Interactive multi-model tracking method based on ellipse constraint and straight line constraint

Abstract

The invention belongs to the technical field of target constraint tracking, and discloses an interactive multi-model tracking method based on elliptical constraint and linear constraint. The method comprises the following steps: generating target motion data, and generating measurement data through a sensor; analyzing constraint conditions met by the elliptical orbit target motion and the linear motion target; establishing a tracking model; applying an interactive multi-model algorithm based on ellipse constraint and straight line constraint; and obtaining an observation result. According to the characteristics of elliptic track motion and linear track motion, an elliptic motion unscented Kalman filtering model and a linear motion unscented Kalman filtering model are designed, and elliptic constraint and linear constraint are introduced into interactive multi-model algorithm filtering. The tracking precision of a target with linear and elliptical track motion characteristics is improved; in addition, a powerful framework is established for introducing constraints into an interactive multi-model algorithm, and model selection of different target motion states is more targeted.

Description

Technical Field

[0001] The present invention mainly relates to the technical field of target constraint tracking, and in particular to an interactive multi-model tracking method based on ellipse constraints and straight line constraints. Background Art

[0002] Interactive multi-model is an adaptive algorithm with Markov chain transition probability. This algorithm assumes that the target motion follows a Markov chain with a known finite transition probability between different models, and considers the interaction of multiple models to estimate the state of the target. In actual target motion, the target often obeys specific constraints, such as the trajectory of vehicles on the road will follow the shape of the road, satellites obey satellite orbit constraints, and the flight path of a specific jammer is runway-shaped or horseshoe-shaped. Introducing constraint information into the interactive multi-model can improve the tracking accuracy of the target.

[0003] Interactive multi-models usually have multiple filters processed in parallel, each filter corresponds to a different state space, and different state spaces describe different target motion patterns. If the prior constraint information can be introduced into the filtering process, the accuracy of target tracking can be improved. Existing interactive multi-model algorithms have done a lot of research on different types of target maneuvers, but have failed to introduce the prior information of constraints into interactive multi-model algorithms.

[0004] Currently, the existing constraint information is mainly linear constraints, which are introduced into the tracking system as pseudo-measurements to improve the accuracy of the target. Linear constraints are too idealistic and are rarely used in actual application scenarios. Nonlinear constraints have constraint relationships that are difficult to accurately express with mathematical relationships, and there are problems such as filter divergence due to strong nonlinearity.

[0005] In the existing interactive multi-model algorithms, different filters process the same target dimension. The actual target motion may have multiple models, and different models have different dimensions. There is no more comprehensive and effective solution for filtering different target state dimensions and weighted fusion to obtain the final estimate. Summary of the invention

[0006] The purpose of the present invention is to provide an interactive multi-model tracking method based on elliptical constraints and straight line constraints. In view of the current lack of trajectory constraints to introduce interactive multi-models, the present invention is designed for scenarios where the target may perform elliptical and straight line motions, and introduces elliptical trajectory constraints and straight line trajectory constraints as pseudo-measurements into the filtering process of the interactive multi-model to improve the targeted selection of models for different motion states and the accuracy of target tracking precision.

[0007] To achieve the above object, the present invention provides an interactive multi-model tracking method based on ellipse constraint and straight line constraint, comprising the following steps: Step 1: Generate motion data of elliptical trajectory targets and straight trajectory targets, and generate measurement data through sensors; Step 2: Analyze the constraints satisfied by the elliptical trajectory target motion and obtain the elliptical trajectory constraint information; Step 3: Introduce the elliptical trajectory constraint information into the tracking system through the augmented measurement equation to establish the tracking model; Step 4: Analyze the constraints satisfied by the straight-line trajectory target motion and obtain the straight-line trajectory constraint information; Step 5: Introduce the linear trajectory constraint information into the tracking system through the augmented measurement equation to establish the tracking model; Step 6: Use the interactive multi-model algorithm based on ellipse constraints and straight line constraints; Step 7: Obtain observation results.

