Accuracy evaluation method of underwater acoustic positioning system with irregular offshore stations

Abstract

The present invention belongs to the field of underwater target navigation and positioning, and discloses a method for evaluating the accuracy of an underwater acoustic positioning system for irregular offshore stations. The specific steps are: 1) setting initial conditions, selecting a single-point, trajectory, or regional evaluation mode, and establishing theoretical values ​​for the positions of the target to be measured and the long-baseline array elements; 2) field information collection and analysis, combining measurement equipment and offshore information to estimate station site errors and ranging errors, and obtaining an estimated maximum effective range through a pull-off test; 3) accuracy simulation evaluation, using the Monte Carlo method to generate station site positions and slant range measurements with errors, performing error transfer simulation and parameter solution under effective range constraints, and performing accuracy estimation using the root mean square error as an indicator. This method has the characteristics of good adaptability to complex and irregular offshore stations, accurate error transfer reproduction, and a wide range of application scenarios, and can provide technical support for high-quality positioning and measurement of underwater targets.

Description

Technical Field

[0001] The present invention belongs to the field of underwater target navigation and positioning, and specifically relates to a method for evaluating the accuracy of an underwater acoustic positioning system for irregular offshore stations. The method is a technology for evaluating the positioning accuracy of a single point, trajectory, area, and other diverse working conditions when a long-baseline underwater acoustic positioning system is used to position and measure underwater targets in an offshore environment. Background Art

[0002] The long-baseline underwater acoustic positioning system is a commonly used technical means for underwater target positioning. It has the advantages of supporting station redundant design, high positioning accuracy, and robust measurement performance. However, it also has problems such as being greatly restricted by the marine environment and the measurement performance being highly dependent on the station geometry. In order to avoid the risk of measurement failure and ensure the high-quality implementation of measurement work in engineering applications, it is very necessary to adopt reliable accuracy assessment technology.

[0003] During offshore operations, the array elements of the long-baseline underwater acoustic system are pre-deployed in a specific geometric configuration in the operating sea area. They receive positioning acoustic signals from the target to be measured (a cooperative acoustic beacon must be installed) and, after signal processing, obtain measurement elements such as slant range and radial velocity. The array element data is then aggregated to a data processing platform for intersection and solution, ultimately yielding the positioning result. In actual operations, the achievable positioning accuracy is often not the equipment's factory calibration accuracy, and the limitations of the on-site environment and operating conditions must be considered. Complex and harsh operating conditions can lead to reduced accuracy, or even undetectable targets and unusable data. To ensure that the underwater acoustic positioning system meets engineering requirements, a preliminary assessment of measurement performance and accuracy must be conducted, taking into account the environmental characteristics of the operating area and the operating conditions.

[0004] There are two main factors that affect the performance of long-baseline underwater acoustic positioning systems in maritime applications: one is the influence of geometric station layout. The seabed topography in the operating sea area usually has certain undulations. When there are multiple types of array elements on the surface and underwater, their geometric configuration may show obvious irregularities, which have an important impact on the positioning characteristics, especially for the measurement of vertically moving targets. When a geometric configuration with coplanar array elements is adopted, the vertical error will increase rapidly as the baseline increases; the second is the influence of the underwater acoustic channel. The sound velocity gradients in different sea areas are significantly different. Even in the same sea area, there are often large seasonal changes, especially in shallow sea areas. The strong attenuation sound channel under thermocline conditions and the surface sound channel under mixed layer conditions can make the effective range of the underwater acoustic positioning system differ by several times.

[0005] In recent years, with the continuous advancement of underwater acoustic technology, technologies such as environmentally adaptive underwater acoustic positioning signal processing, high-precision measurement, and precise calibration of underwater stations have been developed, greatly improving the performance of underwater acoustic positioning systems. However, the complexity of environmental changes and the diversity of offshore operating conditions cannot be solved solely by underwater acoustic technology. Conventional theoretical methods are difficult to effectively estimate the accuracy of underwater acoustic positioning systems, especially in the case of irregular offshore station layouts. It is necessary to combine numerical simulation technology with field observation information to develop underwater acoustic positioning system accuracy assessment technology that meets the requirements of offshore engineering applications and provide technical support for high-quality positioning and measurement of underwater targets. Summary of the Invention

