15 results about "Bernoulli filter" patented technology

The present application relates to a kind of close approach trajectory robust control method in multi-target tracking, it is related to optimal control problem analytical objective function, constraint establishment and robust approach trajectory control scheme, belong to spacecraft situation awareness and control field.The present application first establishes optimal trajectory control problem, considers information gain and security constraint comprehensively, to establish the model basis of robust control instruction.Second, by establishing the analytical expression of Hellinger distance in optimal control problem under the theory of Outer Probability Measures (OPMs), the difference between the prior estimate and the posterior estimate of the target state is effectively evaluated, and then the information gain value under different control instructions is calculated to select the optimal control instruction.At the same time, by establishing the analytical upper and lower bound expression of the void probability in the constraint condition of optimal control problem under the theory of OPMs, the void probability of the approach trajectory at any time under different control instructions can be efficiently calculated, so as to evaluate the safety of the approach trajectory and select the control instruction that meets the constraint condition.Finally, a robust trajectory control and a possibility label multiple Bernoulli filter module are designed to form a close approach trajectory robust control scheme, which realizes the goal of generating optimal control instructions in real time during the approach and tracking process, and achieves the effect of maximizing information gain and ensuring the safety of the approach trajectory.

The application provides an underwater multi-target tracking method based on target state extension, and since a standard random finite set algorithm does not consider non-Gaussian measurement error and detection probability time-varying problems, model mismatch of the multi-target tracking method is caused, thereby causing tracking performance deterioration, and even completely wrong target state estimation is given. By using a generalized labeled multi-Bernoulli filter and a Dirichlet process-hidden Markov chain hybrid model, a target state, a probability density distribution of observation noise and a detection probability are adaptively and real-timely estimated by using variational inference, tracking error is significantly reduced, and track estimation precision and track integrity are improved.

The application discloses a label multi-Bernoulli tracking method for cooperative and non-cooperative unmanned aerial vehicles, and comprises the following steps: acquiring a current time unmanned aerial vehicle measurement set, a cooperative unmanned aerial vehicle report set and a basic state vector; introducing a discrete mode variable to establish a Markov jump mechanism to obtain an extended state and a motion state function thereof; adopting a label multi-Bernoulli filter to track multiple unmanned aerial vehicles to obtain corresponding prediction densities; obtaining a posterior density of the cooperative unmanned aerial vehicle and a posterior density of the non-cooperative unmanned aerial vehicle according to the measurement set, the cooperative unmanned aerial vehicle report set and the prediction densities; introducing a trajectory quality evaluation mechanism to adjust the trajectory quality to obtain a final posterior density of the non-cooperative unmanned aerial vehicle; and taking the posterior density of the cooperative unmanned aerial vehicle and the posterior density of the non-cooperative unmanned aerial vehicle as the output on one hand to obtain a current time unmanned aerial vehicle target set, and taking the combination thereof as the input of a next time prediction step on the other hand to start a new round of filtering circulation.

The invention discloses a Bernoulli filter-based single target detection and DOA tracking method and system, and relates to the field of target tracking. The method solves the problems that a traditional tracking algorithm after detection cannot jointly detect the target when the underwater target repeatedly enters under the conditions of low signal-to-noise ratio and small snapshot, and the tracking effect is unstable. The method comprises the steps of constructing a Bernoulli random finite set model, defining a single target state as a dynamic description target state or a dynamic description target state, predicting the target existence probability by using a Bernoulli Markov process, and generating a prediction state distribution particle set composed of new particles and survival particles through particle filtering. A hydrophone array signal is decomposed into a noise subspace observation model and a signal subspace observation model, a generalized likelihood function after exponential weighting is used for updating a target posteriori existence probability and a particle weight, and an equal-weight particle set is generated through resampling. And judging whether the target exists or not according to the single-target posterior existence probability, thereby realizing single-target DOA joint detection and tracking.

The application relates to a parallel fast label multi-Bernoulli filtering method under a phased array radar system and belongs to the field of radar target tracking. The application comprises the following steps: obtaining a measurement set of a radar, dividing the measurement set by adopting a mean shift algorithm, performing plot condensation on the divided measurement set, performing parallel joint prediction and updating based on the plot condensed measurement set, including LMB state transition, target and measurement grouping, parallel processing of the grouping results, and merging of the group posterior density, pruning and truncating the LMB parameter set after the parallel joint prediction and updating, and estimating the target number and state based on the pruned and truncated LMB parameter set. Through grouping of the target and the measurement, parallel processing of the FLMB filter is realized, the calculation amount is effectively reduced under the condition that the tracking performance loss is small, the real-time performance is improved, and the target detection speed and accuracy are improved.

