A method and device for multi-source real-time positioning of AUVs based on a single seabed beacon and a dual-transducer terminal

Abstract

This invention provides a multi-source real-time positioning method and device for AUVs based on a single seabed beacon and a dual-transducer terminal, belonging to the field of underwater navigation and positioning technology. This invention configures dual acoustic transducers on a rigid AUV carrier, constructs real-time geometric constraints using the acoustic path difference under the far-field plane wave assumption, achieves dimensionality reduction by combining depth information provided by a high-precision pressure sensor, and uses extended Kalman filtering to loosely combine and optimize the geometric positioning results with strapdown inertial navigation data. This method achieves low-cost, wide-range, high-precision, and real-time navigation and positioning for deep-sea AUVs without the need for complex maneuvers and multi-beacon arrays.

Description

Technical Field

[0001] This invention relates to the field of underwater navigation technology, and in particular to an AUV multi-source real-time positioning method and device based on a single seabed beacon and a dual transducer terminal. Background Technology

[0002] As human exploration and utilization of marine resources deepens, the demand for high-precision positioning of Autonomous Underwater Vehicles (AUVs) in large-scale deep-sea operations has become increasingly urgent. Due to the severe attenuation effect of electromagnetic waves in seawater, traditional land-based and air-based positioning technologies such as Global Navigation Satellite Systems (GNSS) cannot be directly applied in underwater environments. This forces AUVs to rely on their onboard autonomous navigation sensors or external underwater acoustic positioning equipment to maintain positional accuracy after diving. Currently, mainstream underwater navigation technologies mainly include Inertial Navigation Systems (INS). Although INS can provide navigation information with high update rates, strong autonomy, and excellent stealth, the inherent drift characteristics of inertial devices cause errors to accumulate and diverge over time, making it impossible to independently meet the requirements of long-endurance, large-scale deep-sea navigation. Therefore, introducing external absolute position observation information (such as underwater acoustic positioning systems) for fusion correction is currently the most crucial and reliable means to suppress the divergence of inertial navigation errors.

[0003] Existing acoustic positioning technologies are mainly divided into long baseline (LBL), short baseline (SBL), and ultra-short baseline (USBL) systems. LBL systems utilize multiple transponder arrays deployed on the seabed for spherical intersection positioning, achieving extremely high accuracy. However, they suffer from severe drawbacks such as extremely high deployment costs, cumbersome array calibration, and difficult recovery operations, significantly limiting their widespread application in rapid maneuvering and ultra-large-scale detection missions. In contrast, SBL systems reduce the number of sensors but significantly decrease positioning accuracy. While USBL systems have high integration, their effective tracking range and positioning accuracy heavily rely on ray length. As operating depth and horizontal distance increase, angular measurement errors are significantly amplified, and underwater acoustic delay becomes increasingly severe. To overcome the high cost and deployment challenges of multi-beacon arrays, utilizing a single acoustic beacon to provide positioning information for AUVs has become an important development trend in recent years. Most existing single-beacon positioning methods are based on range intersection, virtual beacon networks, or filtering estimation algorithms, fusing single-beacon ranging, dead reckoning, and heading information to constrain horizontal positioning errors.

[0004] In the process of realizing this invention, the inventors discovered that there are at least the following problems in the prior art: the existing traditional single beacon technology still has significant technical bottlenecks in terms of deep-sea long-distance positioning, real-time performance and adaptability to long-endurance operations.

[0005] Specifically: First, relying solely on single-point distance measurements suffers from severe positional ambiguity, typically requiring AUVs to perform specific maneuvers to enhance geometric observability. This consumes a significant amount of valuable energy in practical engineering, severely reducing operational efficiency. Second, due to the acoustic ray bending effect caused by the complex sound speed profile (SSP) in the deep-sea environment, directly converting acoustic time into straight-line distance introduces huge nonlinear geometric errors, and existing methods often lack high-precision mechanistic-level compensation for these errors. Third, simple geometric cross-multiplication algorithms, lacking a high-frequency fusion mechanism, are highly susceptible to acoustic multipath effects and outliers, leading to trajectory divergence and failing to guarantee the robustness of the positioning system.

[0006] In summary, there is an urgent need in this field for an underwater combined positioning method that can maintain the advantages of low cost and convenient deployment of a single beacon, while also achieving real-time, high precision, strong robustness, and automatic elimination of abnormal interference, in order to meet the needs of large-scale operations of the next generation of full-ocean-depth AUVs. Summary of the Invention

[0007] To address the shortcomings of existing technologies and practical pain points in engineering applications, this invention aims to provide an AUV multi-source real-time positioning method and device based on a single seabed beacon and a dual-transducer terminal. By configuring dual acoustic transducers on the rigid AUV carrier, real-time geometric constraints are constructed using the acoustic path difference under the far-field plane wave assumption. Dimensionality reduction is achieved by combining depth information provided by a high-precision pressure sensor, and the geometric positioning results are loosely combined with strapdown inertial navigation (SINS) data through extended Kalman filtering (EKF) for optimization. This method achieves low-cost, wide-range, high-precision, and real-time navigation and positioning for deep-sea AUVs without the need for complex maneuvers and multiple beacon arrays.

