Underwater Target Motion Analysis Method and Device

Abstract

The present application relates to a method and device for analyzing the movement of underwater targets. The method includes: using MCMC to sample each parameter of the state vector of the underwater target, and obtaining the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling results; determining the value of the state vector under the current sampling according to the posterior probability density ratio; counting the values of the state vectors corresponding to each sampling in a preset number of samplings, and determining the final estimated value of the state vector of the underwater target according to the statistical results. By sampling each parameter of the state vector of the underwater target through the MCMC sampling method, and determining the value of the state vector by combining the posterior probability density ratio of the state vector after sampling, and then obtaining the final estimated value of the state vector through the statistical analysis of the state vectors corresponding to each sampling in the determined preset number of samplings, it can effectively achieve the state estimation of underwater targets under weak observation conditions.

Description

Technical Field

[0001] The present disclosure relates to the technical field of underwater target motion analysis, and in particular, to an underwater target motion analysis method and apparatus. Background Art

[0002] Underwater target motion analysis is to solve how to estimate the motion elements of an underwater target from the information about the target detected by a detector. To achieve the purpose of covert detection, a common detector is in a passive detection mode, and the information about the detected target usually only includes azimuth information. Therefore, underwater target motion analysis usually estimates the state information of the target, such as distance and speed, from the detected azimuth information.

[0003] In the related art, underwater target motion analysis usually estimates the state vector of the target by iteratively solving the maximum likelihood estimation method. However, the estimability of the target state vector has the characteristics of the Cramer-Rao lower bound. The maximum likelihood estimation method is asymptotically efficient when solving the underwater target motion analysis problem. When the maneuver amplitude of the observer is lower than a certain value, that is, when the observer is in weak observation conditions, the Cramer-Rao lower bound coefficient will increase sharply, making the underwater target motion analysis problem non-estimable. Summary of the Invention

[0004] In view of this, the present disclosure provides an underwater target motion analysis method, which realizes the analysis of underwater target motion even under weak observation conditions.

[0005] According to one aspect of the present disclosure, there is provided an underwater target motion analysis method, including:

[0006] Sampling each parameter of the state vector of the underwater target by using MCMC, and obtaining the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling result;

[0007] Determining the value of the state vector under the current sampling according to the posterior probability density ratio;

[0008] Counting the values of the state vector corresponding to each sampling in a preset number of samplings, and determining the final state vector estimation value of the underwater target according to the statistical result.

[0009] In a possible implementation manner, when determining the final state vector estimation value of the underwater target according to the statistical result, it includes:

[0010] Counting the value with the highest frequency of occurrence of each parameter in the state vector in a preset number of samplings as the final state vector estimation value.

[0011] In a possible implementation, the state vector parameters include at least one of the distance r of the underwater target from the observation platform, the measured azimuth θ of the underwater target, the speed v of the underwater target, and the angle α between the underwater target and the horizontal direction.

[0012] In a possible implementation, when using MCMC to sample the state vector parameters of the underwater target, sampling is performed based on the prior conditions of the underwater target;

[0013] The prior conditions include at least one of the upper and lower limits of the speed of the underwater target; the upper and lower limits of the distance between the underwater target and the observation platform; the interval of the measured azimuth of the underwater target, and the upper and lower limits of the angle between the underwater target and the horizontal.

[0014] In a possible implementation, when obtaining the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling result, it is calculated based on the prior probability distribution of the state vector of the underwater target under the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target under the current sampling.

[0015] In a possible implementation, when calculating the posterior probability density ratio based on the prior probability distribution of the state vector of the underwater target under the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target under the current sampling, it further includes:

[0016] Combining the likelihood estimate of the state vector of the underwater target under the previous sampling of the current sampling and the likelihood estimate of the state vector of the underwater target under the previous sampling of the current sampling to calculate the posterior probability density ratio.

[0017] In a possible implementation, the prior probability distribution of the state vector of the underwater target under the current sampling is obtained by successively performing product operations on the probability distributions of the parameters in the state vector according to the independence of the prior probability distribution.