[0008] Furthermore, in step 2, the constraint conditions satisfied by the elliptical trajectory target motion are analyzed to obtain the elliptical trajectory constraint information, including the following steps: S1: Suppose a target moves in an elliptical trajectory in a two-dimensional plane, and the coordinates of the center of the ellipse are , the semi-major axis and semi-minor axis of the ellipse are a and b respectively, and the angle between the line connecting any point on the ellipse trajectory and the circle and the x-axis is , we can get the standard equation of the ellipse as , the parametric equation of the ellipse is ; S2: Deriving the standard equation and parametric equation of the ellipse respectively, the position constraint expression of the ellipse trajectory is obtained as follows: , the velocity constraint expression of the elliptical trajectory is , Indicates the angular velocity of the target moving along the elliptical trajectory; S3: Arrange the expressions in steps S1 and S2 to obtain the ellipse trajectory constraint information. The three expressions contained in the ellipse trajectory constraint information are: .

[0009] Furthermore, in step 3, the elliptical trajectory constraint information is introduced into the tracking system through the augmented measurement equation to establish a tracking model, which includes the following steps: S4: The constant speed turning motion model is used to describe the target moving along an elliptical trajectory. The target state vector expression under the constant speed turning motion model is: ; The target state equation is , is the state transfer equation, is the noise distribution equation; S5: Use the sensor to observe. The sensor can observe the distance and azimuth of the target, and the sensor measurement data is obtained as ; S6: By expanding the state dimension, the elliptical trajectory constraint information is introduced into the sensor measurement data expression in step S5, and the target state vector is expanded to , the target measurement equation is expanded to ; S7: For targets subject to elliptical trajectory constraints, the noise mean corresponding to the corresponding constraints is zero. Assuming that the target states after dimension expansion are uncorrelated, the corresponding noise covariance matrix is ​​diagonally augmented in the traditional noise covariance matrix. The augmented noise covariance matrix is , R k is the covariance matrix of observation distance noise and observation angle noise, O is a zero matrix, n is the number of constraints, is an n*n zero matrix.

[0010] Furthermore, in step 4, the constraint conditions satisfied by the straight-line trajectory target motion are analyzed to obtain the straight-line trajectory constraint information, including the following steps: S8: A target moving along a straight line in a two-dimensional plane can be represented by a slope and an intercept. The general equation of a straight line is ; S9: By differentiating the general equation of a line, we can obtain the velocity constraint equation for any point: ; S10: Arrange the equations in steps S8 and S9 to obtain the straight line trajectory constraint information. The straight line trajectory constraint information contains two formulas: .

[0011] Furthermore, in step 5, the straight line trajectory constraint information is introduced into the tracking system through the augmented measurement equation to establish a tracking model, which includes the following steps: S11: The uniform motion model is used to describe the target moving along a straight line trajectory. The target state vector expression under the uniform motion model is: ; S12: Expand the dimension of the target state vector expression. The target state vector is expanded to , the target measurement equation is expanded to ; S13: Assume that the target states after dimension expansion are uncorrelated, and the noise covariance matrix is ​​diagonally augmented in the traditional noise covariance matrix. The augmented noise covariance matrix is , R k is the covariance matrix of observation distance noise and observation angle noise, O is a zero matrix, n is the number of constraints, is an n*n zero matrix.

[0012] Furthermore, the interactive multi-model algorithm based on elliptical constraints and straight line constraints in step 6 includes interaction, filtering, model probability update, and state estimation fusion.

[0013] Further, is the state estimate of filter j at time k-1, is the state covariance transfer matrix, is the probability of model j at time k-1; The state estimates of the N filters after interaction are ; The state covariance transfer matrix of the N filters after interaction is .

[0014] Furthermore, the elliptical trajectory tracking model and the straight line estimation tracking model are independently processed by unscented Kalman filtering, and the unscented Kalman filtering processing includes the following steps: S14: Calculate 2nx+1 sampling points first and the corresponding weights , sampling point The expression is , the corresponding weight The expression is ; S15: Using the state equation to obtain the one-step prediction and weight of the sampling point, the state prediction estimate and state prediction covariance can be obtained. The expression of the state prediction estimate is: , the expression of state prediction covariance is ; S16: The update process is: ; ; ; ; ; ; ; ; Where n represents the state vector The dimension, j is calculated as ; Representation Matrix The j-th row or j-th column of , and the calculation formula related to the weight is ,in , and These are all empirical parameters of the sigma point in the unscented Kalman filter.

[0015] Furthermore, the model probability is updated, including the following steps: S17: If the filter residual of model j is , the corresponding covariance is , we can get the model probability of each filter output Updated to , where the filter residual of model j is and covariance The expression is ; S18: According to step S17, the update probability of model j is obtained as follows: .

[0016] Furthermore, the state estimation fusion step is to weight the state estimates of all models according to their latest model probabilities to form a comprehensive state estimate.