The invention specifically relates to a target track-before-detect method suitable for a distributed low-frequency passive sonar, and the method comprises the steps: building a Cartesian coordinate system with the center of an observation region as an original point, carrying out the network division of the observation region, and obtaining a two-dimensional observation grid which comprises a plurality of grid points; the sonar system based on the detection nodes of the observation area obtains the measurement value of each grid point; and inputting the measured value of each grid point into a Bernoulli filter, calculating a target existence probability and a target state probability density function through a Bernoulli filtering algorithm, considering that a target exists when the target existence probability is greater than a given threshold, and outputting a mean value of the target state probability density function as a target state estimation result at the current moment. The method has the advantage of integrated processing of detection and tracking, effectively improves the detection performance and positioning precision of a weak target, and is suitable for underwater target monitoring requirements in a deep sea complex environment.

The invention provides an adaptive detection and tracking method based on a Bernoulli filter, and the method comprises the steps: employing a random finite set to model a target state, and describing the Markov process dynamic characteristics of a target based on a state transition probability; calculating a detection probability and a false alarm rate by using a Newman Pearson criterion, and modeling measurement as a random finite set; according to the existence probability and the space state of the target at the previous moment, combining the birth probability, the survival probability and the state transition model of the target, predicting the existence probability and the space state of the target at the current moment; predicting the GOSPA performance of the Bernoulli filter, modeling a target adaptive detection and tracking problem into an optimization problem taking a false alarm rate as an optimization parameter, and achieving the minimization of the GOSPA performance predicted by the Bernoulli filter by adjusting the false alarm rate; a golden proportion constant is calculated by initializing a false alarm rate search interval and an allowable error, GOSPA performance is iteratively optimized by using a golden section method, and an optimal false alarm rate and a corresponding detection threshold value are obtained; a detection result is obtained by using an optimal detection threshold value, then the detection result is used as input, the target existence probability and the spatial state are recursively updated, the calculation complexity is controlled by cutting low weight components and combining similar components, and efficient and accurate self-adaptive detection and tracking are achieved.

The invention relates to the field of signal processing, in particular to a radar target tracking and recognition method based on multi-dimensional information fusion, which comprises the following steps: performing pulse compression and coherent processing on a sea surface echo signal received by a radar, constructing a distance-Doppler graph, and detecting and positioning a suspected target through a constant false alarm rate. And the radial speed characteristic and the distance dimension contour fitting coefficient based on the weighted least square method are synchronously calculated. Afterwards, the motion and structure features are fused with a traditional energy state, an extended target state with a unique identification tag is constructed, and a probability distribution model of an initial multi-target state is established by adopting a tag multi-Bernoulli random finite set theory; and performing multi-target state estimation under a Bayesian recursion framework by using a generalized label multi-Bernoulli filter. Time domain statistical features are extracted based on a tracking trajectory, a vector machine is adopted to discriminate a target type, and through multi-dimensional information fusion and joint optimization processing, tracking stability and recognition reliability of a marine weak target are improved.

The invention discloses a multi-target tracking method for Poisson multi-Bernoulli filtering based on time matching, which mainly solves the problem of sampling time diversity when a random finite set filter is applied to sensor scanning, and introduces a Poisson multi-Bernoulli filter into a time matching Bayesian filtering framework. A time matched Poisson multi-Bernoulli (TM-PMB) filter is presented. A TM-PMBM filter and a TM-PMB implementation method under KLD divergence approximation are provided under a linear Gaussian dynamic and measurement model, and performance comparison is performed on the TM-PMB filter and other RFS filters based on time matching in a TWS (scanning while tracking) one-way scanning mode. The filter provided by the invention can effectively reduce the influence caused by the diversity of sampling time, improves the estimation precision of the target state and the cardinal number, needs less filtering time, and has very high robustness.

The invention particularly relates to a deep sea underwater target tracking method based on a frequency domain interference sound field. The method comprises the following steps: firstly, reconstructing an array original broadband wave beam output by using a complementary integrated empirical mode decomposition algorithm and mapping the array original broadband wave beam output into a target glancing angle-depth ambiguity plane; and secondly, modeling ambiguity planes at multiple moments into a random finite set form, and filtering and outputting a grazing angle-depth trajectory of the underwater target by using a label multi-Bernoulli filter. And finally, converting the glancing angle of the target into a horizontal distance in combination with a sound field model, and outputting a distance-depth track of the underwater target. According to the method, a grazing angle-depth measurement random finite set of the underwater target is constructed by using frequency domain interference characteristics generated by the moving underwater target in a direct sound area range, and label multi-Bernoulli filtering is performed on the measurement set in combination with a target motion state equation and a sound field model, so that continuous tracking of the underwater moving target is realized.

