Ultra-short baseline positioning method and device, electronic equipment and storage medium

Abstract

The present application relates to the technical field of marine detection positioning, and discloses an ultra-short baseline positioning method and device, electronic equipment and storage medium. The present application constructs an error equation based on an observation equation and an observation value and performs local linearization processing, sets weights for the residual errors of each group of error equations, obtains position correction and velocity correction through weighted least squares method, and updates the current estimated value according to the position correction and velocity correction and executes the next round of iteration to determine the new position correction and velocity correction until the preset convergence condition is met, wherein each group of weights is updated according to the corresponding group of residual errors after standardization in each round of iteration, and the effective target sound source position vector and effective sound signal velocity are output according to the current estimated value, position correction and velocity correction of the last round of iteration. The present application can realize accurate positioning of the target sound source and has high reliability.

Description

Technical Field

[0001] This invention relates to the field of marine exploration and positioning technology, specifically to an ultra-short baseline positioning method, device, electronic equipment, and storage medium. Background Technology

[0002] Ultra-Short Baseline (USBL) positioning systems typically consist of a receiver array of several hydrophones. The direction of the sound source is calculated by the time difference of arrival (TDOA) of the acoustic signal between different hydrophones, and the source location is determined by combining this with propagation delay or geometric relationships. Because the array size is relatively small compared to the distance to the sound source, TDOA observations are highly sensitive to signal detection accuracy, array orientation, noise environment, and multipath effects. Summary of the Invention

[0003] This invention provides an ultra-short baseline positioning method, device, electronic device, and storage medium to solve the problems of ultra-short baseline positioning being susceptible to interference and real-time positioning being unstable and unreliable.

[0004] In a first aspect, the present invention provides an ultra-short baseline positioning method, comprising: The position vectors of multiple hydrophones in the carrier coordinate system are obtained, and the arrival time difference of any two of the multiple hydrophones receiving the sound signal from the target sound source is obtained, so as to determine multiple sets of observation values; Based on each set of observations, corresponding error equations are constructed. Each set of error equations is configured to establish the residual between the corresponding set of observation equations and the observations. The observation equations express the geometric distance difference between the target sound source location vector to be estimated and the two hydrophones in the corresponding set, and the geometric distance difference is divided by the velocity of the sound signal to be estimated to express the calculated value of the arrival time difference of the sound signal received by the two hydrophones in the corresponding set. The initial values ​​of the target sound source position vector and the sound signal velocity are obtained. Each set of error equations is processed by local linearization so that the initial values ​​of the target sound source position vector and the sound signal velocity are used as the current estimated values, representing the target sound source position vector to be estimated and the sound signal velocity to be estimated. The position correction amount and velocity correction amount to be estimated are also introduced. Weights are assigned to the residuals of each set of error equations. The position correction and velocity correction are obtained by weighted least squares. The current estimate is updated based on the position correction and velocity correction and the next iteration is performed to determine the new position correction and velocity correction until the preset convergence condition is met. In each iteration, the weights are updated based on the corresponding set of residuals after standardization. Based on the current estimate and position and velocity corrections from the last iteration, the effective target sound source position vector and effective sound signal velocity are output.

[0005] The above implementation method introduces position correction and velocity correction quantities that can be solved iteratively by constructing an error equation and then performing local linearization. Through continuous iteration, the initial values ​​of the target sound source position vector and the initial values ​​of the sound signal velocity are made close to the effective target sound source position vector and the effective sound signal velocity, which can achieve accurate positioning of the target sound source with high reliability.

[0006] In one optional implementation, when the target sound source is in the near field, the error equation includes: , in, Let m be the residual of the m-th group. Let j be the position vector of the j-th hydrophone. Let i be the position vector of the i-th hydrophone. For the m-th observation, the time difference between the arrival times of the sound signal from the target sound source received by the i-th and j-th hydrophones is given. , Let be the target sound source position vector to be estimated, and c be the sound signal velocity to be estimated.

[0007] The observation equations in the above implementation use a near-field model to meet the localization requirement that the target sound source is in the near field. Accordingly, a distance threshold is set; when the distance between the target sound source and the hydrophone array is less than the distance threshold, the target sound source is considered to be in the near field.

[0008] In one optional implementation, when the target sound source is in the far field, the error equation includes: , in, Let m be the residual of the m-th group. Let j be the position vector of the j-th hydrophone. Let i be the position vector of the i-th hydrophone. For the m-th observation, the time difference between the arrival times of the sound signal from the target sound source received by the i-th and j-th hydrophones is given. , Let be the unit vector corresponding to the position vector of the target sound source to be estimated in the propagation direction. , where c is the velocity of the acoustic signal to be estimated.