[0018] In a possible implementation, when calculating the posterior probability density ratio by combining the likelihood estimate of the state vector of the underwater target under the previous sampling of the current sampling and the likelihood estimate of the state vector of the underwater target under the previous sampling of the current sampling, it is calculated by the following formula:

[0019]

[0020] Where, is the likelihood estimate of the state vector at the previous sampling of the current subsampling; p(r, θ, v, α) (i-1) is the prior probability distribution of the state vector at the previous sampling of the current subsampling, is the likelihood estimate of the state vector at the current subsampling, p(r, θ, v, α) (i) is the prior probability distribution of the state vector at the current subsampling;

[0021] r is the distance between the underwater target and the observation platform, θ is the measured azimuth of the underwater target, v is the speed of the underwater target, and α is the angle between the underwater target and the horizontal direction.

[0022] In a possible implementation manner, when determining the value of the state vector at the current subsampling according to the posterior probability density ratio, it includes:

[0023] Comparing the posterior probability density ratio with the generated random number;

[0024] When the posterior probability density ratio is greater than or equal to the random number, using the sampling value of the current subsampling as the value of the state vector;

[0025] When the posterior probability density ratio is less than the random number, using the sampling value of the previous sampling of the current subsampling as the value of the state vector.

[0026] According to another aspect of the present disclosure, there is also provided an underwater target motion analysis device, including: a sampling module, a probability density ratio calculation module, a state vector determination module, and a state vector estimation module;

[0027] The sampling module is configured to sample each parameter of the state vector of the underwater target by using MCMC;

[0028] The probability density ratio calculation module is configured to obtain the posterior probability density ratio of the state vector of the underwater target at the current subsampling according to the sampling result;

[0029] The state vector determination module is configured to determine the value of the state vector at the current subsampling according to the posterior probability density ratio;

[0030] The state vector estimation module is configured to count the values of the state vector corresponding to each sampling in a preset number of samplings, and determine the final state vector estimation value of the underwater target according to the statistical result.

[0031] By sampling each parameter of the state vector of the underwater target through the MCMC sampling method, and determining the value of the state vector by combining the posterior probability density ratio of the state vector after sampling, and then obtaining the final state vector estimation value through the statistical analysis of the state vector corresponding to each sampling in the preset number of samplings, compared with the method of using the maximum likelihood estimation method in the related technology, it can effectively realize the underwater target state estimation under weak observation conditions.

[0032] Other features and aspects of the present disclosure will become apparent from the following detailed description of exemplary embodiments with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are included in and constitute a part of this specification, illustrate exemplary embodiments, features, and aspects of the present disclosure and are used to explain the principles of the present disclosure together with the specification.

[0034] Figure 1 A schematic diagram showing a scenario of underwater target motion analysis according to an embodiment of the present application;

[0035] Figure 2 A flowchart showing a method for underwater target motion analysis according to an embodiment of the present application;

[0036] Figure 3 A diagram showing the motion trend of an observer and a target during a simulation test using the method for underwater target motion analysis according to an embodiment of the present application;

[0037] Figure 4 A schematic diagram showing the sampling values obtained by sampling the motion elements of a target using the method for underwater target motion analysis according to an embodiment of the present application and the Bayesian interval estimation results of the target state;

[0038] Figure 5 A structural block diagram showing an apparatus for underwater target motion analysis according to an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0039] Various exemplary embodiments, features, and aspects of the present disclosure will be described in detail below with reference to the accompanying drawings. The same reference numerals in the drawings denote elements having the same or similar functions. Although various aspects of the embodiments are shown in the drawings, the drawings are not necessarily drawn to scale unless otherwise specified.

[0040] The term "exemplary" used herein means "serving as an example, embodiment, or illustration". Any embodiment described herein as "exemplary" is not necessarily to be construed as superior to or better than other embodiments.

[0041] In addition, for better illustration of the present disclosure, numerous specific details are given in the following detailed implementation manners. Those skilled in the art should understand that the present disclosure can be implemented without some specific details. In some instances, methods, means, elements, and circuits well-known to those skilled in the art are not described in detail to highlight the gist of the present disclosure.