The present application belongs to the technical field of electronic countermeasure, and particularly relates to a single-radiation-source joint detection and tracking method based on received signal strength. The present application establishes a measurement model containing normal shadow fading parameters based on received signal strength, introduces particle filtering technology to estimate the shadow fading parameters, and designs a Bernoulli filter by taking signal strength as a measurement index, thereby providing a solution for single-target radiation source joint detection and tracking in complex scenes. The present application can realize joint detection and tracking of single-target radiation sources, and uses soft decision, i.e. calculation of existence probability, to replace hard decision to extract states and detect whether a target exists, thereby avoiding intermediate errors generated by the traditional two-step tracking method. The method has strong robustness and good effect.

The application discloses a simultaneous localization and mapping method based on multiple Bernoulli filters, and specifically comprises the following steps: (1) parameter initialization; (2) obtaining input data; (3) obtaining a robot position prediction value; (4) obtaining an observation set of the robot; (5) obtaining a Bernoulli item of the robot for representing a map feature at the kth moment through a potential balance multiple Bernoulli filtering method; (6) target extraction is performed on the obtained Bernoulli item; (7) the number of map features and the pose obtained at the kth moment are recorded; (8) whether a graph optimization process is executed is judged through an adaptive information control method; (9) the robot pose corresponding to the tth moment is obtained through a graph optimization method, and then step (2) is executed; (10) whether the robot pose and the state estimation value of the map feature are output is determined according to whether the maximum running moment number is reached. According to the method, the robot pose estimation precision in the simultaneous localization and mapping method is improved, and the real-time performance is improved.

The application relates to a distributed multi-sensor fusion tracking method and system, which comprises the following steps: acquiring original measurement information of each sensor, and adopting a pre-established extended label multi-Bernoulli filter to obtain multi-sensor local posterior probability density; determining a distributed multi-sensor fusion network topology structure and a relative weight matrix, and fusing the multi-sensor local posterior probability density based on the relative weight matrix to obtain a fusion probability posterior density; and updating a target state and quantity according to the fusion probability posterior density to obtain a distributed multi-sensor fusion tracking result. The application realizes simultaneous estimation of multi-target states and quantities based on the extended label multi-Bernoulli filter, does not need a complex data association algorithm, improves calculation efficiency, and realizes real-time multi-target tracking. Therefore, the application can be widely applied to the environment perception field of intelligent automobiles.

This invention relates to the field of radar signal processing technology, specifically to a radar signal sorting method based on a tag-based multi-Bernoulli filter. The method includes: modeling each potential radar source as a tagged Bernoulli element, which contains information on survival probability and spatial distribution; predicting the state and survival of existing targets based on the target state and motion model of the previous time step; updating the existence probability of the Bernoulli element by calculating the likelihood of each "predicted target-observation" pair and associating it with each target for new measurements; generating new Bernoulli elements with a certain probability for observations not used to update Bernoulli elements; and removing Bernoulli elements with an existence probability below a threshold in the next step. This invention, by applying the tag-based multi-Bernoulli method to radar signal sorting, achieves robustness against missed detections and false alarms, while avoiding the need for complex target management where the survival of a target is entirely determined by its existence probability.

The invention discloses a group target fine tracking method based on cooperative relation speed correction, and the method comprises the steps: firstly obtaining a random finite set model on the premise of supposing to track a dense group target, and building an intra-group multi-target tracking model; and through a generalized label multi-Bernoulli filtering algorithm, forming and predicting a hypothetical track of a dense target in the group, a hypothetical track weight and a target probability density, and updating the hypothetical track and a target state according to the obtained group target measurement. On the basis, a group target fine tracking method for performing speed correction based on a cooperative relationship is provided, and the overall movement speed estimation of the group target is used for updating the individual hypothesis track of the dense targets in the group, so that the phenomenon of track association errors of the targets in the group is remarkably reduced. According to the method, the tracking accuracy of the dense targets in the group in the complex environment is effectively improved, and reliable technical support can be provided for fine reconnaissance and attack application of the group targets.