[0009] The observation equations in the above implementation method apply a far-field model to meet the localization requirement that the target sound source is in the far field. Accordingly, a distance threshold is set. When the distance between the target sound source and the hydrophone array is greater than the distance threshold, it is considered that the distance of the sound source is much greater than the array scale, and the target sound source is in the far field.

[0010] In one alternative implementation, each set of error equations undergoes local linearization, including: Perform local linearization using the following formula: , in, Let m be the residual of the m-th group. For the observation equation of the m-th group, This is the current estimate for the k-th iteration. For the m-th observation, These represent the position and velocity corrections for the k-th iteration.

[0011] Since the above implementation replaces the target sound source position vector and the sound signal velocity to be estimated with the current estimated value, only the position correction amount and velocity correction amount need to be calculated in each iteration to continuously correct the current estimated value, thereby approximating the effective target sound source position vector and the effective sound signal velocity.

[0012] In one optional implementation, each set of weights is updated in each iteration based on the corresponding set of residuals after standardization, including: The residual of this iteration is determined based on the position and velocity corrections obtained in this iteration. A robust scale is obtained by estimating the median absolute deviation of the residuals in the current iteration. When the robust scale is greater than a robust threshold, the original robust scale is maintained. When the robust scale is less than the robust threshold, the robust scale is determined according to the robust threshold. The standardized residual is obtained by dividing the residual of the current iteration by the robust metric. The corresponding weights are determined based on the standardized residuals.

[0013] The above implementation method standardizes each set of residuals in each iteration, thereby avoiding the influence of a few outliers on the robust scaling estimate and preventing the standardized residuals from approaching zero, thus improving the stability of the values.

[0014] In one optional implementation, determining the corresponding weights based on the standardized residuals includes: In response to the standardized residual being in the first interval, a first weight is configured; In response to the standardized residual being in the second interval, a second weight is configured; The value in the first interval is less than the value in the second interval, and the first weight is greater than the second weight.

[0015] The above implementation method adaptively adjusts the weights assigned to the residuals according to the magnitude of the residuals, reduces the weights of larger residuals, suppresses the impact of abnormal observations on the estimation results, and improves the robustness of ultra-short baseline positioning.

[0016] In one optional implementation, the step of waiting until a preset convergence condition is met includes: Stop iteration if one or more of the following convergence conditions are met: The position correction amount is less than the first threshold; The speed correction amount is less than the second threshold; The change in the weight is less than the third threshold.

[0017] The above implementation method obtains the effective target sound source position vector and effective sound signal velocity that best reflect the actual situation through limited iterations, reducing unnecessary iteration cycles in the later stages.

[0018] In a second aspect, the present invention provides an ultra-short baseline positioning device, comprising: The observation acquisition module is used to acquire the position vectors of multiple hydrophones in the carrier coordinate system, and to acquire the arrival time difference of any two of the multiple hydrophones receiving the sound signal from the target sound source, so as to determine multiple sets of observations. The error equation construction module is used to construct corresponding error equations based on each set of observations. Each set of error equations is configured to establish the residual between the corresponding set of observation equations and the observations. The observation equations express the geometric distance difference between the target sound source position vector to be estimated and the two hydrophones in the corresponding set, and the geometric distance difference is divided by the velocity of the sound signal to be estimated to express the calculated value of the arrival time difference of the sound signal received by the two hydrophones in the corresponding set. The initial value acquisition module is used to acquire the initial value of the target sound source position vector and the initial value of the sound signal velocity. Each set of error equations is processed by local linearization so that the initial value of the target sound source position vector and the initial value of the sound signal velocity are used as the current estimated value, representing the target sound source position vector to be estimated and the sound signal velocity to be estimated, and the position correction amount and velocity correction amount to be estimated are introduced. The correction iteration module is used to set weights for the residuals of each set of error equations, obtain the position correction and velocity correction through weighted least squares, update the current estimate based on the position correction and velocity correction and execute the next round of iteration to determine the new position correction and velocity correction until the preset convergence condition is met. In each round of iteration, the weights of each set are updated based on the corresponding set of residuals after standardization. The positioning information output module is used to output the effective target sound source position vector and effective sound signal velocity based on the current estimate and position correction and velocity correction of the last iteration.

[0019] Thirdly, the present invention provides an electronic device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the ultra-short baseline positioning method of the first aspect or any corresponding embodiment described above.

[0020] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to perform the ultra-short baseline positioning method of the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0021] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0022] Figure 1 This is a schematic diagram of the hydrophone array and the target sound source according to an embodiment of the present invention.

[0023] Figure 2 This is a schematic diagram of the first type of ultra-short baseline positioning method according to an embodiment of the present invention.

[0024] Figure 3 This is a schematic diagram of the second process of the ultra-short baseline positioning method according to an embodiment of the present invention.