[0042] First of all, it should be noted that, referring to Figure 1 , the scenario where the underwater target motion analysis method provided by this application conducts underwater target motion analysis can be as Figure 1 shown. When using the underwater target motion analysis method of the embodiments of this application to perform motion analysis on the detected underwater target, it is assumed that the observer's motion is in a two-dimensional plane. The initial position of the underwater target is determined by the initial distance r0 and azimuth θ0, and it moves at a fixed speed v in a direction with an angle α to the horizontal. The target track is represented by the state vector (r, θ, v, α). Let be the noisy azimuth measurement at times {τ1, τ2, …, τ N}}, and obeys independent and identically distributed Gaussian noise with a mean of θ k and a variance of σ 2 . Due to the influence of factors such as the volume and power consumption of the underwater unmanned platform, the maneuverability of the platform is restricted, that is, the platform is under weak observation conditions. Therefore, in this case, this application proposes an underwater target motion analysis method that can be realized under weak observation conditions.

[0043] Figure 2 shows a flowchart of an underwater target motion analysis method according to an embodiment of the present disclosure. As Figure 2 shown, the method includes: Step S100, using MCMC to sample each parameter of the state vector of the underwater target. Here, those skilled in the art can understand that MCMC refers to Markov Chain Monte Carlo sampling. At the same time, combined with the foregoing, in the method of this application, each parameter of the sampled state vector can include at least one of the distance r of the underwater target from the observation platform, the measured azimuth θ of the underwater target, the speed v of the underwater target, and the angle α between the underwater target and the horizontal direction.

[0044] After sampling each parameter of the state vector of the underwater target, through step S200, the posterior probability density ratio of the state vector of the underwater target for the current sampling can be obtained according to the sampling result. Here, it needs to be explained that in the underwater target motion analysis method of the embodiments of this application, when using the MCMC method to sample each parameter of the state vector of the underwater target, the number of samplings is a preset number of samplings. According to the recommendation of the MCMC sampling algorithm, the number of samplings is generally 10 5. Through multiple samplings, and then analyzing the movement of the underwater target based on the results of multiple samplings, the accuracy of the analysis results is ensured, making the analysis results more objective.

[0045] At the same time, it should also be noted that when using the MCMC method to perform multiple samplings on the parameters of the state vector of the underwater target, the sampling times corresponding to each sampling are different.

[0046] Furthermore, through step S300, according to the posterior probability density ratio of the state vector of the underwater target at the current sampling, the value of the state vector at the current sampling is determined. Finally, through step S400, the values of the state vectors corresponding to each sampling in the preset number of samplings are statistically analyzed, and the final state vector estimation value of the underwater target is determined according to the statistical results.

[0047] Therefore, in the method of the embodiment of the present application, when analyzing the movement of the underwater target, the parameters of the state vector of the underwater target are sampled by the MCMC sampling method, and after sampling, the value of the state vector is determined in combination with the posterior probability density ratio of the state vector. Furthermore, the final state vector estimation value is obtained through the statistical analysis of the state vectors corresponding to each sampling in the determined preset number of samplings. Compared with the method of using the maximum likelihood estimation method in the related art, it can effectively realize the state estimation of the underwater target under weak observation conditions.

[0048] At the same time, it should also be noted that since the MCMC method requires the prior information of the target state, therefore, in the method of the embodiment of the present application, when using MCMC to sample the parameters of the state vector of the underwater target, it needs to be based on the prior conditions of the underwater target. In a possible implementation manner, the prior conditions for the underwater target include but are not limited to: the upper and lower limits of the speed of the underwater target; the upper and lower limits of the distance between the underwater target and the observation condition; the measurement azimuth interval of the underwater target, and the upper and lower limits of the angle between the underwater target and the horizontal.

[0049] Specifically, set the upper and lower limits of the target speed to v max , v min ; the upper and lower limits of the target distance are r max , r min ; assume that the measurement azimuth follows a normal distribution with a standard deviation of σ, that is, the interval of the kth measurement azimuth is The parameter m is usually selected as 5; the upper and lower limits of the angle between the target and the horizontal are -π, +π.

[0050] Among them, it should be noted here that the specific values of the upper and lower limits of the target speed in the above prior conditions can be determined by the speed range of the ship in the real world; the specific values of the upper and lower limits of the target distance can be determined by the sonar operating range; the standard deviation of the measured azimuth can be determined by the sonar factory parameters.

[0051] After determining the prior conditions of the underwater target, MCMC can be used to sample the parameters of the state vector of the underwater target. In a possible implementation, when the prior probability distribution of the state vector parameters of the underwater target is missing, in order to simplify the sampling process, a uniform distribution model can be used to describe the parameters of the state vector of the underwater target. That is, a uniform distribution model is used in combination with the set prior conditions of the underwater target to sample the parameters of the state vector of the underwater target.

