A passive sonar target location method based on dual sensors

Abstract

The present application relates to underwater acoustic equipment design and manufacturing technical field, especially a kind of passive sonar target positioning method based on double sensor, comprising the following steps: step 1, the coordinate information of all candidate targets relative to each array at t1 Time is calculated;Suppose the coordinate of a certain array is (x, y), the motion equation of the target about orientation is calculated according to the orientation (θ) of the target y=θ*x+b;The motion equation of all targets about a certain array is calculated, and the intersection point is solved by combining two equations;Step 2, initialize parameters: parameters include: the number of tracking ants required for tracking and positioning, the initial value of objective function, convergence threshold and divergence coefficient;Step 3, randomly place tracking ants on candidate targets at t1 Time.The method of the present application simulates ant colony optimization algorithm, makes reasonable assumption on the motion of target, i.e. assuming uniform linear motion of target in a short time, and the positioning problem of target is abstracted as optimal path problem.

Description

Technical Field

[0001] This invention relates to the field of underwater acoustic equipment design and manufacturing technology, and in particular to a passive sonar target localization method based on dual sensors. Background Technology

[0002] Passive sonar systems have attracted much attention due to their excellent stealth and anti-jamming capabilities, as well as the unique characteristics of the propagation medium in the marine environment. However, since passive sonar does not emit detection signals but relies on the target's own radiated noise for detection, a single array, under fixed conditions, can only obtain the target's azimuth information, which cannot meet the target localization requirements. Compared to single-element detection, multi-array joint detection has significant advantages. Multi-array systems are characterized by strong anti-jamming capabilities and high survivability. Especially in a fixed-element environment, multiple arrays can perform mathematical cross-localization using the obtained target azimuth to obtain the target's coordinate information. However, the target coordinate information obtained by cross-localization includes false target coordinates, see... Figure 1 Effectively removing fake targets remains a challenge.

[0003] Based on the above description, this invention proposes an optimized method for removing false targets to obtain the coordinates of the true targets. Digital signal processing is used to process the array signal to achieve superior beam performance. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention discloses a passive sonar target localization method based on dual sensors. This method mimics the ant colony optimization algorithm, making reasonable assumptions about the target's motion, specifically assuming uniform linear motion over a short period. The target localization problem is thus abstracted into an optimal path problem. This invention uses a dual-array, dual-target approach, with three consecutive frames as a short time interval, to converge the optimal path. Targets not on the optimal path are ultimately identified as false targets.

[0005] This invention is achieved through the following technical solution:

[0006] A passive sonar target localization method based on dual sensors includes the following steps:

[0007] Step 1: Calculate the coordinate information of all candidate targets relative to each matrix at time t1;

[0008] Assuming the coordinates of a certain array are known to be (x, y), calculate the target's motion equation about the azimuth (θ) as y = θ*x + b; calculate the motion equations of all targets about a certain array, and solve for the intersection point of the equations pairwise;

[0009] The intersection points are the candidate targets (x1, y1), (x2, y2), (x3, y3), (x4, y4); similarly, the coordinates of the candidate targets at times t2, t3, etc. are calculated.

[0010] Step 2, Initialize parameters: Parameters include: the number of tracking ants required for tracking and localization, the initial value of the objective function, the convergence threshold, and the divergence coefficient;

[0011] Step 3: Randomly place the tracking ants on the candidate targets at time t1, and calculate the probability that each tracking ant will track a candidate target j at the next time, i.e., formula (2); add up the probabilities that each tracking ant will visit node j. If the sum of probabilities is greater than a random value between (0,1), then j becomes a candidate node; record the nodes that all ants will visit at the current time in this way.

[0012] Step 4: Update pheromones: Update pheromones according to formula (3);

[0013] Step 5: Perform single-step optimization for each objective, that is, find the node combination that minimizes the objective function and find the optimal solution for this iteration; if the convergence domain of the objective function is less than the set value, stop the iteration; otherwise, go back to step 3 and start again.

[0014] Preferably, step 2 further includes defining the objective function ARG:

[0015] Define the objective function:

[0016]

[0017] in, Given the set of tracks that the tracking ants can traverse, for the first T time window, under the condition of a dual array, there are a total of (N) 2 ) T There are 3 paths, where N is the number of nodes that each tracking ant needs to traverse in each iteration, and T is the selected time length. For distance difference, The track l represents the trajectory between the t-th and t-1-th sampling intervals. n The objective function ARG represents the sum of the distance differences between nodes traversed by each tracking ant.