[0008] To achieve the above objectives, the present invention adopts the following technical solution: A multi-source real-time positioning method for AUVs based on a single seabed beacon and dual transducer terminals includes: S1. Obtain the depth information of the AUV and the ray length from a single seabed beacon to the transducer at the stern of the AUV, and project the ray length onto the EN two-dimensional horizontal plane of the local northeast-northeast coordinate system; S2. Obtain the ray distance difference between a single seabed beacon and the two transducers at the bow and stern of the AUV. Based on the plane wave approximation theory and combined with the baseline length of the dual transducers, calculate the auxiliary angle α. S3. Obtain the AUV heading angle ψ provided by SINS, fuse the heading angle ψ with the auxiliary angle α, and calculate the polar coordinate angle θ of the AUV relative to the seabed beacon in the two-dimensional plane; S4. Combine the polar coordinate angle θ and the projected horizontal distance to calculate the geometric positioning two-dimensional coordinates of the AUV; S5. Construct an extended Kalman filter model, and loosely combine the geometric positioning two-dimensional coordinates as observations with the inertial navigation position provided by SINS to obtain the final positioning result of the AUV.

[0009] As a further improvement of the present invention, the specific implementation process of projecting the sound ray length onto the EN two-dimensional horizontal plane of the local northeast-northeast coordinate system in step S1 is as follows: (1) Obtain depth information Z of AUV using a high-precision pressure sensor. AUV By combining the depth of the seabed beacon with geometric calculations, the pitch angle β of the beacon relative to the AUV is obtained. The calculation formula is as follows: ; (2) Based on the pitch angle β, the ray length is projected onto a two-dimensional plane to obtain the horizontal projection distance l1, as shown in the formula: ; Where l1 is the horizontal projection distance and L1 is the ray length.

[0010] As a further improvement of the present invention, the specific implementation process of step S2 is as follows: The baseline length r of the two transducers at the bow and stern in the AUV body coordinate system is a fixed value, and the auxiliary angle α is calculated as follows: When the ray length L1 of the stern transducer is greater than the ray length L2 of the bow transducer, the following calculation formula is used: ; When the ray length L2 of the bow transducer is greater than the ray length L1 of the stern transducer: .

[0011] As a further improvement of the present invention, the specific implementation process of step S3 is as follows: A local northeast-sky two-dimensional coordinate system is established with the seabed beacon as the origin. Using the direction cosine matrix formed by the AUV's heading, pitch, and roll angles provided by SINS, the transducer coordinates are transformed from the volume coordinate system to the navigation coordinate system. Based on the quadrant in which the AUV is located in the local northeast-sky coordinate system, the polar coordinate angle θ is determined through the addition and subtraction relationship between the heading angle ψ and the auxiliary angle α. The formula is as follows: ; The symbols in the formula have the following meanings: dL1 is the ray length from the AUV to the stern transducer, and dL2 is the ray length from the AUV to the bow transducer; q is the quadrant in which the AUV is located in the local northeast-sky (ENU) coordinate system, with values ​​of 1, 2, 3, and 4, corresponding to the first to fourth quadrants respectively; ψ is the heading angle of the AUV provided by the strapdown inertial navigation system, in degrees; α is the auxiliary angle used for polar coordinate angle calculation; θ is the polar coordinate angle of the AUV relative to the reference point in the EN plane, in degrees.

[0012] As a further improvement of the present invention, the specific implementation process of step S4 is as follows: The position of the AUV in the EN plane is calculated based on the polar coordinates θ and the l1 of the EN two-dimensional plane projection, using the following formula: ; In the formula, l1 is the projected length of the distance from the AUV to the reference point on the EN two-dimensional plane; x1 is the position coordinate of the AUV on the eastward axis of the local northeast-northeast coordinate system; y1 is the position coordinate of the AUV on the northward axis of the local northeast-northeast coordinate system.

[0013] As a further improvement of the present invention, the specific process of constructing the extended Kalman filter model and performing loose combination fusion in step S5 includes: (1) Establish a 15-dimensional SINS error state vector, including three-dimensional attitude error, three-dimensional velocity error, three-dimensional position error, three-dimensional gyroscope zero bias and three-dimensional accelerometer zero bias; (2) Establish the continuous-time state equation of the system and discretize it; (3) Convert the geometric positioning two-dimensional coordinates into the observation longitude and observation latitude in the geodetic coordinate system, and subtract them from the indicated longitude and indicated latitude calculated by SINS to construct the observation residual equation; (4) Calculate the Kalman gain and perform measurement updates. Inject the updated error state vector as a feedback compensation quantity into SINS and output the fused navigation and positioning data.