[0025] Figure 4 This is a schematic diagram of the weight distribution of different weight functions in an embodiment of the present invention.

[0026] Figure 5 This is a structural block diagram of the ultra-short baseline positioning device according to an embodiment of the present invention.

[0027] Figure 6 This is a schematic diagram of the hardware structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0028] 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.

[0029] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0030] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0031] A hydrophone is a specialized underwater acoustic sensor, analogous to a microphone in the air, used to collect sound signals propagating in water. For example... Figure 1 As shown, the ultra-short baseline positioning system consists of an array of hydrophones p1, p2, p3, p4, and p5 arranged closely together. The spacing between the hydrophones is typically around ten centimeters, expressed by the array scale. The sound signal emitted by the target sound source d1 will arrive at each hydrophone in the array sequentially, resulting in a small time difference in arrival. Ultra-short baseline positioning uses this time difference between the arrival of the sound signal received by any two hydrophones to determine the location of the target sound source. During the positioning process, the position of each hydrophone is also required. Specifically, the position vector of each hydrophone in the carrier coordinate system is recorded. The carrier coordinate system is a unified coordinate system established based on the carrier on which the hydrophones are mounted; the position vector of each hydrophone is determined by the position of the carrier.

[0032] The location of a target sound source relies on estimating the time difference of arrival (TDOA) of the sound signal between different hydrophones. However, the TDOA detected by the hydrophones as an observation value often contains errors, such as spurious peaks caused by multipath propagation, trigger jitter caused by sudden noise changes, anomalies in individual hydrophone channels, and correlation peak drift. Therefore, during the localization process, it is necessary to avoid abnormal observations dominating the solution direction, ensure that the iteration can converge, and guarantee the accuracy of the localization results.

[0033] According to an embodiment of the present invention, an embodiment of an ultra-short baseline positioning method is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0034] This embodiment provides an ultra-short baseline positioning method, which can be used in positioning devices including hydrophone arrays, etc. Figure 2 This is a flowchart of an ultra-short baseline positioning method according to an embodiment of the present invention, such as... Figure 2 As shown, the process includes the following steps: Step S201: Obtain the position vectors of multiple hydrophones in the carrier coordinate system, and obtain the arrival time difference of any two hydrophones receiving the target sound source sound signal, so as to determine multiple sets of observation values.

[0035] The array can consist of N hydrophones, where N can be at least 5, and their position vectors in the carrier coordinate system are respectively... , , ..., , The position of each hydrophone is represented by a three-dimensional vector pointing to a coordinate point in space. Due to the differences in their positions, all hydrophones will have a time difference in arrival when receiving the target sound source signal. Accordingly, selecting any two hydrophones from N to determine their arrival time differences will generate M observations. , , respectively corresponding to the hydrophone pair , .

[0036] Step S202: Construct corresponding error equations based on each group of observations. Each group of error equations is configured to establish the residual between the corresponding group of observation equations and the observations. The observation equations express the geometric distance difference between the target sound source position vector to be estimated and the two hydrophones in the corresponding group, and express the calculated value of the arrival time difference of the received sound signal by the two hydrophones in the corresponding group by dividing the geometric distance difference by the velocity of the sound signal to be estimated.

[0037] The error equation establishes a linear relationship between the observed values ​​and the parameters to be determined, quantifies the magnitude of the residual, and is used to solve the unknowns and evaluate the accuracy using the least squares method. In this embodiment of the invention, the error equation introduces the position vector of the target sound source to be estimated and the velocity of the sound signal to be estimated using the observation equation, and constructs the geometric relationship and propagation time relationship of the sound signal in the water, thereby forming an expression between the position vector of the target sound source to be estimated, the velocity of the sound signal to be estimated, and the calculated value of the time difference of arrival of the hydrophone pair. The residual of the error equation comes from the error between the calculated value of the time difference of arrival and the observed value.

[0038] Step S203: Obtain the initial value of the target sound source position vector and the initial value of the sound signal velocity. Each set of error equations is processed by local linearization so that the initial value of the target sound source position vector and the initial value of the sound signal velocity are used as the current estimated values, representing the target sound source position vector to be estimated and the sound signal velocity to be estimated. The position correction amount and velocity correction amount to be estimated are also introduced.

[0039] The error equations for local linearization transform the process of estimating the target sound source position vector and the sound signal velocity into solving for the position correction and velocity correction values. The initial values ​​of the target sound source position vector and the sound signal velocity are then used as the current estimates for evaluation. This allows for multiple iterations, using the position and velocity correction values ​​obtained in each iteration to continuously refine the current estimates, thus bringing them closer to the actual values. The initial value of the target sound source position vector, as prior information about the target sound source position, can be roughly estimated using conventional methods, such as empirical values ​​determined based on the detected time difference of arrival. The initial value of the sound signal velocity, as prior information about the underwater sound propagation speed, can be an empirical value around 1500 m / s.

