Long baseline positioning system and fusion algorithm resistant to shallow water acoustic jump interference

Abstract

The application discloses a long baseline positioning system and a fusion algorithm resisting shallow sea sound jump interference, the long baseline array adopts a buoy type, a seat bottom type or other composition forms or a composite form of multiple forms, comprises multiple underwater acoustic beacons and is used for receiving underwater acoustic signals emitted by underwater vehicles; two (or multiple) transducers are installed on the long baseline nodes at different depths (on both sides of a jump layer), meanwhile, an intelligent algorithm is used to select redundant transducers to solve the jump layer interference in the long baseline positioning, and a data fusion algorithm is used to fuse and solve a navigation track, so that the comprehensive underwater acoustic measurement performance of the system in a complex sea condition, the anti-shallow sea sound jump interference capability of the long baseline array and positioning precision can be greatly improved.

Description

Technical Field

[0001] This invention relates to the field of underwater acoustic positioning, tracking, and measurement technology, specifically to a long baseline positioning system and fusion algorithm that resists acoustic abrupt changes in shallow water. Background Technology

[0002] Underwater positioning technology is a crucial component of deep-sea research and seabed missions, primarily relying on acoustic positioning techniques. Among these, the long baseline (LBL) system has demonstrated significant advantages. In applications requiring long-distance, high-precision positioning for underwater vehicles, the LBL offers advantages such as convenient calibration, a large single-array range, and high long-range positioning accuracy. Compared to other positioning methods, the LBL exhibits considerably higher accuracy and reliability.

[0003] In terms of its specific working principle, a long baseline system requires at least three nodes to be deployed on the seabed, forming a seabed node array according to a certain formation. Then, by measuring the propagation time delay of acoustic signals between the underwater vehicle and the seabed nodes, the geodetic coordinates of the underwater vehicle are calculated using the spherical intersection method. Once calibrated, this system can continuously provide accurate positioning data.

[0004] However, long baseline array positioning also presents certain challenges. Due to the stratification of underwater acoustic velocities and the difference in depth between the underwater vehicle and the transducer, especially in shallow seas, lakes, and reservoirs, underwater acoustic signals are often interfered with by transition layers, which affects the accuracy of signal processing and positioning tracking.

[0005] Currently, long baseline positioning is mainly divided into two modes: buoy-based and bottom-based. However, both modes operate based on a single transducer. If there is a significant depth difference between the underwater vehicle and the transducer, especially when above or below a transition layer, the propagation of the underwater acoustic signal will be interfered with by the transition layer. This will cause the spherical intersection method to fail during positioning calculations, thus making accurate positioning impossible.

[0006] Therefore, future research needs to address the interference of the transition layer on signal propagation. Possible measures include improving signal processing techniques or changing the transducer layout to reduce the depth difference with the vehicle. Only in this way can we fully realize the potential of long baseline systems in underwater positioning, providing more accurate positioning services for deep-sea research and seabed missions. Summary of the Invention

[0007] To address the shortcomings and deficiencies of the existing technology, the inventors have improved the design and now provide a new solution. The aim is to avoid the failure of the spherical intersection method during positioning calculations, thereby achieving accurate positioning. Specifically, this invention is implemented as follows:

[0008] A long baseline positioning system for resisting acoustic transition interference in shallow water includes: a long baseline array employing buoy-type, bottom-mounted, or other configurations, or a combination of multiple configurations, comprising multiple underwater acoustic beacons for receiving underwater acoustic signals emitted by an underwater vehicle; the array includes: a buoy connected sequentially to an electronics compartment, at least one upper transducer, and at least one lower transducer via a transmission cable, with an anchor block at the end; a wireless transmission device connected to the electronics compartment; wherein adjacent upper and lower transducers are positioned at different depths with a distance greater than the transition layer depth; capable of receiving and parsing underwater acoustic signals emitted by the underwater vehicle, and transmitting them wirelessly to positioning and tracking measurement software; the underwater vehicle navigating within the measurement area, carrying cooperative beacons to emit underwater acoustic signals within the measurement area; and the positioning and tracking measurement software acquiring positioning data transmitted by the long baseline array, calculating, tracking, and locating the underwater vehicle's trajectory based on the received data, performing data fusion and filtering to further improve positioning accuracy and stability.