[0014] As a further improvement of the present invention, the expression for the 15-dimensional error state vector is as follows: ; In the formula: the superscript T denotes matrix transpose; R15 denotes 15-dimensional real space; This is the three-dimensional attitude error vector of SINS; This is the three-dimensional velocity error vector of SINS; This is the three-dimensional position error vector of SINS; This is the zero bias vector of the three-dimensional gyroscope; This is the zero bias vector of the three-dimensional accelerometer.

[0015] As a further improvement of the present invention, the observation residual equation is: ; In the formula, For the first Position observation residual vector at time; , These are the longitude and latitude calculated from SINS, respectively. , These are the observed longitude and observed latitude obtained through geometric positioning, respectively.

[0016] Another objective of this invention is to provide an AUV multi-source real-time positioning device based on a single seabed beacon and a dual transducer terminal, comprising: The projection module is used to acquire the depth information of the AUV and the ray length from a single seabed beacon to the stern transducer, and to project the ray length onto a two-dimensional horizontal plane. The auxiliary angle calculation module is used to obtain the ray distance difference between a single seabed beacon and the two transducers at the bow and stern of the AUV, and to calculate the auxiliary angle α by combining the baseline length of the two transducers. The polar coordinate calculation module is used to obtain the AUV heading angle provided by SINS, fuse the heading angle with the auxiliary angle α, and calculate the polar coordinate angle θ of the AUV relative to the seabed beacon in a two-dimensional plane. The geometric positioning module is used to calculate the two-dimensional geometric positioning coordinates of the AUV by combining the polar coordinate angle θ and the projected horizontal distance. The model fusion module is used to construct an extended Kalman filter model, which fuses the geometric positioning two-dimensional coordinates with the inertial navigation position provided by SINS to obtain the final positioning result of the AUV.

[0017] Compared with the prior art, the present invention has the following significant advantages: 1. Traditional SBL (Self-Borrowing Baseline) systems rely on a surface vessel to deploy a large-scale acoustic array for three-dimensional rendezvous. This not only results in poor concealment and limited maneuverability but is also highly susceptible to baseline non-rigid errors caused by wave turbulence and hull deformation. This invention directly fixes dual transducers to an absolutely rigid AUV carrier to form a miniature fixed baseline, eliminating measurement errors introduced by baseline deformation at their physical source. Simultaneously, a high-precision pressure sensor is introduced to lock the Z-axis depth, actively reducing and decoupling the complex three-dimensional nonlinear rendezvous to a two-dimensional plane, achieving fully autonomous and covert navigation of underwater targets.

[0018] 2. Traditional USBL systems rely on phase difference angle measurement. As the target acoustic ray length increases, the geometric angle measurement error is significantly amplified, making them unsuitable for large-scale operations in deep seas. This invention establishes a geometric constraint model based on a single beacon dual transducer configuration and introduces the far-field plane wave approximation theory. As the acoustic propagation distance increases, the acoustic rays reaching the dual transducers become increasingly parallel, enabling the system to maintain extremely high positioning accuracy and robustness in deep sea operations.

[0019] 3. Compared with the high cost of multi-node seabed deployment, calibration and recovery of traditional long baseline (LBL) systems, this invention only requires the deployment of a single acoustic beacon on the seabed, achieving high-precision underwater positioning with extremely low hardware configuration costs. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings: Figure 1 This is a schematic diagram illustrating the principle of a single acoustic beacon and dual transducer positioning system in an embodiment of the present invention; wherein, Figure 1 (a) shows the state when the acoustic ray length of the stern transducer is greater than that of the bow transducer. Figure 1 (b) shows the state when the acoustic ray length of the bow transducer is greater than that of the stern transducer.

[0021] Figure 2 This is a schematic diagram illustrating the solution principle for polar coordinate system geometric positioning within a local northeast plane in an embodiment of the present invention.

[0022] Figure 3 This is a flowchart of the algorithm for loosely combining SINS and geometric positioning results based on extended Kalman filtering in an embodiment of the present invention.

[0023] Figure 4 This is a schematic diagram illustrating the geometric angle relationship between the heading angle vector and the sound wave ray propagation direction vector in an embodiment of the present invention.

[0024] Figure 5 This is a schematic diagram illustrating the three-dimensional geometric analysis and derivation of the ray length difference based on the physical structure of the dual transducers in an embodiment of the present invention.

[0025] Figure 6 This is a flowchart illustrating the method of the present invention. Detailed Implementation

[0026] 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.