[0017] Beneficial effects: The present invention provides an interactive multi-model tracking method based on elliptical constraints and straight line constraints. According to the characteristics of elliptical trajectory motion and straight line trajectory motion, an elliptical motion unscented Kalman filter model and a straight line motion unscented Kalman filter model are designed. By introducing the elliptical constraint and the straight line constraint into the interactive multi-model algorithm filtering, the tracking accuracy of the target with the characteristics of straight line and elliptical trajectory motion is improved. In addition, the present invention establishes a powerful framework for introducing constraints into the interactive multi-model algorithm, and the model selection for different target motion states is more targeted. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 is a flow chart of an interactive multi-model tracking method based on ellipse constraints and straight line constraints according to an embodiment of the present invention; Figure 2 The target motion trajectory and tracking trajectory diagram involved in the embodiment of the present invention; Figure 3 is a target tracking position root mean square error curve diagram involved in an embodiment of the present invention; Figure 4 is a target tracking speed root mean square error curve diagram involved in an embodiment of the present invention; Figure 5 is a model probability graph of an interactive multi-model with constraints involved in an embodiment of the present invention; Figure 6 It is a conventional interactive multi-model model probability graph involved in the embodiment of the present invention. DETAILED DESCRIPTION

[0019] The preferred mechanism and implementation method of the present invention are further described below in conjunction with the accompanying drawings and specific implementation methods.

[0020] like Figures 1 to 6 As shown, an embodiment of the present invention discloses an interactive multi-model tracking method based on ellipse constraints and straight line constraints. Figure 1 It is a flow chart of an interactive multi-model tracking method based on ellipse constraints and straight line constraints involved in an embodiment of the present invention.

[0021] In this method, due to the different conditions of elliptical constraints and linear constraints, and the different numbers of constraint equations, the dimension of the state vectors of the two models is inconsistent after expansion. For this reason, the method adopted by the present invention is to reduce the dimension before fusion, that is, only retain the same state components in each model for fusion, and use the estimation results of a single model as the final estimation for other components. The estimation results of a single model are used to expand the mixed initial value before filtering. Example 1

[0022] An interactive multi-model tracking method based on ellipse constraints and straight line constraints comprises the following steps: Step 1: Generate motion data of elliptical trajectory targets and straight trajectory targets, and generate measurement data through sensors; Step 2: Analyze the constraints satisfied by the elliptical trajectory target motion and obtain the elliptical trajectory constraint information; S1: Suppose a target moves in an elliptical trajectory in a two-dimensional plane, and the coordinates of the center of the ellipse are , the semi-major axis and semi-minor axis of the ellipse are a and b respectively, and the angle between the line connecting any point on the ellipse trajectory and the circle and the x-axis is , we can get the standard equation of the ellipse as (Formula 1), the parametric equation of the ellipse is (Formula 2); S2: Deriving the standard equation and parametric equation of the ellipse respectively, the position constraint expression of the ellipse trajectory is obtained as follows: (Equation 3), the velocity constraint expression of the elliptical trajectory is (Formula 4), Indicates the angular velocity of the target moving along the elliptical trajectory; S3: Arrange the expressions in steps S1 and S2 to obtain the ellipse trajectory constraint information. The three expressions contained in the ellipse trajectory constraint information are: (Equation 5), the ellipse trajectory constraint includes the constraint relationship of position, velocity, angular velocity, ellipse center, and major and minor axes of the ellipse.

[0023] Step 3: Introduce the elliptical trajectory constraint information into the tracking system through the augmented measurement equation to establish the tracking model; S4: The constant speed turning motion model (CT model) is used to describe the target motion along an elliptical trajectory. The target state vector expression under the constant speed turning motion model is: (Equation 6), the state of traditional two-dimensional target tracking is position and velocity, and the target moves in an elliptical trajectory. At this time, the state vector of the target increases the angular velocity; The target state equation is (Formula 7), is the state transfer equation, is the noise distribution equation.

[0024] S5: Observe using a sensor. The present invention uses a radar. The sensor can observe the distance and azimuth of the target, and the sensor measurement data is obtained as follows: (Formula 8); S6: The pseudo-measurement method introduces the elliptical trajectory constraint information into the sensor measurement data expression in step S5 through state dimension expansion. The target state vector is expanded to (Equation 9), the target measurement equation is expanded to (Formula 10); S7: For targets subject to elliptical trajectory constraints, the noise mean corresponding to the corresponding constraints is zero. Assuming that the target states after dimension expansion are uncorrelated, the corresponding noise covariance matrix is ​​diagonally augmented in the traditional noise covariance matrix. The augmented noise covariance matrix is (Equation 11), the augmented noise covariance matrix is ​​a 5*5 diagonal matrix, R k is a 2*2 diagonal matrix, is the covariance matrix of the observation distance noise and the observation angle noise, O is a zero matrix, n is the number of constraints, is an n*n zero matrix.