[0006] To address the problem that the measurement performance of long-baseline underwater acoustic positioning systems under irregular station geometry is strongly correlated with the station geometry and significantly affected by the environment, the present invention proposes a method for evaluating the accuracy of long-baseline underwater acoustic positioning systems under irregular offshore station geometry. This method uniformly models underwater targets and measurement arrays under irregular offshore station geometry, estimates site error and ranging error in combination with field information, then simulates error transfer using the Monte Carlo method. The accuracy evaluation result is obtained using the root mean square error as an indicator. This method provides diverse application modes for single-point, trajectory, and regional conditions. It has good adaptability to complex and irregular offshore station configurations, accurate error transfer reproduction, and a wide range of application scenarios. It solves the problem of multi-dimensional positioning accuracy evaluation of long-baseline underwater acoustic positioning systems under irregular offshore station geometry. This method is of certain reference value for pre-discovering the measurement performance and defects of underwater acoustic positioning systems under different station geometry conditions, and then conducting optimization design and performance evaluation. It can provide technical support for high-quality positioning measurement of underwater targets.

[0007] The technical solution of the present invention is:

[0008] A method for evaluating the accuracy of an underwater acoustic positioning system with irregularly distributed offshore stations is proposed. First, initial conditions are set. For underwater targets in different states, one of the single-point evaluation mode, trajectory evaluation mode, or regional evaluation mode is selected to establish the true value of the simulation model and the theoretical value of the long-baseline array element position. Second, on-site information collection and analysis are carried out. The station location error and ranging error are estimated by combining the measurement equipment with test information under offshore conditions, and the estimated maximum effective range is obtained through a pull-off test. Third, an accuracy simulation evaluation is carried out. The station location and slant range measurement values ​​with errors are simulated based on the theoretical values. The nonlinear least squares method is used for iterative calculation and solution to obtain large-sample simulated measurement values. The root mean square error (RMSE) is then used for statistical calculation to obtain the accuracy evaluation results under the three modes of single-point evaluation, trajectory evaluation, or regional evaluation.

[0009] The specific steps are as follows:

[0010] Step 1: Set initial conditions

[0011] The zero point of the theoretical initial position of the target is set as the origin O, and the survey station coordinate system is established, with the OX axis pointing to the east, the OY axis pointing to the north, and the OZ axis pointing vertically upward.

[0012] Step (1.1) Establishing the evaluation model

[0013] Three evaluation modes are provided for underwater targets in different states:

[0014] Single point evaluation mode: evaluates underwater point targets with fixed positions and constant target parameters, which are expressed as X in the measurement coordinate system. P =[x P ,y P ,z P ];

[0015] Trajectory evaluation mode: evaluates the trajectory of underwater targets, whose positions change with time. The target parameters are the sequence of positions over time, which are expressed in the measurement coordinate system as Where l represents the time node number of the trajectory sequence, l = 1, 2, ..., N L , N L is the number of sequence samples;

[0016] Regional assessment mode: The coverage area of ​​underwater target activities is assessed. Its location is grid-based according to the operating sea area, but does not change with time. The target parameters are matrices, which are expressed in the measurement coordinate system as Where m and n represent the horizontal and vertical grid point numbers respectively, m=1,2,…,M A 、n=1,2,…,N A , M A 、N A Represents the number of horizontal and vertical grids respectively.

[0017] Step (1.2) Establishing the measurement geometry

[0018] According to the station layout design, the long baseline array with a certain geometric configuration can be expressed in the measurement coordinate system as follows: Where j is the array element number, j = 1, 2, ..., N S , N S is the total number of array elements, N S ≥3; The geometric configuration of the measurement array is defined by the array elements and constrained by the effective range of each array element, while the measurement geometry is composed of the relative positions of the target and the measurement array.