[0027] The present invention will now be described in further detail with reference to the accompanying drawings: This invention proposes a low-cost, high-precision real-time combined positioning architecture specifically designed for large-scale deep-sea exploration missions. Its core technological idea is to overcome the spatial degree-of-freedom observation limitations of existing single-beacon positioning systems. By configuring two acoustic transducers with a fixed spacing on the rigid carrier structure of an AUV, this method cleverly transforms the Time-Difference-of-Arrival (TDOA) information into a spatial geometric constraint of physical acoustic path difference. Combined with powerful dimensionality reduction and decoupling in the vertical dimension by a high-precision depth sensor, this method reduces the highly complex three-dimensional spatial addressing problem, which is prone to getting trapped in nonlinear local optima, into a deterministic two-dimensional polar coordinate analytical solution problem. Subsequently, using the strapdown inertial navigation system (SINS) onboard the system, the geometric transient solution is smoothed and multi-source information is fused within the extended Kalman filter (EKF) framework.

[0028] To fully present the implementation details and practical effectiveness of this invention, the following will elaborate on it from three core dimensions: overall implementation environment and hardware architecture, core navigation and positioning methods and algorithms, and specific application examples and effect verification.

[0029] Example 1: Overall Implementation Environment and Hardware Architecture To achieve underwater integrated navigation and positioning for AUVs based on the geometric constraints of a single acoustic beacon and dual transducers, this invention first constructs a highly adaptable deep-sea operational environment and a highly integrated hardware payload architecture. The specific implementation environment and hardware architecture configuration are as follows: 1. Implementation environment and coordinate system definition like Figure 1As shown, the implementation environment of this invention is mainly aimed at exploration operations in large-scale open waters of the deep sea and open ocean. In the deep-sea environment, the inhomogeneity of seawater (affected by factors such as temperature, salinity, and hydrostatic pressure) causes dynamic changes in sound speed, thus forming a sound speed profile (SSP) with depth gradient. To unify the positioning benchmark and spatial scale, this invention uses the absolute position of a single acoustic beacon pre-deployed on the seabed as the coordinate origin to establish a local northeast-north sky (ENU) navigation coordinate system (i.e., the n-system). Before the AUV dives, the initial absolute position and quadrant information of the AUV are obtained using a surface global navigation satellite system (GNSS). After entering the water, the AUV performs long-endurance autonomous operations within the three-dimensional space constructed with the aforementioned single acoustic beacon on the seabed as the origin.

[0030] 2. AUV rigid body load and core hardware configuration as support like Figure 2 As shown, the core algorithm for three-dimensional dimensionality reduction and real-time geometric constraints described in this invention is integrated with the following core navigation and underwater acoustic sensor units on the AUV carrier: Dual acoustic transducers: such as Figure 2 As shown, a stern transducer and a bow transducer are mounted on the rigid body of the AUV along its bow and stern axes, respectively. In the AUV's body coordinate system (b-frame), there is a physically fixed and rigorously calibrated baseline length *r* between the stern and bow transducers. In this embodiment, this baseline length *r* is 2.3 meters. This fixed baseline is a physical prerequisite for subsequently constructing a geometric constraint model for deep-sea far-field plane waves.

[0031] High-precision pressure sensor: The AUV is equipped with a high-precision pressure sensor (such as the ImpactSubsea ISD4000 depth gauge, whose measurement accuracy is better than 0.01% of the full range). This sensor is used to obtain high-precision depth information of the water layer where the AUV is located in real time. Its core function is to strictly decouple the complex three-dimensional slant distance calculation and reduce the dimensionality to a two-dimensional horizontal plane, which greatly reduces the mathematical dimension of the positioning problem.

[0032] High-precision strapdown inertial navigation system (SINS): The AUV is rigidly mounted with a high-precision strapdown inertial navigation system (such as fiber-optic SINS). The SINS is mainly used to provide the AUV with high-update-rate three-dimensional attitude information (including heading, pitch angle, and roll angle) and continuous dead reckoning (velocity, position) data in real time. On the one hand, the high-precision attitude data is used to construct the direction cosine matrix (DCM), accurately projecting and aligning the transducer baseline in the volume coordinate system to the global navigation coordinate system; on the other hand, its output heading angle data is used to directly participate in the subsequent calculation of the AUV's polar coordinate angle in the two-dimensional plane.

[0033] Auxiliary Environmental Perception Matrix: Preferably, the AUV is also equipped with a CTD (Conductivity, Temperature, Depth) meter to acquire real-time hydrological information and sound velocity profile (SSP) data of the operating sea area, thereby providing accurate prior environmental parameters for subsequent ray tracing algorithms and sound ray bending error correction. Simultaneously, the system is internally configured with a chip-scale atomic clock (CSAC) to ensure strict time synchronization of data from all sensors.

[0034] By constructing the overall implementation environment and hardware architecture described above, this invention effectively eliminates the large and expensive multi-beacon array required by traditional long baseline (LBL) systems. It utilizes only a minimalist acoustic configuration of "single seabed beacon + carrier dual transducer" and integrates multi-source heterogeneous data from pressure sensors and SINS, providing a complete data input boundary for the implementation of subsequent high-precision fusion positioning methods.

