CN-117761714-B - Target tracking method and device based on probability data association under infrared active interference condition

CN117761714BCN 117761714 BCN117761714 BCN 117761714BCN-117761714-B

Abstract

The invention discloses a target tracking method and device under an infrared active interference condition based on probability data association, which comprises the steps of constructing a model set, determining a target and an infrared active interference movement mode according to the model set, simulating an actual combat simulation scene, establishing a adhesion object measurement model and an infrared sensor error self-adaptive measurement model, estimating the actual size of an adhesion object, preprocessing measurement information of the adhesion object measurement model and the infrared sensor error self-adaptive measurement model, carrying out data association on a plurality of measurement information to obtain the probability of being associated with a track, processing the target movement model and the adhesion object measurement model and the infrared sensor error self-adaptive measurement model to carry out target movement state prediction, updating the target movement state prediction, and obtaining the state estimation and estimation error covariance of the target. The invention solves the problems of inaccurate target measurement and measurement quality uncertainty when the target and infrared active interference are adhered, and improves the measurement precision.

Inventors

LAN JIAN
XU HANCHI
GUO XIAOXIAO
HE JIAXING

Assignees

西安交通大学

Dates

Publication Date: 20260512
Application Date: 20231227

Claims (8)

1. The target tracking method under the infrared active interference condition based on probability data association is characterized by comprising the following steps of: constructing a model set comprising a target motion model and an infrared active interference model, determining a target and an infrared active interference motion mode according to the model set, and simulating an actual combat simulation scene; Based on the simulation scene, establishing a adhesion object measurement model and an infrared sensor error self-adaptive measurement model for a complex scene of the target interference adhesion; Estimating the actual size of the adhesion object according to the bullet distance and the line-of-sight opening angle in the actual combat scene, and preprocessing measurement information of the adhesion object measurement model and the infrared sensor error self-adaptive measurement model by adopting a wave gate self-adaptive method; carrying out data processing on a plurality of measurement information containing targets and interference in a gate threshold by adopting a probability data association algorithm to obtain the probability of association between measurement and a track; Processing the target motion model, the adhesion object measurement model and the infrared sensor error self-adaptive measurement model by adopting a nonlinear filtering algorithm to predict the target motion state, and updating the target motion state prediction; Establishing a adhesion object measurement model for a complex scene of target interference adhesion, comprising: the method comprises the steps of adhering a target with interference or covering the target with the interference to form an adhering object, taking the mass center of a boundary frame of the adhering object, four vertexes of the boundary frame and midpoints of four sides of the boundary frame as adhering measurement characteristics according to uncertainty of measurement of the target and interference information, increasing measurement number, selecting points which are closer to real target measurement in adhering measurement, and forming an adhering object measurement model; The method comprises the following steps of: In the formula, Is an infrared sensor Measuring the angle of the moment; Azimuth and pitch angles measured for the sensor; Is a position component of the target in the geodetic coordinate system; Measuring noise for azimuth angles; Measuring noise for pitch angle; measuring noise variance for azimuth; Measuring noise variance for pitch angle; for the angle of the line of sight of the bounding box, , For the measurement of the pitch angle of the diagonal vertices of the bounding box, , Azimuth angle measurement is performed for the diagonal vertices of the bounding box.
2. The method for tracking the target under the infrared active interference condition based on the probability data association according to claim 1, wherein a target motion model set is constructed by adopting a non-maneuvering target uniform motion model and a three-dimensional uniform turning target motion model; The non-motorized target uniform motion model is as follows: In the formula, Representation of The time of day target state is set, Representation of The position of the time object is in the coordinate system with components of the respective axes, Representation of The components of the speed of the moment target on each axis in the coordinate system; Representing a diagonal matrix; Is a system state transfer function; is system process noise; Is a noise driving matrix; is the sampling interval; the three-dimensional uniform turning target motion model is as follows: , In the formula, Representing the position components along three axes of the navigation system as the object moves in the three-dimensional spatial coordinate system, Representing the velocity components along three axes of the navigation system as the object moves in the three-dimensional spatial coordinate system, Representing the angular velocity components along three axes of the navigation system as the object moves in the three-dimensional spatial coordinate system.
3. The method for tracking the target under the condition of infrared active interference based on probability data association according to claim 1, wherein the method for constructing the infrared active interference model is as follows: In the formula, Is that A time target state; Representation of A time target state; Is a system state transfer function; is system process noise; For input of Is a gain matrix of (a); Is that Inputting a time system; is the sampling interval; is the wind resistance coefficient; is air density; gravitational acceleration; the windward area is an infrared radiation source; Is that The mass of the infrared radiation source at the moment; Is that The velocity of the infrared radiation source at the moment is a component of the respective axis in the coordinate system.
4. The method for tracking the target under the infrared active interference condition based on the probability data association according to claim 1, wherein the method for preprocessing the measurement information of the adhesion object measurement model and the infrared sensor error adaptive measurement model by adopting a wave gate adaptive method adopts the following formula: In the formula, Is the variance gain; A measurement received for the infrared sensor; Is one-step predictive measurement; As the standard deviation of the residual error, Is the gate threshold.
5. The method for tracking an object under an infrared active disturbance condition based on probability data correlation according to claim 1, wherein the probability associated with a track is measured The following are provided: In the formula, To measure Is a track The probability of an effective measurement hypothesis; To measure Is a track The probability of an effective measurement hypothesis of (1), l is the sum variable, N is the on-track A given number of measurements within the wave gate.
6. The method for tracking the target under the infrared active interference condition based on the probability data association according to claim 1, wherein the target motion state prediction is performed by adopting a nonlinear filtering algorithm to process a target motion model, a stuck object measurement model and an error adaptive measurement model, and the method comprises the following steps: a. and (3) carrying out prediction state estimation: In the formula, Is that A time of day prediction state estimate is made, Is that Estimating the motion state of a moment target; Is a system state function; b. calculating a one-step predictive value of the measured value: In the formula, A one-step predictive value for the measured value; is a measurement function.
7. The method for tracking the target under the condition of infrared active interference based on probability data association according to claim 1, wherein updating the prediction of the motion state of the target comprises: Equivalent innovation vector : In the formula, To measure Probability associated with the track; Is that Time measurement N is the number of measurement values given in the aviation wave gate; Numbering the measurements; target state estimation : In the formula, In order to predict the state estimate, Is that Time kalman gain; Is that An innovation vector equivalent to the moment; Estimation error covariance : In the formula, Representing the kalman covariance of a measurement return value when the measurement value is correctly associated with the track, The delta of the influence of the uncertainty association on the covariance.
8. A target tracking device in an infrared active interference condition based on probability data correlation according to any one of claims 1-7, comprising: The simulation module is used for establishing a model set, determining a target and infrared active interference movement mode according to the model set, and simulating an actual combat simulation scene; the construction module is used for establishing a adhesion object measurement model and an infrared sensor error self-adaptive measurement model for a complex scene of target interference adhesion; The preprocessing module is used for preprocessing measurement information of the adhesion object measurement model and the infrared sensor error self-adaptive measurement model by adopting a wave gate self-adaptive method; The association module is used for carrying out data association on a plurality of measurement information containing targets and interference in the gate threshold so as to obtain the probability of association between measurement and flight path; The state prediction updating module is used for processing the target motion model, the adhesion object measurement model and the infrared sensor error self-adaptive measurement model to predict the target motion state, updating the target motion state prediction and obtaining the state estimation and estimation error covariance of the target.