[0019] Step 2: On-site information collection and analysis

[0020] Step (2.1) Estimate the site error

[0021] Under offshore operation conditions, for array elements carried on surface platforms, strapdown installation with satellite navigation and positioning equipment is adopted, and the station error is directly determined by the accuracy of satellite navigation and positioning equipment; for array elements carried on underwater fixed duty platforms, on-site calibration is performed using shipborne calibration equipment, and the station error is determined by the accuracy of shipborne calibration equipment. The station error of each array element in the X, Y, and Z directions is expressed as

[0022] Step (2.2) estimates the ranging error

[0023] Under offshore operating conditions, a set of slant range observation samples is collected under typical measurement distance conditions using the positioning acoustic beacon and receiving array elements of the underwater fixed standby platform or surface ship platform equipped with the hydroacoustic positioning system. N R is the number of sequence samples; at the same time, according to the position information of the underwater fixed guard platform and the surface ship platform, the calculated samples with the same sequence length and time synchronization as the slant range observation samples are obtained. in is the acoustic beacon position at the kth moment, is the receiving array element position at the kth moment; according to the slant range observation sample and calculate samples The residual estimation of the ranging error σ R , the calculation formula is

[0024]

[0025] Step (2.3) Estimate the maximum range

[0026] The distance test was conducted using a surface ship platform and an underwater fixed guard platform to obtain the maximum distance at which the receiving array element can stably detect the positioning sound signal and obtain correct ranging information, which is the maximum effective range of the array element, denoted as R. max .

[0027] Step 3: Accuracy simulation evaluation

[0028] Step (3.1) simulates and generates station location and slant range measurements

[0029] In the simulation scenario, the theoretical value of the target position in the single-point evaluation, trajectory evaluation or regional evaluation mode is taken as the true value. The Monte Carlo method is used for large sample sampling. The station error and ranging error are substituted in a normal distribution manner to obtain the simulated station position and slant range measurement values.

[0030] First, according to the true value of the station location According to normal distribution Generate a set of samples of Nsam The station location observation value sequence of the kth sampling and array element j is expressed as

[0031] Secondly, according to the normal distribution Generate a set of samples of N sam The observed slant range sequence in is the current target position X t =[x t ,y t ,z t ] is the slant distance between array element j, X t Represents the single point evaluation mode (X P ), trajectory evaluation mode or regional assessment model A single value in .

[0032] Step (3.2) Effective array element screening

[0033] by As the criterion for determining the effective array elements, the number of effective array elements is N V , N V ≤N s ; For the kth sampling, the observation sequence at the station location On the basis of V The sequence of the station position observation value sequence with effective action distance constraint is obtained by reorganizing

[0034]

[0035] Step (3.3) Positioning solution

[0036] For the kth sampling, the target position measurement value at the i-th iteration is defined as The target position measurement at the i+1th iteration is N ite is the preset maximum number of iterations. According to the prior information, the initial estimation error of the target position is set And according to the normal distribution Generate a set of samples of N sam The observed slant range sequence That is the initial value of the target position for the kth sampling.

[0037] For the kth sampling, and Drive, use nonlinear least squares method to iteratively calculate and solve, the recursive calculation formula is

[0038]

[0039] in:

[0040]

[0041] when When the preset accuracy is met, the iteration stops. That is the measurement value of the kth sampling, recorded as

[0042] Step (3.4) Error Analysis Evaluation

[0043] For single point evaluation mode, only a single position truth value needs to be calculated; for trajectory evaluation mode or area evaluation mode, repeat steps (3.2) and (3.3) to calculate point by point. sam After the sampling calculation, the root mean square error is used as the indicator for statistical calculation to obtain the accuracy evaluation results in the X, Y, and Z directions under different modes. The calculation formula is:

[0044]

[0045] Beneficial effects of the present invention:

[0046] This method fully considers the potential impact of irregular station geometry, complex underwater acoustic channels, and other environmental factors on measurement performance, achieving accurate simulation of the random error propagation process. This method provides diverse application modes for single-point, trajectory, and regional conditions. It features good adaptability to complex and irregular offshore station configurations, accurate error propagation reproduction, and a wide range of application scenarios. It solves the problem of assessing the positioning accuracy of diverse underwater targets, such as single points, trajectories, and regions, under irregular station geometry. This helps engineers understand the measurement performance and defects under different station geometry conditions in advance, enabling them to conduct optimized design and accuracy assessment, avoid the risk of ineffective offshore operations, and ultimately achieve high-quality positioning and measurement of underwater targets under marine conditions, providing technical support for high-quality positioning and measurement of underwater targets. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is the basic process of the method proposed in the present invention.