[0018] Preferably, step 3 further includes the definition of the heuristic factor and the probability of selecting a node;

[0019] Define heuristic factors: Define heuristic factors It is the reciprocal of the distance between two candidate targets in adjacent time intervals;

[0020] Define the probability of selecting a node:

[0021]

[0022] Where ω k (t) represents the weight of the k-th objective, τ ij k (t) represents the pheromone.

[0023] Preferably, step 4 further includes updating the pheromones, and the updating method is as follows:

[0024] τ ij k (t,n)=(1-ρ)τ ij k (t)+ρτ0 (3)

[0025] Where ρ is the volatility coefficient, τ0 is the initial information content, and n is the nth tracking ant.

[0026] Preferably, the algorithm input is the target's orientation information obtained by each array, assuming that the target moves at a constant speed within a short period of time (less than 6 seconds).

[0027] The present invention has the following beneficial effects:

[0028] Existing passive sonar multi-target localization techniques often utilize measurement information from previous time steps to construct auxiliary measurements for the current time step, and these two measurements exhibit a weak correlation. Therefore, the tracking and localization results are significantly affected by the tracking environment. This invention, however, combines multi-array, multi-target tracking and localization with path optimization, which reduces the impact of the tracking environment on localization effectiveness and also reduces computational load. This algorithm has the following advantages:

[0029] 1) The algorithm has few parameters, is easy to set up, has low computational load, low time and space complexity, and low hardware requirements.

[0030] 2) The target tracking and localization problem is combined with the optimization problem to reduce the impact of the tracking scenario on the results.

[0031] 3) Multi-array joint positioning improves the observability of the target. Attached Figure Description

[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0033] Figure 1This is a schematic diagram of the cross-positioning of the present invention; where 1 and 4 are real targets, and 2 and 3 are false targets.

[0034] Figure 2 This is a simulation diagram of the positioning of the present invention; where the black solid line A is the actual trajectory of target 1, the red solid line B is the actual trajectory of target 2, the red point C is the simulated trajectory of target 1 after the cross positioning of the present invention, and the black point D is the simulated trajectory of target 2 after the cross positioning of the present invention. Detailed Implementation

[0035] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0036] While passive sonar offers excellent stealth capabilities, its ranging performance is relatively poor. In real-world environments, precise target coordinates are often required. Multi-array joint localization obtains target coordinates through azimuth cross-referencing, but this can introduce false targets. This invention proposes an improved ant colony optimization algorithm that transforms the problem of removing false targets into finding the optimal path, reducing computational complexity and improving localization accuracy. This provides a solution for passive sonar target localization.

[0037] like Figures 1-2 This invention provides a passive sonar target localization method based on dual sensors. The algorithm input values ​​are the target's azimuth information acquired by each array. It is assumed that the target moves at a constant speed within a short period of time (less than 6 seconds).

[0038] The first step is to define the necessary parameters.

[0039] Define the objective function:

[0040]

[0041] in, Given the set of tracks that the tracking ants can traverse, for the first T time window, under the condition of a dual array, there are a total of (N) 2 ) T There are 3 paths, where N is the number of nodes that each tracking ant needs to traverse in each iteration, and T is the selected time length. For distance difference, The track l represents the trajectory between the t-th and t-1-th sampling intervals. n The objective function ARG represents the sum of the distance differences between nodes traversed by each tracking ant.

[0042] Defining Heuristic Factors: This invention defines heuristic factors. It is the reciprocal of the distance between two candidate targets in adjacent time intervals.

[0043] Define the probability of selecting a node:

[0044]

[0045] Where ω k (t) represents the weight of the k-th objective, τ ij k (t) represents the pheromone. To prevent premature convergence of the solution, the pheromone is updated as follows:

[0046] τ ij k (t,n)=(1-ρ)τ ij k (t)+ρτ0 (3)

[0047] Where ρ is the volatility coefficient, τ0 is the initial information content, and n is the nth tracking ant.

[0048] The specific implementation process is as follows:

[0049] (1) Calculate the coordinate information of all candidate targets relative to each array at time t1. The calculation process is illustrated using a specific array as an example; the same logic applies to other arrays. Given the array coordinates (x, y), calculate the target's motion equation y = θ*x + b based on the target's azimuth (θ). Calculate the motion equations of all targets relative to a given array. Solve for the intersection points of the equations pairwise; these intersection points represent the candidate targets (x1, y1), (x2, y2), (x3, y3), and (x4, y4), as shown below. Figure 1 As shown. Similarly, the coordinates of the candidate targets at times t2, t3... are calculated.