[0035] Example 2: Core Navigation and Positioning Method and Algorithm The core of this invention lies in achieving high-precision continuous positioning of deep-sea AUVs through acoustic feature extraction, geometric constraint construction, and multi-source data fusion. For example... Figure 3 As shown in the flowchart, the overall algorithm flow comprises two key stages: pure geometric positioning constraints and loose combination optimization using the Extended Kalman Filter (EKF). (Reference) Figure 6 As shown, the specific execution steps of the method of the present invention are as follows: Step S1: Depth Dimensionality Reduction and Horizontal Projection of Acoustic Ray Length. Direct ranging and positioning in the three-dimensional space of the deep sea is prone to introducing huge nonlinear errors. This embodiment adopts a technical solution of dimensionality reduction using high-precision depth information. Specifically, high-precision depth information Z is obtained using a pressure sensor mounted on an AUV. AUV Simultaneously, the system acquires the ray length L1 of a single seabed beacon reaching the AUV's stern transducer (Transducer 1). Combining the depth information and ray length, the pitch angle β of the seabed beacon relative to the AUV is calculated using the following formula: ; Subsequently, the length of the sound ray in three-dimensional space is projected onto a two-dimensional horizontal plane to obtain the horizontal projection distance l1, which is calculated as follows: ; Where l1 is the horizontal projection distance and L1 is the ray length.

[0036] This step strictly constrains the complex three-dimensional spatial positioning problem within a two-dimensional horizontal plane.

[0037] Step S2: Obtaining the acoustic ray distance difference based on the plane wave approximation and solving the auxiliary angle. Under the far-field deep-sea operation conditions, the acoustic wave front arriving at the transducer can be approximated as a plane wave propagating in parallel. In this embodiment, a stratified constant sound speed ray tracing algorithm is introduced to calculate and eliminate the acoustic ray bending error caused by the sound speed profile, and obtain the true acoustic ray distance difference from a single seabed beacon to the bow and stern dual transducers of the AUV. Let the fixed baseline length of the bow and stern dual transducers of the AUV in the vehicle coordinate system be r (2.3 m in this embodiment). As Figure 5 shown, in this embodiment, a three-dimensional geometric analysis and derivation of the acoustic ray difference is carried out based on the physical structure of the dual transducers, clarifying the spatial geometric relationship among the acoustic ray propagation direction, the dual transducer baseline vector, and the acoustic ray difference, providing a strict geometric basis for solving the auxiliary angle under the plane wave approximation. Based on the obtained absolute acoustic ray length difference, an auxiliary angle α between 0° and 90° is introduced. The calculation of this auxiliary angle depends on the relative magnitude of the acoustic ray lengths received by the two transducers. As Figure 1 (a) shows, when the acoustic ray length L1 of the stern transducer (Transducer1) is greater than the acoustic ray length L2 of the bow transducer (Transducer2), that is, L1 > L2, the calculation formula is as follows: ; As Figure 1 (b) shows, when the acoustic ray length L2 of the bow transducer is greater than the acoustic ray length L1 of the stern transducer, that is, L1 < L2: ; Step S3: Constructing a two-dimensional polar coordinate system by fusing the heading angle. A local east-north-up (ENU) two-dimensional coordinate system is established with the seabed beacon as the origin. The high-precision heading angle ψ of the AUV provided by the SINS system is obtained. This heading angle ψ is geometrically fused with the observed auxiliary obtained in Step S2 to solve the polar coordinate θ of the AUV in the two-dimensional plane. As Figure 2 and Figure 4 shown, in a typical geometric configuration (for example, the AUV is in the second quadrant and the heading is parallel to the east-west axis), a generalized calculation model of the polar coordinate angle can be derived based on geometric relationships such as the parallel line theorem. This calculation model adaptively selects and solves the formula according to the quadrant where the AUV is located relative to the seabed beacon in the ENU coordinate system.

[0038] Specifically, combining the direction cosine matrix composed of the heading angle, pitch angle, and roll angle of the AUV provided by the SINS, the transducer coordinates are converted from the body coordinate system to the navigation coordinate system, and according to the quadrant where the AUV is located in the local east-north-up (ENU) coordinate system, the polar coordinate angle θ is determined through the addition and subtraction operation relationship between the heading angle ψ and the auxiliary angle α. The formula is: The symbols in the formula have the following meanings: dL1 is the ray length from the AUV to the stern transducer, and dL2 is the ray length from the AUV to the bow transducer; q is the quadrant in which the AUV is located in the local northeast-northeast (ENU) coordinate system, with values ​​of 1, 2, 3, and 4, corresponding to the first to fourth quadrants respectively; ψ is the heading angle of the AUV provided by the strapdown inertial navigation system (SINS), in degrees; α is the auxiliary angle used for polar coordinate angle calculation; θ is the polar coordinate angle of the AUV relative to the reference point in the EN plane, in degrees.