[0048] Figure 2(a) to Figure 2(c) The distribution diagram of the target and the measuring station positions is shown in Figure 2(a), where Figure 2(b) represents a point target, and Figure 2(c) represents a track target.

[0049] Figure 3(a) and Figure 3(b) are schematic diagrams of on-site ranging sample collection and distance measurement test, where Figure 3(a) is the ranging sample collection scene, and Figure 3(b) is the distance measurement test scene.

[0050] Figure 4 Schematic diagram of a single measurement for underwater positioning of a point target.

[0051] Figure 5(a) and Figure 5(b) are the underwater target positioning results in the single-point evaluation mode, where Figure 5(a) is the three-dimensional positioning result and Figure 5(b) is the positioning result on the XOY plane.

[0052] Figure 6(a) and Figure 6(b) are the position sequence and positioning error curve of the underwater moving vehicle in the trajectory evaluation mode, respectively. Figure 6(a) is the position sequence in the X, Y, and Z directions, with the solid line representing the true value and the dotted line representing the measured value. Figure 6(b) is the RMSE sequence in the X, Y, and Z directions.

[0053] Figures 7(a), 7(b), and 7(c) are the positioning error distribution diagrams in the X, Y, and Z directions under the regional assessment mode, respectively. They are given in the form of contour lines, and the unit of the contour line scale value is m. DETAILED DESCRIPTION

[0054] The following further illustrates the specific implementation of the method proposed in the present invention in combination with the technical solution and the accompanying drawings.

[0055] The present invention provides a method for evaluating the accuracy of an acoustic positioning system with irregular offshore stations. The main steps are as follows: Figure 1 shown.

[0056] (1) Setting initial conditions

[0057] The following working conditions are considered: using ocean buoys and underwater buoys as the carrier platforms of the long-baseline underwater acoustic positioning system array elements, positioning measurements are performed on the on-duty platform or underwater vehicle (equipped with positioning acoustic beacons) in a limited area, and accuracy evaluation is performed using three modes: single-point evaluation, trajectory evaluation, and area evaluation.

[0058] Combine different evaluation modes to establish measurement geometry:

[0059] The ocean buoys are arranged in a regular quadrilateral, with the array elements evenly distributed on a circle with a radius of 200m. The baseline direction is parallel or perpendicular to the X-axis. The receiving array element depth is 5m, and the position of each array element is (j=1,2,3,4); the buoys are arranged in a regular quadrilateral, the array elements are evenly distributed on a circle with a radius of 200m, the baseline direction is at an angle of 45° to the X-axis, the receiving array element depth is 100m, and the position of each array element is

[0060]

[0061] For the single-point evaluation mode (see Figure 2(a)), the positioning performance of the working platform in the on-duty state is mainly examined, and the theoretical value (true value) of the target position to be measured is set to [0, 0, -50] (m).

[0062] For the trajectory evaluation mode (see Figure 2(b)), the positioning performance of the underwater vehicle in the maneuvering state is mainly examined. The theoretical value (true value) of the target trajectory to be measured is set to a track moving along the positive X-axis, with an average depth of 50m and a navigation distance of 30m. The number of data sequence samples is 101.

[0063] For the regional evaluation mode (see Figure 2(c)), the positioning performance of the underwater vehicle at different locations within an 800m×800m area is mainly examined. The theoretical value (true value) of the target grid point is set to range from -800m to 800m along the X-axis with an interval of 20m, and range from -800m to 800m along the Y-axis with an interval of 20m. The number of grid points is 41×41.

[0064] (2) Combine on-site calibration and testing to obtain simulation parameters

[0065] The site error of the ocean buoy is given by the accuracy of the strapdown satellite navigation positioning equipment. The site errors in the X, Y, and Z directions are The station location error of the buoy is given by the accuracy of the shipborne calibration equipment. The station location errors in the X, Y, and Z directions are σ Sx =0.3(m),σ Sy =0.3(m),σ Sz =0.3(m), j=5,6,7,8.

[0066] In the operating sea area, surface vessel 1 and surface vessel 2 were equipped with receiving array elements and positioning acoustic beacons respectively. The relative distance between the two vessels was kept in the range of 300-500m. Their positions were given by strapdown satellite navigation and positioning equipment. Ranging samples were collected on site (see Figure 3(a)). Based on 300 sets of ranging samples, the ranging error σ was obtained according to the method described in step (2.2) of the invention. R =0.5(m).