[0009] Another aspect of the present invention discloses a fusion algorithm for a long baseline positioning system based on the above-mentioned resistance to acoustic abrupt changes in shallow water, comprising the following steps:

[0010] Step S1: Deploy a long baseline array in the measurement water area and calibrate each node of the long baseline array; when installing underwater acoustic transducers to deploy buoy nodes, ensure that the two underwater acoustic transducers are deployed at different depths and the distance between them is greater than the depth of the transition layer; the long baseline array can be buoy type, bottom type, or other composition forms, or a combination of multiple forms.

[0011] Step S2: Use an underwater vehicle to navigate within the long baseline array area, carry a cooperative beacon to transmit underwater acoustic signals in the measurement water area, and have the underwater acoustic transducers at each node of the long baseline array receive and analyze the underwater acoustic signals. Then, send the analysis results to the positioning and tracking measurement software through the data transmission link.

[0012] Step S3: Use positioning and tracking measurement software to perform positioning and tracking measurements on the underwater vehicle, analyze the data of each node, and calculate the underwater vehicle's trajectory through a data fusion algorithm; the positioning and tracking measurement software uses the acquired coordinates of each node and underwater acoustic positioning delay data to calculate the underwater vehicle's trajectory based on the intersection principle, and displays it on the interface.

[0013] The fusion algorithm includes the following steps:

[0014] Step A1: For each node, the underwater acoustic time delay can be received from m transducers. There are n nodes, where m≥2 and n≥3. When all nodes and transducers can receive signals and the time delay can be calculated, there exists at least according to the spherical intersection equation. There are at least a few possible combinations that can form A self-organizing array positioning base array;

[0015] Step A2: Perform time delay filtering on the signals received by each transducer, compare the time delay data with the previously received data, and exclude data that is much larger than the average value;

[0016] Step A3: Perform location calculations on the matrix formed by various combinations according to the intersection equation;

[0017] Step A4: Select the long baseline array nodes and transducers required for the solution based on the underwater vehicle's travel depth to further reduce the dimensionality of the positioning solution results, or weight or reduce the weight of the solution results of the transducers located on both sides of the transition layer at different depths.

[0018] Step A5: Perform data fusion on the various positioning solutions; data fusion algorithms include, but are not limited to, Kalman filtering, weighted fusion, data association, Bayesian probabilistic fusion, and probabilistic multi-hypothesis tracking;

[0019] Step A6: Draw and display the fused flight trajectory in the positioning and tracking measurement software.

[0020] The working principle of this invention mainly includes the following aspects:

[0021] Multi-point deployment: Multiple beacon buoys are deployed on the seabed and arranged in a specific array to form a seabed node array. These beacons can receive underwater acoustic signals emitted by underwater vehicles and transmit the received signals to positioning and tracking measurement software via wireless transmission equipment, thereby enabling the tracking and positioning of underwater vehicles.

[0022] Dual-layer transducer deployment: In the buoy node, the electronics compartment, upper transducer, and lower transducer are connected sequentially via transmission cables. The distance between the upper and lower transducers at different depths is greater than the depth of the transition layer. This is an important design feature to avoid the failure of the spherical intersection method during positioning calculations due to interference from the transition layer.

[0023] Data fusion algorithm: The positioning and tracking measurement software acquires positioning data transmitted by the long baseline array and uses a data fusion algorithm for calculation, tracking, and positioning. Data fusion algorithms can include Kalman filtering, weighted fusion, data association, Bayesian probabilistic fusion, and probabilistic multi-hypothesis tracking, etc. The purpose of data fusion is to integrate data from various beacons to improve the accuracy and stability of positioning.