[0040] Step S204: Set weights for the residuals of each set of error equations, obtain the position correction and velocity correction by weighted least squares, update the current estimate based on the position correction and velocity correction and execute the next round of iteration to determine the new position correction and velocity correction until the preset convergence condition is met. In each round of iteration, the weights of each set are updated based on the corresponding set of residuals after standardization.

[0041] The purpose of setting weights is to control the influence of different groups of residuals when pursuing the comprehensive minimization of the residuals of multiple sets of error equations. In particular, abrupt changes in the arrival time difference between two hydrophones in a corresponding group can significantly increase the residual of that group. Therefore, reducing the weight of a single residual can weaken its impact on the overall estimation. In each iteration, the position and velocity corrections obtained in this round are determined using the least squares method. This continuously corrects the current estimate before inputting it into the next iteration for new position and velocity corrections. The position and velocity corrections obtained in each round, besides adjusting the prior information from the previous round, also determine the residuals of different groups of error equations in this round. The residuals of each group can be used to determine the weights set for each group, thus participating in the least squares estimation in the next round. Accordingly, in the first iteration, the same weight is set for the residuals of each group of error equations, for example, all of them are 1, i.e., assuming that there are no anomalies in any hydrophones. In the early iterations, such as those less than a preset number of iterations, the update magnitude of the weights can be controlled. For example, when updating the weights based on the changes in residual values ​​in each iteration, no adjustments can be made; only the weight values ​​determined in the previous iteration are maintained, thereby improving the convergence speed and obtaining a stable current estimate. Once the preset number of iterations has been exceeded, the weights set for each group of residual values ​​are adjusted normally based on the residual values ​​in each iteration.

[0042] Step S205: Output the effective target sound source position vector and effective sound signal velocity based on the current estimated value and position correction amount and velocity correction amount of the last iteration.

[0043] Through iterative convergence, in the final iteration, the effective target sound source position can be obtained by superimposing the target sound source position vector in the current estimated value with the position correction amount calculated in the current iteration. This position serves as the final localization result. The effective sound signal velocity is obtained in the same way.

[0044] This embodiment provides an ultra-short baseline positioning method, which can be used in positioning devices including hydrophone arrays, etc. Figure 3 This is a flowchart of an ultra-short baseline positioning method according to an embodiment of the present invention, such as... Figure 3 As shown, the process includes the following steps: Step S301: Obtain the position vectors of multiple hydrophones in the carrier coordinate system, and obtain the arrival time difference of the target sound source acoustic signal received by any two of the multiple hydrophones, in order to determine multiple sets of observation values. For details, please refer to [link to relevant documentation]. Figure 2 Step S201 of the illustrated embodiment will not be described again here. Step S302: Construct corresponding error equations based on each group of observations. Each group of error equations is configured to establish the residual between the corresponding group of observation equations and the observations. The observation equations express the geometric distance difference between the target sound source position vector to be estimated and the two hydrophones in the corresponding group, and express the calculated value of the arrival time difference of the received sound signal by the two hydrophones in the corresponding group by dividing the geometric distance difference by the velocity of the sound signal to be estimated.

[0045] Specifically, step S302 includes: Step S3021: When the target sound source is in the near field, the error equation includes: , in, Let m be the residual of the m-th group. Let j be the position vector of the j-th hydrophone. Let i be the position vector of the i-th hydrophone. For the m-th observation, the time difference between the arrival times of the sound signal from the target sound source received by the i-th and j-th hydrophones is given. , Let be the target sound source position vector to be estimated, and c be the sound signal velocity to be estimated.

[0046] Near-field distance refers to the distance between the target sound source and the hydrophone array, such as when it is less than the set distance threshold. Let represent the Euclidean distance between the target sound source and the j-th hydrophone. Let represent the Euclidean distance between the target sound source and the i-th hydrophone. The time difference in arrival of the sound signal received by the j-th hydrophone and the i-th hydrophone from the target sound source originates from the difference in geometric distance between the two hydrophones and the target sound source. Therefore, the calculated value of the time difference in arrival of the sound signal received by the two hydrophones can be expressed by the observation equation, which is formed by dividing the geometric distance difference by the velocity of the sound signal to be estimated. The residual of the error equation is the difference between the calculated value and the observed value in the observation equation.

[0047] Step S3022: When the target sound source is in the far field, the error equation includes: , in, Let m be the residual of the m-th group. Let j be the position vector of the j-th hydrophone. Let i be the position vector of the i-th hydrophone. For the m-th observation, the time difference between the arrival times of the sound signal from the target sound source received by the i-th and j-th hydrophones is given. , Let be the unit vector corresponding to the position vector of the target sound source to be estimated in the propagation direction. , where c is the velocity of the acoustic signal to be estimated.