[0050] (2) Initialization parameters: the number of tracking ants required for tracking and positioning, the initial value of the objective function, the convergence threshold, the divergence coefficient, and other parameters.

[0051] (3) Randomly place the tracking ants on the candidate targets at time t1, and calculate the probability that each tracking ant will track a candidate target j at the next time step, i.e., formula (2). Add up the probabilities that each tracking ant will visit node j. If the sum of the probabilities is greater than a random value between (0,1), then j becomes a candidate node. Record the nodes that all ants will visit at the current time step in this way.

[0052] (4) Update pheromones. Update pheromones according to formula (3).

[0053] (5) Perform single-step optimization for each objective, that is, find the node combination that minimizes the objective function and find the optimal solution for this iteration. If the convergence domain of the objective function is less than the set value, stop the iteration. Otherwise, go to step 3 and start again.

[0054] Simulation data verification:

[0055] The simulation uses two arrays and two targets as an example. Array 1 has coordinates (-1000, 0), and array 2 has coordinates (1000, 0). Target 1 moves parallel to the x-axis from its initial position (-150, 150) in the positive x-direction at a speed of 30 m / s. Target 2 starts at (0, 50) at a 45-degree angle to the coordinate axis and moves towards the first quadrant at a speed of 30 m / s. The initial distance difference is 0, the objective function threshold is 1, the maximum number of iterations is 100, the number of tracking ants is 4, the simulation duration is 30 seconds, and the time interval is 2 seconds. The simulation results are as follows: Figure 2 As shown in the figure, this method can effectively remove false targets and achieve stable tracking with a tracking accuracy of 87.4%.

[0056] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A passive sonar target localization method based on dual sensors, characterized in that, Includes the following steps: Step 1: Calculate the coordinate information of all candidate targets relative to each matrix at time t1; Assuming the coordinates of a certain array are known to be (x, y), calculate the target's motion equation about the azimuth (θ) as y = θ*x + b; Calculate the motion equations of all targets with respect to a given matrix, and solve for the intersection points of the equations pairwise. The intersection points are the candidate targets (x1, y1), (x2, y2), (x3, y3), (x4, y4); similarly, the coordinates of the candidate targets at times t2, t3, etc. are calculated. Step 2, Initialize parameters: Parameters include: the number of tracking ants required for tracking and localization, the initial value of the objective function, the convergence threshold, and the divergence coefficient; Step 3: Randomly place the tracking ants on the candidate targets at time t1, and calculate the probability that each tracking ant will track a candidate target j at the next time, i.e., formula (2); add up the probabilities that each tracking ant will visit node j. If the sum of probabilities is greater than a random value between (0,1), then j becomes a candidate node; record the nodes that all ants will visit at the current time in this way. Step 4: Update pheromones: Update pheromones according to formula (3); Step 5: Perform single-step optimization for each objective, that is, find the node combination that minimizes the objective function and find the optimal solution for this iteration; if the convergence threshold of the objective function is less than the set value, stop the iteration; otherwise, go back to step 3 and start again. Step 3 also includes the definition of the heuristic factor and the probability of selecting a node; Define heuristic factors: Define heuristic factors It is the reciprocal of the distance between two candidate targets in adjacent time intervals; Define the probability of selecting a node: Where ω k (t) represents the weight of the k-th objective, τ ij k (t) represents the pheromone; Step 4 also includes updating the pheromones, which is done as follows: t ij k (t,n)=(1-ρ)τ ij k (t)+ρτ0 (3) Where ρ is the volatility coefficient, τ0 is the initial information content, and n is the nth tracking ant.

2. The passive sonar target localization method based on dual sensors as described in claim 1, characterized in that, Step 2 also includes defining the objective function ARG: Define the objective function: in, Given the set of tracks that the tracking ants can traverse, for the first T time window, under the condition of a dual array, there are a total of (N) 2 ) T There are 3 paths, where N is the number of nodes that each tracking ant needs to traverse in each iteration, and T is the selected time length. For distance difference, The track l represents the trajectory between the t-th and t-1-th sampling intervals. n The objective function ARG represents the sum of the distance differences between nodes traversed by each tracking ant.

3. The passive sonar target localization method based on dual sensors as described in claim 1, characterized in that, The algorithm input is the target's orientation information obtained by each array, assuming that the target moves at a constant speed within a short period of time (less than 6 seconds).

Images