[0067] In the operating sea area, surface ship 1# and surface ship 2# are equipped with receiving array elements and positioning acoustic beacons respectively. Their positions are given by strapdown satellite navigation positioning equipment. Through the distance test between the two ships (see Figure 3(b)), the maximum range R of the receiving array element that can robustly detect the positioning acoustic signal and obtain high-precision ranging information is obtained. max =400(m).

[0068] (3) Accuracy simulation evaluation

[0069] According to the true value of the station position According to normal distribution Generate a set of samples of N sam The sequence of observation values ​​of the station location

[0070] For the single-point evaluation mode, the target truth value is X t =X P , according to the normal distribution Generate a set of samples of N sam The observed slant range sequence For the kth sampling, As the criterion for determining the effective array element, the measurement value of the sampling is obtained by solving the method described in (3.3) of the invention content. (See the measurement diagram Figure 4 ), complete N sam After the sampling calculation, a set of measurement values ​​were obtained (see Figure 5(a) and Figure 5(b)). According to the method described in (3.4) of the invention content, the RMSE calculated values ​​of a single grid point in the X, Y, and Z directions were 0.31m, 0.31m, and 0.88m, respectively.

[0071] For the trajectory evaluation mode, the target truth value is At each sample point, according to the normal distribution Generate a set of samples of N sam The observed slant range sequence For the kth sampling, As a criterion for determining the effective array element, the measurement value of the sampling is obtained by solving the method described in step (3.3) of the invention content. Completed N sam After the sampling calculation, the RMSE calculation value of the single grid point in the X, Y, and Z directions is obtained according to the method described in step (3.4) of the invention content; after traversing all the sample points of the sequence, the error distribution results in the X, Y, and Z directions are obtained, as shown in Figure 6(a) and Figure 6(b). The average errors of the entire sequence in the X, Y, and Z directions are 0.29m, 0.30m, and 0.87m, respectively.

[0072] For the regional assessment model, the target truth value is At each grid point, according to the normal distribution Generate a set of samples of N sam The observed slant range sequence For the kth sampling, As a criterion for determining the effective array element, the measurement value of the sampling is obtained by solving the method described in step (3.3) of the invention content. Completed N samAfter the sampling calculation, the RMSE calculation value of a single grid point in the X, Y, and Z directions is obtained according to the method described in step (3.4) of the invention content; after traversing all grid points, the error distribution results in the X, Y, and Z directions of the entire area are obtained, as shown in Figures 7(a), 7(b), and 7(c).

[0073] Therefore, the results obtained from different evaluation modes can be used to provide auxiliary decision-making information for related applications. Based on the results of single-point evaluation, the exact position and error of the underwater operating platform in the on-duty state can be determined to support the regular inspection, maintenance, and recovery of underwater facilities. Based on the results of trajectory evaluation, the track and error range of the underwater vehicle can be identified to support the design and implementation of experimental test measurements and the evaluation of the vehicle's motion performance. Based on the results of regional evaluation, the positioning error distribution characteristics of the measurement array coverage area can be provided to support the optimization design, plan formulation, and operation implementation of the underwater acoustic positioning system's offshore station.