[0048] The far-field distance is when the target sound source is relatively far from the hydrophone array, much larger than the array size, such as greater than the set distance threshold. In this case, the direction in which the sound signal from the target sound source propagates to each hydrophone is approximately considered to be parallel. This represents the distance between the coordinate projections of the j-th and i-th hydrophones onto a unit vector along the propagation direction, i.e., the geometric distance difference between the two hydrophones and the target sound source. Similarly, an observation equation can be constructed using the ratio of this unit vector to the velocity of the sound signal to be estimated, and an error equation can be built using the observed values. The unit vector must satisfy the unit constraint: .

[0049] Steps S3021 and S3022 are selected based on the actual target sound source distance, which can be roughly estimated using conventional methods.

[0050] Step S303: Obtain the initial value of the target sound source position vector and the initial value of the sound signal velocity. Each set of error equations is processed by local linearization so that the initial value of the target sound source position vector and the initial value of the sound signal velocity are used as the current estimated values, representing the target sound source position vector to be estimated and the sound signal velocity to be estimated. The position correction amount and velocity correction amount to be estimated are also introduced.

[0051] Specifically, step S303 includes: Perform local linearization using the following formula: , in, Let m be the residual of the m-th group. For the observation equation of the m-th group, This is the current estimate for the k-th iteration. For the m-th observation, These represent the position and velocity corrections for the k-th iteration.

[0052] As mentioned above, the observation equation itself is nonlinear, therefore it is necessary to linearize it locally and iteratively approximate the actual value. For ease of description, the target sound source position vector and the sound signal velocity to be estimated in the error equation are expressed as follows: (Near field) or (Far field) This represents the current estimate in the k-th iteration, where the current estimate in the first iteration can be directly substituted with the initial values ​​of the target sound source position vector and the sound signal velocity. The position and velocity correction values ​​for the k-th iteration are used to adjust the current estimate after each iteration, thus updating the estimate for the next iteration. This continuously approximates the actual value. Optionally, the velocity of the acoustic signal to be estimated can be directly substituted with the initial velocity value as a constant in the iteration, meaning it is not part of the current estimate updated with each iteration. Alternatively, updates can be performed only in some iterations, while the initial velocity value or the current estimate from the previous iteration is used in the remaining iterations.

[0053] Multiple sets of error equations after local linearization can be expressed as the following vector equations: , in, For the residual vector, It is a closed difference vector. ; The coefficient matrix, , These represent the position and velocity corrections for the k-th iteration.

[0054] For the far-field model, add a constraint observation to the observation equations: , Add corresponding rows to the coefficient matrix The closed difference vector increases To enhance the constraint effect, a weight much larger than that of the arrival time difference observations can be set for this row (e.g., to (on a certain scale), and after each iteration... Normalization is performed.

[0055] Accordingly, under the near-field model, the observation equations The partial derivatives are as follows: , , 2) Under the far-field model, the observation equation The partial derivatives are as follows: , .

[0056] Step S304: Set weights for the residuals of each set of error equations, obtain the position correction and velocity correction by weighted least squares, update the current estimate based on the position correction and velocity correction and execute the next round of iteration to determine the new position correction and velocity correction until the preset convergence condition is met. In each round of iteration, the weights of each set are updated based on the corresponding set of residuals after standardization.

[0057] The weights assigned to the residuals of each set of error equations can form a weight matrix. It satisfies the following diagonal matrix form, and weights can be assigned to the residuals of each set of error equations by calculation, with the diagonal matrix... , Until This refers to the weights corresponding to different groups, with zero values ​​for positions outside the diagonal.

[0058] , The position and velocity corrections can be obtained using the weighted least squares method, i.e., minimizing... The details are as follows: , The current estimate in this iteration is updated based on the obtained position and velocity corrections to serve as the current estimate for the next iteration, as detailed below: , For the kth round, the current estimate of the kth round can be corrected by the correction amount determined in the kth round, so as to obtain the current estimate for the (k+1)th round. This process is repeated in each iteration.

[0059] If the degrees of freedom meet the conditions, the unit weight variance can be further estimated using the following formula: , Where u is the unit weight variance, M is the number of error equations, and n is the number of unknown parameters.

[0060] The parameter covariance is estimated based on the unit weighted variance, using the following formula: , in, Let covariance be a parameter.

[0061] During the iteration process, the weights of the residuals of each set of error equations can also be updated iteratively. In each round of iteration, the weights are updated according to the corresponding set of residuals after standardization, realizing a two-layer iteration of outer weight iteration and inner weighted least squares iteration. The inner layer solves for parameters under a given weight matrix, and the outer layer updates the weight matrix according to the residuals, so that the outlier observations are automatically reduced in weight in subsequent iterations.