[0052] Specifically, when the parameters of the state vector include the distance r of the underwater target from the observation platform, the measured azimuth θ of the underwater target, the speed v of the underwater target, and the angle α between the underwater target and the horizontal direction, set the total number of MCMC samplings of these four state vector parameters to M, and the i-th sampling is (r, θ, v, α) (i) , and the probability distribution followed by each sampling is:

[0053] p(r) ∼ U[r min , r max ,

[0054]

[0055] p(v) ∼ U[v min , v max ,

[0056] p(α) ∼ U[-π, +π]

[0057] After obtaining the sampling values of the parameters of the state vector of the underwater target through the above sampling method, the posterior probability density ratio of the state vector of the underwater target can be calculated according to the sampling results. Here, it should be explained that the posterior probability density ratio of the state vector refers to: the probability of the underwater target at the previous sampling moment (i.e., the previous sampling) following independent and identically distributed Gaussian noise with a mean of θ k , variance of σ 2 and the probability of the current sampling moment (i.e., the current sampling) following independent and identically distributed Gaussian noise with a mean of θ k , variance of σ 2 ratio of probabilities.

[0058] Among them, the probability of the underwater target at the previous sampling moment following independent and identically distributed Gaussian noise with a mean of θ k , variance of σ 2 is At the current sampling moment, it follows a Gaussian noise with a mean of θ k and a variance of σ 2 The probability of independent and identically distributed Gaussian noise is:

[0059] In a possible implementation, when calculating the posterior probability density ratio of the state vector of the underwater target at the current sampling, it can be calculated based on the prior probability distribution of the state vector of the underwater target at the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target at the current sampling. Among them, the prior probability distribution of the state vector of the underwater target can be calculated from the probability distributions of the various parameters of the sampled state vector of the underwater target.

[0060] Specifically, according to the independence of the prior probability distribution, when calculating the prior probability distribution of the state vector of the underwater target, the multiplication method can be directly used. That is, the probability distributions of the various parameters of the sampled state vector of the underwater target are successively multiplied to obtain the prior probability distribution of the state vector.

[0061] That is, according to the independence of the prior probability distribution, the probability distributions of the samplings of the various parameters of the state vector are successively multiplied to obtain:

[0062]

[0063] where C = 1 / (4mπσ(r min -r min )(v max -v min )) and 1 represents the indicator function.

[0064] Furthermore, when calculating the posterior probability density ratio based on the prior probability distribution of the state vector of the underwater target at the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target at the current sampling, it also includes: combining the likelihood estimates of the state vector of the underwater target at the previous sampling of the current sampling and the likelihood estimates of the state vector of the underwater target at the previous sampling of the current sampling to calculate the posterior probability density ratio.

[0065] In a possible implementation, the likelihood estimate of the state vector of the underwater target can be calculated by the following formula:

[0066]

[0067] where θ k [(r, θ, v, α)] is the accurate azimuth of the target at the kth measurement moment under the assumption of the uniform linear motion of the underwater target, and its calculation formula is:

[0068]

[0069] wherein, v tgt , v obs are the velocities of the target and the observer, respectively.

[0070] After respectively calculating the prior probability distribution and the likelihood estimate of the state vector at the current subsampling (i.e., the current sampling moment) and the prior probability distribution and the likelihood estimate of the state vector at the previous sampling of the current subsampling (i.e., the previous sampling moment before the current sampling moment) through the above formula, the posterior probability density ratio of the state vector of the underwater target at the current subsampling can be carried out. Here, it should be noted that when calculating the posterior probability density ratio of the state vector of the underwater target at the current subsampling, it is carried out under the condition of independent hypothesis.

[0071] Specifically, under the condition of independent hypothesis, the formula for calculating the posterior probability density ratio of the state vector at the current subsampling is as follows:

[0072]

[0073] In addition, in the method of the embodiment of the present application, after calculating the posterior probability density ratio of the state vector of the underwater target at the current subsampling through any of the above-mentioned manners, the value of the state vector at the current subsampling can be taken according to the calculated posterior probability density ratio of the state vector at the current subsampling.