[0025] Step 4: Analyze the constraints satisfied by the straight-line trajectory target motion and obtain the straight-line trajectory constraint information; S8: A target moving along a straight line in a two-dimensional plane can be represented by a slope and an intercept. To describe a straight line perpendicular to the x-axis, the general equation of a straight line can be: (Formula 12); S9: By differentiating the general equation of a line, we can obtain the velocity constraint equation for any point: (Formula 13); S10: Arrange the equations in steps S8 and S9 to obtain the straight line trajectory constraint information. The straight line trajectory constraint information contains two formulas: (Equation 14), the straight line constraint equation constrains the position and velocity relationship of any point on the straight line by introducing three additional parameters.

[0026] Step 5: Introduce the linear trajectory constraint information into the tracking system through the augmented measurement equation to establish the tracking model; S11: The uniform motion model (CV model) is used to describe the target moving along a straight line trajectory. The target state vector expression under the uniform motion model is: (Formula 15); S12: Expand the dimension of the target state vector expression. After the target state vector is expanded, it becomes (Equation 16), the target measurement equation is expanded to (Formula 17); S13: Assume that the target states after dimension expansion are uncorrelated, and the noise covariance matrix is ​​diagonally augmented in the traditional noise covariance matrix. The augmented noise covariance matrix is , R k is the covariance matrix of observation distance noise and observation angle noise, O is a zero matrix, n is the number of constraints, is an n*n zero matrix.

[0027] Step 6: Use the interactive multi-model algorithm based on ellipse constraints and straight line constraints; The interactive multi-model algorithm based on elliptical constraints and straight line constraints in step 6 includes four parts: interaction, filtering, model probability update, and state estimation fusion.

[0028] Part I Interaction Interaction occurs between models, and its purpose is to reasonably distribute the state estimates of each model at the previous moment to each model at the current moment. This process involves probability mixing between models, thereby creating a mixed state estimate and covariance for each model. Specifically, based on the model transition probability of each model and the model probability of the previous step, a mixing probability is calculated, and then this probability is used to weighted average the state and error covariance of each model to generate a new initial condition.

[0029] make is the state estimate of filter j at time k-1, is the state covariance transfer matrix, is the probability of model j at time k-1; The state estimates of the N filters after interaction are (Formula 18); The state covariance transfer matrix of the N filters after interaction is (Formula 19).

[0030] Part 2 Filtering Each model is filtered independently. The filtering used in this method is unscented Kalman filtering. The unscented Kalman filtering process includes the following steps: S14: First calculate 2nx+1 sampling points and the corresponding weights , sampling point The expression is (Equation 20), the corresponding weights The expression is (Formula 21); S15: Using the state equation to obtain the one-step prediction and weight of the sampling point, the state prediction estimate and state prediction covariance can be obtained. The expression of the state prediction estimate is: (Equation 22), the expression of state prediction covariance is (Formula 23); S16: The update process is: (Formula 24); (Formula 25); (Formula 26); (Formula 27); (Formula 28); (Formula 29); (Formula 30); (Formula 31); Where n represents the state vector The dimension, j is calculated as ; Representation Matrix The j-th row or j-th column of , and the calculation formula related to the weight is ,in , and These are all empirical parameters of the sigma point in the unscented Kalman filter.

[0031] Part III Model Probability Update Model probability update is to update the probability of each model after obtaining the filtering results of each model. This step is based on the possibility of each model generating observation data (calculated by likelihood function) and the previous model probability. The probability of each model is updated through the Bayesian rule and the support of each model by the current observation.

[0032] S17: If the filter residual of model j is , the corresponding covariance is , we can get the model probability of each filter output Updated to (Equation 32), where the filter residual of model j is and covariance The expression is (Formula 33); S18: According to step S17, the update probability of model j is obtained as follows: (Formula 34).

[0033] Part IV State Estimation State estimation fusion takes the weighted average of the state estimates of all models according to their latest model probabilities to form a comprehensive state estimate. This comprehensive state integrates the information of all possible models to the greatest extent and provides an optimal estimate that comprehensively considers all possible situations.