[0024] Optimization strategy for depth differences: Since the depth difference between the underwater vehicle and the transducer may be large, this technical solution selects the long baseline array nodes and transducers required for the calculation based on the underwater vehicle's travel depth, further reducing the dimensionality of the positioning calculation results, or weighting or reducing the weighting of the calculation results of the transducers located on both sides of the different depth transition layers, so as to reduce the interference caused by depth differences.

[0025] To resist interference from shallow sea transition layers: Each node can receive the underwater acoustic time delay of m transducers, and n nodes (m≥2, n≥3) are combined for calculation. This can better cope with the interference problem of transition layers in environments such as shallow seas, lakes and reservoirs.

[0026] The beneficial technical effects of this invention include the following:

[0027] 1. High-precision positioning: Through multi-node setup, optimized transducer layout and data fusion algorithm, this invention can achieve high-precision underwater positioning, which greatly improves the positioning accuracy of seabed missions and deep-sea research.

[0028] 2. Resistance to abrupt change layer interference: Through the deep layout of the transducer and data processing technology, this invention can effectively resist abrupt change layer interference in shallow sea environments, thereby improving the reliability of positioning.

[0029] 3. Strong system stability: Through data fusion and filtering, the stability of the positioning system can be further enhanced, reducing the impact of environmental changes on positioning accuracy.

[0030] 4. Flexible application: This invention has considered different deployment modes (such as buoy type, bottom type, etc.), and can be combined and selected in various ways based on actual needs and environmental conditions, which has high flexibility and adaptability.

[0031] 5. Easy to operate: This invention takes into account detailed deployment steps and calculation processes, as well as supporting software, making the operation process simple and easy to understand, and requiring relatively low technical skills from the operator. Attached Figure Description

[0032] Figure 1 This is a flowchart illustrating the long baseline design and positioning fusion algorithm for resisting acoustic abrupt changes in shallow water, as described in this invention.

[0033] Figure 2 This is a schematic diagram of the long baseline array buoy deployment and system of the present invention;

[0034] Figure 3 This is a schematic diagram of the long baseline array buoy structure of the present invention;

[0035] Figure 4 This is a flowchart illustrating the positioning fusion algorithm of the present invention;

[0036] Figure 5 This is a schematic diagram of the extended Kalman state space model in an embodiment of the present invention. Detailed Implementation

[0037] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments and the accompanying drawings. It should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and techniques are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0038] Example 1: A long baseline design and positioning fusion algorithm for resisting acoustic abrupt changes in shallow water, comprising a long baseline array, an underwater vehicle, positioning and tracking measurement software, etc.; in use, the underwater vehicle navigates within the long baseline array area; the long baseline array adopts a buoy-type, bottom-mounted, or other compositional form, or a combination of multiple forms; the positioning and tracking measurement software is used to perform positioning and tracking measurements on the underwater vehicle and calculate its navigation trajectory; the intersection algorithm includes, but is not limited to, spherical intersection, circular intersection, and hyperbolic intersection; the positioning and tracking measurement software uses the coordinates of each node and the underwater acoustic positioning delay data to calculate the underwater vehicle's navigation trajectory based on the intersection principle, specifically including the following steps (e.g. Figure 1 As shown):

[0039] Long baseline arrays need to be deployed in advance in the measurement water area (e.g. Figure 2 (As shown); After deployment, each node of the long baseline array is calibrated, mainly targeting the underwater acoustic transducers located below the transition layer within the nodes; When installing nodes (buoys), it should be ensured that the distance between two underwater acoustic transducers is greater than the depth of the transition layer, and they should be placed at different depths (e.g., Figure 3 (as shown);

[0040] When the system is in operation, the underwater vehicle carries a cooperative beacon and transmits underwater acoustic signals in the measurement area; the underwater acoustic transducers at each node of the long baseline array receive and analyze the underwater acoustic signals, and then send the analysis results to the positioning and tracking measurement software through the data transmission link;

[0041] The positioning and tracking measurement software analyzes the data from each node and uses a data fusion algorithm to calculate the underwater vehicle's trajectory.

