image sharpness evaluation method is constructed by combining direction consistency and contrast normalization.

This invention provides a biomimetic lateral line communication system based on propeller wake field perception, belonging to the field of underwater communication technology. Based on the lateral line perception mechanism of fish, it utilizes the flow field disturbance signal generated by the interaction between the propeller and the flow field as the communication signal to achieve network communication for underwater robots. The flow field disturbance of the main propeller is converted into a carrier signal to increase the transmission distance and intensity, while the disturbance of the modulation propeller is converted into a modulation signal for information transmission. A neural network model and a nonlinear filtering model are used to achieve signal modulation and demodulation, exhibiting good environmental adaptability and improving the working efficiency and accuracy of the detection system. The beneficial effects of this invention are: the signal source used for biomimetic navigation perception is the flow field medium fluctuation signal, which can provide a novel approach and practical detection sensing means for underwater covert detection and guidance tracking technology.

The invention relates to a self-adaptive tracking method and system for a maneuvering magnetic target, and belongs to the technical field of target tracking and magnetic detection. The method comprises the following steps: establishing a target physical model and a motion model, and determining a state variable of a tracked magnetic target according to the models; constructing a state space model, and obtaining a motion state equation and a nonlinear observation equation of the magnetic target according to the determined state variable; constructing an IMM-AKKF-FNN model, comprising the following steps: a maneuvering decision-making layer adopts an IMM framework, runs a plurality of filters for different motion modes in parallel, and performs probability weighting according to the likelihood of each model so as to adaptively cope with macroscopic motion mode switching of a target; the nonlinear filtering layer adopts AKKF as a core state estimator in each model of the IMM framework; the noise sensing layer is provided with a pre-trained FNN for each AKKF. According to the invention, the tracking precision and robustness can be improved.