Claims

1. A method for evaluating the accuracy of an acoustic positioning system for irregularly distributed offshore stations, characterized in that: The specific steps are as follows: Step 1: Set initial conditions The zero point of the theoretical initial position of the target is set as the origin O, and the station coordinate system is established, with the OX axis pointing east, the OY axis pointing north, and the OZ axis pointing vertically upward; Step (1.1) Establishing the evaluation model Three evaluation modes are provided for underwater targets in different states: Single point evaluation mode: evaluates underwater point targets with fixed positions and constant target parameters, which are expressed as X in the measurement coordinate system. P =[x P ,y P ,z P ]; Trajectory evaluation mode: evaluates the trajectory of underwater targets, whose positions change with time. The target parameters are the sequence of positions over time, which are expressed in the measurement coordinate system as Where l represents the time node number of the trajectory sequence, l = 1, 2, L, N L , N L is the number of sequence samples; Regional assessment mode: The coverage area of ​​underwater target activities is assessed. Its location is grid-based according to the operating sea area, but does not change with time. The target parameters are matrices, which are expressed in the measurement coordinate system as Where m and n represent the horizontal and vertical grid point numbers respectively, m = 1, 2, L, M A 、n=1,2,L,N A , M A 、N A Respectively represent the number of horizontal and vertical grids; Step (1.2) Establishing the measurement geometry According to the station layout design, the long baseline array with a certain geometric configuration is expressed in the measurement coordinate system as follows: Where j is the array element number, j = 1, 2, L, N S , N S is the total number of array elements, N S ≥3; the array geometry is defined by the array elements and constrained by the effective range of each element, while the measurement geometry is composed of the relative positions of the target and the array; Step 2: On-site information collection and analysis Step (2.1) Estimate the site error Under offshore operation conditions, for array elements carried on surface platforms, strapdown installation with satellite navigation and positioning equipment is adopted, and the station error is directly determined by the accuracy of satellite navigation and positioning equipment; for array elements carried on underwater fixed duty platforms, on-site calibration is performed using shipborne calibration equipment, and the station error is determined by the accuracy of shipborne calibration equipment. The station error of each array element in the X, Y, and Z directions is expressed as Step (2.2) estimates the ranging error Under offshore operation conditions, a set of slant range observation samples is collected under the condition of measuring distance using the positioning beacon and receiving array elements of the underwater fixed guard platform or the surface ship platform equipped with the hydroacoustic positioning system. k=1,2,L,N R , N R is the number of sequence samples; at the same time, according to the position information of the underwater fixed guard platform and the surface ship platform, the calculated samples with the same sequence length and time synchronization as the slant range observation samples are obtained. in is the acoustic beacon position at the kth moment, is the receiving array element position at the kth moment; according to the slant range observation sample and calculate samples The residual estimation of the ranging error σ R , the calculation formula is Step (2.3) Estimate the maximum range The distance test was conducted using a surface ship platform and an underwater fixed guard platform to obtain the maximum distance at which the receiving array element can stably detect the positioning sound signal and obtain correct ranging information, which is the maximum effective range of the array element, denoted as R. max ; Step 3: Accuracy simulation evaluation Step (3.1) simulates and generates station location and slant range measurements In the simulation scenario, the theoretical target position values ​​of the single-point evaluation, trajectory evaluation, or regional evaluation modes are taken as the true values. A large sample is sampled using the Monte Carlo method. The station location error and ranging error are substituted in a normal distribution to obtain the simulated station location and slant range measurement values. First, according to the true value of the station location According to normal distribution Generate a set of samples of N sam The station location observation value sequence of the kth sampling and array element j is expressed as Secondly, according to the normal distribution Generate a set of samples of N sam The observed slant range sequence in is the current target position X t =[x t ,y t ,z t ] is the slant distance between array element j, X t Represents the single-point evaluation model X defined in step (1.1) P , trajectory evaluation mode or regional assessment model A single value in ; Step (3.2) Effective array element screening by As the criterion for determining the effective array elements, the number of effective array elements is N V , N V ≤N s ; For the kth sampling, the observation sequence at the station location On the basis of, the effective array elements are sorted according to the sequence number j=1,2,L,N V The sequence of the station position observation value sequence with effective action distance constraint is obtained by reorganizing Step (3.3) Positioning solution For the kth sampling, the target position measurement value at the i-th iteration is defined as The target position measurement at the i+1th iteration is N ite is the preset maximum number of iterations; according to the prior information, the initial estimation error of the target position is set And according to the normal distribution Generate a set of samples of N sam The observed slant range sequence That is the initial value of the target position of the kth sampling; For the kth sampling, and Drive, use nonlinear least squares method to iteratively calculate and solve, the recursive calculation formula is in: when When the preset accuracy is met, the iteration stops. That is the measurement value of the kth sampling, recorded as Step (3.4) Error Analysis Evaluation For the single point evaluation mode, only the true value of a single position needs to be calculated; for the trajectory evaluation mode or the regional evaluation mode, repeat the steps (3.2) and (3.3) to calculate point by point; complete all N sam After the sampling calculation, the root mean square error is used as the indicator for statistical calculation to obtain the accuracy evaluation results in the X, Y, and Z directions under different modes. The calculation formula is:

Images