[0042] The localization fusion algorithms mentioned include, but are not limited to, Kalman filtering, weighted fusion, data association, and Bayesian probabilistic fusion, with specific processes (e.g.) Figure 4 As shown below:

[0043] Given that each node can receive underwater acoustic time delays from m (m≥2) transducers, and there are n (n≥3) nodes, when all nodes and transducers can receive signals and the time delays can be calculated, there exists at least... There are several possible combinations (according to the spherical intersection equation), that is, at least one possible combination. A positioning array (self-assembly array); however, since the horizontal spacing between transducers on the same node cannot meet the baseline length requirement of a long baseline array (insufficient positioning accuracy), the combined result can be further reduced;

[0044] The signals received by each transducer are subjected to time delay filtering. The time delay data is compared with the previously received data, and outliers are removed (data that is much larger than the average value is removed).

[0045] The matrix composed of various combinations is located and solved according to the intersection equation;

[0046] Based on the underwater vehicle's travel depth, select the long baseline array nodes and transducers required for the calculation, further reduce the dimensionality of the positioning calculation results, or weight or reduce the weight of the calculation results of transducers located at different depths (both sides of the transition layer).

[0047] Data fusion is performed on the various positioning solutions; data fusion algorithms include, but are not limited to, Kalman filtering, weighted fusion, data association, Bayesian probabilistic fusion, and probabilistic multi-hypothesis tracking.

[0048] The fused flight path is plotted and displayed in positioning, tracking, and measurement software.

[0049] This patent proposes a long baseline array design and positioning fusion algorithm to resist acoustic transition interference in shallow water: two (or more) transducers are installed at different depths (on both sides of the transition layer) on the long baseline nodes. At the same time, an intelligent algorithm is used to select redundant transducers to solve the transition layer interference in long baseline positioning. A data fusion algorithm is used to fuse and solve the navigation trajectory, which can significantly improve the comprehensive underwater acoustic measurement performance of the system under complex sea conditions, the ability of the long baseline array to resist acoustic transition interference in shallow water, and the positioning accuracy.

[0050] Example 2: Actual Operation

[0051] The long baseline array (buoys) is deployed in the measurement waters, and a boat calibration is performed after deployment.

[0052] Install a cooperative beacon on the underwater vehicle and enable it to navigate within the surveyed waters;

[0053] The long baseline array receives and processes underwater acoustic signals and transmits the processing results to the positioning and tracking measurement software;

[0054] The positioning and tracking measurement software uses a data fusion algorithm based on the information from each transducer to calculate the underwater vehicle's trajectory and displays it on the interface.

[0055] Example 3

[0056] The long-baseline array employs the underwater acoustic positioning principle based on range convergence to calculate the target's location. It utilizes the underwater acoustic propagation speed, the time delay between the target's sound source and the receiver, and the array's position parameters to locate the target and calculate its position. Its positioning equations are as follows:

[0057]

[0058] In the formula, (x,y,z) represents the target position to be determined; (x i ,y i ,z i ) represents the coordinates of the array nodes; c represents the speed of sound, t represents the speed of sound. i This represents the time it takes for the underwater acoustic signal to reach each array node.

[0059] After dimensionality reduction, the positioning equations are simplified to the form Ax = b. Since the number of equations is much greater than the number of unknowns, the least squares method is used for optimization. This yields the optimized coordinates of the underwater target: x = (A... T A) -1 A T b. The solution result is represented as a measurement vector from the fusion algorithm, indicating the target's position and velocity, in the form:

[0060]

[0061] The array measurement fusion vector composed of multiple combinations is:

[0062] z = [s1 s2 s] i… ] T (i = 3, 4).