[0062] Specifically, step S304 includes: Step a1: Determine the residual of this iteration based on the position and velocity corrections obtained in this iteration. Specifically, the position and velocity corrections can be substituted into the error equation processed by local linearization for calculation.

[0063] Step a2: Perform median absolute deviation estimation on the residuals of this iteration to obtain a robust scale. When the robust scale is greater than the robust threshold, the original robust scale is maintained. When the robust scale is less than the robust threshold, the robust scale is determined according to the robust threshold.

[0064] To avoid the influence of a few outliers on the scaling estimate, robust scaling is calculated using the following formula: , in, For the robustness scale of the corresponding group, This represents the residuals for the corresponding group.

[0065] Furthermore, to improve the numerical stability of the robust scale, a robustness threshold is set, and the robust scale is adjusted accordingly: , in, For the robustness scale of the corresponding group, This is a robust threshold.

[0066] Step a3: Divide the residual of this iteration by the robust metric to obtain the standardized residual.

[0067] Calculate using the following formula: , in, For the standardized residuals of the corresponding group, For the residuals of the corresponding group, This is the robustness metric for the corresponding group.

[0068] Step a4: Determine the corresponding weights based on the standardized residuals.

[0069] In some alternative implementations, step a4 above includes: Step a401: In response to the standardized residual being in the first interval, configure the first weight; Step a402: In response to the standardized residual being in the second interval, configure the second weight; Step a403: The value of the first interval is less than the value of the second interval, and the first weight is greater than the second weight.

[0070] The obtained standardized residuals are weighted according to their size. For standardized residuals with large values, interference needs to be suppressed by reducing the weight. The following algorithm is used to adaptively adjust the weights, eliminating the need to manually remove outliers, which is suitable for online real-time positioning.

[0071] a. Huber (Huber weight function): , in, The weights are determined based on the standardized residuals. For the standardized residuals, k is the threshold set for dividing the intervals.

[0072] b. IGG-III: , in, The weights are determined based on the standardized residuals. To standardize the residuals, k1 and k2 are the threshold values ​​set for dividing the intervals.

[0073] c. Cauchy (Cauchy weight function): , in, The weights are determined based on the standardized residuals. To standardize the residuals, This is the attenuation scale parameter.

[0074] d. Tukey (illustrated double-squared weight function): , in, The weights are determined based on the standardized residuals. To standardize the residuals, The threshold set for dividing the interval.

[0075] like Figure 4As shown, the weighting trends of Huber, IGG-III, Cauchy, and Tukey weighting functions under different standardized residuals are illustrated. Generally, the weights for larger residuals at the two ends are lower than those for smaller residuals in the middle, thus achieving the effect of weight reduction to suppress interference. However, the weight reduction trends of different weighting functions vary, and the appropriate function can be selected based on the specific noise environment, array size, and signal characteristics. For example, Huber is suitable for suppressing mild anomalies, while IGG-III can directly reduce significant gross errors to near zero.

[0076] In many other examples, adaptive weight adjustment can also be achieved using the following algorithm: e. Fair (Fair weight function): , in, The weights are determined based on the standardized residuals. To standardize the residuals, This is the linear decay boundary scale.

[0077] f. Welsh (Welsh weight function): , in, The weights are determined based on the standardized residuals. To standardize the residuals, This is the Gaussian decay threshold.

[0078] g. Hampel (Hampel's three-segment weight function): , in, The weights are determined based on the standardized residuals. To standardize the residuals, a, b, The threshold set for dividing the interval.

[0079] Each set of residuals is reweighted to obtain new weights, which can then be updated to the following weight matrix: , The updated weight matrix can participate in the next iteration to achieve... Minimize.

[0080] Specifically, step S304 includes: Step b1: Determine if one or more of the following convergence conditions are met, and then stop the iteration: The position correction amount is less than the first threshold; The speed correction amount is less than the second threshold; The change in the weight is less than the third threshold.

[0081] For example, refer to the following formula to determine whether convergence has met the requirements and implement finite iteration.

[0082] , , in, , To set a threshold, for example Magnitude. This represents the weight with the largest adjustment change among the M groups of residuals between the k-th and k+1-th iterations, used to determine whether it is less than the threshold for stopping iterations. This represents the position correction (including the three coordinate corrections in the three-dimensional representation) and all parameter corrections in the velocity correction. The largest of the three (those) is used to determine whether it is less than the threshold for stopping iteration.

[0083] Step S305: Output the effective target sound source position vector and effective sound signal velocity based on the current estimated value and position correction amount and velocity correction amount of the last iteration.