[0034] Step 7: Obtain observation results. Example 2

[0035] The interactive multi-model tracking method based on ellipse constraint and straight line constraint in this embodiment is basically the same as that in embodiment 1, except that: Assume that the motion trajectory of a target conforms to the elliptical trajectory and the straight line trajectory, such as Figure 1 As shown in the figure, the target's motion trajectory is generated first. The position and velocity information of the target is generated in one dimension, and then the one-dimensional information is mapped to a two-dimensional trajectory. The one-dimensional motion noise variance of the target is 0.5 meters. The first part of the trajectory is an elliptical trajectory, and the second part is a straight trajectory. The shape of the target's trajectory is as follows: Figure 2 As shown. The long semi-axis of the elliptical trajectory is 1000 meters, and the short semi-axis is 700 meters. Assume that the sensor used is a radar, the radar is located at the origin of the coordinate system, the radar's distance observation noise variance is 5 meters, and the azimuth observation noise variance is 0.05 degrees. The radar's scanning interval is 0.1 seconds, the simulation time is 300 seconds, and 40 Monte Carlo simulations are performed.

[0036] In order to evaluate the accuracy of the interactive multi-model algorithm with or without constraints, the root mean square error is used to measure the deviation between the true value and the observed value. The calculation formula of the root mean square error (RMSE) is as follows: (Equation 35), where n is the observation step length, is the target true value, Predicted value for the sensor.

[0037] like Figure 3 , Figure 4As shown, the root mean square error of the interactive multiple model algorithm with constraints is smaller than that of the conventional interactive multiple model algorithm, which proves that the interactive multiple model algorithm with constraints has higher accuracy.

[0038] In order to evaluate the impact of constraints on model selection in interactive multiple models, Figure 5 , Figure 6 The probabilities of the two models during the observation process are plotted. When the target moves in an elliptical trajectory, the probability of the model of constant-speed circular motion in the constrained interactive multi-model is high, and the probability of the uniform-speed model in linear motion is high. There is no significant difference between the two model probabilities of the conventional interactive multi-model. It is proved that the interactive multi-model tracking method with elliptical constraints and linear constraints proposed in the present invention can effectively improve the tracking accuracy of the target under elliptical trajectory and linear motion and the accuracy of model selection.

[0039] The present invention provides an interactive multi-model tracking method based on elliptical constraints and straight line constraints. According to the characteristics of elliptical trajectory motion and straight line trajectory motion, an elliptical motion unscented Kalman filter model and a straight line motion unscented Kalman filter model are designed. By introducing the elliptical constraint and the straight line constraint into the interactive multi-model algorithm filtering, the tracking accuracy of the target with the straight line and elliptical trajectory motion characteristics is improved. In addition, the present invention establishes a powerful framework for introducing constraints into the interactive multi-model algorithm, and the model selection for different target motion states is more targeted.

[0040] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or make equivalent substitutions for some of the technical features therein. However, any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. An interactive multi-model tracking method based on ellipse constraints and straight line constraints, characterized in that: The following steps are involved: Step 1: Generate motion data of elliptical trajectory targets and straight trajectory targets, and generate measurement data through sensors; Step 2: Analyze the constraints satisfied by the elliptical trajectory target motion and obtain the elliptical trajectory constraint information; Step 3: Introduce the elliptical trajectory constraint information into the tracking system through the augmented measurement equation to establish the tracking model; Step 4: Analyze the constraints satisfied by the straight-line trajectory target motion and obtain the straight-line trajectory constraint information; Step 5: Introduce the linear trajectory constraint information into the tracking system through the augmented measurement equation to establish the tracking model; Step 6: Use the interactive multi-model algorithm based on ellipse constraints and straight line constraints; Step 7: Obtain observation results.

2. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 1, characterized in that: In step 2, the constraint conditions satisfied by the elliptical trajectory target motion are analyzed to obtain the elliptical trajectory constraint information, which includes the following steps: S1: Suppose a target moves in an elliptical trajectory in a two-dimensional plane, and the coordinates of the center of the ellipse are , the semi-major axis and semi-minor axis of the ellipse are a and b respectively, and the angle between the line connecting any point on the ellipse trajectory and the circle and the x-axis is , we can get the standard equation of the ellipse as , the parametric equation of the ellipse is ; S2: Deriving the standard equation and parametric equation of the ellipse respectively, the position constraint expression of the ellipse trajectory is obtained as follows: , the velocity constraint expression of the elliptical trajectory is , Indicates the angular velocity of the target moving along the elliptical trajectory; S3: Arrange the expressions in steps S1 and S2 to obtain the ellipse trajectory constraint information. The three expressions contained in the ellipse trajectory constraint information are: .

3. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 2, characterized in that: In step 3, the elliptical trajectory constraint information is introduced into the tracking system through the augmented measurement equation to establish the tracking model. The following steps are involved: S4: The constant speed turning motion model is used to describe the target moving along an elliptical trajectory. The target state vector expression under the constant speed turning motion model is: ; The target state equation is , is the state transfer equation, is the noise distribution equation; S5: Use the sensor to observe. The sensor can observe the distance and azimuth of the target, and the sensor measurement data is obtained as ; S6: By expanding the state dimension, the elliptical trajectory constraint information is introduced into the sensor measurement data expression in step S5, and the target state vector is expanded to , the target measurement equation is expanded to ; S7: For targets subject to elliptical trajectory constraints, the noise mean corresponding to the corresponding constraints is zero. Assuming that the target states after dimension expansion are uncorrelated, the traditional noise covariance matrix is ​​diagonally augmented. The augmented noise covariance matrix is , R k is the covariance matrix of observation distance noise and observation angle noise, O is a zero matrix, n is the number of constraints, is an n*n zero matrix.

4. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 1, characterized in that: In step 4, the constraint conditions satisfied by the straight-line trajectory target motion are analyzed to obtain the straight-line trajectory constraint information, which includes the following steps: S8: A target moving along a straight line in a two-dimensional plane can be represented by a slope and an intercept. The general equation of a straight line is ; S9: By differentiating the general equation of a line, we can obtain the velocity constraint equation for any point: ; S10: Arrange the equations in steps S8 and S9 to obtain the straight line trajectory constraint information. The straight line trajectory constraint information contains two formulas: .

5. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 4, characterized in that: In step 5, the straight line trajectory constraint information is introduced into the tracking system through the augmented measurement equation to establish a tracking model, which includes the following steps: S11: The uniform motion model is used to describe the target moving along a straight line trajectory. The target state vector expression under the uniform motion model is: ; S12: Expand the dimension of the target state vector expression. The target state vector is expanded to , the target measurement equation is expanded to ; S13: Assume that the target states after dimension expansion are uncorrelated, and the noise covariance matrix is ​​diagonally augmented in the traditional noise covariance matrix. The augmented noise covariance matrix is , R k is the covariance matrix of observation distance noise and observation angle noise, O is a zero matrix, n is the number of constraints, is an n*n zero matrix.

6. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 1, characterized in that: The interactive multi-model algorithm based on elliptical constraints and straight line constraints in step 6 includes interaction, filtering, model probability update, and state estimation fusion.

7. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 6, characterized in that: make is the state estimate of filter j at time k-1, is the state covariance transfer matrix, is the probability of model j at time k-1; The state estimates of the N filters after interaction are ; The state covariance transfer matrix of the N filters after interaction is .

8. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 6, characterized in that: The elliptical trajectory tracking model and the straight line estimation tracking model are independently processed by unscented Kalman filtering, and the unscented Kalman filtering process includes the following steps: S14: Calculate 2nx+1 sampling points first and the corresponding weights , sampling point The expression is , the corresponding weight The expression is ; S15: Using the state equation to obtain the one-step prediction and weight of the sampling point, the state prediction estimate and state prediction covariance can be obtained. The expression of the state prediction estimate is: , the expression of state prediction covariance is ; S16: The update process is: ； ； ； ； ； ； ； ； Where n represents the state vector The dimension, j is calculated as ; Representation Matrix The j-th row or j-th column of , and the calculation formula related to the weight is ,in , and These are all empirical parameters of the sigma point in the unscented Kalman filter.

9. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 6, characterized in that: Model probability update, including the following steps: S17: If the filter residual of model j is , the corresponding covariance is , we can get the model probability of each filter output Updated to , where the filter residual of model j is and covariance The expression is ; S18: According to step S17, the update probability of model j is obtained as follows: .

10. The interactive multi-model tracking method based on ellipse constraint and straight line constraint according to claim 6, characterized in that: The step of state estimation fusion is to weight the state estimates of all models according to their latest model probabilities to form a comprehensive state estimate.

Images