[0063] One application example of this embodiment is a fusion algorithm based on extended Kalman filtering, whose extended Kalman state-space model in information fusion is as follows: Figure 5 As shown;

[0064] Based on the initial fused data S1 and the state estimate x from the previous time step (k-1) k-1 and error covariance p k-1 Construct the state space and calculate the state estimate x. 0 k And error covariance p k Treat the data S2 to be fused as the observation space, and the state estimate x 0 k And error covariance p k Constructing the observation equation and updating the observation matrix, 1. Calculating the state estimate x 1 k and Kalman gain K1; treat the data to be fused S3 as a new observation space, and the state estimate x 1 kAnd error covariance p k Constructing the observation equation and updating the observation matrix, we calculate the state estimate x. 2 k and Kalman gain K2; continue the fusion calculation of the data to be fused, each time a new data s to be fused is generated. n The state estimate x will be obtained by updating n through the observation matrix. n k and the corresponding Kalman gain K n Ultimately, the Kalman gain K1...K n This constitutes the error covariance matrix at the current moment;

[0065] After all the data to be fused has completed the above steps, the state estimate x k And error covariance P k All data are updated to accurately reflect the current state at time k; the state estimate x after fusion of all data is also updated. n k The final output fusion result at the current time k will be used as the state estimate at the next time k+1, and together with the error covariance matrix, it will be used as the input parameters for the fusion solution at the next time k+1.

[0066] The initial data and the data to be fused in the figure are vectors to be fused.

[0067] z = [s1 s2 s] i … ] T (i = 3, 4, ...), that is, the positioning results calculated by various combination matrix solutions.

[0068] s i (i = 1, 2, 3) … A set of ).

[0069] The output fusion result is the optimal estimate of the final positioning result.

[0070] The observation space model constructed from the positioning results of each array is as follows:

[0071]

[0072] In the formula, This represents the system state transition observation coefficients at time k for different fused data.

[0073] In the fusion localization process based on Extended Kalman Filter (EKF), the localization results of one array are first selected as the initial fusion data to construct the state-space equations. Then, the localization results of other arrays are treated as the observation space to construct the observation equations. This enables multi-array fusion localization based on EKF.

[0074] It should be understood that the specific embodiments described above are merely illustrative or explanatory of the principles of the invention and do not constitute a limitation thereof. Therefore, any modifications, equivalent substitutions, improvements, etc., made without departing from the spirit and scope of the invention should be included within the protection scope of the invention. Furthermore, the appended claims are intended to cover all variations and modifications falling within the scope and boundaries of the appended claims, or equivalent forms of such scope and boundaries.

Claims

1. A long baseline positioning system against shallow water sound jump interference, characterized in that The long baseline array includes a plurality of underwater acoustic beacons arranged in a floating ball, a seat bottom or other forms, or a combination of multiple forms, and is used to receive underwater acoustic signals transmitted by underwater vehicles. The floating ball is connected to the electronic cabin, at least one upper transducer, at least one lower transducer, and an anchor block at the end through a transmission cable; a wireless transmission device is connected to the electronic cabin; adjacent upper and lower transducers are arranged at different depths and the distance between them is greater than the depth of the jump layer; the underwater acoustic signals transmitted by the underwater vehicle can be received and analyzed, and the wireless transmission device can send the data to the positioning and tracking measurement software; The underwater vehicle is used to navigate in the measurement water area and emit underwater acoustic signals in the measurement water area with a cooperative beacon; The positioning and tracking measurement software is used to obtain positioning data transmitted by the long baseline array, calculate, track and locate the motion trajectory of the underwater vehicle based on the received data, perform data fusion and filtering processing, and further improve the positioning accuracy and stability. The method comprises the following steps:

2. The fusion algorithm of a long baseline positioning system against shallow water acoustic jump interference according to claim 1, characterized in that Step S1, arranging a long baseline array in the measurement water area and calibrating each node of the long baseline array; when installing underwater acoustic transducers for floating ball node arrangement, ensure that two underwater acoustic transducers are arranged at different depths and the distance between them is greater than the depth of the jump layer; the long baseline array adopts a floating ball type, a seat bottom type, or other forms, or a combination of multiple forms; Step S2, using an underwater vehicle to navigate in the long baseline array area and emit underwater acoustic signals in the measurement water area with a cooperative beacon; the underwater acoustic transducers of each node of the long baseline array receive and analyze the underwater acoustic signals, and then send the analysis results to the positioning and tracking measurement software through a data transmission link; Step S3, using the positioning and tracking measurement software to locate and track the underwater vehicle, analyze the node data, and calculate the navigation trajectory of the underwater vehicle through a data fusion algorithm; the positioning and tracking measurement software uses the obtained node coordinates and underwater acoustic positioning time delay data to calculate the navigation trajectory of the underwater vehicle according to the intersection equation, and displays the result on the interface. The fusion algorithm further comprises the following steps:

3. The fusion algorithm of claim 2, wherein, Step A2, performing time delay filtering on the signals received by each transducer, comparing the time delay data with the previously received data, and excluding data much larger than the average value; Step A1, each node can receive the underwater acoustic time delay of m transducers, and there are n nodes, where m≥2, n≥3; when all nodes and transducers can receive signals and can solve the time delay, at least a combination mode, at least can constitute a self-organizing array positioning base array; Step A3, performing positioning calculation on the base array composed of various combinations according to the intersection equation; Step A4, selecting the required long baseline array nodes and transducers according to the navigation depth of the underwater vehicle to further reduce the dimension of the positioning calculation result, or weighting or de-weighting the calculation results of the transducers located on both sides of the jump layer at different depths; Step A5, performing data fusion on each positioning calculation result; the data fusion algorithm includes but is not limited to Kalman filtering, weighted fusion, data association, Bayesian probability fusion and probability multiple hypothesis tracking; Step A6, displaying the navigation trajectory after data fusion in the positioning and tracking measurement software. ​ 4. The fusion algorithm of claim 3, wherein, The long baseline array adopts a hydroacoustic positioning principle based on ranging intersection to realize positioning calculation of the target, utilizes a hydroacoustic propagation speed, a time delay of a measured target sound source to a receiver and a position parameter of the array to realize positioning of the target, and calculates a target position; a positioning equation set thereof is in a form of: wherein is the target position to be determined; is the array node coordinate; c is the sound speed, is the time of arrival of the underwater acoustic signal at each array node.

5. The fusion algorithm of claim 4, wherein, The positioning equation set is solved by using a least square method to obtain optimized coordinates of the underwater target to be measured The solution result is expressed as a measurement vector of the fusion algorithm, indicating the position and speed of the target, and is in the form of A measurement fusion vector of the array composed of multiple combinations is in a form of: 。 6. The fusion algorithm of claim 5, wherein, The fusion algorithm is a fusion algorithm based on an extended Kalman filter, and includes the following steps: The initial fusion data S1 and the state estimation value x k-1 and error covariance p k-1 at the last time (k-1) are used to build a state space and calculate the state estimation value x 0 k and error covariance p k The to-be-fused data S2 is regarded as an observation space, and the state estimation value x 0 k and error covariance p k are used to build an observation equation and update the first group of observation matrices, and the state estimation value x 1 k and Kalman gain K1 are calculated; the to-be-fused data S3 is regarded as a new observation space, and the state estimation value x 1 k and error covariance p k are used to build an observation equation and update the second group of observation matrices, and the state estimation value x 2 k and Kalman gain K2 are calculated; the fusion calculation of the to-be-fused data is continuously performed, and each new to-be-fused data s n is used to obtain the state estimation value x n k and the corresponding Kalman gain K n ; finally, the Kalman gain K1...K n is used to form the error covariance matrix at the current time. After all the data to be fused complete the above steps, the state estimation value x k and the error covariance P k are updated to accurately reflect the state at the current time k; the state estimation value x n k is fused as the final output at the current time k and is taken as the state estimation value at the next time k+1, and the error covariance matrix is taken as an input parameter for the fusion calculation at the next time k+1; The initial data, the data to be fused are vectors to be fused .

7. The fusion algorithm of claim 6, wherein, An observation space model constructed by positioning results of each array is in a form of: wherein represents the system state transition observation coefficient of different fused data at time k.

Images