[0084] After the iteration ends, the effective target sound source position vector and the effective sound signal velocity can be calculated using the following formula. Calculate, where, This is the current estimate for the last round (round k). These are the position and speed corrections determined in the final round.

[0085] According to embodiments of the present invention, an ultra-short baseline positioning device is also provided, which is used to implement the above embodiments and preferred embodiments, and will not be repeated hereafter. As used below, the term "module" can be a combination of software and / or hardware that implements a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0086] like Figure 5 As shown, this embodiment provides an ultra-short baseline positioning device, including: The observation acquisition module 501 is used to acquire the position vectors of multiple hydrophones in the carrier coordinate system, and to acquire the arrival time difference of any two of the multiple hydrophones receiving the sound signal from the target sound source, so as to determine multiple sets of observations. Error equation construction module 502 is used to construct corresponding error equations based on each group of observations. Each group of error equations is configured to establish the residual between the corresponding group of observation equations and the observations. The observation equation expresses the geometric distance difference between the target sound source position vector to be estimated and the two hydrophones in the corresponding group, and the geometric distance difference is divided by the speed of the sound signal to be estimated to express the calculated value of the arrival time difference of the sound signal received by the two hydrophones in the corresponding group. The initial value acquisition module 503 is used to acquire the initial value of the target sound source position vector and the initial value of the sound signal velocity. Each set of error equations is processed by local linearization so that the initial value of the target sound source position vector and the initial value of the sound signal velocity are used as the current estimated value, representing the target sound source position vector to be estimated and the sound signal velocity to be estimated, and the position correction amount and velocity correction amount to be estimated are introduced. The correction iteration module 504 is used to set weights for the residuals of each set of error equations, obtain the position correction and velocity correction through the weighted least squares method, update the current estimate based on the position correction and velocity correction and execute the next round of iteration to determine the new position correction and velocity correction until the preset convergence condition is met. In each round of iteration, each set of weights is updated based on the corresponding set of residuals after standardization. The positioning information output module 505 is used to output the effective target sound source position vector and the effective sound signal velocity based on the current estimated value and position correction amount and velocity correction amount of the last iteration.

[0087] The ultra-short baseline positioning device provided in this embodiment of the invention can execute the ultra-short baseline positioning method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the various modules and units described above are the same as in the corresponding embodiments described above, and will not be repeated here.

[0088] Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention.

[0089] The following is a detailed reference. Figure 6 This diagram illustrates a suitable structural design for implementing an electronic device according to embodiments of the present invention. The electronic device may include a processor (e.g., a central processing unit, graphics processor, etc.) 601, which can perform various appropriate actions and processes based on a program stored in read-only memory (ROM) 602 or a program loaded from memory 608 into random access memory (RAM) 603. RAM 603 also stores various programs and data required for the operation of the electronic device. The processor 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to the bus 604.

[0090] Typically, the following devices can be connected to I / O interface 605: input devices 606 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 607 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 608 including, for example, magnetic tapes, hard disks, etc.; and communication devices 609. Communication device 609 allows electronic devices to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 6 Electronic devices with various devices are shown, but it should be understood that it is not required to implement or have all of the devices shown, and more or fewer devices may be implemented or have instead.

[0091] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 609, or installed from a memory 608, or installed from a ROM 602. When the computer program is executed by the processor 601, it performs the functions defined in the ultra-short baseline positioning method of the embodiments of the present invention.

[0092] Figure 6 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0093] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as recordable on a storage medium, or implemented as computer code downloaded over a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and subsequently stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the ultra-short baseline positioning method shown in the above embodiments is implemented.

[0094] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A method for positioning with ultra-short baselines, characterized in that, The method includes: The position vectors of multiple hydrophones in the carrier coordinate system are obtained, and the arrival time difference of any two of the multiple hydrophones receiving the sound signal from the target sound source is obtained, so as to determine multiple sets of observation values; Based on each set of observations, corresponding error equations are constructed. Each set of error equations is configured to establish the residual between the corresponding set of observation equations and the observations. The observation equations express the geometric distance difference between the target sound source location vector to be estimated and the two hydrophones in the corresponding set, and the geometric distance difference is divided by the velocity of the sound signal to be estimated to express the calculated value of the arrival time difference of the sound signal received by the two hydrophones in the corresponding set. The initial values ​​of the target sound source position vector and the sound signal velocity are obtained. Each set of error equations is processed by local linearization so that the initial values ​​of the target sound source position vector and the sound signal velocity are used as the current estimated values, representing the target sound source position vector to be estimated and the sound signal velocity to be estimated. The position correction amount and velocity correction amount to be estimated are also introduced. Weights are assigned to the residuals of each set of error equations. The position correction and velocity correction are obtained by weighted least squares. The current estimate is updated based on the position correction and velocity correction and the next iteration is performed to determine the new position correction and velocity correction until the preset convergence condition is met. In each iteration, the weights are updated based on the corresponding set of residuals after standardization. Based on the current estimate and position and velocity corrections from the last iteration, the effective target sound source position vector and effective sound signal velocity are output.