Description

Target tracking method and device based on probability data association under infrared active interference condition Technical Field The invention relates to a target tracking technology under an infrared active interference condition, in particular to a target tracking method and device under the infrared active interference condition based on probability data correlation. Background With the great use of infrared guided weapons, corresponding infrared interference techniques are continually evolving. Infrared guidance and infrared interference continue to evolve in this interrelated relationship. Infrared antagonism essentially begins in two ways. On one hand, the method is used for inhibiting infrared radiation emitted by a target, namely infrared stealth, and on the other hand, generating infrared active interference, and utilizing a stronger infrared radiation source emitted by an airplane to cause a false target to attract the missile to get off the target. In the existing research of infrared active interference resistant technology, the technology is concentrated in the signal processing layer and the image processing field, and the technology is used for resisting the radiation intensity, the wave band radiation, the terminal imaging and the like. And the current mainstream infrared active interference resistant algorithm mainly considers the situation after the separation of the target and the interference. In practical situations, however, there are a variety of complex infrared active interference phenomena. Because the infrared active interference is complex in form, the interference time, the interference frequency and the target motion state are different, various infrared interference scenes, such as overlapping of the infrared active interference and the target, and adhesion of the infrared active interference and the target into a plurality of complex situations such as a straight shape, a herringbone shape and the like, appear. Because the infrared active interference and the target have similar radiation characteristics, the infrared active interference and the target cannot be directly distinguished through gray scale on an infrared image, under the condition of adhesion between the target and the interference, measurement is seriously deteriorated, all measurement deviates from a real target, so that measurement accuracy is reduced, and aiming at the problems, the problems of initial infrared active interference and target tracking under the complex scene of the adhesion between the target and the interference are considered. Disclosure of Invention In order to solve the defects in the prior art, the invention aims to provide a target tracking method under the infrared active interference condition based on probability data association, which aims at solving the problems that the infrared active interference is complex in form, different in target and interference adhesion shape and uncertain in measurement quality. The invention is realized by the following technical scheme. According to one aspect of the invention, there is provided a target tracking method under an infrared active interference condition based on probability data correlation, comprising: constructing a model set comprising a target motion model and an infrared active interference model, determining a target and an infrared active interference motion mode according to the model set, and simulating an actual combat simulation scene; Based on the simulation scene, establishing a adhesion object measurement model and an infrared sensor error self-adaptive measurement model for a complex scene of the target interference adhesion; Estimating the actual size of the adhesion object according to the bullet distance and the line-of-sight opening angle in the actual combat scene, and preprocessing measurement information of the adhesion object measurement model and the infrared sensor error self-adaptive measurement model by adopting a wave gate self-adaptive method; carrying out data association on a plurality of measurement information containing targets and interference in a gate threshold by adopting a probability data association algorithm to obtain probability of association between measurement and a track; and processing the target motion model, the adhesion object measurement model and the infrared sensor error self-adaptive measurement model by adopting a nonlinear filtering algorithm to predict the target motion state, updating the target motion state prediction, and obtaining the state estimation and estimation error covariance of the target as a final filtering result. Preferably, the building of the adhesion object measurement model for the complex scene of the target interference adhesion includes: And (3) the object is stuck with or covered by the interference to form a stuck object, the mass center of the boundary frame of the stuck object, the four vertexes of the boundary frame and the midpoints of the four si