[0039] Step S4: Calculation of Geometric Rectangular Coordinates After determining the polar coordinate angle θ, and combining it with the planar projection distance l1 obtained in Step S1, calculate the position of the AUV in the EN plane. The formula is: ; In the formula, l1 is the projected length of the distance from the AUV to the reference point on the EN two-dimensional plane; x1 is the position coordinate of the AUV on the eastward axis of the local northeast-northeast (ENU) coordinate system; y1 is the position coordinate of the AUV on the northward axis of the local northeast-northeast (ENU) coordinate system.

[0040] Step S5: To suppress the accumulation of inertial navigation system errors over time and overcome noise fluctuations in pure geometric positioning, this embodiment constructs an EKF loose combination optimization model. The specific implementation steps are as follows: (1) First, define a 15-dimensional state vector X containing the total SINS error. Specifically, it includes: a three-dimensional attitude error vector (roll, pitch, and yaw angle errors), a three-dimensional velocity error vector (east, north, and yaw velocity deviations), a three-dimensional position error vector (latitude, longitude, and altitude errors), and a three-dimensional gyroscope zero-bias vector and a three-dimensional accelerometer zero-bias vector; the state vector expression is: ; In the formula: the superscript T denotes matrix transpose; R 15 Represents a 15-dimensional real number space; This is the three-dimensional attitude error vector of SINS; This is the three-dimensional velocity error vector of SINS; b is the three-dimensional position error vector of SINS; g b is the zero bias vector of the three-dimensional gyroscope; a This is the zero bias vector of the three-dimensional accelerometer.

[0041] (2) Establishing the system state equations and discretization prediction. First, establish the continuous-time system state equations as follows: ; In the formula, The time derivative of the state vector; Here is the state transition matrix for a continuous-time system; This is the system noise driving matrix; This is the system process noise vector.

[0042] Discretizing the state transition matrix using a first-order Taylor expansion yields the discrete-time state prediction equation: ; In the formula, For the first The state vector at time 1 is predicted in one step. For the first Update the state vector value at time step; This is the discretized state transition matrix.

[0043] The corresponding covariance prediction equation is: ; In the formula, For the first The one-step prediction of the covariance matrix at time t; For the first The updated value of the covariance matrix at time t; Let be the system process noise covariance matrix.

[0044] (3) Construct the position observation residual equation. Convert the obtained geometric positioning two-dimensional coordinates into the observation longitude in the geodetic coordinate system. and observation latitude This is compared with the indicated longitude calculated using SINS mechanical choreography. and indicating latitude Differences are calculated to construct the position observation residual equations. : ; In the formula, For the first Position observation residual vector at time; , These are the longitude and latitude calculated from SINS, respectively. , These are the observed longitude and observed latitude obtained through geometric positioning, respectively.

[0045] (4) Perform Kalman gain calculation and measurement update. First, calculate the Kalman gain... Kalman gain at time step : ; In the formula, H k R is the observation matrix at time k; k For the observation noise covariance matrix; P k|k-1Let P represent the prior error covariance matrix at time k. In this invention, this matrix is ​​a 15×15 dimensional square matrix, corresponding to the 15-dimensional system state vector defined in this invention, which includes attitude error, velocity error, position error, gyroscope bias, and accelerometer bias; it is composed of the posterior error covariance matrix P at time k-1. k|k-1 The state transition matrix is ​​obtained through iterative calculation via a time update step. The symbol T is the matrix transpose operator, which represents performing a linear algebraic operation to interchange the rows and columns of the matrix immediately to its left.

[0046] The error state vector is updated using the aforementioned observation residuals: ; In the formula, For the first The state vector update value at time 1.

[0047] Measurement update of the covariance matrix: ; In the formula, For the first The updated value of the covariance matrix at time t.

[0048] The updated error state vector is injected into SINS as feedback compensation to complete the closed-loop correction of system errors and output continuous and smooth AUV fusion positioning results.

[0049] Example 3: Specific Application Examples and Effect Verification To fully verify the effectiveness and robustness of the underwater integrated navigation and positioning method described in this invention under real marine environments and extreme water depth conditions, this embodiment uses a verification framework that combines real-world testing and numerical simulation.

[0050] 1. Implementation environment and configuration of the flight test platform During the deep-sea operational test in the South China Sea in July 2024, the seabed depth in the operating area was 2,388 meters. The test platform mother ship was equipped with a high-precision ultra-short baseline system (USBL, with a nominal accuracy of 0.06% slant range and incorporating fiber optic SINS for motion compensation) as the ground truth reference for the measured positioning. During this period, the AUV conducted extensive submersible navigation within a water depth range of 130 meters to 380 meters, and the three-dimensional acoustic ray length between itself and the monoacoustic beacon fixed on the seabed covered a range of 8 to 16 kilometers.