2. The ultra-short baseline positioning method according to claim 1, characterized in that, When the target sound source is in the near field, the error equation includes: ， in, Let m be the residual of the m-th group. Let j be the position vector of the j-th hydrophone. Let i be the position vector of the i-th hydrophone. For the m-th observation, the time difference between the arrival times of the sound signal from the target sound source received by the i-th and j-th hydrophones is given. , Let be the target sound source position vector to be estimated, and c be the sound signal velocity to be estimated.

3. The ultra-short baseline positioning method according to claim 1, characterized in that, When the target sound source is in the far field, the error equation includes: ， in, Let m be the residual of the m-th group. Let j be the position vector of the j-th hydrophone. Let i be the position vector of the i-th hydrophone. For the m-th observation, the time difference between the arrival times of the sound signal from the target sound source received by the i-th and j-th hydrophones is given. , Let be the unit vector corresponding to the position vector of the target sound source to be estimated in the propagation direction. , where c is the velocity of the acoustic signal to be estimated.

4. The ultra-short baseline positioning method according to claim 1, characterized in that, Each set of error equations undergoes local linearization, including: Perform local linearization using the following formula: ， in, Let m be the residual of the m-th group. For the observation equation of the m-th group, This is the current estimate for the k-th iteration. For the m-th observation, These represent the position and velocity corrections for the k-th iteration.

5. The ultra-short baseline positioning method according to claim 1, characterized in that, Each set of weights is updated in each iteration based on the corresponding set of residuals after standardization, including: The residual of this iteration is determined based on the position and velocity corrections obtained in this iteration. A robust scale is obtained by estimating the median absolute deviation of the residuals in the current iteration. When the robust scale is greater than a robust threshold, the original robust scale is maintained. When the robust scale is less than the robust threshold, the robust scale is determined according to the robust threshold. The standardized residual is obtained by dividing the residual of this iteration by the robust metric. The corresponding weights are determined based on the standardized residuals.

6. The ultra-short baseline positioning method according to claim 5, characterized in that, The step of determining the corresponding weights based on the standardized residuals includes: In response to the standardized residual being in the first interval, a first weight is configured; In response to the standardized residual being in the second interval, a second weight is configured; The value in the first interval is less than the value in the second interval, and the first weight is greater than the second weight.

7. The ultra-short baseline positioning method according to claim 1, characterized in that, The process of until the preset convergence condition is met includes: Stop iteration if one or more of the following convergence conditions are met: The position correction amount is less than the first threshold; The speed correction amount is less than the second threshold; The change in the weight is less than the third threshold.

8. A short baseline positioning device, characterized in that, The device includes: The observation acquisition module is used to acquire the position vectors of multiple hydrophones in the carrier coordinate system, and to acquire the arrival time difference of any two of the multiple hydrophones receiving the sound signal from the target sound source, so as to determine multiple sets of observations. The error equation construction module is used to construct corresponding error equations based on each set of observations. Each set of error equations is configured to establish the residual between the corresponding set of observation equations and the observations. The observation equations express the geometric distance difference between the target sound source position vector to be estimated and the two hydrophones in the corresponding set, and the geometric distance difference is divided by the velocity of the sound signal to be estimated to express the calculated value of the arrival time difference of the sound signal received by the two hydrophones in the corresponding set. The initial value acquisition module is used to acquire the initial value of the target sound source position vector and the initial value of the sound signal velocity. Each set of error equations is processed by local linearization so that the initial value of the target sound source position vector and the initial value of the sound signal velocity are used as the current estimated value, representing the target sound source position vector to be estimated and the sound signal velocity to be estimated, and the position correction amount and velocity correction amount to be estimated are introduced. The correction iteration module is used to set weights for the residuals of each set of error equations, obtain the position correction and velocity correction through weighted least squares, update the current estimate based on the position correction and velocity correction and execute the next round of iteration to determine the new position correction and velocity correction until the preset convergence condition is met. In each round of iteration, the weights of each set are updated based on the corresponding set of residuals after standardization. The positioning information output module is used to output the effective target sound source position vector and effective sound signal velocity based on the current estimate and position correction and velocity correction of the last iteration.

9. An electronic device, characterized in that, include: A memory and a processor are communicatively connected, the memory stores computer instructions, and the processor executes the ultra-short baseline positioning method according to any one of claims 1 to 7 by executing the computer instructions.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing the computer to perform the ultrashort baseline positioning method according to any one of claims 1 to 7.

Images