[0074] That is to say, in the underwater target motion analysis method of the embodiment of the present application, by calculating the posterior probability density ratio of the state vector of the underwater target at the current subsampling, the value of the state vector at the current subsampling is determined, which ensures the accuracy of the sampling value used for the motion analysis of the underwater target, and further improves the accuracy of the underwater target motion analysis result.

[0075] In a possible implementation manner, when determining the value of the state vector according to the calculated posterior probability density ratio of the state vector of the underwater target at the current subsampling, it can be determined according to the magnitude of the calculated posterior probability density ratio. Specifically, it can be determined by the magnitude relationship between the calculated posterior probability density ratio and the generated random number. Here, it should be pointed out that the generated random number is a uniformly distributed random number between (0, 1).

[0076] More specifically, first, compare the posterior probability density ratio of the state vector of the underwater target at the current sampling with the generated random number. When it is compared that the posterior probability density ratio is greater than or equal to the random number, the parameters of each state vector sampled at the current sampling are used as the value of the state vector. When it is compared that the posterior probability density ratio is less than the random number, the parameters of each state vector sampled at the previous sampling of the current sampling are used as the value of the state vector.

[0077] That is,

[0078]

[0079] where R is the posterior probability density ratio of the state vector of the underwater target at the current sampling time (i.e., the current sampling), and x is a uniformly distributed random number between (0, 1). (r, θ, v, α) (i) is the value of the state vector of the underwater target determined at the current sampling time i.

[0080] After obtaining the values of the state vectors of the underwater target at each sampling time through any of the above-described methods, statistical analysis can be performed on the values of the state vectors determined for each sampling in the preset number of samplings, and then the final state vector estimate of the underwater target can be obtained based on the results of the statistical analysis.

[0081] In a possible implementation, when performing statistical analysis on the values of the state vectors determined for each sampling in the preset number of samplings, it can be statistically analyzed according to the occurrence frequencies of the values of the parameters in the state vectors of the preset number of samplings, and then the values of the parameters in the final state vector estimate can be determined based on the occurrence frequencies of the statistically analyzed parameters. Here, it should be noted that the occurrence frequencies of the values of the parameters in the state vectors of the preset number of samplings can be characterized by the number of occurrences of each parameter in M samplings.

[0082] For example, the values of the state vectors of the underwater target collected in the preset number of samplings M are: (r1, θ1, v1, α1), (r2, θ2, v2, α2), ……, (r i , θ i , v i , α i ), ……, (r M , θ M , v M , α M ). Among them, (r1, θ1, v1, α1) is the value of the state vector of the first sampling, (r2, θ2, v2, α2) is the value of the state vector of the second sampling, (r i , θ i , v i , α i) is the value of the state vector for the i-th sampling, (r M , θ M , v M , α M ) is the value of the state vector for the M-th sampling.

[0083] Separate the counts of the occurrences of each value such as r1 to r M from the 1st to the M-th samplings, and select the value of r with the highest occurrence count as r1; separate the counts of the occurrences of each value such as θ1 to θ M from the 1st to the M-th samplings, and select the value of θ with the highest occurrence count as θ3; separate the counts of the occurrences of each value such as v1 to v M from the 1st to the M-th samplings, and select the value of v with the highest occurrence count as v6; separate the counts of the occurrences of each value such as α1 to α M from the 1st to the M-th samplings, and select the value of α with the highest occurrence count as α9. Then, based on the above statistical results, take the state vector (r1, θ3, v6, α9) as the final state vector estimated value.

[0084] In addition, what those skilled in the art can also point out is that when performing statistical analysis on the values of the state vector of an underwater target under a preset number of samplings and determining the final state vector estimated value based on the statistical results, in addition to determining by the method of the occurrence frequencies of the various parameters of the state vector in the preset number of samplings, it is also possible to calculate the mean values of the various parameters of the state vector in the preset number of samplings and take the mean values of the various parameters of the calculated state vector as the final state vector estimated value.

[0085] For example, calculate the mean values of r, θ, v, and α in the M samplings respectively to obtain r 平均 , θ 平均 , v 平均 and α 平均 . Then, take (r 平均 , θ 平均 , v 平均 , α 平均 ) as the final state vector estimated value.

[0086] Here, it should also be pointed out that what those skilled in the art can understand is that other logical statistical analysis methods can also be performed on the values of the state vector of the preset number of samplings, including but not limited to the above two methods, and no further examples will be given in this application.