[0051] 2. Performance Analysis of Complex Trajectory Tracking In this sea trial, the system's positioning accuracy was quantitatively evaluated by designing different combinations of maneuvering conditions, including straight lines, broken lines, and curves. The system's performance indicators driven by measured data are shown in Table 1. Table 1: Statistical Distribution of Positioning Errors in Multi-Trajectory Working Conditions Analyzing the measured data of the aforementioned polygonal trajectory and conducting statistical analysis using the Pearson correlation coefficient method reveals a positive correlation between depth variation amplitude perturbation and the horizontal two-dimensional error envelope, confirming the impact of the dimensionality reduction uncertainty of depth measurement on the final two-dimensional fusion positioning accuracy. Further global regression analysis based on the two-dimensional positioning error of the entire flight shows that the positioning degradation rate of the system of this invention exhibits a linear evolution law of approximately 0.46 m / km. This means that in large-scale mapping operations with acoustic ray lengths extended to 20 kilometers, the positioning error of this invention can still be controlled within approximately 10 meters.

[0052] 3. Simulation and Robustness Verification of Extreme Water Depth Span To comprehensively evaluate the theoretical positioning boundary of the present invention under extreme deep-water conditions, this embodiment utilizes a high-performance computing cluster to design simulation experiments with seabed depth gradients of 500 meters, 1000 meters, 1500 meters, 2500 meters, and up to 3500 meters.

[0053] In the simulation test conditions, a residual error of 1.5 m / s sound velocity calibration was actively introduced into the system, and Gaussian white noise was superimposed on the measurement. Under this extreme test setting, the system performance is shown in Table 2: Table 2: Statistics on Extreme Positioning Accuracy Control (RMSE / meter) under Different Water Depth Gradients As shown in Table 2, the accuracy control of this invention is stable in the mainstream operating depth of 500 to 1500 meters. However, under the extreme water depth of 3500 meters, due to the deterioration of the geometrical precision attenuation factor (GDOP) of the horizontal plane and the amplification of the sound velocity residual caused by the near vertical incidence of seabed sound waves, the 15-dimensional multi-source loose combination filter model proposed in this invention still maintains the global convergence of the trajectory envelope by relying on the covariance constraint mechanism, without the occurrence of filter divergence, demonstrating extremely strong robustness in deep-sea operations.

[0054] In summary, this invention effectively avoids the high deployment cost of traditional long baseline (LBL) multi-beacon networks by fusing a dual-transducer far-field geometric prior difference model, implementing three-dimensional dimensionality reduction calculations based on high-precision pressure sensors, and combining a ray tracing iterative algorithm with a 15-state EKF closed-loop compensation mechanism. It also overcomes the positional ambiguity caused by traditional single-beacon addressing. This invention balances extremely low hardware configuration costs with high positioning robustness in deep-sea ultra-long-range environments, meeting the needs of large-scale, full-ocean-depth underwater integrated positioning technology.

[0055] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A multi-source real-time positioning method for AUVs based on a single seabed beacon and a dual transducer terminal, characterized in that, Two acoustic transducers are fixedly connected to the rigid carrier of the AUV along the bow and stern axis. The baseline length between the two transducers is a fixed calibration value. The method includes: S1. Obtain the depth information of the AUV and the ray length from a single seabed beacon to the transducer at the stern of the AUV, and project the ray length onto the EN two-dimensional horizontal plane of the local northeast-northeast coordinate system; S2. Obtain the ray distance difference between a single seabed beacon and the two transducers at the bow and stern of the AUV. Based on the plane wave approximation theory and combined with the baseline length of the dual transducers, calculate the auxiliary angle α. S3. Obtain the AUV heading angle ψ provided by SINS, fuse the heading angle ψ with the auxiliary angle α, and calculate the polar coordinate angle θ of the AUV relative to the seabed beacon in the two-dimensional plane; S4. Combine the polar coordinate angle θ and the projected horizontal distance to calculate the geometric positioning two-dimensional coordinates of the AUV; S5. Construct an extended Kalman filter model, and loosely combine the geometric positioning two-dimensional coordinates as observations with the inertial navigation position provided by SINS to obtain the final positioning result of the AUV.

2. The method according to claim 1, characterized in that, The specific implementation process of projecting the sound ray length onto the EN two-dimensional horizontal plane of the local northeast-northeast coordinate system in step S1 is as follows: (1) Obtain depth information Z of AUV using a high-precision pressure sensor. AUV By combining the depth of the seabed beacon with geometric calculations, the pitch angle β of the beacon relative to the AUV is obtained. The calculation formula is as follows: ； (2) Based on the pitch angle β, the ray length is projected onto a two-dimensional plane to obtain the horizontal projection distance l1, as shown in the formula: ； Where l1 is the horizontal projection distance and L1 is the ray length.