[0087] To more clearly illustrate that the method of the embodiments of the present application can realize the analysis of the target motion under weak observation conditions when performing underwater target motion analysis, the following is its simulation experiment.

[0088] See Figure 3 , for the observer and the target motion situation as Figure 3 shown, where the underwater target moving speed is 5 m / s, the platform moving speed is 2.3 m / s (i.e., the observer moving speed), the true value of the distance between the two is 9136.1 m, and the number of MCMC samplings is taken as M = 10 5 .

[0089] Based on the underwater target motion analysis method of the application, the motion elements of the target are sampled, and the sampling results are as Figure 4 shown, where the solid curve represents the upper and lower limits of the Bayesian interval, and the black solid line represents the true value. It can be seen that due to the observer being in weak observation conditions, theoretically, the motion elements are unsolvable, so there are deviations in the target heading and left-right fuzzy estimations. However, the probability density sampling method still makes the best estimation of the distance, and the optimal estimation result of its Bayesian interval is a speed of 5.18 m / s and a distance of 9552.6 m, which is very close to the true value.

[0090] Therefore, for the underwater target motion analysis of the embodiment of the present application, by using the MCMC method to sample the parameters of the state vector required for target analysis, it can not only achieve the underwater target motion analysis even under weak observation conditions, but also combine the determination of the value of the state vector according to the posterior probability density ratio of the state vector, and through the statistical analysis of the values of the state vector with a preset number of samplings, and then determine the final estimated value of the state vector according to the statistical results, which further ensures the accuracy of the final target analysis result.

[0091] It should be noted that although Figures 1 to 4 is used as an example to introduce the underwater target motion analysis method as described above, those skilled in the art can understand that the present disclosure should not be limited to this. In fact, users can flexibly set the specific processes of each step according to personal preferences and / or actual application scenarios, as long as the parameters of the state vector can be sampled by the MCMC method, so as to achieve the target motion analysis even under weak observation conditions.

[0092] Correspondingly, based on the underwater target motion analysis method described in any of the foregoing, the present application also provides an underwater target motion analysis device. Since the working principle of the underwater target motion analysis device provided by the present application is the same as or similar to the principle of the underwater target motion analysis method of the present application, the repeated parts will not be elaborated here.

[0093] See Figure 5, the underwater target motion analysis device 100 provided by this application includes a sampling module 110, a probability density ratio calculation module 130, a state vector determination module 140, and a state vector estimation module 150. Among them, the sampling module 110 is configured to sample each parameter of the state vector of the underwater target by using MCMC. The probability density ratio calculation module 130 is configured to obtain the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling result. The state vector determination module 140 is configured to determine the value of the state vector under the current sampling according to the posterior probability density ratio. The state vector estimation module 150 is configured to count the values of the state vectors corresponding to each sampling in the preset number of samplings, and determine the final state vector estimation value of the underwater target according to the statistical result.

[0094] The embodiments of the present disclosure have been described above. The above description is exemplary and not exhaustive, and is not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements to the technologies in the market, or to enable other ordinary technical personnel in the technical field to understand the embodiments disclosed herein.

Claims

1. An underwater target motion analysis method, characterized in that, Including: Using MCMC to sample each parameter of the state vector of the underwater target, and obtaining the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling result; Determining the value of the state vector under the current sampling according to the posterior probability density ratio; Counting the values of the state vector corresponding to each sampling in a preset number of samplings, and determining the final state vector estimation value of the underwater target according to the statistical result; When obtaining the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling result, it is calculated according to the prior probability distribution of the state vector of the underwater target in the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target in the current sampling; When calculating the posterior probability density ratio according to the prior probability distribution of the state vector of the underwater target in the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target in the current sampling, it further includes: Combining the likelihood estimation of the state vector of the underwater target in the previous sampling of the current sampling and the likelihood estimation of the state vector of the underwater target in the current sampling to calculate the posterior probability density ratio; When combining the likelihood estimation of the state vector of the underwater target in the previous sampling of the current sampling and the likelihood estimation of the state vector of the underwater target in the current sampling to calculate the posterior probability density ratio, it is calculated by the following formula: ； Among them, is the likelihood estimate of the state vector at the previous sampling of the current subsampling; is the prior probability distribution of the state vector at the previous sampling of the current subsampling, is the likelihood estimate of the state vector at the current subsampling, is the prior probability distribution of the state vector at the current subsampling, is the probability of independent and identically distributed Gaussian noise with mean and variance at the previous sampling time of the underwater target, is the probability of independent and identically distributed Gaussian noise with mean and variance at the current sampling time, is the noisy bearing measurement at time , is independent and identically distributed Gaussian noise with mean and variance ; is the distance of the underwater target from the distance observation platform, is the measured azimuth of the underwater target, is the speed of the underwater target, is the angle between the underwater target and the horizontal direction.