3. The method according to claim 1, characterized in that, The specific implementation process of step S2 is as follows: The baseline length r of the two transducers at the bow and stern in the AUV body coordinate system is a fixed value, and the auxiliary angle α is calculated as follows: When the ray length L1 of the stern transducer is greater than the ray length L2 of the bow transducer, the following calculation formula is used: ； When the ray length L2 of the bow transducer is greater than the ray length L1 of the stern transducer: 。 4. The method according to claim 1, characterized in that, The specific implementation process of step S3 is as follows: A local northeast-sky two-dimensional coordinate system is established with the seabed beacon as the origin. Using the direction cosine matrix formed by the AUV's heading, pitch, and roll angles provided by SINS, the transducer coordinates are transformed from the volume coordinate system to the navigation coordinate system. Based on the quadrant in which the AUV is located in the local northeast-sky coordinate system, the polar coordinates θ are determined through the addition and subtraction relationship between the heading angle ψ and the auxiliary angle α, as shown in the formula: ； The symbols in the formula have the following meanings: dL1 is the ray length from the AUV to the stern transducer, and dL2 is the ray length from the AUV to the bow transducer; q is the quadrant in which the AUV is located in the local northeast-sky (ENU) coordinate system, with values ​​of 1, 2, 3, and 4, corresponding to the first to fourth quadrants respectively; ψ is the heading angle of the AUV provided by the strapdown inertial navigation system, in degrees; α is the auxiliary angle used for polar coordinate angle calculation; θ is the polar coordinate angle of the AUV relative to the reference point in the EN plane, in degrees.

5. The method according to claim 1, characterized in that, The specific implementation process of step S4 is as follows: The position of the AUV in the EN plane is calculated based on the polar coordinates θ and the l1 of the EN two-dimensional plane projection, using the following formula: ； In the formula, l1 is the projected length of the distance from the AUV to the reference point on the EN two-dimensional plane; x1 is the position coordinate of the AUV on the eastward axis of the local northeast-northeast coordinate system; y1 is the position coordinate of the AUV on the northward axis of the local northeast-northeast coordinate system.

6. The method according to claim 1, characterized in that, The specific process of constructing the extended Kalman filter model and performing loose combination fusion in step S5 includes: (1) Establish a 15-dimensional SINS error state vector, including three-dimensional attitude error, three-dimensional velocity error, three-dimensional position error, three-dimensional gyroscope zero bias and three-dimensional accelerometer zero bias; (2) Establish the continuous-time state equation of the system and discretize it; (3) Convert the geometric positioning two-dimensional coordinates into the observation longitude and observation latitude in the geodetic coordinate system, and subtract them from the indicated longitude and indicated latitude calculated by SINS to construct the observation residual equation; (4) Calculate the Kalman gain and perform measurement updates. Inject the updated error state vector as a feedback compensation quantity into SINS and output the fused navigation and positioning data.

7. The method according to claim 6, characterized in that, The expression for the 15-dimensional error state vector is: ； In the formula: the superscript T denotes matrix transpose; R 15 Represents a 15-dimensional real number space; This is the three-dimensional attitude error vector of SINS; This is the three-dimensional velocity error vector of SINS; This is the three-dimensional position error vector of SINS; This is the zero bias vector of the three-dimensional gyroscope; This is the zero bias vector of the three-dimensional accelerometer.

8. The method according to claim 6, characterized in that, The observation residual equation is: ； In the formula, For the first Position observation residual vector at time; , These are the longitude and latitude calculated from SINS, respectively. , These are the observed longitude and observed latitude obtained through geometric positioning, respectively.

9. An AUV multi-source real-time positioning device based on a single seabed beacon and a dual transducer terminal, characterized in that, include: The projection module is used to acquire the depth information of the AUV and the ray length from a single seabed beacon to the stern transducer, and to project the ray length onto a two-dimensional horizontal plane. The auxiliary angle calculation module is used to obtain the ray distance difference between a single seabed beacon and the two transducers at the bow and stern of the AUV, and to calculate the auxiliary angle α by combining the baseline length of the two transducers. The polar coordinate calculation module is used to obtain the AUV heading angle provided by SINS, fuse the heading angle with the auxiliary angle α, and calculate the polar coordinate angle θ of the AUV relative to the seabed beacon in a two-dimensional plane. The geometric positioning module is used to calculate the two-dimensional geometric positioning coordinates of the AUV by combining the polar coordinate angle θ and the projected horizontal distance. The model fusion module is used to construct an extended Kalman filter model, which fuses the geometric positioning two-dimensional coordinates with the inertial navigation position provided by SINS to obtain the final positioning result of the AUV.

Images