2. The method according to claim 1, characterized in that, When determining the final state vector estimation value of the underwater target according to the statistical result, it includes: Counting the value with the highest frequency of occurrence of each parameter in the state vector in a preset number of samplings as the final state vector estimation value.

3. The method according to claim 1, characterized in that, The state vector parameters include the distance of the underwater target from the observation platform , the measured azimuth of the underwater target , the speed of the underwater target , and the angle between the underwater target and the horizontal direction and at least one of them.

4. The method according to claim 1, characterized in that, When using MCMC to sample the state vector parameters of the underwater target, sampling is performed based on the prior conditions of the underwater target; The prior conditions include at least one of the upper speed limit and the lower speed limit of the underwater target; the upper distance limit and the lower distance limit between the underwater target and the observation condition; the measurement azimuth interval of the underwater target, and the upper angle limit and the lower angle limit between the underwater target and the horizontal angle.

5. The method according to claim 1, characterized in that, The prior probability distribution of the state vector of the underwater target in the current sampling is obtained by successively performing product operations on the probability distributions of the parameters in the state vector according to the independence of the prior probability distribution.

6. The method according to any one of claims 1 to 5, characterized in that, When determining the value of the state vector under the current sampling according to the posterior probability density ratio, it includes: Comparing the posterior probability density ratio with the generated random number; When the posterior probability density ratio is greater than or equal to the random number, taking the sampling value of the current sampling as the value of the state vector; When the posterior probability density ratio is less than the random number, taking the sampling value of the previous sampling of the current sampling as the value of the state vector.

7. An underwater target motion analysis device, characterized in that, Including: A sampling module, a probability density ratio calculation module, a state vector determination module, and a state vector estimation module; The sampling module is configured to use MCMC to sample each parameter of the state vector of the underwater target; The probability density ratio calculation module is configured to obtain the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling result; The state vector determination module is configured to determine the value of the state vector under the current sampling according to the posterior probability density ratio; The state vector estimation module is configured to count the values of the state vector corresponding to each sampling in a preset number of samplings, and determine the final state vector estimation value of the underwater target according to the statistical result; When obtaining the posterior probability density ratio of the state vector of the underwater target under the current sampling according to the sampling result, it is calculated according to the prior probability distribution of the state vector of the underwater target under the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target under the current sampling; When calculating the posterior probability density ratio according to the prior probability distribution of the state vector of the underwater target under the previous sampling of the current sampling and the prior probability distribution of the state vector of the underwater target under the current sampling, it further includes: Combining the likelihood estimation of the state vector of the underwater target under the previous sampling of the current sampling and the likelihood estimation of the state vector of the underwater target under the current sampling to calculate the posterior probability density ratio; When combining the likelihood estimation of the state vector of the underwater target under the previous sampling of the current sampling and the likelihood estimation of the state vector of the underwater target under the current sampling to calculate the posterior probability density ratio, it is calculated by the following formula: ； Among them, is the likelihood estimate of the state vector at the previous sampling of the current subsampling; is the prior probability distribution of the state vector at the previous sampling of the current subsampling, is the likelihood estimate of the state vector at the current subsampling, is the prior probability distribution of the state vector at the current subsampling, is the probability of independent and identically distributed Gaussian noise with mean and variance at the previous sampling time of the underwater target; is the probability of independent and identically distributed Gaussian noise with mean and variance at the current sampling time; is the noisy bearing measurement at time ; is independent and identically distributed Gaussian noise with mean and variance . is the distance of the underwater target from the distance observation platform, is the measured azimuth of the underwater target, is the speed of the underwater target, is the angle between the underwater target and the horizontal direction.

Images